Search About Donate
The California Experiment

The Great California Prison Experiment

Crime is up. The mystery is why.

Shattered glass from a vehicle near the Palace of Fine Arts in San Francisco in December.
Shattered glass from a vehicle near the Palace of Fine Arts in San Francisco in December.
Shattered glass from a vehicle near the Palace of Fine Arts in San Francisco in December.

Over the past decade, California has led the nation in reducing its prison population. After the Supreme Court ruled in 2011 that conditions in the state’s overcrowded penitentiaries had “fallen short of minimum constitutional requirements,” lawmakers and voters enacted a series of measures to reduce sentences for many crimes and divert offenders to the authority of the counties.

This story was published in collaboration with the Los Angeles Times.

From the outset, critics have warned that putting fewer offenders in prison would lead to a crime wave. Now the first serious attempt is underway to roll back the reforms, backed by law enforcement groups and already planted on the 2020 ballot. Sponsors claim the prison downsizing has “threatened the public safety of Californians and their children from violent criminals.”

In fact, violent crime has increased in the years since 2011, when California embarked on its campaign—called “realignment,” beginning with Assembly Bill 109—to make the state’s prisons more humane.

The mystery is why.

An analysis by The Marshall Project and the Los Angeles Times found:

Property crimes per 100,000 people

4000

3,687

AB 109

Prop 47

U.S.

3500

3,118

3000

2,497

California

2500

2,344

2000

2000

2005

2010

2015

4000

3,687

AB 109

Prop 47

U.S.

3500

3000

3,118

California

2,497

2500

2,344

2000

2000

2005

2010

2015

4000

3,687

AB 109

Prop 47

U.S.

3500

3000

3,118

California

2,497

2500

2,344

2000

2000

2005

2010

2015

Source: FBI
Note: U.S. rate excludes California

Violent crimes per 100,000 people

593

AB 109

Prop 47

600

California

412

458

400

U.S.

344

200

0

2000

2005

2010

2015

593

AB 109

Prop 47

600

California

412

458

400

U.S.

344

200

0

2000

2005

2010

2015

593

California

AB 109

Prop 47

600

412

400

458

U.S.

344

200

0

2000

2005

2010

2015

Source: FBI
Note: Excludes rape. U.S. rate excludes California

As Congress and many other states look for ways to reform their criminal justice practices, California’s prison downsizing experiment holds lessons for lawmakers.

The prison downsizing plan went into effect during a long, steady decline in crime, both in California and nationally. Overall, the state’s crime levels remain at historic lows, comparable to rates seen in the 1960s.

Dubbed “the great American crime decline” by Franklin Zimring, a University of California, Berkeley criminologist and law professor, the trend doesn’t have a single, clear explanation. Some of it—experts dispute how much—was the result of so many criminally inclined people being locked up. Researchers also have credited everything from the obvious (the hiring of more police, a booming economy) to the more obscure (legalized abortion resulting in fewer unwanted children, limits on lead in gasoline reducing cognitive damage).

Violent crimes per 100,000 people

1200

1000

722

800

California

600

412

U.S.

494

400

344

200

0

1985

1990

1995

2000

2005

2010

2015

1200

1000

722

800

California

600

412

U.S.

494

400

344

200

0

1985

1990

1995

2000

2005

2010

2015

1200

1000

722

California

800

600

412

U.S.

400

494

344

200

0

1985

1990

1995

2000

2005

2010

2015

Source: FBI
Note: Excludes rape. U.S. rate excludes California

California’s experience during the years of prison downsizing is similarly confounding. Advocates and critics—along with the leading researchers on the state’s crime and corrections—agree there is no simple explanation for California’s numbers. What single factor can explain the fact that violent crime went up 6 percent last year in Los Angeles, but fell 6 percent in Sacramento? What explains the fact that homicides in the state fell 6 percent while robberies increased 3 percent?

One factor influencing the crime trends may be major differences in how counties invested the billions in state money appropriated to implement the new measures. Some jurisdictions focused on building jails, others focused on recruiting and deploying police, and still others experimented with collaborative courts and reentry programs. (Another story in this series will examine the role of counties in carrying out the prison downsizing.)

Prisoners in the North Block at California's San Quentin State Prison.

To complicate matters, specific crimes come with their own caveats. Reports of rape have increased nationally since 2013, for example, but sexual assaults have traditionally been under-reported and in recent years there has been a concerted effort to encourage survivors of rape to come forward. (Part of the increase stems from the FBI’s decision to broaden the definition of rape in 2013, and our analysis takes this into account by omitting rape reports.)

“Causation is very, very difficult,” said Michele Hanisee, head of the Association of Deputy District Attorneys for Los Angeles County and a leader in the campaign to roll back some of the reductions in the prison population. “The question that I think would be better to ask is ‘Is Prop 47 successful?’ Did it accomplish what it set out to accomplish?”

But proponents of the rollback assert that Assembly Bill 109, Proposition 47 and other recent laws have created a permissive climate that makes policing harder and weakens the deterrent effect of a possible prison sentence.

“There’s no accountability,” said Assemblyman Jim Cooper, D-Elk Grove. “People know they can get away with things. That’s contributed to it. That’s really been a big source of frustration. No one’s going to jail anymore.”

Assemblyman Jim Cooper, D-Elk Grove, supports a statewide initiative that would toughen supervision of parolees and disqualify some prisoners from early release.

Cooper, a retired Sacramento sheriff’s captain, has been a leading voice in a coalition of prosecutors and law enforcement groups pushing back against realignment and Proposition 47.

The forthcoming ballot measure would reverse some provisions of Proposition 47. It would also toughen supervision of parolees and disqualify some prisoners from early release.

Backers of the proposed rollback claim the state’s drug courts, intended as an alternative to criminal courts, are seeing fewer people because prosecutors can no longer force someone into treatment with the threat of a felony. (No statewide database exists to track the number of people who participate in drug courts, but some counties, including San Diego, have reported decreases in participation since Prop 47.)

Those who favor toughening the law also claim counties are struggling to supervise offenders with violent criminal records.

After a recent mass shooting in Bakersfield, the Kern County seat, some local leaders blamed California’s prison downsizing for the county’s increase in murders. Kern County has one of the state’s highest homicide rates.

The shooting suspect, 54-year-old Javier Casarez, who killed himself as law enforcement tried to apprehend him, had never been in prison, according to the California Department of Corrections and Rehabilitation, and had not benefited from realignment or Proposition 47.

Kern County Sheriff’s Lt. Mark King said that there was no direct link between the measures and the shooting, but said he was concerned that the new laws are a type of “social engineering” that has removed the consequences for low-level crimes.

“Younger people are growing up making poorer decisions,” King said. “They lack respect for property and life...When you change laws, you change things down the road.”

Supporters of the prison downsizing measures strenuously dispute any link between the new laws and an increase in crime. They point out that the recent rise in crime is exaggerated when the baseline is 2014, the year with the fewest crimes reported in the state since the 1960s.

“To look at it from a year-to-year basis is very short-sighted," said Michael Romano, the director of the Three Strikes Project at Stanford Law, who helped write Proposition 47. “We really have had a sustained downward trend over the past decade or two.”

Property crimes per 100,000 people

5,753

6000

California

5000

U.S.

4,514

4000

3000

2,497

2,344

2000

1985

1990

1995

2000

2005

2010

2015

5,753

6000

California

5000

U.S.

4,514

4000

3000

2,497

2,344

2000

1985

1990

1995

2000

2005

2010

2015

5,753

6000

California

5000

U.S.

4,514

4000

2,497

3000

2,344

2000

1985

1990

1995

2000

2005

2010

2015

Source: FBI
Note: U.S. rate excludes California

He said that it’s unlikely any single factor led to an increase in crime, but rather a combination of issues like poverty and unemployment in different counties throughout the state.

Californians for Safety and Justice, a group that co-authored Prop 47, points out that several other states saw larger increases in violent crime than California from 2016 to 2017. (An analysis by the Marshall Project found 20 states with larger increases in violent crime rates.) They note that none of the recent laws changed penalties for violent crimes, focusing instead on non-violent offenders and low-level drug and property crimes.

“There’s no link between reform and violent crime,” said Lenore Anderson, founder and executive director of Californians for Safety and Justice. “What we have seen repeatedly are wide variations at the local level, and wide variations at the local level point to the importance of local practices.”

Several academic studies have examined the question of whether or not the new laws led to crime increases. None has so far found a link between increases in violent crime and the prison downsizing measures.

In 2013, the non-partisan Public Policy Institute of California found that the first major law, realignment, had no impact on violent crime but had led to an increase in auto thefts. In 2016, a prestigious social science journal reached a similar conclusion.

Under realignment, people convicted of auto theft, a nonviolent felony, serve sentences in their local jails and are released under local supervision.

PPIC researcher Magnus Lofstrom said that people convicted of auto theft have a high rate of new crimes after release, perhaps because auto theft requires skills, tools and a network of associates, making it more like a career than most crimes.

Two studies published this summer—one by a University of California, Irvine criminologist and another by PPIC—found no link between Prop 47 and increases in violent crime. Both noted a possible link between the initiative and increases in larceny, although the Irvine study found those links too tenuous to conclude Prop 47 was to blame.

The fatal shooting of a Whittier police officer in February 2017 galvanized law enforcement groups to support a ballot measure to roll back some of the state's new laws aimed at prison-downsizing. The suspect in the shooting, Michael Christopher Mejia, had served time in state prison but had not been released early because of the new laws.

After national crime data for 2017 released this fall showed California departed from the national trend—violent crime in California ticked up slightly while it fell slightly across the other 49 states—researchers said they planned revisit the question of a link between Prop 47 and violent crime, particularly robberies.

“It is troubling and deserves more attention,” Lofstrom said.

Several counties have reported outbreaks of car break-ins in the last few years, but one county accounted for much of the increase: San Francisco. Thefts from vehicles nearly tripled in the city from 2011 to 2017, when they hit 29,851.

It’s become common for San Francisco residents to feel the crunch of car window glass underfoot and to see cars with signs alerting thieves that there’s nothing valuable inside. Stickers throughout the city warn, “Park smart. If you love it, don’t leave it!”

Shattered glass on a San Francisco street this month.

About a year ago, Bob Hinckley and his wife, Tina, moved into a home near the top of the city’s famously curvy Lombard Street. Since then, they estimate they’ve seen dozens of break-ins, mainly of unsuspecting tourists who leave valuables in full view.

Hinckley said he once saw police try to stop a suspect, who simply drove onto the sidewalk to avoid capture.

“He would have killed anyone on the sidewalk,” he said. “Why do they come over here to break into cars? Because that’s where the tourists are with their bags.”

Case in Point

An examination of a single case that sheds light on the criminal justice system

The Hinckleys said they sense a lack of political will in San Francisco to focus more on policing.

An analysis of break-ins confirmed that these thefts were concentrated near tourist hotspots—a stretch of road near museums in Golden Gate Park, a viewpoint overlooking the Pacific Ocean, a street near the Palace of Fine Arts—or in large, public parking garages.

San Francisco Police Lt. Michael Nevin, who until November oversaw car break-in investigations in a district that includes many of the tourist hot spots, said he had noticed a sharp spike in auto burglaries in 2013, before Prop 47. Nevin said break-ins are often charged as “auto burglary,” which was unaffected by the ballot measure. (Auto burglary is a “wobbler,” which means it can be charged as a misdemeanor or a felony.) He said the break-ins are typically committed by organized theft rings that work in groups of two or three.

“I think it’s important to point out that an auto break-in was a felony before [Prop 47], and it’s a felony now,” he said.

When asked whether new criminal justice measures have hampered his daily policing work, he said, “Absolutely not.”

Car
owners placed notes in their vehicles to try to prevent break-ins in the Alamo Square neighborhood of San Francisco in December.
Car
break-ins have been concentrated near tourist hotspots

The break-ins have become a hot-button issue in San Francisco, particularly because District Attorney George Gascón co-authored Prop 47.

In September, Suzy Loftus, a former prosecutor who now works as a lawyer for the San Francisco Sheriff’s Department, announced she would challenge Gascón with a focus on the car break-ins. Shortly after, Gascón announced he would not seek reelection.

Loftus doesn’t blame Prop 47 or other prison downsizing laws for the crime increase. She said, if anything, it shows a failure by San Francisco leaders to adapt to needed criminal justice changes.

“I think the idea that when these crimes stopped being felonies—There’s nothing you can do? There’s plenty you can do,” she said. “It’s easy to say the changes in our system are causing this. It’s harder to say, ‘Hey, we needed those changes, and we’ve got to work together better locally.’”

Another crime that has become central to the backlash against the new criminal justice measures is shoplifting.

Proposition 47 raised the value of stolen goods necessary to charge a felony from $400 to $950. Retailers complained that organized theft rings were taking advantage of the reduced punishments to steal misdemeanor levels of merchandise from store after store.

“It’s gotten far worse in the last few years—there are no consequences,” Rudy Ruiz, owner of World 1-1 Games, a video game store in a mall about an hour north of Santa Barbara. “Whatever numbers you’re using, I can guarantee they’re much, much higher.”

At first, he would call police, but they would take three or four hours to respond, Ruiz said.

So Ruiz tried to run down shoplifters himself and to shame suspects on social media. A few months ago, a group of thieves stole a video game controller. He started chasing after them, and they dropped it. Another time, he caught a woman slipping an item into her baby stroller.

“I’ll personally go after people,” he said. “I’m not a million-dollar corporation. If somebody steals $100, that’s a lot for me.”

Shoplifting rates soared in many counties in 2015, with Los Angeles and San Bernardino seeing the most dramatic increases in the total number of shoplifting reports. But then they dropped statewide over the next two years, falling to their lowest level in more than a decade.

To head off a frontal assault on Proposition 47, legislators created a new category of crime, “organized retail theft.” That new law, signed by Governor Jerry Brown in September, gives prosecutors discretion to charge shoplifting as a felony if two or more people are acting together to steal.

The bill’s author, Assemblyman Reggie Jones-Sawyer, D-Los Angeles, said the bill’s aim was to crack down on organized shoplifting rings without retreating from the reforms or criminalizing people who are stealing because they are hungry, homeless or addicted.

Cooper, the Prop 47 critic, said it would do little to curb shoplifting because it places too high a burden on prosecutors to prove a conspiracy.

In San Joaquin County, which has long had one of the state’s highest crimes rates, property crimes have fallen by 25 percent since 2011. In 2017, it saw the lowest number of shoplifting reports in 15 years.

When Eric Jones, chief of police in Stockton, the San Joaquin County seat, first saw the local crime numbers, he worried that perhaps the drop in crime meant fewer people were reporting to police. The department double-checked the figures by looking at auto thefts—a crime that’s nearly always reported—to verify their data.

He said he’d been skeptical when realignment and Proposition 47 went into effect.

“I was concerned when those passed about the impact those were going to have on my community,” Jones said. “Would crime go up? And it’s just not that simple. I mean, look at our county—we’ve seen some success.”


How We Computed the Crime Metrics

For our joint investigation with the Los Angeles Times, we used crime data from two sources: The Open Justice portal from the California Department of Justice, which provides crime data for the Golden State alone, and the FBI's Uniform Crime Reporting program, which provides data for all 50 states.

To compute crime rates—defined as number of crimes per 100,000 inhabitants—we needed population figures. The UCR includes population numbers in its crime data, but the California Department of Justice does not, so we obtained the state's population and that of its 58 counties from the U.S. Census' annual population estimates.

The crime data on California obtained from the UCR should, in theory, be the same as those provided by that state’s Department of Justice, since it is this agency that submits crime data to the FBI in the first place. There are always minor differences, however, likely generated by the standardizing of the data by the FBI to make it consistent across the 50 states.

We confirmed these differences are not significant enough to alter the conclusions of the analysis. To be consistent, however, we use UCR data when we draw comparisons between the Golden State and the rest of the country, and data from the California Department of Justice when we analyze what has happened inside California alone.

All of the data and code used for our analysis can be found in this GitHub repository.

The Marshall Project receives funding from the California Endowment and other organizations that favor efforts to reform California’s criminal justice system. Under terms of its funding, the Marshall Project has sole editorial control of its news reporting.