At the beginning the COVID-19 pandemic, the criminal justice system was affected by public health measures such as the statewide shelter-in-place order, as well as state and local directives that altered interactions between law enforcement officers and the public, resulted in the early release of inmates, and modified bail procedures. Importantly, this research does not establish definitive causal links between specific policies and changes in arrest trends. It outlines trends in arrests and discusses pandemic-era factors that should be more definitively explored in future research.

Like the rest of the nation, California experienced overall decreases in crime in 2020, although there were increases in some crime categories, including homicides and auto thefts. The causes of these trends are not easy to discern, given the highly unusual and challenging context of the pandemic.

Arrests declined at the beginning of the pandemic

Early in the pandemic, as Californians began to limit their public movement, law enforcement officers were directed to modify their interactions with civilians to limit exposure to the virus. They were also instructed to avoid arrests and bookings when possible, to reduce the number of people entering county jails. In addition, some inmates were released early from jails and prisons. Zero-bail orders, which set bail at zero dollars for most misdemeanors and lower-level felonies—as well as a handful of more serious felonies—were issued by a few county superior courts before the Judicial Council of California enacted a statewide emergency zero-bail order in mid-April.

Arrests began to fall precipitously in March 2020, driven by reductions in misdemeanor arrests. Though some COVID-related criminal justice policies and measures expired after a few months, California experienced persistent declines of 5 percent for felony arrests and 40 percent for misdemeanor arrests until at least July 2021—resulting in a rare near-convergence of these two arrest types. The largest declines were in arrests for lower-level offenses related to drugs and driving.

The partial rebound in arrests after April 2020 occurred similarly across Californian counties, regardless of whether the counties decided to maintain zero bail after the state order was rescinded in June. Further investigation of the impact of zero-bail policies is needed.

Declines in arrests contributed to a 30 percent reduction in new admissions into jail and a 17 percent reduction in the jail population that persisted until at least December 2021. While it is clear that multiple factors contributed to this sustained decline, the impact of any one policy is difficult to quantify.

Mobility patterns were the clearest factor in pandemic arrest trends

In 2020, reductions in the number of civilians walking, driving, and using transit were linked to decreases in arrests, while upticks were consistent with rebounds. Arrest and mobility patterns were mostly aligned with fluctuations in the number of police stops. But the mobility-arrest relationship was not as close in 2021, when arrests did not fluctuate as much.

Reductions in police stops and formal enforcement also contributed to declines in arrests. Among the state’s largest law enforcement agencies, police stops declined by around 35 percent until at least the end of 2020. Police also altered their handling of the stops they did make: officers were more likely to let people go without formal enforcement in the first few months of the pandemic; they later transitioned to citing and releasing stopped individuals in the field.

Zero bail and other pandemic policies raised concerns about a “revolving door” effect—whereby people are repeatedly detained and released—that could have an impact on public safety. To assess these concerns, we tracked the percentage of arrests that were re-arrests during the early pandemic. Re-arrests did not change significantly in the context of key pandemic policies or events. However, the felony re-arrest share remained higher than the misdemeanor arrest share from late June 2020 to at least July 2021. We also find that the proportion of re-arrests dropped sharply in June 2021. The determinants of this decline are unclear, and we do not yet know whether it continued into 2022.

An important takeaway for policymakers seeking to understand arrest trends in this unprecedented period is the substantial association between civilian movement and arrests, which seemingly eclipsed other factors. However, it is possible that zero bail, cite-and-release orders, and early releases from jails and prisons were also influencing arrests and re-arrests. It will be important for future research to isolate the impact of these policies.