Originally posted at the Public Policy Institute of CA.
By Ryken Grattet, Sonya Tafoya, Mia Bird, and Viet Nguyen.

California’s county jails have been profoundly affected by several reforms over the last decade. Most importantly, in 2011, public safety realignment shifted responsibility for large numbers of non-serious, non-violent, and non-sexual offenders from state prisons to county correctional systems. This lowered the state prison population—allowing prisons to prioritize beds for more serious offenders—but increased county jail populations. Three years later, Proposition 47 downgraded a range of drug and property offenses from potential felonies to misdemeanors. The reduced population pressure has allowed jails to prioritize beds for more serious drug and property offenders who are no longer eligible for prison.

Despite the growing importance of jails, little is known about the basic characteristics of jail populations. In this report, we analyze state and local data on individuals moving through county correctional systems. Using data from 11 counties, we find that:

  • Reforms altered the offender composition of the jail population, especially among those held on drug and property crimes. After three years under realignment, the number of drug and property offenders in jails increased by 55 percent and 40 percent, respectively. One year after the passage of Proposition 47, the number of drug and property offenders fell by 35 percent and 13 percent, respectively.
  • Length of stay for felony drug and property offenders increased after realignment. For example, median time served for felony drug offenders released in October 2011 was 45 days, compared to 98 days for those released in October 2015. However, length of stay for people who served time for misdemeanors and felony crimes against persons has remained stable.
  • Releases due to overcapacity rose under realignment and dropped after Proposition 47, when jail population pressure eased.
  • The demographic composition of jails has largely remained stable. But the age distribution does show modest signs of change: the share of those ages 18–21 in jail has decreased slightly, as the share of those in their 30s has increased.

As jail populations shift toward more serious drug and property offenders, counties and the state will need to consider how jail security and rehabilitative programs might be made more effective. While researchers and policymakers continue to examine the longer-term effects of realignment and Proposition 47, it is also important to keep in mind that the recent reprioritization of jail beds may have implications for crime and recidivism.

[divider] [/divider]

Introduction

County jails have been characterized as the “front door” to the US correctional system (Subramanian et al. 2015). Each year, there are 12 million entrances into the nation’s jails, and roughly 1 million of those entrances take place in California.1 Historically, jails have served a different function from prisons, which tend to receive more policy and research attention. As multipurpose institutions, jails hold a variety of inmates whose offenses range in severity, from infractions, to misdemeanors, to felonies. Jails hold both sentenced inmates, who are serving time, and unsentenced inmates, who are awaiting the outcome of their court cases. Jails also hold Immigration and Customs Enforcement detainees, transferees en route to other correctional facilities or agencies, and probation and parole violators.

Understanding changes in California’s county jail populations is critical because jails have been significantly affected by two recent and historic correctional reforms. First, Assembly Bill 109 (Public Safety Realignment or realignment) shifted incarceration and supervision responsibility for many non-serious, non-violent, and non-sexual offenders from the state prison and parole systems to county sheriffs and probation departments. Jails became the presumptive incarceration site for a large number of drug and property offenders who used to be eligible for prison, and the place where probation and parole violators must now serve their revocations. Realignment had immediate impacts on jail populations, causing them to rise markedly during the first year of implementation (Lofstrom and Martin 2015).

In 2014, the second major reform, Proposition 47 (Prop 47), downgraded a range of drug and property offenses from potential felonies to misdemeanors. Realignment had put many jail systems under strain, as reflected in the sharply rising population levels and the increased number of early releases due to overcapacity (known as capacity releases). Prop 47 eased jail population pressure (Bird et al. 2016). In the process, the priorities of the correctional system shifted again: the reduction in population pressure allowed overcrowded jail systems to prioritize jail beds for individuals held for more serious drug and property crimes.

Although jails are an increasingly important part of California’s correctional system, our understanding of the basic characteristics of the state’s jail population—who is in jail, why they are being held, how long they stay, and how they are released—is limited by available data. In this report, we analyze these characteristics using newly available data collected as part of a collaboration between the California Board of State and Community Corrections (BSCC), PPIC, and 12 partner counties, known as the BSCC–PPIC Multi-County Study (MCS). This county group constitutes roughly two-thirds of the state’s jail population during the period under study.

The BSCC–PPIC Multi-County Study

The Multi-County Study is a collaborative effort between the Public Policy Institute of California (PPIC), the California Board of State and Community Corrections (BSCC), and our partner counties throughout the state. The BSCC–PPIC Multi-County Study is an ongoing effort to collect and merge California state and county criminal justice data. The results from this study will assist counties and the state in better using their custodial resources in the fight against crime by

  • evaluating the effects of key criminal justice reforms and
  • identifying effective recidivism-reduction policies and practices.

This work is also conducted in collaboration with the California Department of Corrections and Rehabilitation and the California Department of Justice. Participating counties include Alameda, Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus.

Realignment Stressed Jails, Prop 47 Provided Some Relief

The statewide jail population increased steadily in the first year following realignment—straining jail facilities—and then fell sharply under Prop 47. Before realignment took effect in 2011, the statewide jail population was at the lowest it had been in 10 years. In fact, in the four years preceding realignment, the jail population had gradually decreased from a record high in 2007. But after one year under realignment, the statewide jail population grew 12 percent, from an average daily population of 72,285 to 80,941. In the next two years, the population continued to creep up.

By June 2014, five months prior to the passage of Prop 47, a majority of jail facilities were at or above 90 percent of their rated capacity; and 20 counties had court-ordered population caps (Lofstrom and Martin 2015). The jail population had reached 83,280, just below the all-time high of 84,046 in 2007.2 After the passage of Prop 47, the jail population immediately began to drop, decreasing 11 percent from 82,005 to 73,253 in four months.

This statewide pattern of the jail population rising under realignment and then falling under Prop 47 is paralleled in the group of counties participating in the Multi-County Study.3 As Figure 1 shows, the MCS jail population increased in the year following realignment and then decreased sharply within the first few months under Prop 47.4

Figure 1. Changes in statewide jail populations are reflected in the MCS counties

Figure 1. Changes in statewide jail populations are reflected in the MCS counties

SOURCE: Authors’ calculations based on the BSCC–PPIC Multi-County Study data (2011–2015).

NOTE: “MCS counties” refers to the following counties: Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus.

Using MCS data, we find that these population trends were driven by increases and decreases in people held for felony offenses under realignment and Prop 47 (see technical appendix Figure A1).5 Although Prop 47 converted several drug and property offenses from potential felonies to misdemeanors, the number of people held on misdemeanors rose only slightly after Prop 47, which is likely due to the fact that bookings for individuals arrested for Prop 47 offenses decreased (Bird et al. 2016). And when booked, individuals whose offenses were converted to misdemeanors under Prop 47 stayed for shorter periods of time.

Realignment put considerable upward pressure on jail populations. In such circumstances, county correctional systems can use their discretion to manage jail populations through various kinds of releases from custody, such as pretrial release and release to alternative custody programs outside the jail. Additionally, in counties facing court-ordered population caps, capacity releases—either during the pretrial phase or after sentencing—may be used when the population approaches or exceeds the cap.

Figure 2 shows that capacity releases increased in the first year after realignment and then decreased after Prop 47. These trends were driven by the five MCS counties that both faced crowding and relied on capacity releases to comply with court-ordered population caps: Fresno, Los Angeles, Kern, San Bernardino, and Stanislaus.6

Figure 2. Capacity releases increased sharply after realignment, but dropped under Prop 47

Figure 2. Capacity releases increased sharply after realignment, but dropped under Prop 47

SOURCE: Authors’ calculations based on the BSCC–PPIC Multi-County Study data (2011–2015).

NOTE: This figure includes data from the following counties: Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus. “Pretrial release” includes cite and release, bail, own recognizance, pretrial supervision, and pretrial release. “Capacity release” includes unsentenced and sentenced capacity releases.

Figure 2 also shows that immediately after Prop 47 was passed, the number of pretrial releases dropped. However, this decline was driven largely by an overall decline in bookings into jail (Bird et al. 2016). In subsequent months, the number of pretrial releases rose again. Releases from jail to alternative custody programs—which typically involve some combination of electronic monitoring, home detention, and supervision—increased after realignment from 353 in October 2011 to 613 one year later. By the end of October 2015, one year after Proposition 47, releases to alternative custody programs had risen to 732, though the use of alternatives remains at a low level (Martin and Grattet 2015).

Drug and Property Offenders Drive Population Shifts

Both realignment and Prop 47 were aimed at non-serious, non-violent, and non-sexual drug and property offenders. While realignment shifted a segment of this population from state prison to county jails, Prop 47 reduced the penalties for certain lower-level drug and property crimes. Our findings show that the offender composition of jails largely changed in ways targeted by the reforms.

Under realignment, the number of people held in jails for property and drug crimes grew, while the number of people held on crimes against persons—of which violent crimes are a subset—remained steady.7 As Figure 3 shows, after three years under realignment, the number of people held on property crimes rose from 8,135 to 11,426 (or by 40%) and the number of people held on drug crimes increased from 7,347 to 11,415 (55%). Conversely, the number of people held in jails whose most serious offense was a crime against persons remained relatively stable in the first three years under realignment, hovering around 13,000. Because individuals held on crimes against persons tend to spend more time in custody compared to drug and property offenders, they contribute more to the average daily population in jails.

By reducing the penalties associated with certain offenses, Prop 47 aimed to reduce the number of lower-level drug and property offenders held in custody. The initiative had its intended effect. From the time the proposition passed in November 2014 to October 2015, the number of inmates held for property crime decreased from 11,426 to 9,886 (or by 13%), and the number of inmates held on drug offenses decreased from 11,415 to 7,394 (35%). In contrast, the number of inmates held on crimes against persons grew from 13,335 to 14,561 (9%) between October 2014 and October 2015.

Figure 3. The number of people held for drug and property crimes rose under realignment and declined after Prop 47

Figure 3. The number of people held for drug and property crimes rose under realignment and declined after Prop 47

SOURCE: Authors’ calculations based on the BSCC–PPIC Multi-County Study data (2011–2015).

NOTE: This figure includes data from the following counties: Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus. The figure omits crimes classified as “other,” which includes driving under the influence, driving with a suspended license, and supervision violations without accompanying new charges.

When examining individuals held for different crimes as a share of the total jail population, we find that under realignment, as the total jail population grew, the share of individuals held on property and drug crimes also increased—from 19.6 percent to 23 percent for property crimes and from 17.7 percent to 23 percent for drug crimes between October 2011 and October 2014. As the number of people held on crimes against persons remained steady, their share of the jail population began to drop, from 32.3 percent to 26.8 percent over the same three-year period.

Under Prop 47, the share of the jail population held for property and drug crimes decreased as the share of those held for crimes against persons rose. Between October 2014 and October 2015, the share of the jail population held for property offenses dropped from 23 percent to 21.6 percent, and the percentage of inmates held for drug offenses decreased from 23 percent to 16.1 percent. The share of people held on crimes against persons increased by 5 percentage points over this period, to 31.8 percent in October 2015, returning to where it was in the first month of realignment.

The MCS data also allow us to investigate how jail populations changed in terms of offense categories under the two reforms (Table 1). Similar patterns emerged for those held on common types of drug and property crimes:

  • Under realignment, the number of individuals held or serving time for the most common drug offenses, including controlled-substance possession and controlled-substance possession for sale, increased markedly. Under Prop 47, we see declines in the number of individuals held for these offenses.
  • Under realignment, the number of people held or serving time for common types of property crime, including burglary and petty theft with a prior, grew. Under Prop 47, we see declines in the number of people held or serving time for the majority of common property crimes, with the exception of motor vehicle theft, forgery, and fraud.

Table 1. The number of people held for common crimes has changed under realignment and Prop 47

Table 1. The number of people held for common crimes has changed under realignment and Prop 4

SOURCE: Authors’ calculations based on the BSCC–PPIC Multi-County Study data (2011–2015).

NOTE: Table shows changes in the number and percent of individuals held for the specified offenses between October 2011 and October 2013 (change under realignment), and October 2014 and October 2015 (change under Prop 47). This figure includes data from the following counties: Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus.

Table 1 also shows that trends are somewhat mixed for those held on robbery and assault with a deadly weapon—two common crimes against persons. Under realignment, the change in the number of individuals held for crimes against persons varied depending on the specific offense. The number of individuals held for robbery decreased from October 2011 to October 2013, while the number of individuals held for assault with a deadly weapon increased. Under Prop 47, there appears to be a clearer trend: the number of individuals held or serving time for crimes against persons has increased. This may be the result of an increased number of offenders. Crime reports for 2015 show that both aggravated assaults and robberies have increased from 2014 to 2015 in California (Lofstrom, Bird, and Martin 2016; Lofstrom and Martin2016).

Reforms Have Affected Length of Stay

Thus far, we have considered how the offender composition of the jail population has changed in the context of recent reforms. However, jail population pressure is driven not only by how many and what types of offenders are incarcerated, but also by how long they stay in custody. Below, we turn to the issue of how long inmates stay in jail.

We focus specifically on length of stay for convicted inmates who served jail time or were released due to capacity constraints.8 The analysis compares inmates’ median length of stay for offenses broken down by level (misdemeanor or felony) and type (drug crime, property crime, or crime against persons). We compare cohorts released in October of each year from 2011 to 2015 to examine changes that occurred under realignment and Prop 47.

As Table 2 shows, length of stay for each type of misdemeanor declined slightly as realignment was implemented. Thus, the median length of stay for each type of misdemeanor in October 2011 is longer than in the subsequent three years under realignment. For the cohort released under Prop 47 in October 2015, the length of stay for each type of misdemeanor increased. This pattern suggests that as realignment increased the felony segment of the jail population, length of stay for each type of misdemeanor declined. Once Prop 47 took effect and jail population pressure eased, the length of stay for misdemeanants returned to and occasionally surpassed the level found in the early days of realignment.

Table 2. Length of stay (in days) for misdemeanors decreased under realignment, then increased under Prop 47

Table 2. Length of stay (in days) for misdemeanors decreased under realignment, then increased under Prop 47

SOURCE: Authors’ calculations based on the BSCC–PPIC Multi-County Study data (2011–2015).

NOTE: Median length of stay in days for each offense level and type using October releases. People released in October 2011 were likely to have been convicted and sentenced for offenses prior to realignment, and thus their length of stay represents a pre-realignment baseline. This is less true for misdemeanants; but even then, given the medians are 13 days and above, a large segment of releases in October would have entered jail before October 1, 2011, when realignment officially commenced. People released in October 2012, 2013, and 2014 were most likely sentenced under realignment, and those released in October 2015 were likely sentenced under Prop 47.This figure includes data from the following counties: Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus.

Table 2 also shows that changes in length of stay for felony offenders are less straightforward. Under realignment, median length of stay for felony drug offenses rose substantially from 45 days in 2011 to 73 days in 2014. The substantial increase is likely due to more felony drug offenders being mandated to serve time locally, as they were no longer eligible to be sent to state prison. This meant that overall the population of drug offenders in jail under realignment included a greater portion of individuals whose offenses were more serious. Under Prop 47, the median length of stay for individuals released from jails for felony drug offenses remained high because the proposition reduced the lower-level segment of the drug population, leaving behind only individuals whose drug offending was serious enough to remain a felony.

Median length of stay for property offenses also rose, from 66 days in 2011 to 71 days in 2014, and was even higher in the intervening years. Like felony drug offenses, the increase in time served for felony property offenders released under realignment is likely due to individuals held for more serious property crimes remaining in local custody rather than being sent to state prison. However, unlike drug offenses, length of stay for property offenses continued to rise under Prop 47. This may be due to the increasing seriousness of the offense charges of those coming into jails in 2015. Alternatively, it could be caused by reduced reliance on capacity releases for these offenders in counties with population caps, meaning that jails can hold people in custody longer.

Median length of stay for felony crimes against persons was 90 days for individuals released in October 2011. This decreased to 76 days for the cohort released in October 2012, but then increased in the following years under realignment to 85 days. Under Prop 47, median time served remained at roughly the same level. This suggests that the easing of population pressure on jails due to Prop 47 did not change how long individuals serving time for felony crimes against persons stayed in custody.

Demographic Composition Remains Stable

Criminal justice reforms like realignment and Prop 47 often result in changes in the racial/ethnic, gender, and age composition of those held in custody. Some forecasts suggested that these reforms might change the demographic composition of jails and that the proportion of female and older offenders held locally might increase (Abcarian 2014; Maciag2014). But so far, the demographic composition of the jail population has remained largely unchanged.

Figure 4 shows that nonwhites—especially Latinos and African Americans—make up the majority of jail inmates. These percentages held steady from 2011 to 2015, indicating that thus far recent reforms have had little impact on the racial/ethnic composition of jails in the state. Across racial/ethnic groups, African Americans are the most overrepresented in the jail population. African Americans comprise only 6 percent of California’s population, but make up 25 percent of California’s jail population. Latinos also make up a disproportionate share of the jail population (44%) relative to their share of the state population (39%). The overrepresentation of African Americans in jails is less extreme than in prisons, where African Americans made up 29 percent of the prison population as of December 31, 2013.

Figure 4. African Americans are disproportionately represented in California’s jails and prisons

Figure 4. African Americans are disproportionately represented in California’s jails and prisons

SOURCE: Authors’ calculations based on the BSCC–PPIC Multi-County Study data (2011–2015), California Department of Finance (2015), California Department of Corrections and Rehabilitation (2014).

NOTE: This figure includes data from the following counties: Contra Costa, Fresno, Humboldt, Kern, Los Angeles, Orange, Sacramento, San Bernardino, San Francisco, Shasta, and Stanislaus.

Because both reforms targeted lower-level offenders, and female offenders tend to commit lower-level offenses, it seemed plausible that under realignment the proportion of female inmates in jails might increase and that under Prop 47 the proportion of female inmates might decrease. This hypothesis was borne out in counties that were not under a court-ordered population cap. In these counties, the share of women in jail increased to 14.8 percent under realignment and fell to 13 percent in the post-Prop 47 period. In contrast, the percentage of women in county jails with court-ordered population caps held stable at 13 percent. Overall, statewide data show a fairly consistent proportion of women in jails since 2002. In state prisons, the number of female inmates did, in fact, decrease after realignment, dropping from 9,295 in September 2011 to 5,441 in December 2015—from 6 percent of the prison population to 4.4 percent.

Similarly, because realignment requires many non-serious, non-violent, and non-sexual felony offenders to serve their sentences in local jails, some suspected that the jail population would begin to tilt toward older inmates. Thus far there have been minimal changes in the distribution of ages in the jail population (see technical appendix Figure A2). The segment of the population ages 18–21 decreased from October 2011 to October 2015 by 4 percentage points, while the population ages 31–40 grew by 2 percentage points. All other age groups were fairly stable. Although these trends are relatively modest, changes in the age composition in jail may take more time to manifest and should be watched closely in the coming years.

Conclusion

Newly available data from the BSCC–PPIC Multi-County Study permit a unique view into how county jails are changing under California’s historic correctional reforms. Realignment reprioritized prisons as institutions where offenders serve time for convictions of serious, violent, and sexual offenses. Non-serious, non-violent, non-sexual drug and property offenders remain in the counties for supervision and incarceration. Prop 47 prompted a second wave of reprioritization—this time concerning jail beds. Counties appear to be concentrating resources on more serious felony drug and property offenders, holding them in custody for longer periods of time.

The trajectory of people held on felony drug charges is particularly clear. Realignment drove an increase in the number and share of individuals held or serving time in jail on felony drug charges. After Prop 47, people held or serving time for drug offenses decreased. However, those drug offenders who remain in local custody tend to stay for longer periods, suggesting that jails have prioritized more serious types of drug offenders.

Alongside these changes, the post-realignment increases in jail population coincided with increased use of release mechanisms to relieve population pressure. Both pretrial and capacity releases increased after realignment. The use of these types of releases then declined after Prop 47, although the number of pretrial releases appears to be trending upward again. Alternative custody releases remain at a low level, but are slowly increasing as well. These trends are important to monitor because, unless additional jail space is created, releases and alternatives to custody will remain necessary to manage overcrowding.

There is not much evidence that realignment and Prop 47 are dramatically changing the demographic composition of jails. Realignment did increase the share of women in the jail population in counties that were not under court-ordered population caps, but under Prop 47 the share of women returned to earlier levels. In counties with population caps, the share of women in jails has remained stable for the period under study. However, small changes are emerging in the age distribution of the jail population and, if they continue, could lead to the “greying” of jails that some have predicted.

As the population shifts toward more serious drug and property offenders, security inside jails may need to be reconsidered for the safety of both inmates and staff. Since jail populations increasingly comprise longer-term inmates, there will also be greater opportunities and need for programming. Many jails around the state are already pursuing resources to expand their provision of services. However, it remains the case that most individuals entering jail stay for periods of time that are too short for programmatic interventions. As a result, it makes sense to focus on building rehabilitative programs that span the jail walls and better integrate incarceration and community reentry. Given the volume of lower-level offenders moving through jails, some counties may want to consider expanding the range and types of alternatives to custody that provide a measure of punishment without the negative consequences of incarceration.

Going forward, many questions remain about the impacts of realignment and Prop 47 on local communities. Although the question of how realignment affected crime has received attention, the effect of Prop 47 on crime remains unknown (Lofstrom and Martin 2016). Examining the recidivism of people released from jail is also important, as investigations thus far have only provided a partial view of how major reforms have affected high rates of recidivism in the state (Lofstrom, Raphael, and Grattet 2014; Bird and Grattet 2014). A more comprehensive understanding of the long-term effects of realignment and Prop 47 will assist the state and counties in using their custodial resources to ensure public safety.