RESEARCH BRIEF
The Changing Use of Jails in Safety and Justice Challenge Counties
Unpacking Progress Made Towards More Fair and Equitable Use of Jail
Introduction

The majority of people in local jails around the country are awaiting their criminal trial. This means they have not been convicted of the crimes that brought them in. Many also do not pose a danger to public safety, nor a significant risk of flight, but remain in custody because they cannot afford bail or bond—making incarceration a fact of financial ability, not of safety.

In fact, keeping those in jail who could be better served in the community can cause long-term instability. Well-established research demonstrates the harms of pretrial detention, including worse criminal legal system outcomes, disruption in employment, poorer mental and physical health, and strained social and familial relationships (Turney and Conner 2019). Unnecessarily overcrowding jails also means those who actually do need attention—especially for risk factors such as mental health, substance use, and housing instability—aren't given the resources they need to avoid landing back in jail.

Given this research, cities and counties involved in the John D. and Catherine T. MacArthur Foundation’s Safety and Justice Challenge (SJC) have engaged in collaborative, multi-agency efforts to safely shrink their local jail populations and increase equity across the system. This includes using resources to promote fairer and more just pretrial decision-making, address the needs of system-involved people, and establish a robust array of alternative options.

These efforts have yielded impressive results: in these SJC cities and counties, nearly 18,000 fewer people are in jails today compared to the start of the initiative. This reduction was driven largely by fewer people being booked into jail (Measuring Progress: Jail Populations, 2024). Progress has been slower on reducing racial and ethnic disparities, though in many sites there are fewer people of color in jail than there were prior to the start of the initiative (see Measuring Progress: Disparities, 2024). Importantly, SJC research has established that throughout nearly a decade of the initiative, people released pretrial were no more likely to return to jail than before the SJC, including for violent crime (see Measuring Progress: Returns to the Community, 2024)—demonstrating that data-driven and cross-agency collaborative planning efforts can lead to positive outcomes for individuals while keeping communities safe.

These efforts have yielded impressive results: in these SJC cities and counties, nearly 18,000 fewer people are in jails today compared to the start of the initiative.

Since the SJC launched in 2015, the Institute for State & Local Governance at the City University of New York (CUNY ISLG) has served as the national intermediary for the Initiative, as well as the primary data and analytic partner to participating cities and counties. Through this work, CUNY ISLG has collected rich data from SJC sites that inform effective, data-driven strategies that support the goal of reducing the misuse and overuse of jails and eliminating racial and ethnic disparities in jails. Overall, many jurisdictions have been successful at safely reducing their jail population and, subsequently, reducing the harm that comes with incarceration across all racial and ethnic groups. Despite this, disparities persist, as many jurisdictions saw declines in jail population that were steeper for white individuals compared to Black, Indigenous, and People of Color (BIPOC).

While metrics focused on reductions in the overall jail population are important, they alone do not present the full picture of progress in these jurisdictions. Equally important is understanding how the population has changed —that is, who gets booked into jail and how long they stay—as cities and counties implement strategies to advance their goals. At the core of the SJC mission is the idea that not only should jail incarceration be reduced, but that it should be used only when necessary. This means reserving pretrial detention for people who either pose a significant risk to community safety or seek to evade justice; and for those individuals, ensuring that their criminal cases move through the court as efficiently and swiftly as possible.

Equally important is understanding how the population has changed —that is, who gets booked into jail and how long they stay—as cities and counties implement strategies to advance their goals.

So, to answer this important question of how jail reduction strategies have changed the way SJC cities and counties1 use jails beyond the overall volume of people who are incarcerated, CUNY ISLG analyzed individual-level jail data from five participating counties to explore trends in three key indicators:

  1. Who gets booked into jail;
  2. How they are released from custody (e.g., money bond, pretrial supervision);
  3. How long they stay.

We also examine whether change in these domains varied for different racial and ethnic groups. The sites included in this analysis are Allegheny County, PA; Charleston County, SC; Palm Beach County, FL; New Orleans, LA; and Pima County, AZ. This brief presents the findings from the analysis.

These five sites were chosen based on pre-COVID patterns demonstrating large reductions in jail bookings but stagnant or increasing overall lengths of stay among those released from jail. CUNY ISLG wanted to explore these trends more deeply in a subset of sites. Palm Beach County's participation in the SJC ended on October 7, 2025. However, the data presented in this brief covers data through April 2023 during which Palm Beach County was a participating SJC jurisdiction.

Jails were being used more judiciously, with substantially fewer people booked for lower-level and/or administrative charges.
Compared to 2015 (the year before SJC cities and counties began fully implementing strategies to reduce the misuse and overuse of jail), fewer individuals were booked into jails for administrative reasons with no new criminal charges, such as missing an appointment with a probation officer or not showing up for a court hearing. As a result, the share of bookings that have a new criminal charge increased over time (from between 64 percent-82 percent across sites pre-SJC to between 72 percent-91 percent in 2023). Further, the share of bookings involving felony crimes increased over time in most study sites (up by nearly 7 percent, on average, from pre-SJC), as jurisdictions found more effective and judicious ways of handling lower-level, non-violent crimes.

Among those who were booked, a greater proportion of people were released while their court cases were pending.
 Over time, the proportion of people released from custody on a pretrial status—as their cases made their way through court—increased from 52 percent to 59 percent of all people released. This increase was likely influenced by the implementation of pretrial strategies to reduce the reliance of monetary payment where community safety and flight risk were not present, and to increase pretrial supervision programs to allow for some higher-risk individuals to be released with additional community supports and oversight. These types of policy and practice changes allowed more individuals to return to their communities, families, employment, schools, and local service providers regardless of financial ability.

People who were booked into jail on felony charges stayed for shorter periods of time.
In most study sites, those booked into jail on felony charges in 2023, after SJC implementation, spend less time in jail on average than those booked before the SJC began. Increasing the ability to safely release people to the community lessens the harms and burdens that unnecessary detention has on individuals. While the number of people booked for misdemeanor charges declined over the course of the SJC, their time spent in custody remained fairly steady over time.

Racial and ethnic disparities persist even as bookings declined across racial and ethnic groups.
 Broadly, white people saw steeper reductions in the number of bookings over time than Black people. As a result, Black people represent a growing share of the jail population in some sites. However, there were some findings indicating that disparities were lessening, with data showing increases in the proportion of Black individuals booked on new criminal charges (indicating fewer admissions into jail for administrative violations alone) and steeper decreases in time spent in jail custody compared to white individuals.



This study demonstrates that SJC counties have made progress towards fairer and more equitable uses of their jails by: reserving jail space for those who may pose a threat to community safety or are flight risks; finding safe ways to release more individuals while their cases are pending; and decreasing time spent in jail. While these positive trends were broadly experienced by all racial and ethnic groups, findings indicate that on average, progress was more pronounced for white individuals, resulting in continued racial and ethnic disparities in bookings and length of stay. This illuminates the need for more intentional, data-driven, and targeted work to better identify disparity inflection points and develop strategies that specifically center impact on the racial and ethnic population(s) that is overrepresented.

Methodology and Context

CUNY ISLG collects case-level data on all cases and people that pass through the jails in SJC implementation sites (see CUNY ISLG brief ‘Turning Local Data into Meaningful Reforms’ here for more details). Researchers examined five diverse participating jurisdictions for this brief: Allegheny County, PA; Charleston County, SC; Palm Beach County, FL; New Orleans, LA; and Pima County, AZ. The analysis for each site used longitudinal jail booking and release data, drawing six-month samples of bookings into custody in two-year increments to examine how populations evolve over time. In each period, researchers examined bookings between January 1st–June 30th in each of the following years: 2015, 2017, 2019, 2021 and 2023.2 Each time frame reflects different “stages” of participation in the SJC initiative—pre-reform, early implementation, later implementation, mid-COVID-19 pandemic, and post-pandemic (see Implementation Timeline). These data explore how the use of jails—that is, who gets booked and how long they stay—has changed as sites have undertaken jail decarceration reform.

These data explore how the use of jails—that is, who gets booked and how long they stay—has changed as sites have undertaken jail decarceration reform.
Allegheny’s first (Pre-SJC) booking cohort began in 2017, as they were part of the third wave of jurisdictions to join SJC, which began implementation in 2018. New Orleans’s most current booking cohort runs through 2022, not 2023, as jail data for 2023 was not available at the time of this report.

General trends in bookings and length of stay

Prior to analyzing the ways in which uses of jails have changed in these five sites, this section presents trends in overall jail bookings and average length of stay (ALOS) across the five sites in the context of ongoing efforts of these SJC cities and counties to reduce the misuse and overuse of jails. Over the course of the SJC initiative, jail populations in all sites dropped, driven largely by steep declines in the volume of bookings over time (Figure 1) rather than changes to time spent in jail custody (Figure 2). Drops in the volume of bookings into jails began after their participation in the SJC—and were accelerated by the 2020 pandemic. Those decreases have been sustained across multiple years following the end of the pandemic crisis period: in the latest 2023 booking cohort, all five study sites had maintained jail booking numbers well below pre-pandemic periods.3

As of the most recent Measuring Progress report, using quarterly aggregate jail data from SJC sites, bookings remain below pre-pandemic levels in four of these five sites (August 2023).

The change in time spent in jail custody was less pronounced and varied across these study sites.4 Figure 2 reveals that by 2019, most of the five sites saw no change, or even slight increases. However, by 2023, trends across sites were more varied: two study sites saw notable declines in ALOS, two remained fairly stable, and one saw an increase in overall ALOS.5

Length of stay in this study is based on time from jail booking to release within each booking cohort in order to examine whether people booked during those periods experience different treatment with respect to length of stay compared with those booked in other periods. Methods for measuring length of stay in jail custody vary and can produce quite different results depending on the sample of individuals used in analysis. See Appendix A for more details.
Allegheny’s overall ALOS is much higher than the other four sites in part due to their substantial detainer population.

The varied and less steep changes in lengths of stay compared to the stark decrease in volume of bookings across sites motivates the exploration of who is being booked into jails. For instance, if the decline in bookings over time was primarily driven by reductions in bookings for low level offenses—cases more likely to result in relatively quick pretrial release on nonfinancial conditions or with low bail—sites might expect to see less change in overall average lengths of stay, as the remaining jail population would increasingly consist of bookings for more serious criminal allegations that are less likely to obtain pretrial release. Using the case-level data from these five SJC jurisdictions, CUNY ISLG sought to better understand the effects of jail reduction strategies on the composition of each site’s jail population and on ALOS for various booking and charge types.

Finding I: Fewer people were booked for lower-level and administrative charges.
Key Takeaways

Over time, in these five SJC study sites, the share of bookings involving new criminal offenses increased, as fewer people were being booked into jail for administrative reasons alone. The composition of those booked changed, too: the share of felony-related bookings increased over time (both non-violent and violent), as fewer individuals with lower-level charges were booked into jail. The consistency and timing of these changes in the composition of jail bookings likely reflects more judicious decision making regarding who gets booked into jail custody, with jurisdictions finding alternative ways to handle lower-level crimes and violations, such as missed court dates or failed drug tests while on community supervision.

Jails were being used more judiciously, with substantially fewer people booked for lower-level and/or administrative charges.

SJC cities and counties have sought more effective and fair ways to respond to both lower-level crimes such as trespassing and public disorder—where substance abuse or mental health disorders are more likely to be driving factors—and administrative-related violations like failing to make a court appearance or failing a drug test while on community supervision. While these behaviors represent violations of the law, detention in jail may not be the most effective response for individuals who are not a flight risk or do not put public safety in jeopardy.

Cities and counties have designed and implemented strategies to better address these issues without resorting to booking people into jail, including misdemeanor arrest diversion and deflection programs, substance use and mental health services, resolving outstanding warrants without arrest, and programs aimed at preventing violations from occurring in the first place (e.g., text reminders for court appointments; transportation and drug testing subsidies). See Appendix B for a list of strategies each of these SJC sites have implemented.

There were fewer jail admissions for administrative-related reasons alone. Sites reduced the share of bookings for administrative violations including failure to make a court appearance or violating a condition of community supervision. This in turn led to an increase in the share of bookings associated with new criminal charges over time by an average of seven percentage points across jurisdictions. Simply put, admissions to jail for charges that were not a threat to public safety—administrative violations—declined following SJC reform efforts.

Felony bookings made up a larger share of the total jail population following reforms. In four out of the five sites, the share of bookings associated with felony charges notably increased between pre-SJC cohort and the 2023 cohorts, as jurisdictions found more effective and judicious ways to handle lower-level, non-violent crimes. Across these five sites, the share of bookings into jail associated with a felony charge rose by an average of about nine percentage points between the pre-SJC implementation and 2023 cohorts. The largest increase was in Pima County, where felony bookings represented 38 percent in 2015 and 60 percent in 2023. A similar pattern was found when isolating violent felony crimes, which rose by an average of about eight percentage points across the study sites.

Trends among misdemeanor bookings were more varied. Pima and Charleston counties showed notable decreases in the proportion of misdemeanor-related bookings, Palm Beach and Allegheny showed slight decreases, and New Orleans showed a slight increase. New Orleans’ increase in both the proportion of felony and misdemeanor bookings was due to a notable decrease in their “other” charge type category (i.e., sub-misdemeanors, such as ordinance violation or administrative charges; data not shown).

  • There were no clear trends across these five sites when examining specific types of administrative bookings (e.g., violations, warrants, and holds) or for sentence-related bookings (usually those entering to serve a local jail sentence). However, there were interesting findings within specific study sites.  

    Charleston County: The proportion of bookings involving a warrant-related charge (e.g., failure to appear or comply) decreased by almost half between the Pre-SJC and 2023 booking cohorts (19 to 10 percent). Although some emergency measures aimed at mitigating the spread of the COVID-19 virus in jails and courts, such as a temporary halt on issuing bench warrants for failure to appear/comply at the onset of the pandemic, may have affected the decline between 2019 and 2021, the share of warrant bookings had started a downward trend well before COVID-19, namely in 2017 and continued falling with a slight uptick in the 2023 cohort. Despite the decrease in the volume and proportion of warrant bookings over the years, the ALOS for this type of booking increased between Pre-SJC to 2023 from 21 to 32 days.   

    Allegheny County: Allegheny County experienced a substantial jump in the share of bookings associated with warrants throughout the SJC periods increasing from eight to 27 percent between the Pre-SJC and 2023 cohorts. Moreover, the increase in the proportion of warrants-related bookings started well before the COVID-19 pandemic (it had already reached 19 percent in 2019) and continued increasing thereafter. The ALOS for warrant-related bookings also decreased over these time cohorts from 38 to 27 days (Pre-SJC to 2023).

  • Charleston: The share of misdemeanor-level bookings in Charleston County’s jail population declined by 10 points between pre-SJC and 2023 (71 to 61 percent), as the share of felony bookings grew. The average number of days spent in custody among individuals booked on misdemeanors remained fairly steady at around eight days, with the exception of an increase to 12 days in 2019 before declining again. The County implemented two law enforcement strategies that likely contributed to the decrease in the share of misdemeanor bookings. The first was a diversion strategy implemented by the largest law enforcement agencies in the County which by and large targeted misdemeanor charges associated with warrants. The second was a cite and release strategy aimed at reducing bookings for some lower-level offenses.  

    New Orleans: In New Orleans, there was an increase over time in the share of both felony bookings (up 8 percent) and misdemeanor bookings (up 4 percent). However, the share of individuals booked on “other” charge types (neither felony nor misdemeanor)—largely municipal criminal offenses such as property damage, trespassing, and public intoxication—decreased from nearly a quarter of all bookings pre-SJC to 11 percent in 2022. The average number of days spent in custody among individuals booked on felonies decreased from 42 to 37 days by 2022, while the misdemeanor ALOS showed more fluctuation, with a low of four days in 2019, a high of 13 days in 2021, and falling to seven days by 2022. The “other” charge type ALOS remained fairly steady across time.     

Finding II: More people were released while their case was pending.
Key Takeaways

Each of the five sites saw an increase in the proportion of individuals released on a pretrial status—meaning that they were able to return to their communities, families, employment, schools, and local service providers while their criminal cases were pending. Further, in three of these five study sites there was an increase in the proportion of pretrial releases that were nonfinancial—meaning that defendants did not need to post bail or bond to return to their communities. Together, these trends help minimize the disruptions to daily life that incarceration causes and allow individuals to navigate their criminal cases while staying connected to important family and neighborhood resources.

Among those who were booked, a greater proportion of people were released while their court cases were pending, as jurisdictions found ways to reduce the reliance on monetary payment to secure release for individuals where risk to community safety and absconding were low.

In addition to implementing strategies to reduce admissions into jails, SJC cities and counties have also focused on ensuring that individuals booked into jail are released from custody as expeditiously as possible – either through pretrial release or resolution of their case. To do this, jurisdictions implemented a variety of reforms to better inform when and how people were released.

People are released from jail for an array of reasons, including release while their criminal case is pending, either with or without financial or supervision conditions; transfers to another agency, including to state or Federal prisons, hospitals, treatment facilities, or to other county jails; completion of their local jail sentence; or a full dismissal of their criminal case, among others. Examining why individuals are released from jail custody provides insight into changes in pretrial release practices. This is particularly important because people being held pretrial are the largest proportion of those held in jail and length of stay itself can lead to negative impacts. For example, individuals who remain detained in jail pretrial for longer periods may be more likely to plead guilty to minimize the time they stay in jail (Peterson, 2019). This matters especially for people held on monetary bail or bond—the most common pretrial release condition—who may not have the financial resources to pay for their release.

Jurisdictions that expand pretrial release and lessen the financial burdens tied to release should expect to see an increase in the proportion of the jail population released on pretrial status. All five study sites implemented pretrial reform strategies aimed at expanding pretrial release through measures like expanding pretrial risk screening to identify more individuals to safely release, having a public defender at first appearance to argue for reasonable bail, and implementing jail population review teams to regularly review who in jail might be a good candidate for release (e.g., bond modification, connections to community services and supports).

Pretrial release increased following reform efforts. In all five study sites, the proportion of pretrial releases increased from pre-SJC to 2023. On average, pretrial release increased by six percentage points across these sites, with the smallest change occurring in Pima and the largest change occurring in New Orleans. On average, across sites, the proportion of pretrial releases involving felony charges increased from 41 to 49 percent, while the misdemeanor proportion decreased slightly from 52 to 48 percent—likely reflecting shifts in the underlying jail population toward those booked on felony-level charges (data not shown).

All study sites saw increases in nonfinancial releases by 2021. By 2023, most saw declines from this peak.6 Three sites (Allegheny, Palm Beach, and New Orleans) saw increases in nonfinancial releases during early SJC implementation—prior to the pandemic—and during the mid-COVID period. However, only Allegheny and Palm Beach counties sustained those increases by 2023. In Pima and Charleston counties, the increase in nonfinancial releases only occurred during the mid-COVID 2021 booking cohort, most likely a reflection of pandemic-related mitigation efforts in the jail.

Not all study sites were able to fully isolate all types of nonfinancial releases. For example, New Orleans only includes releases on own recognizance, while others capture RoR and pretrial supervision releases where monetary conditions were not involved.

  • Pima: While nonfinancial pretrial releases increased from 43 percent of all releases in 2015 to almost half (49 percent) in 2021, it quickly declined back to the same level by 2023. Moreover, not all nonfinancial subset categories followed the same pattern. The proportion of releases representing those released on recognizance declined from 37 to 25 percent between the pre-SJC and 2023 cohorts, while the share of supervised pretrial releases increased from six to 19 percent in the same period. The associated ALOS for nonfinancial pretrial releases remained fairly steady at between three and five days. The increase in pretrial supervision releases was likely impacted by pretrial service strategies implemented as part of the SJC initiative to expand the use of the pretrial screenings and assessments (often with supervision recommendations attached).   

    New Orleans: Pretrial releases where bond or cash bail was set to secure release from custody increased from 27 percent of all releases in the Pre-SJC cohort to nearly half (48 percent) in the 2022 cohort. This occurred as releases resulting in no further action (e.g., charges dismissed, court order release) sharply declined from about half of all releases in the pre-SJC cohort to under a third in the 2022 cohort. The increase in bail/bond releases may have been influenced by the expansion of the use of their pretrial risk assessment to include all felony cases. Prior to this, judges did not have a tool to rely on to assess risk and often set high bail/bond amounts for felony cases which often resulted in pretrial detention rather than release.   

    The associated ALOS for both bail/bond and no action releases did not show notable changes when comparing pre-SJC and 2022, however those released on bail/bond in the 2019 booking cohort spent an average of 12 days in jail, compared to between seven and eight days in all other time periods.    

Finding III: People with felony bookings had shorter jail stays.
Key Takeaways

In all five study sites, those booked into jail with new criminal charges in 2023 spent less time, on average, in jails than those booked before SJC reforms. The same was true for those booked on felony charges in four of five sites. The ALOS of those booked for misdemeanor charges remained fairly steady over time. This may indicate that jurisdictions are having success in their efforts to implement more effective and efficient release policies and practices—all focused on maintaining community safety. At the same time, the proportional increase of bookings for more serious charges has led to an increase in time spent in jail for those held pretrial following SJC reforms in four of five sites. Given that the composition of jails is changing to reflect more complex charges, this trend is not surprising.

People who were booked into jail on felony charges stayed for shorter periods of time.

Time spent in jail custody, particularly when awaiting adjudication of a criminal matter or transfer to another authority, is a critical metric to examine when assessing a local criminal legal system. Most jurisdictions participating in the SJC have focused reform efforts on front-end solutions—i.e., finding ways to reduce the volume of people coming into their jail—rather than strategies aimed at reducing lengths of stay. However, some sites have included strategies to reduce length of stay for those in custody such as expanding pretrial release, reducing administrative or case processing delays for certain cases (e.g., probation or parole violations), and accelerating case resolution. Where such strategies are in place that more directly target reducing lengths of stay, jurisdictions might expect to see change over time in the average length of stay of like cases (e.g., felony cases over time). To determine this, CUNY ISLG researchers examined lengths of stay across different subpopulations of the jail to see whether the ALOS among like cases saw more notable changes than the overall length of stay.

Individuals booked into jail on new criminal charges following reforms had shorter stays. The ALOS for the group of individuals booked on new criminal charges decreased between the pre-SJC period and 2023 in four of the five study sites; reductions ranged between 2 and 27 fewer days spent in custody, on average. In other words, those booked on new criminal charges prior to the implementation of SJC strategies stayed longer in jail, on average, than those who were booked in 2023.

Among those booked on a felony, time spent in jail decreased following the implementation of the SJC. The ALOS for individuals booked on felony-related charges dropped in all five study sites from pre-SJC to 2023. This occurred even as the share of felony bookings increased over time in four out of the five study sites. People booked in 2023 were released from jail between 4 and 45 days faster, on average, than those who were booked in the pre-SJC cohort.

The impact of declines in ALOS for felony bookings is somewhat obscured when examining changes in the overall ALOS, which show much smaller declines (Figure 2). This may be explained by the shifting composition of who gets booked into jail, and how this influences measures of overall length of stay. In general, felony-level bookings have longer lengths of stay, and as these types of bookings become a greater proportion of those booked in to jail, overall ALOS increases. Misdemeanor and lower-level offense bookings have comparatively shorter lengths of stay, on average, as these individuals can more easily secure release on nonfinancial or low-bail pretrial conditions or have their cases dismissed. Since lower-level bookings have become less common, the remaining bookings into jail reflect more serious cases involving more severe offenses. This dynamic likely explains the smaller changes in overall ALOS, relative to the declines seen for felony-level bookings.

Since lower-level bookings have become less common, the remaining bookings into jail reflect more serious cases involving more severe offenses.

Trends in misdemeanor ALOS were more stable over time. ALOS for misdemeanors remained fairly steady (+/- 1 day over time) in four of these sites, with ALOS ranging from 7-8 days in two sites and 4-5 days in two others between pre-SJC and 2023. However, Allegheny saw a decrease in ALOS from 36 to 31 days between the pre-SJC and 2023 cohorts.

Time spent in jail among those held pretrial increased somewhat in most of the sites – in part reflecting a population with more serious and complex charges. ALOS among pretrial releases increased over time in four of the five study sites, ranging from a one-day increase (Palm Beach) to a six-day increase (Allegheny). Average length of stay among pretrial releases declined by three days in New Orleans. Increases in ALOS occur for a variety of reasons, including delays in releasing individuals from jail while their cases are pending and delays in court case processing for those held in jail, either because they are deemed a threat to the community or cannot afford bail/bond. However, as the composition of jail becomes more serious and complex, which the overall findings here suggest, the ALOS for the pretrial population that remains in custody may be higher as fewer individuals/short-stayers enter the front door on lower-level charges.

Additional Information on Transfers

Figures 10 and 11 Individuals who are being released or transferred to a third party (such as to a prison or another county) tend to make up a relatively small proportion of all releases in jails. However, these releases are often associated with very long lengths of stay in custody. These longer lengths of stay are often driven by administrative delays like waiting for a third party to pick up the individuals from the jail for transfer to prison, or another county, for example. These delays may have adverse impacts for those awaiting transfer to state prison or hospitals, as jails often have limited rehabilitative services and supports available compared to other treatment or state correctional institutions that serve sentenced populations. For local jurisdictions, delays in transfer take up scarce jail bed space, which lowers local jurisdictions’ ability to serve incoming populations effectively.

While the proportion of transfer releases in these sites remained relatively stable across the time cohorts, their associated ALOS dropped substantially in four of the five study sites by 2023. For example, transfer releases represented less than 10 percent of all releases in Charleston County over the time cohorts, but the ALOS for people released to other agencies decreased by more than two months—from 126 days in the pre-SJC booking cohort to 62 days in the 2023 cohort.

Finding IV: Racial and ethnic disparities persist.
Key Takeaways

Findings in this section reflect a mix of encouraging progress and persistent, deeply rooted disparities. This highlights how, even when sites explicitly commit to narrowing racial and ethnic disparities, inequities remain deeply entrenched in the system. For example, while overall bookings declined proportionally more for white individuals than Black individuals, Black individuals also had steeper increases among those booked on new criminal charges, as fewer entered the jail on administrative charges alone. Further, in most sites, there were steeper declines in time spent in jail for Black versus white individuals. However, Black individuals continued to have lengths of stay much higher than their white counterparts and were still more likely to be booked for felony charges.

Racial and ethnic disparities persist even as bookings decline across racial and ethnic groups, driven by longer average time spent in custody for Black people compared with white people.7

While results of this study point to encouraging findings on the use of jail space—fewer administrative bookings, fewer lower-level charge bookings, and a growing proportion of pretrial releases—it is essential to examine whether these benefits were equitable across racial and ethnic groups, given the overrepresentation of BIPOC in the criminal legal system. This is particularly true for Black individuals, who are more than three times more likely to be booked into jail than white individuals across the U.S. (Zeng, 2023).

Reducing these disparities is a key SJC focus, which happens in tandem with ongoing efforts to achieve a criminal legal system that delivers both fairness and safety. While participating sites have made efforts towards reducing disparities, it remains a challenge. As previous research conducted by CUNY ISLG demonstrated, despite positive trends in reducing jail populations across many of the SJC cities and counties, racial and ethnic disparities have persisted and even worsened in some sites, particularly during the COVID-19 pandemic period (Measuring Progress, 2024).

Using the most recent population and jail data, CUNY ISLG researchers found that in all five study sites, Black people are overrepresented in the jail population, relative to their share of the county population: in Allegheny County, Black people are overrepresented in the jail by a factor of nearly 4.4; in Charleston, Pima, and Palm Beach counties, the proportion of Black people in jail custody is more than double that of the larger county population; in New Orleans, Black people are over-represented in the jail by a factor of 1.4.

Here we examined the extent to which there have been changes in the racial and ethnic composition of those entering the jail across these booking cohorts and within booking types, charge severity, release reasons, and ALOS.

Cross-site trends here are limited to comparing Black and white individuals, given that not all of these five sites had sizable Latine and Indigenous populations to draw broad conclusions. See site-specific examples later in this section for trends in other applicable racial/ethnic groups in each site.

Declines in total bookings were larger for white people than for Black people across sites, while ALOS remained higher for Black people than for white people. Bookings decreased for all racial and ethnic groups in these sites; however, the volume of bookings declined more for white individuals than for Black individuals (see Figure 12). The trends in ALOS are promising, as the decline in ALOS among Black individuals outpaced the decline for white individuals (see Figure 13). Despite this encouraging finding, ALOS continues to remain higher for Black individuals compared to white individuals across all five studies sites.

In most sites, Black individuals saw a larger increase than white individuals in the proportion of bookings that involved new criminal charges from pre-SJC to 2023. While both groups experienced a shift away from bookings involving only administrative matters or technical violations, the shift was more pronounced among Black people between the pre-SJC period and 2023—with the exception of Allegheny County.

Bookings involving felony-level charges increased more for Black people than for white people across most sites. In three of the five sites (Charleston, New Orleans, and Palm Beach), the proportion of bookings involving felony charges increased more for Black people than for white people between pre-SJC and 2023. In Pima, the increases were comparable across groups; in Allegheny, no change was seen over time across groups. Similarly, these findings coincided with steeper declines in misdemeanor bookings for Black people, on average, compared to white individuals (ISLG’s analysis showed misdemeanor bookings declined by an average four percentage points across sites for white individuals, compared to an average decline of seven percentage points for black individuals).

The proportion of those released on a pretrial status increased among both Black and white people in most sites, but with variable patterns. In two of five sites (Charleston and New Orleans), Black people saw greater increases in the proportion released on a pretrial status compared to white people, while in two other sites (Allegheny and Pima) no change in between-group differences was seen. In Palm Beach, the proportion of white people released on a pretrial status grew faster than the proportion of Black people. Data not shown.

  • Palm Beach: The proportion of Latine individuals being booked into jail has increased slightly from 13 percent in the pre-SJC booking cohort to 17 percent in the 2023 cohort. However, the average number of days in custody for this group of individuals declined by more than two weeks—from 38 to 22 days between the earliest and latest periods. The Latine population also experienced an increase in pretrial releases over time from 50 percent pre-SJC to 62 percent in the 2023 booking cohort—the largest increase among all race/ethnicity groups.  This occurred as the proportion of Latine people released for transfer to another authority (e.g., state prison) decreased from 25 percent pre-SJC to 14 percent in the 2023 cohort.

    Pima: Pima County had the most diverse population of these five sites, including both a notable Latine and Indigenous population. While these two groups did not show changes in their overall representation in bookings over these time cohorts (about 42 percent Latine and seven percent Indigenous), the average length of stay that these groups spent in custody did increase more than the change in the overall ALOS. The ALOS for Latine individuals increased from 23 to 33 days between the pre-SJC and 2023 cohort and the ALOS for Indigenous people doubled—from 16 to 33 days. Further, when examining booking trends by top charge, Indigenous people saw the largest decline in the proportion of misdemeanor bookings across all available race and ethnicity groups—decreasing from 75 percent Pre-SJC to 44 percent in the 2023 cohort. Indigenous people saw a decrease from four to 15 percent, while their associated ALOS increased from six to 14 days.  

Conclusion
CUNY ISLG researchers found that jurisdictions have reduced populations primarily by shifting away from using jails for those with low-level charges and administrative violations, and by expanding pretrial release.

Looking at data across eight years since the SJC initiative began, SJC cities and counties have made great progress towards reducing the footprint of jails in communities across the county. In this brief, CUNY ISLG explored how the use of jails has changed over the course of the initiative—examining both who gets booked into jail and how long they stay. Although reducing overall jail populations is a crucial metric of success for the initiative, it is equally important that those reductions are achieved in a just, fair, and equitable manner, and in such a way that reserves scarce jail resources for those who are at most risk of harm to themselves or others.

To that point, it is critical to understand how these jurisdictions are doing the work. CUNY ISLG researchers found that jurisdictions have reduced populations primarily by shifting away from using jails for those with low-level charges and administrative violations, and by expanding pretrial release. It is important to note that system changes implemented under the SJC to increase pretrial releases were all designed with community safety at the forefront. Importantly, related CUNY ISLG research finds that even as more individuals were released prior to their cases being resolved, they were no more likely to be rebooked into jail on a new crime than those who were released prior to SJC reforms being implemented—indicating that SJC reform efforts focused on reducing the footprint of jails are able to do so without jeopardizing public safety.

While these findings represent promising strides in reducing the footprint and associated harms of jail detention on individuals and communities, there are still challenges that face these SJC jurisdictions and communities like theirs around the nation. Disparities persist even as sites experienced reductions in bookings for all racial and ethnic groups. Broadly, white individuals saw steeper declines than Black individuals in bookings, both overall and within most subcategories examined here, resulting in jail populations with a higher proportion of Black individuals and a shrinking white population. More intentional, data-driven, and targeted work is needed to better identify disparity inflection points and develop strategies that specifically center impact on the racial and ethnic population(s) that is overrepresented, while continuing to maintain community safety. Additionally, given the systemic nature of inequities and the interconnectedness of the jail’s policies to the policies, processes, and decisions made outside jail walls, meaningful and sustainable change requires collaboration across criminal legal system agencies and with community members, stakeholders, service providers, and organizations.

Finally, while this brief drew out compositional trends that were similar across multiple sites, there were many important and interesting trends specific to individual sites, as exemplified in Boxes 1-4. Therefore, as cities and counties work to improve their local criminal legal systems, it is essential to examine data and policies at the local level to better understand the drivers of change and identify where future efforts will be most impactful towards a more just and fair system.

Acknowledgements

We want to extend thanks to our partners at the MacArthur Foundation, especially those who work on the Safety and Justice Challenge (SJC). This project would not be possible without their commitment to using data to safely reduce the mis- and over-use of jails across the country.

We are also grateful to the criminal legal stakeholders in SJC cities and counties who we closely collaborate with to develop, track, and evaluate reform strategies and jail data.

PROJECT CREDITS

Writing: Emily West, Senior Research Associate; Rebecca Tublitz, Senior Research Associate; Otgonjargal Okhidoi, Research Associate; Brandon Martinez, Research Associate

Research and Advisement: Stephanie Rosoff, Associate Research Director; Reagan Daly, Research Director

Editing: Alisa Orlowsky, Communications Associate; Carla Sinclair, Senior Communications Associate

Web Development and Communication Design: Alisa Orlowsky, Communications Associate

Data Visualization: Emily West, Senior Research Associate; Brandon Martinez, Research Associate; Otgonjargal Okhidoi, Research Associate

Web Design Template: Design for Progress

Back to Resources