Jail Inmates in 2023 - Statistical Tables

Zhen Zeng, PhD,
BJS Statistician

Highlights

• At midyear 2023, local jails held 664,200 persons in custody, mirroring figures from midyear 2022 and marking a 10% decrease in inmate population compared to a decade ago.
• Between 2020 and 2023, the population of jail inmates ages 35 and over increased 35%, while those ages 18–34 saw a 9% rise.
• At midyear 2023, there were 198 persons incarcerated in jail per 100,000 U.S. residents, down from 231 inmates per 100,000 at midyear 2013.
• From July 2022 to June 2023, people admitted to local jails spent an average of 32 days in custody before release, 9 days longer compared to 10 years prior (23 days).
• At midyear 2023, there were a total of 915,800 jail beds in the United States and 73% of the beds were occupied.
• The number of inmates held in local jails for federal and state prison authorities declined by 32,600 (down 26.5%) from 2013 to 2023.
• After a 12% increase from 2015 to 2019, the number of staff employed at local jails decreased to 211,700 by midyear 2023, falling below the 2015 level.

Findings in this report are based on the 2013–2018 and 2020–2023 Annual Survey of Jails (ASJ) and the 2019 Census of Jails (COJ). See Methodology for information about the data collections and study design. A public-use 2023 ASJ data file is available for download at National Archive of Criminal Justice Data. 

In June 2024, BJS published Preliminary Jail Data Release for 2023. This report provides an update to the preliminary release.  

Introduction

At midyear 2023, local jails held 664,200 persons in custody, mirroring figures from midyear 2022 and marking a 10% decrease in inmate population compared to a decade ago. The rate of incarceration stood at 198 persons per 100,000 U.S. residents, a 14% decline from 231 jail inmates per 100,000 ten years earlier. From July 2022 to June 2023, local jails nationwide recorded 7.6 million admissions. While this represented a 4% increase from the 7.3 million admissions the year before, annual admissions were 35% lower than a decade ago when admissions totaled 11.7 million.

Demographic characteristics of inmates

At midyear 2023, local jails housed 95,100 females, representing 14% of the total jail inmate population (table 2). Reflecting the impact of the COVID pandemic, the female jail population experienced a 37% drop from 2019 to 2020. However, it rebounded quickly, reaching 86% of its pre-pandemic level by midyear 2023.

The number of older adults incarcerated in local jails has increased at a higher rate compared to younger adults since 2020, when BJS started collecting age group data through the Annual Survey of Jails. Between 2020 and 2023, the population of jail inmates ages 35 and over increased 35%, while those ages 18–34 saw a 9% rise (table 3). 

The demographic shift is also evident in the declining number of individuals aged 17 or younger in jail, which plummeted by more than half from 4,400 in 2013 to 2,000 in 2023 (table 4). As of midyear 2023, 0.3% of incarcerated persons in local jails were 17 or younger.  

The racial and ethnic breakdown of the jail population remained relatively steady over the last decade. Approximately 47% of incarcerated individuals were white, 36% were black, and 14% were Hispanic. Other racial and ethnic groups, including American Indian or Alaska Native individuals, Asian individuals, Native Hawaiian or Other Pacific Islander individuals, and those identifying with two or more races, collectively comprised 3% of the total jail population (table 5). The percentage of jail population that was black was on a downward trend, dropping from 36% in 2013 to 33% in 2018, before rising back to 36% again in 2023. 

Jail incarceration rates

At midyear 2023, there were 198 persons incarcerated in jail per 100,000 U.S. residents, down from 231 inmates per 100,000 at midyear 2013. The incarceration rate for males was notably higher, reaching 343 per 100,000 male U.S. residents, which was over six times the rate for females at 56 per 100,000 female U.S. residents (table 6).

In 2023, individuals aged 25–34 exhibited the highest jail incarceration rate at 480 per 100,000, approximately 21 times the rate for those aged 65 or older, which stood at 22 per 100,000 (table 7). Older adults experienced a faster growth in their incarceration rate than younger adults. From 2020 to 2023, jail incarceration rate increased 30% for individuals aged 35 and over, compared to 10% for those aged 18-34.

The jail incarceration rate for black U.S. residents was 552 per 100,000, 3.4 times the rate for white U.S. residents (162 per 100,000) at midyear 2023 (table 8). American Indians and Alaskan natives had the second highest rate at 425 per 100,000. Hispanic persons (143 per 100,000) were incarcerated at a rate comparable to whites (155 per 100,000).

Conviction status and offense characteristics

At midyear 2023, 70% (467,600) of the jail population were unconvicted, awaiting court action on a current charge or held in jail for other reasons. The remaining 30% (196,600) were convicted, either serving a sentence or awaiting sentencing on a conviction (table 9). Between 2013 and 2023, the number of convicted persons in jail decreased by 31%, while the number of unconvicted persons increased by 3%.

An estimated 75% (500,300) of the jail population at midyear 2023 were held for felony offenses, marking an increase from 68% at midyear 2015 (table 10). A total of 129,600 persons were held in jail for a misdemeanor at midyear 2023, down from 193,100 at midyear 2015.

At midyear 2023, 96,100 individuals were held in jail for probation violations, accounting for 15% of the 14% of inmate population. About 5% (30,900) of the inmates were in jail for parole violations (table 11).

Persons held for federal, state, or tribal authorities

At midyear 2023, local jails held 41,500 persons for federal authorities, 48,400 for state authorities, and 500 for tribal governments (table 12). Overall, in 2023, 90,500 of all jail inmates were held for federal, state, or tribal authorities, marking a 26% decline from a decade ago. The number of persons held in jail for U.S. Immigration and Customs Enforcement experienced a decrease of 59% during this period.

ADP, admissions, and average time in jail

During the 12 months ending on June 30, jails processed 7.6 million admissions and accommodated an average daily population (ADP) of 664,800. Within that timeframe, females constituted 23% of all admissions and 14% of the ADP (tables 13 and 14).

From July 2022 to June 2023, people admitted to local jails spent an average of 32 days in custody before release, 9 days longer compared to 10 years prior (23 days) (table 15). The average duration of jail incarceration rose significantly during the COVID-19 pandemic, increasing from 26 days in 2019 to 33 days in 2021. Since 2021, the average jail time has stabilized over the subsequent three years. From July 2022 to June 2023, males were incarcerated for 36 days and females 19 days on average, up from 27 days and 16 days 10 year prior, respectively.

On average, persons admitted to jail jurisdictions with an ADP of 2,500 or more inmates were detained for 43 days, longer than those admitted to jurisdictions with an ADP of 2,499 or fewer inmates (30 days) (table 16).

Capacity and staffing

At midyear 2023, there were a total of 915,800 jail beds in the United States and 73% of the beds were occupied (table 17). From 2013 to 2023, the total number of jail beds increased 5%, while the occupancy rate decreased from 84% to 73%. As of midyear 2023, 12% of the jail jurisdictions were operating above their rated capacity.

About 5 in 10 beds in jail jurisdictions with an ADP of fewer than 50 persons were occupied at midyear 2023, compared to over 7 beds in jail jurisdictions with an ADP of 50 or more persons (table 18).

Local jails employed 164,800 correctional officers and 47,000 other jail staff, including administrators, clerical and maintenance staff, educational staff, and professional and technical staff in 2023. Throughout the period from 2015 to 2023, about 8 in 10 staff members were correctional officers (table 19). After experiencing a 12% increase from 2015 to 2019, the number of staff employed at local jails decreased to 211,700 by midyear 2023, falling below the 2015 level.

At midyear 2023, there were 4 inmates for every correctional officer in local jails, similar to the year before (table 20).

Persons supervised outside jail and those serving weekend sentences

In addition to supervising inmates in custody, local jails operate various programs such as electronic monitoring, home detention, day reporting, community service, alcohol or drug treatment programs, and other pretrial supervision and work programs outside of a jail facility. At midyear 2023, local jails supervised 50,100 persons in such programs outside of jail (table 21).

About 1,200 persons served weekend-only sentences on the weekend before the last weekday in June 2023, reflecting an 89% decreased from the 11,000 recorded in 2013 (table 22).

Data tables

This zip archive contains tables in individual .csv spreadsheets from Jail Inmates in 2023 – Statistical Tables.

Terms and definitions

A | C | J | M | P | R | Y 

Admissions

All persons booked into and housed in jail facilities by formal legal document and the authority of the courts or some other official agency, including repeat offenders booked on new charges and persons sentenced to weekend programs or entering the facility for the first time. They exclude inmates reentering the facility after an escape, work release, a medical appointment, a stay in a treatment facility, and a bail or court appearance.

Average daily population (ADP)

The sum of all inmates in jail each day for 1 year, divided by the number of days in the year.

Average time in jail

The average time in jail is estimated by dividing the ADP by the number of annual admissions, then multiplying the product by the number of days in a year.

Confined population

The number of inmates in the physical custody of local jails, regardless of conviction status or which authority has jurisdiction over the inmate.

Jail

A confinement facility generally operated under the authority of a sheriff, police chief, or county or city administrator. Jails house individuals before and after arraignment, usually for a period exceeding 72 hours, and those sentenced to incarceration of 1 year or less. 

Jail incarceration rate

The number of inmates held in the custody of local jails per 100,000 U.S. residents.

Jail jurisdiction

A county (parish in Louisiana) or municipal government that administers one or more local jails and represents the entity responsible for managing jail facilities under its authority. Most jail jurisdictions consist of a single facility, but some have multiple facilities or multiple facility operators (e.g., a jail jurisdiction consisting of a county jail and a private jail operated under county authority).

Midyear population

The number of inmates held in custody on the last weekday in June.

Percent of capacity occupied at midyear

The midyear population divided by the rated capacity.

Persons under jail supervision but not confined

All persons in community-based programs operated by jail facilities, including electronic monitoring, house arrest, community service, day reporting, and work programs. This group excludes persons on pretrial release who are not in community-based programs run by jails; persons under supervision of probation, parole, or other agencies; persons on weekend programs; and persons who participate in work release programs and return to jail at night.

Rated capacity

Set by a rating official, rated capacity is the maximum number of beds or inmates that a facility can hold, excluding separate temporary holding areas.

Yearend population

The number of inmates held in custody on December 31. The yearend population is typically smaller than the midyear population.

Methodology

Findings in this report are based on the 2013–2018 and 2020–2022 Annual Survey of Jails (ASJ) and the 2019 Census of Jails (COJ). The Bureau of Justice Statistics (BJS) periodically conducts the COJ, a complete enumeration of local jail facilities and Federal Bureau of Prisons (BOP) detention facilities, to collect data on inmate population and jail programs. The 2019 COJ was the eleventh collection in the series since 1970. Since 1982, BJS has administered the ASJ in the years when the COJ is not conducted. The ASJ samples approximately one-third of the nation’s jails to provide national estimates on the number and characteristics of the jail inmate population. The COJ and ASJ collections are currently conducted through web-based surveys. About 94% of the ASJ respondents complete the web-based survey each year, while the remaining respondents submitted data by email, fax, or phone. 

Local jails covered by the COJ and ASJ

Jails are confinement facilities generally operated under the authority of a sheriff, police chief, or county or city administrator; a small number are privately operated. Regional jails include two or more jail jurisdictions with a formal agreement to operate a jail facility.  

Jails house individuals before and after arraignment, usually for a period exceeding 72 hours, and those sentenced to incarceration of 1 year or less. Separate temporary holding facilities (such as drunk tanks and police lockups) that do not hold persons after they have been formally charged in court are not considered jails for the purpose of COJ and ASJ collections. Jail facilities include detention centers, county or city correctional centers, and special jail facilities (such as reception centers, medical or treatment centers, and prerelease centers). Jails are intended for adults but can hold juveniles before or after their cases are adjudicated.

Jails serve a variety of functions, including— 

• holding inmates who are sentenced to jail facilities and usually have a sentence of 1 year or less
• receiving individuals pending arraignment and hold them as they await trial, conviction, or sentencing
• readmitting probation, parole, and bail bond violators and absconders
• detaining juveniles pending their transfer to juvenile authorities
• holding mentally ill persons pending their movement to appropriate mental health facilities
• holding individuals for the military, for protective custody, as witnesses for courts, and for contempt of court
• releasing convicted inmates to the community upon completion of sentence
• transferring inmates to federal, state, or other authorities
• housing inmates for federal, state, or other authorities due to crowding of their facilities
• operating community-based programs as alternatives to incarceration.

The COJ and ASJ cover all local jail jurisdictions (including county, city, regional, and privately operated jail facilities) in 45 states. Combined jail and prison systems in Alaska, Connecticut, Delaware, Hawaii, Rhode Island, and Vermont are excluded. These combined systems are operated by state departments of corrections and are included in BJS’s National Prisoner Statistics program. However, 15 independently operated jails in Alaska are in scope for the COJ and ASJ.

BOP detention facilities that function as jails are included in the COJ but excluded from the ASJ. In this report, BOP facilities are excluded.

Sampling design of the Annual Survey of Jails

The ASJ uses a stratified probability sampling design based on jail population data from the most recent COJ. The sample is drawn at the jail jurisdiction level. A jail jurisdiction is a legal entity that has responsibility for managing jail facilities. Jail jurisdictions typically operate at the county or municipal level, with a sheriff ’s office or jail administrator overseeing the local facilities. Most jail jurisdictions consist of a single jail operator, but some have multiple operators (called reporting units), such as a sheriff’s office, a detention center, a correctional institution, and a privately-operated jail. When a jail jurisdiction with multiple jail operators is sampled for the ASJ, data are collected from all reporting units within that jail jurisdiction. 

The ASJ sample is refreshed every 5-6 years. Jails in the ASJ sample are surveyed annually until the next sample is drawn. The most recent sample refresh occurred in 2020, when a sample of 899 jail jurisdictions was selected to represent the 2,848 jail jurisdictions nationwide. In selecting jails, all jurisdictions were grouped into 10 strata based on their inmate population and the presence of juveniles at midyear 2019. In 8 of the 10 strata, a random sample of jail jurisdictions was selected. The remaining two strata were designated as certainty strata in which all jail jurisdictions were selected. One certainty stratum consisted of all jails that were operated jointly by two or more jurisdictions. The other certainty stratum consisted of large jail jurisdictions that on June 28, 2019, held either 750 or more adult inmates, or at least one juvenile inmate and a total of 500 or more inmates.

Response rate and nonresponse adjustment

The 2023 ASJ sample consisted of 887 active jail jurisdictions, represented by 927 reporting units. Twelve of the 899 sampled jail jurisdictions were either permanently or temporarily closed at midyear 2023. The survey achieved a response rate of 92%. Nonresponse weighting was implemented to account for unit nonresponse. The nonresponse weighting adjustment factor was calculated for jails in each weighting class h as:  

where—

n_h = number of jurisdictions sampled in weighting class h 

p_hi = sampling weight for jurisdiction i in weighting class h 

A_hi = active status indicator for jurisdiction i in weighting class h (1 = active, 0 = out of scope) 

R_hi = response indicator of jurisdiction i in weighting class h (1 = respondent, 0 = nonrespondent) 

In this report, standard errors for estimates from the ASJ were estimated using the Taylor-linearized variance method. Standard errors are not calculated for estimates from the COJ because the COJ is a complete enumeration of all jails.

Item nonresponse imputation

Item response rates ranged from 91% to 100% for the 2023 ASJ. For responding jail jurisdictions that were unable to provide requested items, missing data were imputed (except for weekender programs, where offenders serve their sentences of confinement only on weekends). This imputation was done in two steps: first, a last observation carried forward (LOCF) procedure was used to replace the missing values with recent data from the same jails; next, for cases with no available recent data from the same jails, a weighted sequential hot deck (WSHD) procedure was implemented to impute the remaining missing data by borrowing values reported by similar jails in the current year. 

Using the LOCF as the first step of imputation increases imputation accuracy because jail population counts and characteristics typically remain relatively stable from year to year. This is especially helpful when imputing missing data for the largest jails that have no comparable-size donor jails. In the LOCF procedure, donor data consisted of reported data from the same jails in the 2020–22 ASJ and the 2019 COJ, adjusted for average year-to-year changes. The adjustment factor was calculated as the ratio of the average of current-year to prior-year values in each weighting class. Only data from jails that provided data in both years were included in the calculation. By applying the average year-to-year adjustment factor, it was assumed that jails with missing items experienced the average year-to-year percentage change as other jails in the same weighting class. Missing data for inmate subpopulations (e.g., female adults) were imputed based on prior-year population distribution and current-year reported or imputed inmate population for the same jails.  

In the second step, the donor for each missing item in the WSHD procedure was randomly selected from a set of similar jails, sorted by related auxiliary population values. Donor pools, also referred to as imputation classes, were formed by sampling stratum and jail size. Within each imputation class, jails were sorted by a variable related to the imputed items (e.g., jails were sorted by the percentage of the confined population that was male when imputing male and female ADP). 

Reference dates and seasonal variations in jail populations

Prior to 2015, the ASJ asked jails to report total and detailed inmate counts on the last weekday in June (the midyear reference date). In 2015 and 2016, the ASJ collected the total confined population at midyear, but detailed inmate counts by demographic and criminal justice characteristic (e.g., sex, race/ethnicity, conviction status, and offense severity) on December 31 (the yearend reference date). Starting with the 2017 collection, the ASJ reverted to the midyear reference.

Comparisons of yearend data with midyear data need to consider seasonal variations, as jails typically hold fewer inmates at yearend than at midyear. For example, in 2015 and 2016, the yearend population was 5% less than the midyear population. To adjust for seasonal variation, yearend inmate counts by demographic and criminal justice characteristics in 2015 and 2016 were multiplied by the ratio of the midyear total population to the yearend total population of the corresponding year. The standard errors for the 2015 and 2016 counts were similarly adjusted. Seasonal adjustment was also applied in calculating the 2015 and 2016 jail incarceration rates by demographic characteristics and their standard errors.

The reference periods for annual admissions and average daily population are the 12 months from July 1 of the previous year to June 30 of the current year. However, in 2015 and 2016, the reference periods were based on the calendar year. BJS started collecting 12-month admissions in 2015. Before 2015, admissions were reported for the last 7 days of June. 

How to cite this report

Bureau of Justice Statistics, "Jail Inmates in 2023 - Statistical Tables," May 24, 2023:
https://bjs.ojp.gov/jail-inmates-2023-statistical-tables-full-report