Who is jailed, how often, and why: Our Jail Data Initiative collaboration offers a fresh look at the misuse of local jails

Using a novel data source, we examine the flow of individuals booked into a nationally-representative sample of jails along lines of race, ethnicity, sex, age, housing status, and type of criminal charge.

by Emily Widra and Wendy Sawyer

Millions of people are arrested and booked into jail every year, but existing national data offer very little information about who these people are, how frequently they are jailed, and why they are jailed. Fortunately, we now have new data through a collaboration with the Jail Data Initiative to help answer these questions: In 2023, there were 7.6 million jail admissions; but 1 in 4 of these admissions was someone returning to jail for at least the second time that year. Based on the Jail Data Initiative data, we estimate that over 5.6 million unique individuals are booked into jail annually and about 1.2 million are jailed multiple times in a given year. Further analysis reveals patterns of bookings — and repeat bookings in particular — across the country: The jail experience disproportionately impacts Black and Indigenous people, and law enforcement continues to use jailing as a response to poverty and low-level “public order” offenses.

We first looked into these questions about repeat jail bookings in our 2019 report, Arrest, Release Repeat: How police and jails are misused to respond to social problems. In particular, our 2019 analysis found that repeated arrests are related to race and poverty, as well as high rates of mental illness and substance use disorders. While these new data don’t include as many contextual details as the survey we used in that analysis, the Jail Data Initiative data offer a more accurate accounting of unique versus repeat jail bookings — a number that the Bureau of Justice Statistics still does not collect or report. Our two analyses are not directly comparable: Our 2019 report uses data from a public health survey, the National Survey of Drug Use and Health, while the Jail Data Initiative data we use here are collected directly from online local jail records and updated daily. As a result, these reports have different strengths: this analysis can provide more accurate jail booking data because of the much larger sample size, while Arrest, Release, Repeat offers more descriptive data about people who have been jailed, including details regarding health, education, and income.

More than 1 in 5 people are jailed multiple times a year

While unique jail admissions (the number of individual people admitted to jail) account for three-quarters of jail bookings, more than 1 in 5 people (22%) booked into jail are booked again within 12 months. People who are jailed multiple times a year inevitably face exacerbated consequences of incarceration: As we have discussed before, there is no “safe” way to jail a person, nor is there an amount of time a person can be detained without escalating short- and long-term risks to themselves, their families, and their communities, including rearrest, legal debt, missed work, lost jobs, and health risks.

Racial disparities in jail admissions extend to repeat bookings

Black people are overrepresented in every part of the criminal legal system including jails, and this new data reveal that not only are Black people jailed at alarmingly high rates, but they are jailed again and again. Relative to their share of the total US population (14%), Black people are greatly overrepresented among the unique jail admissions in this sample (32%) and the people booked multiple times in a year (29%), while white people are underrepresented in both populations. This is consistent with what we know about the over-incarceration of Black people in this country, but the rebookings data add another layer of detail about their experiences with law enforcement, which often targets Black communities.

Indigenous people account for only 1% of the total U.S. population, but 3% of the incarcerated population, with incarceration rates between two and four times higher than that of white people. In the Jail Data Initiative data, we find that Indigenous people are especially likely to be booked into jail multiple times: 33% of Indigenous bookings were people who had been booked at least once in the past 12 months, compared to 18-22% among other racial and ethnic groups.

Women are funneled into jails

Women make up about a quarter of individuals booked into jail each year, which is in line with annual arrest data showing 27% of arrests in 2023 were of women. While we found no significant difference between men and women in terms of multiple jail admissions, we do know that the jail incarceration of women is growing: From 2021 to 2022, the number of women in jail increased 9% while the number of men in jail increased only 3%. The jailing of women has a devastating “ripple effect” on families: At least 80% of women booked into jail are mothers, including over 55,000 women who are pregnant when they are admitted. Beyond having to leave their children in someone else’s care, these women are impacted by the brutal side effects of going to jail: aggravation of mental health problems, a greater risk of suicide, and a much higher likelihood of ending up homeless or deprived of essential support and benefits. So while women may account for a relatively small share of people booked into jails, those jail admissions have serious and long lasting consequences for the women, their families, and their communities.

1 in 10 people booked are 55 years or older

Older adults account for one in ten of all jail bookings, but a slightly smaller share of rebookings (7%). This is consistent with existing arrest data that show an increasing proportion of older adults caught up in the criminal legal system: In 2021, people 55 years or older accounted for 8% of all arrests, a four-fold increase from their share of arrests in 1991. The Bureau of Justice Statistics only began publishing the age ranges of people in jails in recent years, but from 2021 to 2022, the jail incarceration rate of people 55 and older increased by 8%, compared to a 3% increase in jail incarceration rates across all other age groups. Considering most older adults are arrested for low-level, non-violent offenses like trespassing, driving offenses, and disorderly conduct, it is likely that the older adults admitted to jail are in need of other systems of support outside of the criminal legal system, like substance use treatment, accessible medical care, and behavioral health services.

More than 40% of unhoused people booked into jail were booked again within the year

Poor people in the United States are a primary target for policing, especially those forced to live on the streets: In a 2022 analysis of Atlanta city jail bookings, we found that 1 in 8 admissions involved people experiencing homelessness, a proportion more than 30 times greater than the city’s total unhoused population. In this analysis of the Jail Data Initiative data, we find that across the 140 jails that include housing status on their online rosters, 4% of individuals booked are explicitly listed as unhoused, although this is almost certainly a significant undercount. While this is a relatively small portion of all bookings, unhoused people were the most likely to be jailed multiple times across all the demographic categories we looked at: Over 40% of unhoused people booked into jail were booked more than once in a twelve month period. This finding adds to the existing evidence of law enforcement’s ineffective but disproportionate and deliberate targeting of people experiencing homelessness.

Most people are jailed for public order, property, or drug charges — not “violent” charges

The Bureau of Justice Statistics last collected charge data for jail populations in their 2002 Survey of Inmates in Local Jails. Given that the most recent jail offense data is over 20 years old, the Jail Data Initiative dataset offers a rare opportunity to analyze the top charges that people are booked under nationwide. Of course, the difference in data sources makes a fully apples-to-apples comparison of the 2002 data and the more recent Jail Data Initiative data impossible. The data provided in the Bureau of Justice Statistics survey reflects self-reported information from people detained in a sample of local jails on a single day in June 2002, while the Jail Data Initiative data is based on jail bookings across a two-year time period and relies on administrative data. Nevertheless, the overall trends since 2002 offer some valuable insights into the reasons people are detained in jails today:

• Drug charges appear to play a smaller role now than they did two decades ago, when the “war on drugs” was in full effect. In 2002, a quarter of people in jail were held for drug charges, compared to 14% of people admitted to jail in our 2021-2023 sample.
• Property charges also appear to represent a smaller portion of the jail population now than they did in 2002: Property charges are the top charge for 19% of jail admissions, compared to 24% of the jail population in 2002.
• In 2002, public order charges were the top charge for 25% of people in jail, but now, 31% of people admitted to jail are booked for a most serious charge related to public order, such as disorderly conduct, loitering, and public intoxication.
• We see very little change in the proportion of people in jail for violent charges: in 2002, 25% of people were in jail for a violent charge and in our analysis of more recent jail bookings, about 26% of jail bookings were for violent charges.

In our 2021-2023 sample, only one-third (33%) of people booked multiple times had a top charge categorized as violent in the study window, suggesting that the vast majority of people admitted to jail — including people booked repeatedly — are not accused of violent charges like homicide, assault, robbery, or sexual assault.

In their own analysis of a similar sample of the dataset, researchers at the Jail Data Initiative, Orion Taylor and Anna Harvey, found that rebooking rates vary by the type of initial booking charge. Looking closely at the rates of people returning to jail on serious violent charges within six months of their release from jail, they found the average rebooking rate for a serious violent charge was only 2% for people initially booked into jail on any other kind of top charge. The rate was only slightly higher (9%) if they were also jailed on a serious violent charge initially.

Conclusion

The Jail Data Initiative offers a sorely-needed alternative source of information about jail admissions and about people who are jailed repeatedly. In many ways, our findings from this analysis support what we already know: people who are arrested and booked more than once per year often have other vulnerabilities, including homelessness, in addition to the serious medical and mental health needs of this population that we discussed in our 2019 analysis based on public health data. In addition, this dataset fills a serious gap in our knowledge about the demographics and charges of people booked into jails, given that comparable data has not been collected or published from the Bureau of Justice Statistics in over twenty years.

***

For more information, including detailed footnotes and our methodology, see the full version of this briefing on our website.