The most recent data collected and published by the federal government found 75,505 individuals in solitary confinement in the nation’s prisons—a significantly higher number than other recent surveys.
The figure comes from the latest Census of State and Federal Adult Correctional Facilities, typically conducted every five to seven years by the Bureau of Justice Statistics (BJS), an agency of the U.S. Department of Justice. It reflects the facilities’ count of people held on a single day in “restricted housing,” which is “defined as a placement requiring confinement of a prisoner to a cell for at least 22 hours per day.”
The BJS Census data was collected in 2019 and published at the end of 2021. The 2019 Census was the first in 14 years to ask facilities about their use of solitary confinement. The preceding BJS Census to do so, conducted in 2005, found 81,622 people held in restricted housing. The 2000 Census found 80,870 people—a 40 percent increase over the 1995 Census, when 57,591 people were found to be held in restricted housing. (During the same period of time, the overall prison population grew by 28 percent.)
The BJS Census figures include only state and federal prisons. They do not account for jails, where the only available data—which is from 2012 and is extrapolated from a sample rather than based on a complete count—found just under 20,000 people held in solitary in local and federal jail facilities. There are no reliable counts for immigrant or juvenile detention facilities, but the existing figures suggest that if all of these facilities were counted, as many as 100,000 people could well be in solitary confinement in the United States on any given day.
Even the most recent BJS figures pre-date COVID-19. The pandemic initially caused an explosion in the use of solitary, with numbers reaching at least 300,000, largely due to widespread lockdowns. No data exists to show whether the number of individuals in solitary confinement has returned to pre-pandemic levels, or remains higher than the 75,505 found in 2019.
The figure of 75,505 far exceeds the number of people in restricted housing found in the latest survey of prisons by the Correctional Leaders Association (CLA) and the Liman Center for Public Interest Law at Yale Law School. This widely cited survey was also conducted in 2019, with the resulting report published in September 2020.
The CLA/Liman survey received responses from only 39 states representing 57.5 percent of all people held in state and federal prisons. Based on these responses, the authors estimated that “between 55,000 and 62,500 prisoners were held in-cell for twenty-two hours or more per day on average for fifteen days or more.”
An important distinction exists between the data collected by the BJS and CLA/Liman. The latter asked jurisdictions to report only those individuals held in restricted housing for 15 days or more, while the BJS Census asked for all people in restricted housing.
This fact, combined with the fact that the BJS’s figure is a hard number (not an estimate) based on a far higher response rate (91 percent of all facilities provided data for the Census), means that the BJS’s number of 75,505 is the most comprehensive, and should be cited in the future by journalists, researchers, advocates, policymakers, and others who may have previously relied on CLA/Liman for total numbers of people in solitary in U.S. prisons.
This is significant in part because the BJS numbers also show a relatively small reduction in solitary in recent years—from 81,622 in 2005 to 75,505 in 2019. The CLA/Liman reports, in contrast, show a larger drop—from an estimated 80,000 to 100,000 in 2013 to approximately 68,000 in 2015, to 61,000 in 2017, to between 55,000 and 62,500 in 2019.
Overall, the BJS Census figure of 75,505 is surprisingly high, given that the total prison population has fallen incrementally in recent years, and a number of states had already begun to reduce their solitary populations in the years preceding 2019. It may be that populations have shifted from longer stretches in solitary to shorter stints (which would not be reflected in CLA/Liman numbers). It is also likely that new legislation and policy changes in the past three years have brought the total number down. But without more detailed breakdowns, such possibilities remain purely speculative.
This points to the value of the high level of detail found in the CLA/Liman reports. The BJS Census includes only two questions on restrictive housing. In contrast, CLA/Liman’s surveys have broken down individuals in solitary by gender, race/ethnicity, age, length of time in solitary (beyond 15 days), and jurisdiction. (The BJS Census told Solitary Watch in an email that state-by-state numbers from the Census will be released later this year.)
The data variables and analysis in the CLA/Liman reports reveal trends in the use of solitary confinement (or at least, solitary beyond two weeks). For example, they show that Black men—and, even more dramatically, Black women—are overrepresented in solitary confinement even more than in the prison population in general.
The CLA and the Liman center have also consistently collected their data on solitary every two years since 2013, during a period when the federal government allowed 14 years to pass without a full count. (The 2012 BJS Census did not request information on the use of restricted housing.)
The shortcomings of existing data underscore the need for far more frequent and comprehensive data collection and dissemination by the federal government. Although all self-reported data is subject to manipulation by the prison systems who collect it, the federal government has more access to and authority over sources of data on carceral facilities than anyone else does. It should assume corresponding responsibility for thorough collection and prompt dissemination of this information.
A bipartisan bill introduced in Congress just this week seeks to create a structure for the federal government to take up this responsibility. The Solitary Confinement Study and Reform Act would establish a commission to study the use and effects of solitary confinement and issue recommendations for national standards for how federal, state, and local governments can significantly reduce the practice.
Among the areas to be covered is the monthly collection and publication of data by federal, state, and local governments. This includes the average daily number of people in all forms of restrictive housing, the number of placements in each form, and the reasons for placements; the duration of time in restrictive housing and the duration of out-of-cell time and congregate programming while there; and data on self-harm. All of this would also be broken down along demographic characteristics.
The bill, which is supported by the advocacy groups that make up the Federal Anti-Solitary Task Force (FAST), recognizes that accurate information is critical to effective policy making and envisions data collection as the first step toward meaningful change on the issue of solitary confinement.