Skip to main content Skip to office menu Skip to footer
3 golden objects Minnesota Legislature

Office of the Legislative Auditor - Program Evaluation Division

Safety in State Correctional Facilities

February 2020

Evaluation Report Summary

Key Facts and Findings:

  • The Department of Corrections (DOC) operates 11 state correctional facilities housing approximately 9,200 prisoners. DOC prisons employ about 3,700 staff. (pp. 1, 10)
  • DOC cannot state how safe its correctional facilities are for staff and prisoners; its data on violent events are inconsistent and incomplete. (pp. 14-16)
  • Violence between prisoners appears to have decreased slightly over the last four years. However, violent incidents between prisoners are underreported. (pp. 16-21)
  • Reported assaults on prison staff spiked in 2018, then declined. (pp. 23-24)
  • At some prisons, sexual offenses against female staff by prisoners occur frequently, but are often ignored or downplayed by supervisors and coworkers. (p. 25)
  • One in three prison staff said that bullying and harassment among staff is a problem at their prison. (pp. 43-45)
  • Chronic shortages of correctional officers and the increasing use of overtime at several prisons have affected the safety of prisoners and staff. (pp. 31-40)
  • Staffing shortages have also led many prisons to curtail activities for prisoners, increasing tensions and reducing safety. (pp. 52-55)
  • Outdated design features at the St. Cloud and Stillwater prisons create dangers for staff and prisoners. (pp. 63-66)
  • Unlike county jails, which are inspected and licensed by the state, there is little external oversight of safety in DOC prisons. (pp. 77-82)

The Department of Corrections should take additional steps to protect staff and prisoners.

Key Recommendations:

  • DOC should transform its data collection processes so it has better data about violent events, staffing shortages, overtime usage, and prisoner discipline. It should then use that data to improve safety. (pp. 17-18)
  • DOC should ensure that supervisors take sexual offenses against female staff seriously and discipline prisoners when appropriate. (p. 26)
  • DOC should develop additional strategies to reduce bullying and harassment among its staff. (pp. 45-46)
  • DOC should continue its efforts to hire sufficient correctional officers to staff state prisons. (p. 37)
  • DOC should present to the Legislature long-term plans for rehabilitating or replacing the residential units at St. Cloud and Stillwater. (p. 66)
  • The Legislature should require regular external oversight of prison safety procedures, either through licensure or by putting into law and strengthening DOC’s existing “security audits.” (pp. 82-83)

Report Summary

The Department of Corrections (DOC) operates 11 state prisons confining prisoners convicted of serious crimes. The department is responsible for the safety of approximately 9,200 prisoners and 3,700 prison staff. The majority of staff working at prisons are security staff, such as corrections officers and lieutenants.

State prisons have four custody levels: Levels 2, 3, 4, and 5. Generally, prisoners at the lower custody levels live with fewer restrictions, while prisoners at higher custody levels have more. Level 4 prisons are located at Rush City, St. Cloud, and Stillwater. Oak Park Heights is the state's only Level 5 (maximum security) prison.

Prison leaders should do more to address harassment of staff by both prisoners and coworkers.

Violence between prisoners has slightly declined in recent years, but prisoner assaults on staff spiked in 2018.

Although data are limited, multiple sources indicate that prisoner violence against other prisoners slightly declined over the last four years. In a survey we conducted of prisoners in Level 3, 4, and 5 settings, most said they felt somewhat safe in DOC prisons, despite believing that violence among prisoners occurs frequently.

Violence between prisoners is far more common than violence against staff. However, prisoner assaults on staff increased dramatically during calendar year 2018, driven by sharp increases at Level 4 and Level 5 prisons. Convictions in DOC’s internal discipline system for assaults on staff increased from 112 in 2017 to 149 in 2018, before dropping again in 2019. Worker’s compensation claims for prison staff due to conflicts with prisoners also rose steeply and then fell.

DOC documented few cases of staff-against-prisoner physical violence or sexual assault during fiscal years 2016 through 2019. Further, prisoners we interviewed complained more about staff actions that affected their safety indirectly, rather than physical abuse. For example, prisoners said some staff label prisoners as informants, putting them at risk of assault from other prisoners. On the other hand, about one-third of prisoners responding to our survey said that officers or other staff physically harm prisoners “sometimes” or “very often.”

Prison administrators have not done enough to address sexual offenses by prisoners against staff.

In some state prisons, female staff endure repeated sexual offenses by some male prisoners, who catcall, verbally threaten them with sexual assault, or masturbate in front of them. Female staff said some supervisors and coworkers expect them to tolerate this behavior, and that prisoners frequently receive no disciplinary consequences.

Even if these offenses were routinely punished, DOC disciplinary charges do not distinguish sexual misconduct against staff from other infractions, so it would be very difficult to count them. DOC should create a separate disciplinary charge for sexual misconduct against staff and should ensure supervisors support staff that encounter such offenses.

Bullying and harassment between staff is a pervasive issue in DOC prisons.

In a survey we conducted of DOC staff working in prisons, one in three respondents described unprofessional work relationships as an ongoing problem in DOC prisons. Staff told us about different experiences depending on the prison or their role. For example, some staff described a top-down culture of bullying by supervisors, while others described sexual harassment.

Many staff do not believe that their coworkers or supervisors take harassment seriously. Some staff told us they had experienced retaliation from coworkers or supervisors for reporting wrongdoing by other staff members. For example, one staff person told us that officers refused to respond when that person was working alone in a prisoner living unit and called for assistance.

DOC should take strong action to address workplace culture issues. DOC has recently taken a good first step by establishing a new Office of Professional Accountability; it is too soon to evaluate whether this initiative will be successful.

Chronic shortages of correctional officers and increasing overtime usage have reduced safety for both staff and prisoners.

High turnover rates have led several prisons to fall below their budgetary allocations of correctional officers. For example, during Fiscal Year 2019, Stillwater averaged a shortage of 25 officers under its allocated 314 correctional officers. Although DOC recruited a similar number of new officers in Fiscal Year 2019 as it had in previous years, those staff were not enough to fill the increased vacancies.

To address these shortages, DOC almost quadrupled its use of overtime for corrections officers between Fiscal Year 2013 and Fiscal Year 2019. When prisons lack enough volunteers to work overtime shifts, prison administrators often require staff to work overtime. DOC does not track how often it forces officers to work overtime. We estimated that roughly 15 to 20 percent of instances when officers worked overtime in the last year were forced.

Large majorities of DOC staff we surveyed said that staffing shortages and heavy overtime usage create safety challenges for staff and prisoners. For example, staff may not have enough time to perform routine security tasks, or may need to work alone or in small numbers in settings with many prisoners. Staff tired from working excessive overtime may be less alert, less responsive in emergency situations, or more short-tempered.

Prison administrators and staff told us that short-staffing also leads prisons to frequently suspend prisoner activities such as therapy, employment, education, and recreation. Yet research indicates that providing structured prisoner activities leads to reduced violence in prisons. DOC does not track how often its prisons suspend prisoner activities.

DOC should track the extent of its staffing shortages, use of forced overtime, and suspensions of prisoner activities. It should also continue its efforts to hire enough correctional officers to alleviate its staffing shortages.

Line staff often do not trust decisions made by prison administrators.

Most staff at Level 4 and Level 5 prisons doubted that prison leaders do all they can to reduce violence by prisoners against staff. Although staff responding to our survey agreed that structured activities for prisoners help reduce violence, many staff complained that prison administrators often prioritize such activities over measures that would protect safety. Staff told us that administrators often make decisions arbitrarily, without seeking staff input.

Our interviews with administrators suggest that they juggle many different priorities.  However, a lack of transparency around decisions feeds distrust and lowers morale.  Prison administrators should improve their communication and consultation with line staff.

Persistent staffing shortages have threatened safety.

DOC does not systematically assess the level of safety at its prisons.

Decisions made to improve safety are often implemented without any formal assessment of whether they make a difference. Prison and central office administrators instead make decisions by relying on informal impressions or reacting to major incidents.

Taking a more systematic approach to protecting safety is currently challenging because DOC’s data on violent events in prisons are inadequate. Much of the information DOC collects is narrative and difficult to aggregate. The aggregated data the department does collect is often incomplete. For example, DOC’s biennial reports to the Legislature include counts of prisoners administratively charged with “assault” but not prisoners that are charged with “fighting.” Either charge can reflect a violent conflict in which prisoners were hurt.

Additionally, violence among prisoners is likely underreported across all data sources. Staff acknowledged they may not observe some conflicts between prisoners.

DOC’s data on its disciplinary actions for prisoners who violate prison rules have flaws. DOC has faced particular challenges in providing accurate data about the number of prisoners in restrictive housing (isolation from the general prisoner population) and the length of time they spend there.

DOC should transform how it gathers and uses data. Rather than adding data-gathering tasks to the work it already does, the department should explore ways to restructure its processes so that data are gathered automatically. Once better data are available, prison leaders should use that data to make more evidence-based safety decisions.

DOC should develop a plan for the future of its oldest prisons.

There is limited external oversight of safety in state prisons.

Although DOC is subject to oversight from several external entities, each oversees individual components of safety rather than safety as a whole. For example, audits conducted for the federal Prison Rape Elimination Act focus on sexual assault and harassment of prisoners, while Minnesota’s Occupational Safety and Health Administration focuses mostly on occupational injuries.

State law requires that county jails be licensed and inspected, but DOC claims that its prisons are exempt from this requirement. DOC has voluntarily conducted “security audits,” or peer reviews of prisons’ security procedures. However, these audits occur infrequently—St. Cloud has not had a security audit since 2011—and there are no consequences should prisons fail to adopt security audit recommendations.

The Legislature should require that DOC regularly and systematically evaluate safety at state prisons according to defined security standards, either through licensure and inspection or by adding security audits to state law.

The prisons at St. Cloud and Stillwater—both built over 100 years ago—have design features that are outdated and unsafe.

The residential units in these prisons present security challenges, such as the danger of falling or being pushed over railings from several stories in the air. The layout of these residential units also makes it difficult for staff to monitor prisoners. Some key infrastructure elements, such as door locking mechanisms, are no longer manufactured and DOC must fabricate replacement parts as needed.

DOC should develop and present to the Legislature a long-term plan for rehabilitating or replacing the living units at the St. Cloud and Stillwater prisons. At some point, the state will have to substantially reinvest in these prisons if it is to keep using them.

Summary of Agency Response

In a letter dated February 21, 2020, Department of Corrections Commissioner Paul Schnell wrote that “we concur in whole” with the report’s findings and recommendations. He wrote that the department’s new leadership has spent the last year engaging with its staff and has heard much about “the deeply troubling realities pertaining to the safety of our facilities.” He stated that “we have already begun implementing a number of the recommendations highlighted in your report” and continued, “your comprehensive look back at the safety of our state’s correctional facilities confirms, supports, and underscores the significance of the work ahead of us.”

More Information

The Program Evaluation Division was directed to conduct this study by the Legislative Audit Commission in April 2019. For a copy of the full report, entitled “Safety in State Correctional Facilities,” 106 pp., published in February 2020, please call 651/296-4708, e-mail Legislative.Auditor@state.mn.us, write to Office of the Legislative Auditor, Room 140, 658 Cedar St., St. Paul, MN 55155, or go to the web page featuring the report. Staff who worked on this project were David Kirchner (project manager), Tavis Leighton, and Caitlin Zanoni, with assistance from Andrew Duncan.