The Florida Legislature

The Florida Legislature OFFICE OF PROGRAM POLICY ANALYSIS AND GOVERNMENT ACCOUNTABILITY RESEARCH MEMORANDUM Options for Reducing Prison Costs March 3, 2009 Chapter 2009-15, Laws of Florida, directs OPPAGA to examine options for reducing prison costs by expanding alternative inmate placements. This memorandum presents four options for reducing prison costs: two that would serve inmates already in prison, and two that would divert low-risk offenders from being sentenced to prison. Expand non-institutional prison beds for low-risk inmates Expand maximum allowable gain time credits for low-risk inmates who exhibit good behavior Modify drug possession penalties Expand alternative sanctions for probation violations Florida s inmate population (approximately 100,000) is currently growing by 4% annually, which will require the construction of up to four new prisons each year. These options would free up existing beds and temporarily delay the need to construct as many new prisons as would otherwise be needed to accommodate the annual increase in the inmate population. Each option would produce savings, but is subject to caveats because they would place offenders in less secure environments, which could increase public safety risks. Expand Non-Institutional Prison Beds for Low-Risk Inmates Florida statutes permit the Department of Corrections (DOC) to operate non-institutional facilities to divert less serious offenders from secure prison beds into facilities with lower inmate per diem rates. In addition to these reduced operational costs, the construction costs for these beds are lower than the cost of constructing prison beds. 1 Therefore, using these options for low-risk inmates can result in significant cost savings over the alternative of building additional secure prisons. The department s non-institutional facilities currently include work/forestry camps, road prisons, and work release centers. DOC proposed expanding the number of available beds within these existing facilities and creating new supervised re-entry and work prioritized inmate beds. Because these beds are less secure and require inmates to go outside prison gates, they would be most appropriate for lower-risk inmates who have not committed violent crimes and have exhibited good behavior during their time in prison. 1 The department reported that to expand non-institutional beds it would build open bay dorms to expand capacity at its existing prison sites. The department has indicated that it would use a blueprint open bay dorm model to construct any of the non-institutional options discussed here. The department estimates that the construction costs for this type of facility would be approximately $10 million to $12 million. Conversely, the department estimates that it typically costs over $100 million to construct a major institution, which can house about 1,335 inmates. The dorm costs do not include site acquisition costs and planning and designing costs. Gary R. VanLandingham, Ph.D., Director 111 West Madison Street Room 312 Claude Pepper Building Tallahassee, Florida 32399-1475 850/488-0021 FAX 850/487-9083 www.oppaga.state.fl.us

Page 2 Work release centers house community custody inmates who work in the community at paid employment as well as minimum custody inmates who stay in the work release center performing tasks such as providing food and laundry services for the other inmates housed at the center. Inmates must remain at the center when they are not working or attending programs such as Alcoholics Anonymous. As of June 30, 2008, there were 33 work release centers that housed 3,355 inmates who were nearing release from prison. Work release centers house about 3% of the prison population at any given time. The yearly operational cost for work release centers is $8,694 per offender, or about half the $16,410 cost for housing offenders in regular adult male prison beds (based on Fiscal Year 2007-08 costs). 2, 3 The department reported that as of June 30, 2008, it held 3,757 inmates who were eligible for work release. 4 While the 2008 Legislature authorized DOC to construct 600 additional work release beds, this will only partly address this population. The additional beds funded by the 2008 Legislature will house up to 1,440 of these inmates. 5 The Legislature could consider further expanding this program. If the Legislature funded an additional 965 work release beds, all of the offenders who were eligible as of June 30, 2008, could be served in the program, producing savings of up to $7.4 million in yearly operational costs. This step could also produce long-term cost savings by reducing recidivism, as DOC reports that recidivism rates for persons who participated in work release were 5% lower than rates for comparable inmates who did not participate in the program. Another benefit is that the state does not have to incur health care costs, as inmates on work release are responsible for their own medical expenses after they secure employment. In contrast, the department must pay for all medical services provided to inmates in secure and other non-institutional prison beds. Work/forestry camps and road prisons are surrounded by perimeter fences and are designed for minimum to medium custody inmates. DOC currently operates 38 work/forestry camps and five road prisons. Inmates are usually transferred to these facilities after completing part of their sentences at a correctional institution and demonstrating satisfactory adjustment. Most of these facilities are located next to correctional institutions, enabling the sharing of facilities like laundry and health services. The inmates housed at these facilities may be assigned to community and public work squads, where they may perform work under the supervision of correctional officers such as cleaning up roadways and rights-of-way, grounds and building maintenance, painting, building construction projects, moving state offices, and cleaning up forests. About 14.9% of the prison population resides in work/forestry camps and road prisons, which collectively served 14,913 inmates in Fiscal Year 2007-08. DOC has begun modifying three of its existing work camps to create a type of non-institutional beds called work prioritized inmate beds. These beds will serve low-risk inmates who would ideally live in the community where the program is located once released from department custody. These beds differ from traditional work and forestry camps in that inmates with shorter sentences (e.g., year and a day or fewer than 18 months) would be assigned to the facilities soon after they are admitted to prison rather than near the end of their sentences. The work prioritized inmate camps would serve only minimum and community custody inmates over 21 years old with no medical, psychological, special education, or substance abuse needs. The department reports that as of December 31, 2008, it had 515 inmates that met these criteria. Additionally, DOC estimates that based on current admission trends, it will receive approximately 110 additional inmates per month who will fit these criteria. 2 These figures only include operational costs and would be higher for newly constructed work release centers. 3 The cost for private work release centers is $7,347 per offender per year. 4 The 2007 Legislature authorized the construction of 600 additional work release beds. However, the department s invitation to negotiate did not result in a contract award; therefore, the 2008 Legislature directed the beds to the department. The department reports that these beds will not be on line until August 2010. 5 Work release centers housed 7,813 inmates during Fiscal Year 2007-08. Bed capacity ranged from 3,145 at the start of the fiscal year to 3,316 as new beds came on line midyear. Beds serve an average of 2.4 inmates per year; the 600 new beds could serve 1,440 inmates per year.

Page 3 Supervised re-entry is not a current DOC program but has been proposed by the department. Supervised re-entry is a form of work release that would allow inmates to remain in the community at night instead of going back to a work release program facility. These inmates would be placed with their families or community sponsors whenever possible, but could also be placed in appropriate treatment centers or halfway houses. As proposed, offenders served in supervised re-entry would be required to meet with corrections and probation officers no fewer than six times per month, including at least one home and one work visit; undergo regular drug testing, background screening, and verification of employment; and attend and pay for any necessary treatment. In a November 2007 analysis, the department indicated it had 500 communitybased custody inmates eligible for work release who could be placed into supervised re-entry, and it anticipated that an additional 500 inmates could be added per month until the program served 3,000 inmates, the level that it would maintain thereafter. As participating inmates would be housed in the community, the program would be appropriate for only those offenders with less serious, non-violent criminal histories. Creating a supervised re-entry program would delay the need to construct additional prison beds; DOC concluded that a program that served 3,000 inmates in Fiscal Year 2008-09 would have delayed the need for $301.2 million in fixed capital outlay spending. 6 DOC indicates that the program would require $3.7 million to hire additional classification officers (22), correctional probation officers (30), and other support staff. Therefore, the net operational savings would have been approximately $22.2 million for Fiscal Year 2008-09. Fiscal impact. At least 3,757 inmates could be diverted from secure prison beds to non-institutional alternatives. These beds are less expensive than secure prison beds. However, the department reports that it cannot calculate the annual operating costs of these beds with the exception of work release. In order to assist the Legislature in evaluating the total savings from funding these alternatives, the department should begin calculating and reporting costs of work/forestry camps and road prisons on an annual basis. Caveats. While expanding non-institutional beds could produce cost savings, this option has three important caveats: (1) the potential for increased escapes and criminal incidents; (2) potential impacts on other department programs; and (3) the need to maintain contingency plans should contracted providers fail to perform. Potential for increased inmate escapes and criminal incidents. There is the increased likelihood of inmate escapes in the lower security environments of non-institutional beds. 7 All 128 inmate escapes from DOC custody in Fiscal Year 2007-08 were from work release centers and work squads. Of these inmates, 56 were recaptured or returned within 24 hours. The number of escapes would likely increase if non-institutional beds were expanded, and there is a risk that such inmates could commit additional crimes. Although the Department of Corrections carefully screens inmates under stringent criteria, the potential for the commission of serious crimes involving work release inmates exists and has occurred in the past. 8 6 For Fiscal Year 2008-09, the department originally submitted a legislative budget request for $649.3 million for prison construction and planning. 7 An escape is defined as any unauthorized absence from the designated facility boundary or absence from any official assignment outside this boundary. 8 In 1990, inmate Donald Dillbeck, a convicted murderer serving a mandatory sentence, was housed at the Quincy Annex and enrolled in a cooking class. Although he was not at a work release center, the inmate was assigned to a cooking crew and assisted at a community function preparing food. The inmate escaped and two days later he murdered a woman in the parking lot of a Tallahassee mall. As a result, the Governor ordered any inmate serving a mandatory sentence be housed within a secure correctional perimeter. Approximately 2,000 inmates were removed from work release centers and returned to state prisons. In 1989, two inmates on work release in Bradenton sexually assaulted women in the community. As a result, all inmates housed at work release centers convicted of false imprisonment or kidnapping were returned to state prisons and the Bradenton Work Release Center was closed.

Page 4 Impact on other programs. Expanding non-institutional beds would have an impact on other DOC programs that serve the same pool of eligible inmates, including PRIDE and various prison educational programs and treatment programs. 9 It would be important for the department to quickly replace inmates who are assigned to prison work squads if these inmates are placed in other non-institutional beds because the department is contractually obligated to provide work squads to the Department of Transportation and local governments. Need for contingency plans. DOC would need to maintain a supply of contingency placements for inmates in non-institutional facilities run by private contractors. As some vendors could fail to meet program requirements or terminate their contracts, the department would need to relocate the inmates to other placements or quickly replace the vendor. Expand Maximum Allowable Gain Time Credits for Low-Risk Inmates Who Exhibit Good Behavior In 1995, the Legislature began requiring inmates sentenced to prison to serve a minimum of 85% of the sentence imposed. This requirement allows inmates to reduce their full sentence by up to 15% by earning gain time. There are three types of gain time: basic gain time, which is an entitlement offered to all inmates but not currently in use; incentive gain time, which is based on good behavior; and meritorious gain time, which is awarded to inmates who perform an outstanding deed such as saving a life. Changes in gain time for certain classes of offenders could produce substantial savings. For example, the Legislature could increase the maximum allowable gain time to about 35% for low-risk non-violent inmates who exhibit good behavior, such as inmates in prison for low-level drug possession; this could be done by allowing these inmates to earn an additional five days of incentive gain time credits each month (increasing from 10 to 15 days per month served). The Office of Economic and Demographic Research projected that under this scenario, the department would have released an additional 564 offenders in Fiscal Year 2008-09 for a savings of $10.7 million. These early releases would also reduce the need to build new prison beds, saving an additional $42.2 million. The disadvantage of this option is that it would reduce the length of time that offenders are incarcerated, which could increase public safety risks as some released persons could re-offend. If the Legislature wishes to pursue this option, it could manage these risks by requiring such inmates to serve a term of community supervision with or without electronic monitoring to replace the time they would not spend in prison. This would increase DOC Community Corrections (probation) caseloads and require additional staffing, but would still produce substantial savings due to the lower cost of community supervision and avoidance of prison construction costs. Supervising an inmate in the community costs about $2,000 per year compared to the $19,000 per year average cost of incarceration. The additional cost of electronic monitoring ranges from $2 per day for radio frequency monitoring to $9 per day for active-gps monitoring. 9 PRIDE (Prison Rehabilitative Industries and Diversified Enterprises, Inc.) is a non-profit corporation that operates Florida s prison industries program. PRIDE provides inmates with vocational and on-the-job training that can lead to meaningful employment after release; it also provides limited post-release job placement and support.

Page 5 Modify Drug Possession Penalties Some states have changed their sentencing laws to modify penalties for certain drug offenses to reduce incarceration expenses. These states have reclassified certain felony drug penalties as misdemeanors. 10 Florida could reduce prison costs by reclassifying low-level marijuana possession offenses as misdemeanors for first- or second-time offenders who have never committed violent, forcible, or sexual offenses. 11 Instead of being eligible for prison, these offenders could receive sanctions such as fines, probation or community supervision, and incarceration in a county jail for up to one year. The Office of Economic and Demographic Research reported that 43,056 offenders were sentenced for drug possession as their primary offense in Fiscal Year 2007-08, of whom 5,298 (12.3%) were sentenced to prison. Most of these persons (57.3% or 3,037) who were sentenced to prison scored less than the 44-point threshold for a mandatory prison sentence, and 1,265 scored fewer than five prior record points, meaning that they likely did not have significant prior offenses. 12 Assuming that half of these 1,265 offenders qualified for prison diversion, the state would save $10.4 million annually. There are two primary caveats to this alternative. First, it may not free up very many additional prison beds, because offenders in prison for marijuana possession may actually have pled down from a more serious drug trafficking charge. Second, it could increase public safety risks as some drug offenders who otherwise would be sent to prison could commit additional offenses while under community supervision. This option would also limit judicial discretion in sentencing, and could be seen as softening public disapproval of drug offenses. Expand Alternative Sanctions for Probation Violations When offenders on community supervision violate the terms of probation set forth by a judge at sentencing, they can be arrested and sent or returned to prison. These violations can include committing new crimes as well as committing technical offenses such as missing curfew or not keeping appointments with their probation officer. Historically, a significant number of offenders who commit technical violations are returned to prison. The Office of Economic and Demographic Research projects that approximately 1,400 inmates will be admitted into Florida s prisons in Fiscal Year 2008-09 for technical probation violations. The Legislature could consider several options to expand alternative sanctions for low-risk, non-violent offenders in lieu of incarceration, including day reporting centers, graduated sanctions, and house arrest with electronic monitoring. These alternatives would produce savings, as their costs are lower than the $16,410 cost for housing offenders in regular adult male prison beds (based on Fiscal Year 2007-08 costs). 13 Day Reporting Centers are highly structured non-residential sanction and intervention programs that provide intensive treatment for offenders who have not responded to more traditional supervision and treatment efforts. Programs may include components such as substance abuse counseling, cognitive restructuring, adult basic education, employment enhancement, intensive supervision, community service, and an aftercare component may follow the on-site programming. Failure at a day reporting center results in 10 For example, in November 2008, Massachusetts voters approved a referendum to decriminalize the possession of marijuana; under this referendum any person caught with less than an ounce of marijuana is punishable by a civil fine of $100. New York and Mississippi have recently decriminalized marijuana and have established civil fines for marijuana possession. Several states and municipalities have also decriminalized non-medical cannabis. However, cannabis is illegal under federal law. Gonzales v. Raich (2005) held in a 6-3 decision that the commerce clause in the United States Constitution allowed the federal government to ban the use of cannabis, including medical use even if local laws allow it. 11 To do so, the Legislature would need to revise s. 893.13, F.S., to classify these offenses as a misdemeanor, and would need to establish an amount threshold and ensure that penalties for other drug offenses such as sale and trafficking remain intact. 12 Florida statutes require that offenders who score 44 points on the sentencing score sheet serve a mandatory prison sentence; however, there is no minimum score for prison and judges have discretion to sentence any offender to prison. 13 For more information on incarceration of offenders on community supervision for technical violations, see Department of Corrections Zero Tolerance Policy Increases Offender Scrutiny But Is Not Based on Risk to Public Safety, OPPAGA Report No. 07-13, February 2007.

Page 6 revocation proceedings against the offender. Day reporting centers are currently used by some Florida county corrections programs such as Broward and Orange, and in states such as Georgia, West Virginia, and South Carolina. The costs of day reporting centers vary: West Virginia s program costs about $1,800 per inmate annually, South Carolina s program costs about $2,300; and Georgia s program costs about $4,300. Graduated Sanctions enable probation officers to apply a continuum of administrative sanctions when offenders on community control commit technical violations; these sanctions may include loss of privileges, increased reporting and drug testing, the imposition of curfew, and community service. Guidelines governing use of the graduated sanctions take into consideration the determined risk level of the individual, previous violations, the nature of the technical violation, and the range of potential responses. Graduated sanctions offer a less costly alternative to automatic arrest and incarceration for technical violations and would significantly reduce the number of technical violations processed by the courts, the Parole Commission, state attorneys, and public defenders. This approach is currently used in states such as Arizona, Colorado, Georgia, and Indiana. House Arrest with Electronic Monitoring would allow offenders with technical probation violations to remain at home instead of being returned to prison while being monitored using an electronic sensor strapped to their ankle. Supervising an inmate in the community costs about $2,000 per year compared to the $16,410 per-year average cost for a secure prison bed. The additional cost of electronic monitoring ranges from $2 per day for radio frequency monitoring to $9 per day for active-gps monitoring. Fiscal impact. Assuming that 25% of the 1,400 technical violators projected to be sentenced to prison were alternatively sanctioned, the state could save $5.7 million annually. 14 Caveats. Available research indicates that while alternative sanctions are less expensive to incarceration for technical violations, it is unclear whether offenders subject to alternative sanctions are more likely to re-offend than those returned to prison. (1) Creating day reporting centers would require additional resources, including facilities, and probation and treatment staff. (2) A significant limitation in using graduated sanctions is that they cannot be used if the offender does not admit guilt and waive right to a formal hearing, and judges would need to specify the list of graduated sanctions that probation officers could apply in each case at the time of sentencing. (3) Also, while the department has statutory authority to place offenders on electronic monitoring, it has been reluctant to do so unless stipulated in an original court order. As a result, the department has historically underutilized its existing electronic monitoring resources, and has returned these funds back to general revenue. To control potential public safety risks, if the Legislature authorizes prison alternatives to probation violations it should direct the department to continue to arrest offenders who commit new crimes, are violent or dangerous, pose a risk to public safety, or are deemed to pose a risk to flee from supervision. The department should also increase its oversight of offenders who commit technical violations that are not reported to the courts, such as by requiring these persons to have additional monthly contacts with probation officers. 14 The cost of the alternatives (e.g., community supervision) would have to be deducted to calculate net savings.