Review of Bay Correctional Facility and Moore Haven Correctional

by OPPAGA

Review of Bay Correctional Facility and Moore Haven Correctional Facility

April 1998

Office of Program Policy Analysis and Government Accountability Office of Program Policy Analysis and Government Accountability
Report No. 97-68 Report No. 97-68 1
Post Office Box 1735
Tallahassee, Florida 32302
111 West Madison Street, Room 312, Claude Pepper Building
Tallahassee, Florida 32301
850/ 488-0021 SUNCOM 278-0021 FAX 850/ 487-3804

The Florida Legislature

OFFICE OF PROGRAM POLICY ANALYSIS AND GOVERNMENT ACCOUNTABILITY

John W. Turcotte, Director

April 1998

The President of the Senate, the Speaker of the House of Representatives, and the Legislative Auditing Committee

As requested by the Legislative Auditing Committee, I have directed that a review be made of Bay Correctional Facility and Moore Haven Correctional Facility. The results of this review are presented to you in this report. This review was conducted by Lee Cobb and Charles Barrett under the supervision of Byron Brown.

We wish to express our appreciation to the staff of the Correctional Privatization Commission, the Department of Corrections, the Corrections Corporation of America, and the Wackenhut Corrections Corporation for their assistance.

Sincerely,

John W. Turcotte
Director

i
Executive Summary Report No. 97-68
Bay Correctional Facility and Moore Haven Correctional Facility

In the summer of 1995, two private prisons opened under contract with the state through the Correctional Privatization Commission: Bay Correctional Facility, operated by Corrections Corporation of America and Moore Haven Correctional Facility, operated by Wackenhut Corrections Corporation. As required by s. 957.11, F. S., this review evaluates the costs and benefits of those contracts and the performance of the private contractors and recommends whether the contracts should be continued.

Prison privatization is often advocated as a means to reduce corrections costs. When it awarded contracts in 1994, the Correctional Privatization Commission projected that each contract would meet the 7% cost savings requirement established in law.

The first two private prisons built through the commission differ from most public prisons in Florida, partly because the Legislature intended the private prisons to provide more programs than are typically offered in public prisons. In addition, the private prisons are smaller than most public prisons and house inmates who are less likely to escape and less likely to have medical and psychological restrictions than inmates in public prisons. To analyze the construction and operating costs and the performance of Bay and Moore Haven in comparison to public prisons, we selected the most comparable public prison( s), and then made adjustments as necessary to improve comparability.

Although vendor performance during the 1996-97 fiscal year was satisfactory, the private prisons are not providing the state with the level of overall cost savings initially projected by the commission. The Bay Correctional Facility was more costly to construct than the most comparable public prison, and did not provide operating cost savings to the state during the 1996-97 fiscal year. The cost of constructing Moore Haven Correctional Facility was within the range of public prison construction costs, and provided a savings of about $480,000 during the 1996-97 fiscal year, which is about 4% less than the most comparable public prison.

Background
Conclusions
Scope

ii

Comparisons of the construction costs paid by the state for public and private prisons show that private prison site acquisition and preparation costs were lower than the public prison costs of site preparation on donated land. However, the public prison building costs were lower than those of private vendors, partly due to the state's use of inmate labor. Differences in the size of public and private prisons constructed in 1995 limit conclusions about comparisons of construction costs to the state.

The commission did not structure the contracts to ensure that the projected level of cost savings was achieved. Three major factors reduced the level of operating cost savings achieved by the private prisons during the 1996-97 fiscal year from the level of savings initially projected. These factors included the addition of indirect state costs to the contracted costs of private prisons, the Legislature's use of canteen profits and telephone commissions to pay for some of the costs of operating public prisons, and an unfavorably structured contract between one of the vendors and the commission. Additional savings could be achieved by increasing the size of the private prisons and by using the private prisons to house higher custody inmates.

The private prisons have introduced some different methods of construction and operation into the state correctional system, such as a more compact prison design and greater use of technology such as cameras in the housing units. However, it is too early to determine if implementation of these alternatives would improve existing department practices.

We recommend that the Legislature authorize the Correctional Privatization Commission to renew its contracts with the Corrections Corporation of America for the operation of Bay Correctional Facility and with Wackenhut Corrections Corporation for the operation of Moore Haven Correctional Facility. However, certain modifications should be negotiated as a part of the contract renewal, such as requiring that canteen profits and telephone commissions be deposited with the state and that performance-based program budgeting measures be included in the contracts. Requiring canteen profits and telephone commissions to be deposited with the state to offset the costs of private prisons would provide an annual savings to the state of about $600,000 for these two private prisons.

Recommendations

iii

We also recommend that the commission renegotiate the contract for Bay to reduce the inmate per diem rate for inmates in excess of 90% capacity. The department should assign only enough inmates to Bay to keep its population at 90% of capacity until the contract has been renegotiated. Placing inmates in those prisons when space is available at less cost, would save the state approximately $400,000 per year.

Representatives of the Correctional Privatization Commission, the Department of Corrections, the Corrections Corporation of America, and the Wackenhut Corrections Corporation raise a number of challenges to the methodology used by OPPAGA in arriving at the conclusions presented in this report. We have reviewed all of these challenges, and concluded that they do not warrant changes in the methodology or conclusions presented in our report.

The commission and the Corrections Corporation of America reject the conclusions of our report. They assert that public and private prisons cannot be compared, but they both also assert that prison privatization is providing the state with savings in excess of 7% per year.

The department disagrees with our methodology, arguing that if we had used multiple public prisons as a basis for comparison, we would have found that public prisons were at least 11% less costly than private prisons.

The response from Wackenhut Corrections Corporation highlights the accomplishments achieved at Moore Haven Correctional Facility. With the exception of one point, Wackenhut did not take issue with our methodology or conclusions.

All four responses substantially agree with the recommendations included in our report.

Agency Responses

Chapter 1: Introduction

Purpose

In the summer of 1995, two private prisons opened under contract with the state through the Correctional Privatization Commission: Bay Correctional Facility, operated by Corrections Corporation of America, and Moore Haven Correctional Facility, operated by Wackenhut Corrections Corporation. As required by s. 957.11, F. S., this review evaluates the costs and benefits of those contracts and the performance of the private contractors and recommends whether the contracts should be continued.

Background

The Legislature established the Correctional Privatization Commission in 1993 for the purpose of entering into contracts for the design, construction, and operation of private prisons. 1 The commission, located within the Department of Management Services, issued requests for bid proposals, selected vendors, and entered into contracts for two 750-bed prisons.

The practice of governments entering into contracts with private companies to build and/ or operate correctional facilities is a relatively new and growing practice. According to industry sources, the international inmate capacity of privately operated prisons increased from approximately 3,000 in 1987 to over 84,000 in 1996.

Prison privatization is often advocated as a means to reduce corrections costs. Although arguments have been advanced that correctional privatization can lead to improved quality, increased flexibility and accountability, and decreased liability to the state, cost is the primary issue upon which the success of privatization is to be evaluated. In Florida, the Legislature requires that the commission enter into a contract for the construction and operation of a private prison only if the contract is projected to result in at least 7% savings over public provision of a similar facility.

1 Chapter 93-406, Laws of Florida.

When it awarded contracts in 1994, the Correctional Privatization Commission projected that each contract would meet the 7% cost savings requirement established in law. These initial projections were based on costs at three public prisons during the 1992-93 fiscal year as certified by the Auditor General and adjusted by the commission. The commission based its cost savings projection on a combination of first-year operating costs at 95% capacity and annualized construction costs.

In this review, we analyze actual construction costs and compare actual operating costs of the public and private prisons for Fiscal Year 1996-97, the first full year of operation for the private prisons. In addition, we provide preliminary data regarding public and private prison performance.

The first two private prisons built through the commission are not directly comparable to the public prisons operated by the department, partly because the Legislature intended the private prisons to provide more programs than are typically offered in public prisons. In addition, the private prisons are not directly comparable to most public prisons because of differences in size, type of inmates housed, and age of prison. Although the department operates 55 prisons, most house more than 1,000 inmates and have a greater percentage of close custody inmates and inmates with medical and psychological restrictions than do private prisons. Appendix A contains a list of public prisons for adult males that are most similar in security level to Bay and Moore Haven.

To analyze the construction and operating costs of Bay and Moore Haven in comparison to public prisons, we selected the most comparable public prison( s) and then made adjustments to the costs as necessary to improve comparability. Appendix B contains a list and explanation of the adjustments that we made to compare the operating costs of the public and private prisons. We did not make adjustments to construction costs or performance data due to large differences in available data on construction costs and performance for public prisons.

This report has three major sections addressing:

· costs, which compares construction and operating costs of Bay and Moore Haven with the department's reported costs for the most comparable public prisons and discusses the implications of this data; 2

· performance, which compares the performance of Bay and Moore Haven with public prisons and discusses approaches that offer alternatives to current department practices; and

· conclusions and recommendations, which discusses the benefits and drawbacks of privatization and provides recommendations for modifying future contracts with the Corrections Corporation of America (CCA) and Wackenhut Corrections Corporation (Wackenhut).

2 The department reports inmate per diem rates for each public prison in its annual reports. On February 18, 1998, the department informed us that it had discovered accounting errors in the inmate per diems reported in its 1996-97 Annual Report. Certain transfers of inventory had been improperly counted as expenses. In this review, we have used the department's revised inmate per diem rates, which are lower than those reported in the 1996-97 Annual Report. The department also provided us with calculations of indirect costs. Although neither the public prison costs nor the indirect costs have been audited, we reviewed the cost data to ensure that the costs reported for the public prison were reasonable and that the department had accounted for all of its expenses for the 1996-97 fiscal year in its reported costs.

Chapter 2: Costs

The construction and operation of two 750-bed private prisons by the Correctional Privatization Commission has not provided the state with the level of cost savings initially projected by the commission. Only one of the two private prisons provided any construction or operating cost savings to the state.

The state pays each private prison vendor separate contracted rates for construction and operating costs. The commission structured the contracts so that construction costs are paid by the state over a 20-year period in equal annual amounts, while the contracted operating costs are increased each year for inflation, as provided for in the contracts. Although the vendors were required to submit bids estimating their costs for various components of prison construction and operation, the state's payments are based on the contracted rates rather than actual vendor costs. For the purposes of this review, we have compared the actual contracted constructing and operating costs of the Bay and Moore Haven private prisons with the department's reported actual costs of constructing and operating public prisons.

Construction Costs

Contracted private prison site acquisition and preparation costs were lower than comparable costs for public prisons, but private prison building costs were higher. Factors affecting variations in costs include location, inmate labor, and economies of scale.

Construction costs can generally be grouped into two categories: site acquisition and preparation costs, and building costs. To conduct our analysis, we used actual construction cost data for the three public prisons opened in 1995 (Everglades, Taylor, and Washington Correctional Institutions). 3 For Bay and Moore Haven, we used the contracted costs for site acquisition and preparation, and building costs. 4 (See Exhibit 1.)

3 See Appendix A for a discussion of the selection of these prisons. These three public prisons have much larger prison capacities than the private prisons, thus limiting the extent to which comparisons of construction costs can be conclusive. Our analysis did not include estimates of future costs for repair or renovation.

Exhibit 1: Two Public Prisons Were Less Costly Per Bed to Construct Than Smaller Private Prisons

Moore Haven Bay Taylor Washington Everglades
Cost Category (Private) (Private) (Public) (Public) (Public)
Number of Beds 750 750 1,122 1,126 1,538
Site Costs $ 1,340,000 $ 2,580,085 $ 6,294,774 $ 5,955,579 $18,400,000
Building Costs 16,715,500 21,428,187 17,756,753 17,536,076 30,217,288
Total Cost $18,055,500 $24,008,272 $24,051,527 $23,491,655 $48,617,288
Cost Per Bed $24,074 $32,011 $21,436 $20,863 $31,611 1

The Bay prison included more classroom and program space (22,000 square feet) than Moore Haven (12,000) and the public prisons (15,400 square feet each).

Source: OPPAGA analysis of data provided by the Department of Corrections and the Correctional Privatization Commission. (See Appendix C for more details.)

Despite the fact that public prisons used donated sites, site acquisition and preparation costs were substantially lower for private vendors, partly because they built more compact prisons.

Contracted site acquisition and preparation costs were 11% or less of the total construction costs for the two private prisons. By contrast, public prison site preparation costs were at least 25% of total costs. On a per bed basis, the private vendors acquired and prepared the prison sites at a much lower cost than the state prepared donated sites for the public prisons.

Exhibit 2: Private Prison Site Acquisition and Preparation Costs Were Lower Than Public Prison Costs(Omitted)

4 Each private vendor's bid designated a certain amount for these categories. The payment to the vendor is based on the total bid, however, not the individual categories. Therefore, these bid amounts may not represent actual vendor expenses.

Given the significantly lower acquisition and preparation costs of the private prisons, it is ironic that each of the public prisons used donated land, whereas each of the private vendors purchased land. 5 The donated Everglades site in particular incurred unusually high preparation costs totaling $18.4 million, largely due to its location and terrain.

Private prisons' lower site acquisition and preparation costs are due in part to the vendors' use of a compact prison design. The private prisons are constructed with all facilities under a single roof, while department prisons typically include several buildings spread out in a "campus" design. The compact design requires less acreage to be acquired and prepared for construction. (See Appendix C for details on site acreage.)

Building costs were substantially lower for public prisons, partially as a result of using inmate labor.

In contrast to the site acquisition and preparation costs, the building costs for public prisons were substantially lower than the private prisons. The building costs per bed were lower for all three of the public prisons in comparison with the two private prisons. (See Exhibit 3.)

Exhibit 3: Private Prison Per Bed Building Costs Were Higher Than Public Prison Building Costs

Total Beds, Per Bed Building Cost

Moore Haven (private) 750 $22,287
Bay (private) 750 28,571
Taylor (public) 1,122 15,826
Washington (public) 1,126 15,574
Everglades (public) 1,538 19,647

Source: OPPAGA analysis of data provided by the Department of Corrections and the Correctional Privatization Commission

One major factor that reduces public prison building costs is the department's use of inmate labor in some construction, a benefit not available to the vendors that built Florida's private prisons. Typically, the department uses inmates to construct some of the buildings on a prison campus. The inmates are not paid for this labor. In 1993, the Auditor General estimated that not using inmate labor would have increased public prisons' total construction costs by about 16%.

5 The department often builds prisons on land donated by local government. Site preparation costs include the costs of roads, sewage treatment, and other infrastructure for the site.

Looking at construction costs as a whole, we noted the large difference in the contracted cost of the two private prisons. The total construction cost for Bay was 33% more than the total construction cost for Moore Haven. Although CCA and Wackenhut responded to the same request for proposals, CCA built Bay with all inmates housed in two-person cells, whereas Wackenhut used a combination of dormitories and cells in Moore Haven. Cells are generally more expensive to construct than dormitory housing.

The state's cost for constructing Moore Haven was within the range of public prison costs. Moore Haven was less expensive to build than one of the three public prisons opened in 1995, even without making adjustments for size, location, or classroom space. 6 However, the cost to the state of constructing Bay was 1% higher than the most expensive public prison, 53% higher than the nearest public prison, and 33% higher than the other private prison. Although qualified by the lack of adjustments for variations between private and public prisons, we conclude that Bay did not provide the state with any savings in construction costs.

Operating Costs

The commission has not structured the contracts to ensure that the state achieves the level of cost savings initially projected. As a result, private prisons are not providing the state with the level of cost savings initially projected by the commission. During the 1996-97 fiscal year, Bay Correctional Facility did not provide operating cost savings to the state and Moore Haven Correctional Facility provided approximately 4% operating cost savings.

Prison operating costs essentially include all non-construction costs associated with a prison. 7 The state pays each private vendor a per diem rate for each inmate that is housed in the prison. That payment represents the cost to the state of operating the private prison. Therefore, our comparison focuses on the cost to the state of operating private and public prisons rather than a comparison of what it cost the vendor. 8

6 Because of the variations in the construction costs of the three public prisons combined with the differences between the private and public prisons, we concluded that an attempt to adjust private prison costs for size, location, and classroom space would result in spurious conclusions regarding construction cost savings.

7 Prison operating costs include the cost of security, food service, medical services, prison administration, utilities, prison maintenance, classification, substance abuse and educational programming, religious services, and recreation. Operating costs also include all personnel costs of the prison and all costs directly related to inmate management. Operating costs do not include construction, fixed capital outlay, or major maintenance and repair costs.

Construction of Bay Did Not Provide Savings to the State

Costs of operating private prisons did not meet the expected level of savings for the 1996-97 fiscal year.

The commission initially projected that both Bay and Moore Haven would provide the state with cost savings in excess of 10% over the public prison costs originally certified by the Auditor General, as adjusted by the commission. 9 However, for the 1996-97 fiscal year, Bay did not provide the state with any cost savings, and Moore Haven provided the state with approximately 4% in cost savings. (See Exhibit 4.) The adjusted operating costs for Moore Haven were $481,581 less than the total adjusted costs of operating the most comparable public prison.

Exhibit 4: Private Prisons Did Not Achieve the Level of Operating Cost Savings Projected by the Commission

Bay, Moore Haven, Lawtey

Unadjusted Per Diem $44.22, $45.39, $45.37
Net of Adjustments (. 06), (3.03), (2.96)
Indirect Costs 1.92, 1.82, 3.57
Total $46.08, $44.18, $45.98
Estimated Savings (0.2)% 3.9%
Annualized Savings $( 26,754) $481,581

Source: OPPAGA analysis of data provided by the Department of Corrections and the Correctional Privatization Commission. (See Appendices B and C for more detail.)

We determined that Lawtey Correctional Institution was the public prison most comparable to Bay and Moore Haven. Lawtey is most similar in size and inmate custody level to the private prisons. Major differences between Lawtey and the private prisons include the number of inmates with medical restrictions, and the amount of education and substance abuse treatment programming. As a result, the medical, education, and state costs at Lawtey are not representative of actual public prison costs to provide services similar to those provided at the private prisons.

8 In their bids, vendors assigned cost amounts to various categories such as personnel, health services, education, substance abuse, food service, and indirect costs (of the company). The state pays only the total cost of the contract, however, and does not monitor whether the vendor actually spends the amounts designated in the contracts. Therefore, we did not make detailed analyses of cost component variations between public and private prisons.

9 Although the commission included construction costs in its initial projections of cost savings, all of the commission's projected savings for the Bay facility and most of the projected savings for Moore Haven were to come from lower operating costs.

To determine actual public prison costs for providing medical, education, and substance abuse services and programs similar to Bay and Moore Haven, we identified public prisons that had services and programs similar to Bay and Moore Haven in each of the three categories. For medical services, we identified the four prisons closest in size with similar inmate medical profiles. For education and substance abuse programs, we identified at least two prisons that served similar numbers of inmates through their educational and substance abuse programs. Using the actual costs for medical services, education programs, and substance abuse programs at these other public prisons, we calculated the cost per inmate if similar services and programs were provided at Lawtey, and made the appropriate adjustments.

To correct for other differences between the public and private prisons, we made adjustments to actual costs for population, public work squads, area salary differences, and major maintenance and repair costs. We also adjusted private prison costs to reflect direct costs incurred by the department on behalf of the private prisons for classification staff and medical services, as well as the direct costs incurred by the commission in its oversight responsibilities. 10

Funds returned to the state treasury from prison operations reduce the cost to the state of both public and private prisons. For public prisons, the profits received from commissary operations and inmate telephone commissions are placed in the state treasury for appropriation by the Legislature, and thus effectively subsidize part of public prison costs. 11 For private prisons, payments of state sales taxes and of corporate income taxes by vendors can be considered as credits against the cost of private prisons.

CCA and Wackenhut provided estimates of their payments of state sales tax for the 1996-97 fiscal year of $56,153 and $142,141, respectively. However, after reviewing documentation provided by the vendors, we concluded that $40,000 was a more reasonable estimate of the sales taxes paid by these vendors. This resulted in a credit of $. 15 per inmate per day for each prison. We did not accept vendor estimates of sales tax payments for supplies that may not have been taxed or that may have been purchased out of state. 12

10 Florida law requires contractors to reimburse the commission for the salaries and expenses of the contract monitors. In Fiscal Year 1996-97, the commission deducted a total of $148,949.41 from the contracted per diem rates for Bay and Moore Haven combined, but paid a total of $94,885.21 in monitor salary and expense.

11 Commissary profits and telephone commissions from the private prisons are retained by the vendors in a special fund and cannot be spent without the approval of the commission's executive director. See additional discussion of the Inmate Welfare Trust Fund on page 13.

We did not make any adjustments for corporate income tax payments. CCA estimated that approximately $19,000 in Florida corporate income taxes for the 1996-97 fiscal year could be attributed to Bay. However, CCA's tax return provided as documentation did not differentiate income derived from the Bay facility from income from other CCA operations in Florida. Wackenhut reported that it paid no Florida corporate income tax for the 1996-97 fiscal year.

Three major factors reduced the level of operating cost savings achieved by the private prisons during the 1996-97 fiscal year from the level of savings initially projected. For each of these factors, the commission did not structure its calculation of cost savings or the contract with the vendors in a way that maximized savings for the state. These factors include:

· the calculation of state indirect costs;

· different uses of canteen profits and telephone commissions; and

· the structure of per diem rates within the contracts.

Two other factors, increasing the size of the private prisons and using the private prisons to house higher custody inmates, could be used by the state to potentially increase the level of savings achieved through privatization.

12 If the highest vendor estimate of sales taxes paid were verified, it would have resulted in a credit of an additional $0.38 per inmate per day. This would increase the level of savings by less than 1% for each vendor.

The inclusion of indirect state costs in the calculation of private prison costs is one reason that a lower level of cost savings was achieved than initially projected by the commission.

The law requires that a determination of indirect administrative costs, or overhead, be included in the analysis of public and private prison costs. Section 957.07, F. S., directed the Auditor General to include "all administrative costs associated with central administration" in the certification of public prison costs. In 1993, the Auditor General calculated that a cost of $3.15 per inmate per day should be included in public prison costs used for comparison with the private prisons.

In its initial projections of cost savings to be achieved by the private prisons, however, the commission did not estimate its own costs as a part of the indirect costs associated with private prisons, nor did it estimate any department costs that would be associated with the private prisons. This omission led the commission to overestimate the projected savings to be achieved through these two contracts. 13

In our comparison of operating costs for the 1996-97 fiscal year, we have included calculations of indirect state costs for both the public and the private prisons. Based on information provided by the department and the commission, we estimate that indirect state costs added approximately $500,000 to the contracted costs of private prisons during the 1996-97 fiscal year. These costs include the costs associated with the administration of the commission and those department costs that can be reasonably identified as providing an indirect benefit to the private prisons, such as Management Information Services and admissions and release.

The inclusion of state indirect costs in our comparison increased the private prison inmate per diems by about 4% ($ 1.92 per inmate per day for Bay and $1.82 per inmate per day for Moore Haven). Indirect state costs for public prisons for the 1996-97 fiscal year were over $950,000, increasing the public prison per diem by about 8% ($ 3.57 per inmate per day).

13 In recognition of this omission, the commission has added some indirect costs to private prison operating costs in its subsequent efforts to compare the cost of public and private prisons.

Indirect State Costs Not Included in Original Projection of Private Prison Costs

Although it is proper accounting procedure to include indirect overhead costs in unit cost analyses, these costs can be misleading in determining actual savings. Indirect administrative costs are real costs to the state, but they do not increase proportionately as direct costs do. Instead, indirect costs generally act similarly to fixed costs. When a new prison is authorized, the Legislature generally authorizes funds necessary to cover the direct costs, but does not authorize a corresponding increase in indirect costs. The new cost to the state of establishing a prison is primarily reflected in its direct costs. The state's adjusted direct costs for the 1996-97 fiscal year do not reflect a cost savings for Bay ($ 11.8 million) or Moore Haven ($ 11.3 million) in comparison to Lawtey ($ 11.3 million).

The commission has allowed the vendors to retain possession of the earnings of the inmate canteens and telephone commissions, rather than using those funds to reduce state costs as is done in the public prisons.

The Inmate Welfare Trust Fund consists of profits from the department's inmate canteens and commissions collected on long distance phone calls made by inmates. In 1994, the Legislature changed the way this fund is used, so that profits are now used to pay certain prison operating costs. For the private prisons, these profits and commissions are maintained in a separate fund and are not used to reduce the operating costs of the private prisons to Florida taxpayers. 14

The commission has allowed vendors to retain possession of the earnings of the Inmate Welfare Trust Fund in interest-bearing accounts. Any purchase from this fund other than the purchase of items for resale requires the approval of the executive director of the commission.

During the 1996-97 fiscal year, Bay and Moore Haven reported profits (after the cost of sales and operating expenses) of $272,813 and $355,957, respectively, for their Inmate Welfare Trust Fund accounts. The executive director has authorized each vendor to spend less than $10,000 on special activities such as graduation exercises for prison education programs. The purpose of the remaining funds has not yet been determined, although the commission is considering a proposal to use Moore Haven's fund to construct a building for a prison industry enhancement program. As of February 19, 1998, almost all of the profits and commissions had not been expended. These potentially could be reverted to the state to subsidize the cost of the private prisons, similar to the use of the Inmate Welfare Trust Fund for public prisons. 15

14 The original contract with vendors stated that the commissary would be operated in accordance with s. 945.215, F. S. After the Legislature amended that provision of law in 1994, preventing the department from spending commissary profits independent of legislative appropriation, the commission amended its contracts with the vendors to allow the vendors to retain possession of the Inmate Welfare Trust Fund.

The Legislature has used the earnings of the Inmate Welfare Trust Fund from public prisons to pay for a variety of prison operating components, such as educational programs, libraries, and religious programs. If the canteen profits and telephone commissions from private prisons were also used to subsidize private prison costs, then state costs would be reduced by over $600,000 per year for these two prisons. If these amounts were used to subsidize private prison per diems, Bay's per diem would have decreased by $1.06 (an additional cost savings of 2.3%) and Moore Haven's per diem would have decreased by $1.38 (an additional cost savings of 3.1%).

Calculations of average prison costs are highly sensitive to variations in prison population, and contracts that are not structured favorably, as in the case of the Bay Correctional Facility, lead to excess costs to the state.

Average prison costs are calculated by distributing total prison costs equally for each inmate housed in the prison. However, because both fixed and variable costs are included, the average prison cost can fluctuate significantly with the addition of a few inmates. A prison's fixed costs, such as personnel and most utilities, will be incurred at similar levels each month, regardless of fluctuations in the population. Certain variable costs, such as food and medical supplies, will change as the population varies. The department estimates that, once its fixed costs have been covered, each additional inmate costs the department $6.63 per day.

In its private prison contracts, the commission structured the funding to pay the vendors a guaranteed per diem rate for inmates up to 90% of capacity. The state is obligated to pay this guaranteed rate even if the population falls below 90%, thus providing a guarantee to the vendors that the state will cover their fixed costs. The vendor is also paid a second, lower per diem rate for each inmate in excess of 90% capacity.

The contract with Wackenhut Corrections Corporation for operating Moore Haven prison specifies a guaranteed rate for 675 inmates of $46.71 per inmate per day. For every additional inmate up to 750 inmates, the state pays only $8.87 per day. This contract reflects the smaller marginal costs associated with housing additional inmates over the guaranteed rate.

15 Commission staff have asserted that the private prison's Inmate Welfare Trust Fund is state money because it cannot be spent without the approval of the executive director. However, the comptroller for Corrections Corporation of America told us that CCA pays corporate income tax on the profits of the fund. In contrast, the audited financial statements of the Moore Haven Inmate Welfare Trust Fund account states that the fund is exempt from income taxes, as they are held in trust for the benefit of facility inmates. As long as the fund remains under the control of the vendor, and outside the state treasury, we can not consider it as a credit against private prison costs.

By contrast, the commission's contract with the Corrections Corporation of America for operating Bay does not accurately reflect the decreased cost of housing inmates over the guaranteed rate. The Bay contract calls for guaranteed payments of $44.37 for the first 675 inmates, and payments of $41.60 for each inmate in excess of 90% capacity. (See Exhibit 5.) Although Bay's guaranteed rate is lower than Moore Haven's, the marginal rate makes it unfavorable to the state to house inmates at Bay in excess of 90% capacity if vacant beds are available at other prisons. 16

Exhibit 5: Orginal Contract Rate for Bay Is Unfavorable

Private Prison Guaranteed Rate (for each inmate up to 90% of capacity)

Marginal Rate (for each inmate over 90% capacity)

Marginal as Percent of Guaranteed

Bay $44.37 $41.60 94%
Moore Haven 46.71 8.87 19%

Source: OPPAGA analysis of data provided by the Correctional Privatization Commission

As a result, the payment for inmates in excess of 90% capacity at Bay is higher than the comparable cost of housing those inmates at other prisons. In Fiscal Year 1996-97, the state housed an average of 708 inmates per day at the Bay prison, 33 inmates over 90% capacity. The state paid over $500,000 to the Corrections Corporation of America to house inmates that could have been housed at Moore Haven or at department facilities for about $100,000 or less. 17 Limiting the population of Bay to 90% of capacity would have resulted in savings for the state of approximately $400,000 during the 1996-97 fiscal year. If the Bay contract had used a marginal rate similar to Moore Haven, it would have lowered Bay's adjusted operating costs by about 5.6% per inmate per day. 18

16 The commission accepted these rates as proposed by CCA in its response to the request for proposals.

17 During the 1996-97 fiscal year, the department maintained Bay and Moore Haven at 94% capacity, while the public prison system operated at an average of 89% of capacity. Therefore, adequate bed space was available at other prisons for the 33 inmates assigned to Bay in excess of 90% capacity.

The department houses most of its inmates in prisons that are larger than Bay and Moore Haven, thus achieving economies of scale that provide an average savings of more than 9% in operating costs. Similar savings could be achieved by constructing and operating larger private prisons.

Of the 26 public prisons that are most similar in security level to Bay and Moore Haven, only 4 have capacities below 800 inmates; the rest have capacities ranging from 1,000 to 1,700 inmates. Since most inmates are housed in prisons with larger capacities than Bay and Moore Haven, we also compared department general operating costs to house similar inmates in larger prisons with the cost of Lawtey.

The unadjusted inmate per diem rate for these 22 institutions with higher capacities was more than 9% lower than the general operating cost for Lawtey, even though all these prisons housed some close custody inmates. (See Exhibit 6.) Therefore, housing inmates in larger prisons provides the state with greater savings than that provided by Bay or Moore Haven.

Exhibit 6: Operating Costs for Larger Public Prisons Are About 9% Lower Than the Costs for Lawtey 26 Public Prisons

Most Comparable in Security to Private Prisons

Unadjusted Average Operating Cost Per Diem 1

Percent Difference From Lawtey

22 Prisons >1,000 capacity 28,583 inmates $32.96 -9.4%

Lawtey 788 capacity 733 inmates 36.36 Basis for Comparison 3 Prisons <700 capacity 1,246 inmates 41.33 +13.7% 1 Does not include medical or educational programming costs. Source: OPPAGA analysis of data from 1996-97 Department of Corrections Annual Report.

(See Appendix A for more detail.)

When the Correctional Privatization Commission was initially authorized in 1993 to request proposals for two 750-bed private prisons, it was not as unusual for the department to operate prisons of similar size. In the 1992-93 fiscal year, the department was operating nine prisons with capacities in the 700-to 900-inmate range. The department increased the capacities of eight of those nine prisons in subsequent years by adding work camps and/ or dormitories. 19 Lawtey is the only adult male prison remaining with a capacity between 650 and 900 inmates.

18 Commission staff suggested that CCA would propose to lower the marginal rate while increasing the base per diem rate for inmates up to 90% capacity. If base rates were increased, the potential cost savings we have identified would be negated.

The department used these increases in prison size as a means of containing prison costs. The department's average inmate per diem for all adult male prisons increased a total of 9% from the 1992-93 to the 1996-97 fiscal year. During the same period, the average inmate per diem for the eight prisons that increased in size decreased by 1%.

Private vendors can also reduce inmate per diem costs as prison size increases. For example, South Bay Correctional Facility, another private prison operated by Wackenhut in south Florida, has a contracted per diem rate that is approximately 11% less than the rate at Moore Haven. Its capacity is 1,318 inmates. Therefore, increases in the inmate capacity of Bay and Moore Haven should lead to inmate per diem rates at the private prisons that would be competitive with the lower inmate per diem rates at the larger public prisons.

By housing a large number of minimum custody inmates in the private prisons, the state is not receiving the full benefit of the security available at these prisons.

The Bay and Moore Haven prisons were each constructed for the purpose of housing medium custody inmates. (See Exhibit 7 for an overview of inmate custody levels.) However, the department's practice is to assign both minimum and medium custody inmates to prisons that are designed to house medium custody inmates. During the 1996-97 fiscal year, over 47% of the inmates assigned to Bay and Moore Haven by the department were minimum custody. The department assigns a mix of inmates to the private prisons that is similar to the mix at public prisons, as required by law. 20 As a result, the secure housing at Bay and Moore Haven is used to house inmates who could be housed in less secure and less costly housing.

19 A work camp is a lower cost facility that the department often builds adjacent to its major institution. Because it does not have the same perimeter as the main prison, the work camp houses only minimum and medium custody inmates. Some of the savings achieved by the department in expanding prison size is through the use of this less costly housing.

20 Section 957.08, F. S., requires that the prisoners transferred by the department to the private prisons represent a cross section of the general inmate population, based on custody grade or conviction offense, at the most comparable facility operated by the department. The department does not have any prisons that house only medium custody inmates. The closest comparable public prison, Lawtey, had 63% minimum custody inmates as of December 31, 1996.

Larger Private Prison Is Also Less Costly

Exhibit 7: Minimum Custody Inmates Require Less Intense Supervision Than Medium or Close Custody Inmates

Overview of Custody Levels

Types of Classification · Minimum, Medium, Close and Maximum (Maximum custody refers to all inmates sentenced to death)

Statewide Inmate Population -December 30, 1996

· Minimum custody = 15.1%
· Medium custody = 31.6%
· Close custody = 52.6%
· Maximum custody = 0.7%

Factors that determine custody

· Length of sentence remaining
· Severity of crime
· Prior criminal history
· Prior escape history
· Positive or negative behavior

Supervision Requirements

Minimum custody Inmate can be outside a secure perimeter without constant supervision.

Medium Custody Inmate requires constant sight or sound supervision when outside secure perimeter.

Close custody Inmate requires armed supervision when outside the secure perimeter.

Source: Department of Corrections

Minimum custody inmates are the inmates the department determines are the least likely to attempt an escape or to pose a danger to staff, other inmates, or the public. Minimum custody inmates do not require constant supervision. The department typically assigns a small number of minimum custody inmates to even the most secure prisons to provide a group of inmates who can fill certain inmate jobs that require a degree of trustworthiness. Minimum custody inmates are in high demand by the department because they can usually be sent on outside work squads to work in the community. In addition, minimum custody inmates can be housed in work camps, a less expensive form of housing than the cells or dormitories in the private prisons.

Because more than half of the inmate population is classified as close custody, the department has a limited need for beds for minimum custody inmates. Assigning almost 50% of the secure housing at Bay and Moore Haven to minimum custody inmates is not the best use of the security features of those prisons. The housing areas of the private prisons are secure enough to house close custody inmates, although the perimeter security does not meet the department's standards for housing close custody inmates.

Minimum Custody Inmates Can Be Housed in Less Expensive Facilities Than Medium or Close Custody Inmates

It would be more cost effective to use the Bay and Moore Haven prisons to house a higher percentage of medium custody inmates, or to modify the perimeters of the prisons to permit housing close custody inmates. Maintaining a high percentage of minimum custody inmates at these secure private prisons is consistent with department practice at its lower security prisons, but it is not the best practice for these private prisons.

If the commission were to address the issues raised by our analysis in renegotiating the contracts with Bay and Moore Haven, each prison could achieve at least 7% savings in operating costs.

Although Bay's total adjusted operating costs for the 1996-97 fiscal year were slightly higher than the most comparable public prison, we estimated that Moore Haven produced cost savings of $481,581. (See Exhibit 8.) The state could increase the level of cost savings achieved by requiring vendors to return canteen profits and telephone commissions to the state and by properly structuring the per diem rates for the contracts. These measures would have increased the cost savings achieved during the 1996-97 fiscal year to 7.7% for Bay and 6.9% for Moore Haven, or $948,586 and $851,151, respectively.

Exhibit 8: OPPAGA Identified Potential Savings for Bay and Moore Haven

Bay, Moore Haven, Lawtey
Total Adjusted Direct Cost $11,814,787; $11,333,206; $11,346,583
Allocated State Indirect Cost 513,686; 486,932; 955,136
Total Adjusted Cost (Per Diem Rate x 365 days x 733 inmates) $12,328,473; $11,820,138; $12,301,719
Total Estimated Savings $( 26,754) (0.2)% $481,581 3.9%
Other Potential Savings: Inmate Welfare Trust Fund (for 733 inmates) $282,446 2.3% $369,570 3.0% Restructure Contract (for 733 inmates) 1 692,894 5.6% -----------

Total Potential Savings Identified $948,586 7.7% $851,151 6.9% 1

Our savings calculation assumes that the commission will renegotiate a contract with the Corrections Corporation of America maintaining a similar base per diem rate for inmates under 90% capacity.

Source: OPPAGA analysis of commission and department data

Lower state indirect costs account for about half of these potential savings. While possible reductions in state indirect costs are desirable, we believe that the commission should seek to achieve 7% cost savings in direct costs.

Additional savings could be achieved by increasing the size of the private prisons and by using the private prisons to house higher custody inmates. The actual savings achieved through such measures would depend upon the new per diem rates negotiated by the commission with the vendors and upon the per diem rates at the most comparable public prisons.

Correctional privatization has the potential to save more than was achieved for the 1996-97 fiscal year. As noted in OPPAGA Report No. 95-12, the commission's selection of the "most qualified vendors" resulted in the selection of bids that were substantially more expensive to the state than the bids of other qualified vendors. In fact, the CCA and Wackenhut bids were the first and third most costly of 12 bids received by the commission. One qualified vendor submitted proposals for two prisons, each of which would have been about 20% less costly to the state over the first three years of the contract than either the Bay or Moore Haven contracts. Subsequent to our report, the Legislature passed Ch. 96-312, Laws of Florida, to require the commission to maximize the cost savings of private prisons in its vendor selection process.

Chapter 3: Vendor Performance

The Corrections Corporation of America and the Wackenhut Corrections Corporation have performed satisfactorily in managing the two private prisons, based upon selected performance data. In addition, these private vendors have introduced several different approaches to the construction and operation of prisons in Florida.

The commission's contracts with the private vendors require the vendors to operate, maintain, and manage the prison in compliance with applicable federal and state requirements, laws, and court orders, and with required American Correctional Association Standards. Both prisons received accreditation after reviews by the American Correctional Association in 1996. The commission's contract monitors routinely review private prison operations to assure that the vendors are in compliance with applicable requirements and standards.

The Government Accountability Act requires each state agency to identify the outputs produced by each approved program, the outcomes resulting from each approved program, and baseline data associated with each performance measure. The Department of Corrections has submitted a proposed list of such measures for consideration by the 1998 Legislature.

The performance measures under development by the department provide a useful framework for evaluating the performance of the private vendors, and should be used to supplement the measures of compliance by the American Correctional Association and the contract monitors. For the purposes of our evaluation, we obtained data for some performance measures that can be measured on the institution level. As the Department of Corrections begins operating under performance-based program budgeting in Fiscal Year 1998-99, more complete performance data will become available and provide a basis for more detailed performance comparisons.

To review the performance of the private vendors in operating the Bay and Moore Haven prisons, we conducted site visits to private and public prisons, obtained available data from the commission and the department, and reviewed reports made by other entities regarding the performance of the vendors. We determined that vendor performance is generally consistent with performance reported by public prisons.

Performance Measures Provide Framework for Performance Comparisons

Performance outputs and outcomes for the Corrections Corporation of America and the Wackenhut Corrections Corporation compare favorably with the department.

Measures of safety, medical health services, and program services generally indicate that private prison performance compares favorably with public prison performance. However, we have not adjusted data to account for differences that could affect performance, such as differences in inmate custody level.

Bay and Moore Haven's performance on measures of public safety, staff safety, and inmate safety indicate assault and disciplinary report rates that are lower than the statewide average for public prisons. Reviews of medical services by the Correctional Medical Authority identified a small number of Level 1 (potentially life threatening) citations at Bay and Moore Haven. Bay, however, did not correct its major mental health service citations in a timely fashion. 21 Both Bay and Moore Haven reported higher program completion rates for GED, vocational education, and substance abuse treatment programs than the average rates for public prisons. (See Appendix E for data on selected performance measures.)

Bay and Moore Haven have received a variety of performance reviews from various entities. The results of these reviews have been generally satisfactory for both Bay and Moore Haven. In addition to the review of medical services by the Correctional Medical Authority, the commission has contracted for an annual off-site monitoring report on each prison, and receives monthly reports from on-site monitors. The department has also conducted reviews of security at Bay Correctional Facility.

One area of performance that merits comment is the level of participation by inmates in the education and substance abuse programs at both Bay and Moore Haven. (See Exhibit 9.) The commission reduced its performance expectations for both private prisons after both vendors experienced problems meeting the original contract requirements for these programs. As detailed in Exhibit 9, the commission amended the contracts with each vendor to reduce the number of inmates that the vendors were required to have in class at any one time. Despite these modifications in contractual requirements, which also allowed some reductions in private prison program staffing, the amendments did not provide any corresponding reduction in state payments to the vendors. 22 The commission allowed the vendors to redirect those resources to other aspects of the prison, such as security enhancements. While the original contract requirements were unrealistic and warranted modification, adjustments in contract requirements should result in reduced costs to the state.

21 We compared the CMA reviews of Bay and Moore Haven to reviews of two department prisons that were also opened during 1995. For this measure, we believe that similar opening dates provide a more important basis for comparison than size or types of inmates served, because it compares medical units that have been operating for similar lengths of time. Public prisons that had been previously reviewed by the CMA would have had the benefit of previous inspections to correct deficiencies that had been identified.

Exhibit 9: Modification of Contracts Between the Commission and the Vendors Reduced the Requirements for Inmate Participation in Educational and Substance Abuse Programs(omitted)

1 Bay's original contract provided the average daily enrollment numbers with no indication as to whether these would be full-time or part-time students. In the contract revision, these numbers are designated as full-time, with an indication that the requirement could be achieved through twice as many part-time participants.

2 For Moore Haven, the original contract specified the number of staff and the number of daily contact hours that would be achieved. The contract amendment reduced the number of staff from 26.5 to 21 and the total number of daily contact hours from 3,424 to 2,384.

3 In adding the numbers in this column together, the vendor and the commission counted this item (25/ month) for substance abuse intervention as equal to a daily enrollment of one full-time equivalent (FTE).

Source: Correctional Privatization Commission

22 Program participation and completion data is one area where the commission has not established good controls to ensure that the vendors maintain and report reliable data. The private vendors provided inconsistent data in response to our requests for information. Program participation and completion data should be maintained and monitored by the commission to ensure that vendors are complying with the terms of their contract. 30 30 Page 31 32 24 Another important measure of performance is the extent to which the private prisons are able to reduce inmate recidivism. In January 1998, the commission issued a review of recidivism of inmates released from private prisons. This study, conducted by staff of the Private Corrections Project from the University of Florida, provides an initially favorable analysis of one-year recidivism data for inmates released after less than a year in private prison, but the data must be considered preliminary. Future reviews need to improve methodology by analyzing whether public prison inmates used for comparison participated in or completed educational or substance abuse treatment programs, and by comparing sample populations on length of incarceration and type of post-release supervision.

No conclusions about the private prisons' effect on recidivism should be made until data covering two years after release is available on inmates who were housed in private prisons for more than one year. The two-year period after release allows more time for cases to be processed through the court system and more information on inmate post-release behavior to be collected.

The private vendors have introduced several alternatives to department operations.

One advantage of prison privatization is the opportunity to introduce different and sometimes innovative correctional practices that may provide a basis for re-examining public prison practices. We identified four major differences in the private prisons: a more programmatic approach to inmates, a more compact construction design, air-conditioned housing units, and the use of civilian staff and technology to reduce prison staffing needs.

The private prisons use a programmatic approach, a program-rich environment in which most inmates are assigned to academic or vocational education programs and substance abuse treatment programs. The department has used a similar programmatic approach at its youthful offender prisons and at its drug treatment centers. However, adult male prisons in Florida typically provide a limited number of academic and vocational education programs and substance abuse treatment that can serve only a small proportion of the population at any given time.

Bay and Moore Haven designed their facilities to use less space. Each is enclosed under one roof so that the entire facility is interconnected and covers a relatively small area of land. Moore Haven's facility does not include a dining hall, but serves meals to inmates in the day rooms of their dorms. As discussed previously, reducing space can also reduce construction costs, and may also reduce operating costs.

Both Bay and Moore Haven constructed air-conditioned housing units for the inmates. Private prison officials stated that air conditioning improves the working environment for the correctional officers supervising the inmates, and may help alleviate prison tensions among inmates during hot summer months. Although 7 of the department's 55 prisons have some air-conditioned housing units, the department does not build new prisons with air-conditioned housing units.

The private prisons have used civilian staff rather than correctional officers in positions such as mailroom attendant, thereby reducing the number of correctional officers needed on the day shift. Bay and Moore Haven have also used technology to reduce staffing requirements. Each has installed advanced electronic lock and movement systems within the prison and use cameras more extensively than the department to monitor a number of areas both inside and outside the prison. Private prison staff located in isolated locations use body alarms so that the control room can be quickly notified in emergencies.

Some of these technological approaches may be worthy of further consideration by the department. For example, approaches that rely more heavily on technology or that decrease the need for trained correctional officers may provide ways for the department to reduce costs. The compact prison design offers a way of reducing construction costs or locating prisons in areas where smaller parcels of land are available.

Although private prison officials stated that the use of air-conditioned housing for inmates has some advantages, it also has certain disadvantages. For example, air-conditioned housing creates the perception that inmates may be comfortable in prison rather than giving the public the sense that inmates are being punished for their crimes.

The Legislature has placed an emphasis on educational and substance abuse programming within the private prisons. Over the next three years, it will be important to determine whether inmates in the "program-rich" environment are more likely to complete programs and whether those programs appear to have positive effects on inmate behavior and recidivism.

Some Private Prison Alternatives Reduce the Need for Additional Correctional Officers

Chapter 4: Conclusions and Recommendations

Although vendor performance during the 1996-97 fiscal year was satisfactory, the costs of constructing and operating the private prisons is not providing the state with the level of cost savings initially projected by the commission. Private prison site preparation costs were lower for private prisons, while building costs were higher. Private and public construction costs differed as a result of location, inmate labor, and economies of scale. One of the private prisons, Bay, was constructed at a much higher cost to the state than Moore Haven.

Achieving savings in operating costs is a complex issue. Bay did not provide any savings to the state for the 1996-97 fiscal year while Moore Haven's adjusted costs were approximately 4% lower than the adjusted costs of operating a comparably sized public prison. The private prisons did not achieve the originally projected level of savings partly due to the addition of indirect state costs to private prison costs and to changes in legislative directions regarding the use of the Inmate Welfare Trust Fund by the public prisons.

The savings achieved by privatization could be increased if the commission:

· required vendors to return the profits of the Inmate Welfare Trust Fund to the state, resulting in an additional savings of up to 3.1%, or $600,000 per year at these two prisons; and

· included a more favorable marginal per diem rate at the Bay prison, resulting in an additional savings of up to 5.6%, or approximately $400,000 per year.

Additional savings could also be gained for the state by expanding the size of the private prisons and by using the private prisons' secure and costly housing for higher custody inmates, allowing minimum custody inmates to be assigned to less costly work camps.

We believe that the 7% savings goal can be achieved for operating costs by renegotiating contracts with existing vendors. At a minimum, these renegotiations should provide that:

· canteen and telephone commission profits be used to offset private prison operating costs;

· the marginal rate for inmates in excess of 90% of capacity be set at a level that reflects the actual costs of each additional inmate; and

· contract costs reflect the reductions in program requirements made to the original contract.

The two private prisons are performing satisfactorily in the areas of custody and control, health services, and inmate programs. As the Department of Corrections begins operating under performance-based program budgeting during the 1998-99 fiscal year, complete performance data should be available to provide more detailed performance comparisons between public and private prisons.

One performance concern noted for both private prisons was that neither vendor met the original contract requirements for education and substance abuse treatment program. The commission amended both contracts to reduce the contract requirements, but did not reduce the cost of either contract.

The private prisons have introduced some alternatives in prison design, use of technology, and approaches to prison staffing into the state correctional system. Some of these merit consideration and review by the department.

We recommend that the Legislature authorize the Correctional Privatization Commission to renew its contracts with the Corrections Corporation of America for the operation of Bay Correctional Facility and with Wackenhut Corrections Corporation for the operation of Moore Haven Correctional Facility with the following modifications.

· The commission should negotiate new per diem rates for operations that will reflect at least 7% savings from the department's reported costs for a similar sized institution such as Lawtey.

· Canteen profits and telephone commissions should be deposited with the state to offset the costs of the private prisons.

· The commission should require vendors to collect and report data for selected performance measures established by the Legislature for the department in the General Appropriations Act in the areas of custody and control, health services, and offender work and training;

· Bay and Moore Haven should propose modifications to the prison perimeters to increase custody status to a level that can house close custody inmates to allow the assignment of a certain level of program-eligible close custody inmates.

· Bay and Moore Haven should propose plans to expand capacity to over 1,000 beds. The cost of construction expansion should be significantly lower than the initial per bed construction cost, and the proposed inmate operating per diem after expansion should be substantially lower than the current per diem.

Furthermore, we recommend that the contract with Corrections Corporation of America be renegotiated to provide a per diem rate for all inmates housed in excess of the guaranteed 90% capacity be set at a level that is more reflective of the marginal costs of keeping each additional inmate.

According to current estimates of the Criminal Justice Estimating Conference and current construction plans within the Department of Corrections, expansion of Bay and Moore Haven will not be needed prior to Fiscal Year 2000-2001. Therefore, we recommend that the commission submit expansion proposals for Bay and Moore Haven to the 1999 Legislature for consideration and action.

We also recommend that the department:

· assign only enough inmates to Bay Correctional Facility to keep its population at 90% of capacity until the contract with CCA is renegotiated; and

· increase the proportion of medium custody inmates assigned to Bay and Moore Haven.

For future contracts with private vendors, we recommend that the Legislature direct the commission and the department to develop a plan regarding the possible use of inmate labor in the construction of private prisons.

To ensure that the state achieves operating cost savings from prison privatization, we recommend that the Legislature revise Ch. 951.07, F. S., with two provisions. First, the Legislature should require that the contracted rate for future private prisons provide 7% cost savings over the certified state direct costs for a similar prison. In determining state direct costs, any indirect state costs for activities or functions that are not assumed by the vendor would be excluded. For example, indirect costs associated with the department's oversight and operation of the statewide inmate information system would be excluded from the cost calculations because neither the vendor nor the commission is asked to assume similar functions. Second, the Legislature should require the commission to consider the effect of proposed annual contract increases in evaluating the projected cost savings and, if possible, to limit annual contract increases to increases in the department's per diem rates.

Appendices

Appendix A: Selection of Comparable Public Prisons

The department has no prison that provides a basis for a comprehensive comparison of construction and operating costs. We identified 26 prisons that were most similar to Bay and Moore Haven in terms of security level. Several factors can reduce the comparability of these public prisons with the private prisons. These factors include:

· age of facility (the public prisons have been opened from 1934 to 1996);

· size (public prisons ranged in average population for the 1996-97 fiscal year from 352 to 1,719);

· type of facility (many of the department prisons include work camps that are adjacent to the prison and are included within the population count of the prison);

· type of inmates (most public prisons house close custody inmates, and many house a higher proportion of medically restricted or psychologically restricted inmates);

· location (some of the public prisons are located in areas where officers receive a "competitive area differential" added to their salaries); and

· number of programs (most public prisons offer education and substance abuse programming to small percentages of the inmate population).

For comparisons of construction costs, we compared the costs for the public prisons that were also opened in 1995. Of these 26 prisons, three, Everglades Correctional Institution, Taylor Correctional Institution, and Washington Correctional Institution, were also opened during 1995.

For comparisons of operations, only one prison, Lawtey Correctional Institution, had an average inmate population within 25% of the populations of Bay and Moore Haven for the 1996-97

fiscal year. Like the private prisons, Lawtey also houses only minimum and medium custody inmates.

For our analysis of performance, we used the same four public prisons (Lawtey, Washington, Taylor, and Everglades) as well as department-wide data to make some preliminary observations 40 40 Page 41 42 34 about performance comparisons between public and private prisons.

Overview of Public Prisons Closest in Security Level to Bay and Moore Haven

Highest Custody Grade Accepted Year Opened

Medically /Psychologically Restricted Inmates 1

Competitive Area Differential Added to Salary?

Work Camp Included in Facility?

Average Population Fiscal Year 1996-97

Private Prisons

Bay Medium 1995 Less Than 10% No No 708 Moore Haven Medium 1995 Less Than 10% No 2 No 706

Public Prisons

Lawtey Medium 1977 27% Medical No No 733 Everglades Close 1995 30% Psychological Yes No 1,242 Washington Close 1995 39% Psychological No No 1,031 Taylor Close 1995 12% Medical No No 1,003 New River Close 1982 18% Medical No No 1,719 Sumter Close 1965 13% Medical No Yes 1,629 Apalachee Close 1949 27% Psychological No No 1,542 Century Close 1991 Less than 10% No Yes 1,501 Liberty Close 1988 12% Medical No Yes 1,450 Holmes Close 1988 11% Medical No Yes 1,406 Hendry Close 1979 Less than 10% Yes Yes 1,342

Tomoka Close 1981 33% Psychological/ 25% Medical Yes Yes 1,325

Madison Close 1989 14% Medical No Yes 1,321 Polk Close 1978 18% Medical Yes Yes 1,307 Jackson Close 1991 14% Medical No Yes 1,306 Marion Close 1959 14% Medical No Yes 1,273 Calhoun Close 1988 Less than 10% No Yes 1,265 Baker Close 1978 13% Medical No Yes 1,257 Columbia Close 1992 26% Medical No Yes 1,251 Avon Park Close 1957 Less than 10% No Yes 1,239 Walton Close 1991 23% Medical No Yes 1,214 Cross City Close 1973 Less than 10% No Yes 990 Okaloosa Close 1983 Less than 10% No Yes 970 River Junction Medium 1974 Less than 10% No No 478 Putnam Close 1984 Less than 10% No No 416 Quincy Close 1973 Less than 10% No No 352

1 This represents the percentage of inmate population indicated by department records as having moderate or severe medical restrictions, or mild or moderate psychological impairments, as of December 31, 1996. 2 Moore Haven is located in an area where a competitive area differential would be paid by the department for a public prison. Although Moore Haven's correctional officers do not receive a salary differential, the entry-level salaries at Moore Haven are higher than Bay's and competitive with the area in which Moore Haven is located. Source: OPPAGA analysis of department and commission data 41 41 Page 42 43 35 Appendix B Adjustments Made and Considered in Comparison of Public and Private Prison Operating Costs

Both the commission and the department have proposed adjustments in their own cost comparisons. We reviewed their proposed adjustments and accepted or modified some and rejected others. We also identified additional adjustments as appropriate.

Adjustments Explanation Comments Population Adjustment To equalize the effect of size, populations of Bay and Moore Haven were hypothetically increased to 733 inmates, matching Lawtey's average for the year; we used contracted per diem rates to make the adjustment.

We matched on size rather than on percentage of capacity due to differing use of the term "design capacity" between public and private prisons.

On-site Department Classification Staff

The department assigns two classification officers to each private prison to perform classification functions that cannot be delegated to private prison staff.

The commission questions the need for these officers.

Hospital Cost Limit

The contracts set ceilings of $7,500 for inmate medical costs; department pays excess. This adjustment adds the cost of that excess to the private prison cost.

The ceiling protects private vendors from high medical costs involving any individual.

Payments to Maintenance Reserve Fund

Vendors place $75,000 per year into a reserve fund to be expended for major repair/ replacement; these would be fixed capital outlay type expenses, not operating expenses.

The department does not include major maintenance and repair costs in its inmate per diem.

State Sales Tax Payments

Vendors often pay state sales tax on the purchase of items that the state purchases without paying this tax. Since the sales tax is returned to the state treasury, this is considered a credit against state prison costs.

Neither CCA nor Wackenhut keeps records of the amount of sales tax paid. Both vendors provided estimates of sales tax payment that were not documented. We included a credit for those items that we concluded were reasonable.

Public Work Squads

Lawtey costs include five officers to supervise public work squads; private prisons have nothing comparable.

Moore Haven occasionally sends one of its officers into the community with an inmate work squad.

Inmate Welfare Trust Fund

The legislature uses trust fund profits and telephone commissions to pay department operating costs; private prison trust fund profits not returned to the state to be applied to operating costs. (See discussion on page 13).

A change in law affected public prisons after contract with private prisons was established.

No corresponding change was made to private prison administration.

Medical Cost Lawtey had high medical costs reflecting an inmate population with greater medical needs. The weighted average per diem for four public prisons closest in size to private prisons with similar medical populations was $5.15 per inmate per day.

The adjusted public prison medical cost ($ 5.15) is lower than the contracted medical cost for Bay ($ 5.46) or Moore Haven ($ 6.63).

Education Program

Lawtey had a small educational program using two teachers at a reported cost of $0.45 per inmate per day. The total cost from public prisons providing education programs to similar number of inmates as the private prisons, averaged based on Lawtey population, results in an education cost of $2.40 per inmate per day, requiring an adjustment of $1.95 to the per diem rate.

The adjusted public prison education cost ($ 2.40) is higher than the contracted education cost for Bay ($ 2.11) and Moore Haven ($ 2.36).

Substance Abuse Treatment

Lawtey substance abuse program serves fewer inmates than private prisons at a reported cost of $0.26 per inmate per day. The total cost from public prisons serving similar number of inmates, averaged on Lawtey population, was $.79 per inmate per day, requiring an adjustment of $0.53 to the per diem rate.

The adjusted substance abuse cost is higher than the contracted substance abuse treatment cost for Bay ($ 0.21) and lower than the contracted substance abuse treatment cost for Moore Haven ($ 2.79).

Salary Adjustment for Location

The department pays a "competitive area differential" to correctional officers and classification officers in counties that have been identified as needing a salary adjustment. Moore Haven is located in a county (Glades) in which the department pays a competitive area differential to eligible officers. Therefore, we have adjusted the Moore Haven costs to reflect the value of this salary adjustment to the department.

Moore Haven's starting salary for correctional officers is almost $5,000 higher than Bay's. The department would pay a competitive area differential of $3,800 per year for correctional officers in Glades County compared to Bay County.

Indirect Costs -Department

Almost all department costs that are not allocated to public prisons are bundled as indirect overhead costs and allocated on the basis of staffing levels and inmate days.

The allocation for the private prisons is assigned on a pro-rated basis depending upon the benefit accruing to the private prison as a result of that administrative service.

The difference in Bay and Moore Haven increases results from variation in the number of staff, the criteria used for determining level of allocation.

Indirect Costs -Commission

All commission costs were allocated to the private prisons based on the number of inmate days.

Indirect Costs -State

Costs of other state agencies such as the Department of Management Services, the Comptroller, and the Attorney General are allocated by the state.

Other Adjustments Considered But Not Made Adjustments Explanation Comments Age of Facility Lawtey, constructed in 1975, may have higher utility costs. It is difficult to identify data that would show comparable utility rates if Lawtey were a new facility.

Unfunded Liability of the State Retirement System

Public prison costs include payments to reduce the unfunded liability of the state retirement system in the amount of 5.78% of payroll. This amount was paid in the 1996-97 fiscal year to help pay off liabilities incurred by the state in previous years.

This was not adjusted because it is still a state cost to the taxpayers. Privates do not pay this and thus should be able to reduce costs in comparison to state by at least 5.78% of payroll.

Corporate State Income Tax Payments

Corporate income taxes paid by vendors on the profits from prison operation are paid into the state treasury and are a credit against the state's cost for the private prison.

Wackenhut reported that it paid no corporate income tax for the 1996-97 fiscal year. CCA paid corporate income tax in the 1996-97 fiscal year, but did not provide documentation as to the portion of that tax paid for operating Bay.

Under Utilization of Public Prisons

During the 1996-97 fiscal year, the department filled only 89% of its beds throughout the state due to lower prison populations than had previously been expected.

We focused on actual population rather than percent of capacity in determining our adjustments. Lawtey was maintained at a higher percentage of capacity than most public prisons. Work Camps The department houses many of its minimum and medium custody inmates in work camps that are often located adjacent to the main prison plants. These work camps are typically less expensive to construct and operate due to fewer security features. In its comparison of costs, the commission adjusted public prison costs by backing out work camp costs.

By selecting Lawtey, a public prison without a work camp, there was no need to adjust for work camp costs. Also, the department does not separate work camp costs from main plant costs in its accounting system.

Staff Experience Level Personnel costs may increase as staff become more experienced. Staff at an older prison, such as Lawtey, may be more experienced than staff at a newer prison, such as the private prisons.

It is difficult to place a value on staff experience.

Source: OPPAGA analysis of information provided by the department, the commission, and other sources 43 43 Page 44 45 37 Appendix C Comparison of Construction Costs: Private and Public Prisons 1

Privatization Commission Private Prisons Department of Corrections Public Prisons Measures for Comparison Bay Moore Haven Everglades Taylor Washington

Descriptive Measures Year Opened 1995 1995 1995 1995 1995 Number of Beds: Lawful Capacity 750 750 1,538 1,122 1,126

Constructed for: Gender Male Male Male Male Male Population Type Adult Adult Adult Adult Adult Security Level 3— Medium 3— Medium 5— Close 5— Close 5— Close Construction Period: Appropriation to Opening Month 2 14 13 23 18 18

Total Prison Acreage 100 52 114.8 257.03 273.5 Total Prison Building Area Square Footage 199,767 172,440 385,640 279,800 279,800

Construction Cost Measures Site Acquisition $ 510,500 $ 350,000 $ 0 $ 0 $ 0 Site Preparation $ 2,069,585 $ 990,000 $18,400,000 $ 6,294,774 $ 5,955,579 Total Site Acquisition and Preparation $ 2,580,085 $ 1,340,000 $18,400,000 $ 6,294,774 $ 5,955,579

Buildings, Equipment, and Other $21,428,187 $16,715,500 $30,217,288 $17,756,753 $17,536,076

Total Cost of Prison $24,008,272 $18,055,500 $48,617,288 $24,051,527 $23,491,655 Cost of Site Acquisition and Preparation Per Acre $ 25,801 $ 25,769 $ 160,279 $ 24,490 $ 21,775

Buildings, Equipment, and Other Cost Per Square Foot of Buildings $107.27 $96.94 $78.36 $63.46 $62.67

Total Prison Cost Per Bed $ 32,011 $ 24,074 $ 31,611 $ 21,436 $ 20,863 1 Although the construction of Bay and Moore Haven were financed by the state to be paid over a 20-year period, we did not include financing

costs in our analysis of construction costs. The Legislature's decision to finance the construction of a prison is separate from the decision as to whether a prison should be public or private. 2 Construction period does not include time spent developing request for proposals, or receiving and reviewing proposals.

Source: Data provided by the department and the commission 44 44 Page 45 46 38 Appendix D Operating Costs Comparison of Bay and Moore Haven Private Prisons With Hypothetically Similar Lawtey State Prison, Fiscal Year 1996-97

Private Prisons State Prison Bay Moore Haven Lawtey Total due the vendors based on contracted per diem rates 1 $11,723,543 $12,122,766 Less: Deduction for on-site monitor salaries (74,543) (74,406) Deduction for property taxes included in contract but not paid2 (223,757) (347,699) Medical co-payments collected from inmates retained by the vendors (2,975) (1,471)

Total Payments to Vendor $11,422,268 $11,699,190 $12,135,469

1996-97 Per Diem: Operating $36.65 $36.40 $36.36 Health services 5.46 6.63 8.56 Education programs 2.11 2.36 0.45

Total 1996-97 Unadjusted Per Diem $44.22 $45.39 $45.37

Adjustments Adjust private prisons to Lawtey's average population of 733 inmates Bay (708): Add 25 inmates @ $41.60/ day (0.14) Moore Haven (706): Add 27 inmates @ $8.87/ day (1.41)

On-site department classification staff at private prisons 0.30 0.32 Hospital Cost Limit (>$ 7,500) for private prisons 0.21 0.30 Sales tax adjustment 3 (0.15) (0.15) Payments to Major Maintenance Reserve Fund for private prisons (0.28) (0.28) Salary adjustment for location of Moore Haven (1.81) Credit for cost of supervising public work squads (0.62) Credit for Inmate Welfare Trust Fund net revenue 4 (1.41) Adjust medical costs for higher medical grade inmates at Lawtey [$ 8.56 -$5.15] (3.41) Adjust education programs at Lawtey [$ 0.45 -$2.40] 1.95 Adjust substance abuse treatment programs at Lawtey [$ 0.26 -$0.79] 0.53 Net Total of Adjustments $( 0.06) $( 3.03) $( 2.96)

Indirect Cost Allocation From: Department of Corrections 0.98 0.89 3.33 Correctional Privatization Commission 0.89 0.89 Other state agencies 0.05 0.04 0.24

Total Indirect Allocation $1.92 $1.82 $3.57

Total Adjusted Per Diem $46.08 $44.18 $45.98

Estimated Cost Savings (Per Diem) $( 0.10) $1.80 Percentage Cost Savings Based on State Cost (0.2)% 3.9% Annualized Savings $( 26,754) $481,581 1 The contracted inmate per diem rates for Bay and Moore Haven were each increased by $0.12 in September 1996. This increase, which was paid from general

revenue, was used to pay the salaries of two on-site monitor positions retained by the commission when two private youthful offender prisons were transferred to the Department of Juvenile Justice for the 1996-97 fiscal year. The commission subsequently reclassified these positions as an attorney and a monitor coordinator within their central administrative office. 2

The commission deducts the amounts included in the contracted per diem for property taxes from the payments the state makes to the vendors. The commission determined that private vendors are not required to pay property taxes to counties for privately operated state prisons after the initial bids were submitted.

However, in Fiscal Year 1996-97, the Legislature appropriated $355,000 that was paid to Glades County in lieu of property taxes for the Moore Haven Correctional Facility.

3 Neither vendor produced records as to the amount of sales tax paid to the State of Florida. Both vendors estimated sales tax payments based upon expenditures.

Our review of documentation provided by the vendors led us estimate $40,000 per vendor in sales tax payments for the 1996-97 fiscal year. 4

State public prisons remit net profits from commissary, vending, and telephone commission revenues to the state for appropriation to pay for certain specific activities in the operation of state prisons. The private prisons reported receiving similar net revenues as the state prisons. No adjustment was made for the net revenues at private prisons because the vendor has retained the funds. If the commission applied s. 945.215, F. S., to the private prisons, these funds could have been returned to the state and used to provide additional cost savings to the state of up to 3.0%.

Appendix E: Comparison of Performance on Selected Measures: Private and Public Prisons

The Department of Corrections has identified three program areas for performance-based program budgeting that relate to the prison operations: Custody and Control, Health Services, and Offender Work and Training. We selected certain measures for which data were available to provide some preliminary observations regarding private vendor performance for Fiscal Year 1996-97. The data provided here is as reported by the department, the commission, or the private vendors.

Within the public prisons, we noted wide variation in the reported rates on some custody and control measures. (See Table E-1.) For example, Taylor and Washington represent extreme variations, although the two prisons are fairly similar. We did not attempt to determine the cause of this variation.

Table E-1(Omitted)

1 Department data may not be accurate for all institutions. For example, Moore Haven monitoring reports indicate that there have been incidents of inmate on inmate assaults.

Source: Developed by OPPAGA from data provided by the Department of Corrections

For Health Services performance measures, we focused on those measures related to the Correctional Medical Authority (CMA) reviews of public and private prisons, since CMA provides a professional and independent review. Medical services at Everglades, a public prison, are provided through a contract with a private health services provider. Department officials stated that problems at Everglades have resulted in a plan to change the vendor providing health services at that prison.

Table E-2:Selected Medical Services Performance Measures Deficiencies Cited by the Correctional Medical Authority(Omitted)

Source: Developed by OPPAGA from data provided by the Correctional Medical Authority

No adult male prison provides the same level of educational and substance abuse programs as Bay and Moore Haven. Therefore, we did not compare program completion rates directly with individual prisons, but rather with the department's cumulative rate, which includes youthful offenders.

Table E-3: Selected Education Performance Measures 1

Bay Moore Haven All Major Public Prisons

Number of Academic Program Completers 39 8 2,457 Percentage of Participants Completing 46% 100% 49%

Number of GED Program Completers 201 100 1,088 Percentage of Participants Completing 31% 18% 10%

Number of Vocational Education Program Completers 110 156 1,755 Percentage of Participants Completing 39% 30% 28%

Number of Substance Abuse Treatment Program Completers 254 380 3,097 Percentage of Participants Completing 53% 58% 32% 1

According to records provided by the commission, Bay Correctional Facility issued 169 vocational certificates during Fiscal Year 1996-97, and Moore Haven issued 185 vocational certificates and 150 GEDs. We did not determine the cause of this discrepancy in the data.

Source: Developed by OPPAGA from data provided by the Department of Corrections

Appendix F: General OPPAGA Comments Regarding the Responses and Responses From the Correctional Privatization Commission, the Department of Corrections, the Corrections Corporation of America, and the Wackenhut Corrections Corporation

In accordance with the provisions of s. 11.45( 7)( d), F. S., a list of preliminary and tentative review findings was submitted to the executive director of the Correctional Privatization Commission; the secretary of the Department of Corrections; and the chief executive officers of Corrections Corporation of America and Wackenhut Corrections Corporation for their review and response.

The written responses of each have been reproduced herein beginning on page 43.

General OPPAGA Comments Regarding the Responses

Representatives of the Correctional Privatization Commission, the Department of Corrections, the Corrections Corporation of America, and the Wackenhut Corrections Corporation raise a number of challenges to the methodology used by OPPAGA in arriving at the conclusions presented in this report. We have reviewed all of these challenges, and concluded that none of the challenges warrant changes in the methodology or conclusions presented in our report. In response to the issues raised by these challenges, we have two general comments.

First, the executive director of the commission stated that all parties have previously stipulated that comparisons of costs between the private and public prisons are impossible. OPPAGA has not made any such assertion. In OPPAGA Reports 96-69 and 95-12, we identified several difficulties with comparing costs for public and private prisons. However, we never stated that it would be impossible, nor suggested that we would not comply with the law that required us to make the comparison. In addition, both the commission and the department have previously published public and private prison cost comparisons that present data favoring their particular points of view. We believe that valid cost comparisons can be made and are essential to determining the success of prison privatization.

Second, both the executive director of the commission and the secretary of the department questioned our selection of Lawtey Correctional Institution as the public prison serving as the basis for comparison with the private prisons. In Report 95-12, we recommended that the commission and the department agree upon comparable prisons. They have not reached such an agreement. In the absence of an agreement between these parties, it was necessary for us to select the most comparable public prison( s) to comply with the requirements of s. 957.11, F. S.

OPPAGA stands by its identification of Lawtey as the most similar public prison for comparison with Bay and Moore Haven private prisons for Fiscal Year 1996-97. OPPAGA disagrees with CCA's statement that "neither the two privately operated prisons nor the publicly operated facility to which they are compared are similar in any substantial way." Table F-1 illustrates several important similarities between Lawtey, Bay, and Moore Haven, as represented by data from December 31, 1996.

Table F-1 Lawtey Is Similar to Bay and Moore Haven

Comparison Characteristic Bay Moore Haven Lawtey Secure correctional prison Yes Yes Yes

Work camp attached No No No Type of inmates Adult male Adult Male Adult Male Custody rating of the facility Medium Medium Medium Lawful capacity of the facility 750 750 788 Size of inmate population 708 706 733 Percentage of minimum custody inmates 60.1% 43.0% 62.9% Percentage of non-violent offenders 73.5% 78.6% 69.6% Average age of inmates 33 34 37 Source: Institution profiles from the Department of Corrections

As the department suggests, the alternative to using Lawtey as a basis for comparison would be to use some aggregate group of public prisons. Unfortunately, as we indicate in Appendix A of our report, this would only increase the dissimilarities between the public and private prisons. For example, any group of several public prisons will include prisons with larger populations, with substantial numbers of close custody inmates, and with work camps adjacent to the main facility that are included in the department's prison costs. Section 957.07, F. S., specifically requires that cost comparisons be made with public prisons of similar size, type, and location.

CORRECTIONAL PRIVATIZATION COMMISSION
4030 Esplanade Way, Suite 315
Tallahassee, Florida 32399-0950
904/ 921-4034
FAX 904/ 922-7594

March 18, 1998
John W. Turcotte, Director
Office of Program Policy Analysis and Government Accountability
111 West Madison Street, Room 312
Tallahassee, FL 32302

RE: Commission's Response to OPPAGA Report No. 97-68

Dear Mr. Turcotte:

Since 1995 the Correctional Privatization Commission (Commission) has committed countless man-hours to address precisely the same topic that is focused on by OPPAGA Report No. 97-68 (Report), so we appreciate the literal impossibility of OPPAGA satisfying the mandate imposed on it by §957.07 and §957.11. While we sharply disagree with many of the conclusions reached in the Report we do respect their effort to achieve an objective that all parties (including OPPAGA) previously have stipulated to be impossible. Respect notwithstanding, the Commission is obliged to explain why we believe the Report is as defective as it is and why it provides such weak guidance to both the Commission and the Legislature.

The root problem with this report is that there is simply not a Department of Corrections (Department) correctional institution that is comparable to these two private correctional institutions. This is a problem that both the Commission and the Department have experienced in previous attempts to produce similar cost studies. The problem is then compounded by OPPAGA's use of hypothetical assumptions, questionable adjustments, and system wide averaging as well as partial system averaging creating a piece meal approach to producing a fantasy correctional institution to be used for comparison.

The Fundamental Flaw in the Report is lack of Comparability

The critical flaw in the Report is in no manner attributable to OPPAGA. The flaw, clearly unintended by the Legislature, is to be found in §957.07. At the heart of that statutory provision is the requirement for evidence of "at least a 7 percent cost savings over the public provision of a similar correctional institution." The requirement presupposes that there will be at least one public correctional institution that is substantially similar to each Commission correctional institution. The Commission has determined that there are no correctional institutions that are substantially similar to either of the correctional institutions on which the Report focuses. The Department has acknowledged that there are no sufficiently public comparable correctional institutions. The arithmetic contortions found in the Report make it clear that OPPAGA could find no comparable public correctional institutions.

The effect of all concerned parties stipulating to there being no comparable public correctional institutions that are substantially similar to the Bay Correctional Institution (Bay) or to the Moore Haven Correctional Institution (Moore Haven) followed by OPPAGA trying valiantly to create artificial similarity when actual similarity does not exist is to foolishly favor form over substance. This fundamental fact of accounting life has been recognized time and time again most recently by the United States General Accounting Office (GAO) in its 1996 report to Congress on privatization.

OPPAGA encourages readers to accept the premise that the Lawtey Correctional Institution (Lawtey) is similar enough to both Bay and Moore Haven that the comparison is sufficiently meaningful for government cost accounting purposes. The premise is false as we have outline below:

1. Our belief is that the design capacity of Lawtey is 555 and that the 733 population figure in the Report implies that Lawtey was being operated at 132 percent of its design capacity. This is below the lawful capacity of Lawtey (i. e., design capacity multiplied by 1.50). However, major economies of scale materialize under such a circumstance that distort the validity of comparisons with correctional institutions that were operating at less than 100 percent of their design capacities.

OPPAGA Comment: The state of Florida does not house inmates based on the "design capacity" of prisons, but on the "lawful capacity" of those prisons.

2. Our belief is that there are fundamental dissimilarities between the design of Lawtey and the designs of the two Commission correctional institutions. In the aggregate, for example, 67 percent of the housing capacity of the private correctional institutions is in cell-designed units. Our belief is that approximately 75 percent of the housing capacity at Lawtey is provided by dormitories.

3. Recent data reveal that the prisoner population at Lawtey contains a substantially higher percentage of minimum security prisoners and a substantially lower percentage of medium security prisoners than the Commission correctional institutions.

Notwithstanding OPPAGA's own admission that "minimum custody inmates can be housed less expensively than medium custody inmates", OPPAGA did not make any cost adjustment for this population discrepancy.

OPPAGA Comment: The executive director essentially suggests that we give the vendors a credit for using more expensive housing than necessary. The private prisons are designed to house medium custody offenders, and were not required to be totally cell-designed units. During the 1996-97 fiscal year, Moore Haven housed a somewhat lower percentage of minimum custody inmates than either Lawtey or Bay. See Table F-1 on page 42. However, we did not make an adjustment for this difference because neither the department nor the commission calculates or varies per diem rates based upon custody levels.

4. The same recent data reveal that the prisoner population at Lawtey includes a higher proportion of prisoners age 50 and older than do the Commission correctional institutions.

5. The same recent data reveal that the medical and psychological grades of prisoners at Lawtey differ significantly from those in the Commission correctional institutions.

OPPAGA attempted to make adjustments for some but far from all of these significant differences.

Questionable Adjustments

Members of the Commission staff have spent a considerable amount of time discussing technical details associated with the cost comparison OPPAGA attempted to make. We appreciate the willingness of OPPAGA to cooperate with our staff. For the record, however, we must voice our disagreement with several adjustments OPPAGA made:

a) An adjustment was made for on-site Department classification staff. Although we believe the adjustment accurately reflects expenditures, we also believe the Report should have noted that there is no statutory requirement for any on-site Department personnel.

b) It is obvious from the analysis provided by the Report that the sharp downward adjustment in the medical costs incurred at Lawtey was both arbitrary and speculative. This adjustment for medical costs attests to the lack of comparability problem.

c) We believe the adjustment made for educational and substance abuse programs was both arbitrary and speculative. Additionally, quantitative evidence of variations in the quality of programs in these areas is readily available to OPPAGA. The Report fails to take caliber of program differences into account.

OPPAGA Comment: Our adjustments were neither arbitrary nor speculative. All parties agree that Lawtey differs from the private prisons in the areas of medical services, education programs, and substance abuse treatment programs.

Our methodology was the most reasonable approach to measure the actual state cost for delivering services similar to those provided by the private prisons.

d) OPPAGA estimates that $. 98 per prisoner per day is expended in the form of central office Department costs. Given the exceedingly modest role played by the Department in the oversight of Commission correctional institutions, the Commission believes that so high an allocation of Department overhead costs cannot be justified. Further, a significant but unknown proportion of actual Department overhead costs could be translated into cost savings. Much of the Department involvement is redundant given the ample oversight of Commission correctional institutions that is provided by the Commission itself. There is no evidence with which we are familiar that even implies that redundant oversight by the Department yields any cost or performance benefits whatsoever.

e) Had the account balances in the two inmate welfare funds been given an appropriate credit, the estimated cost savings for both Commission correctional institutions would have risen by approximately $600,000.

OPPAGA Comment: See OPPAGA's comments about the Inmate Welfare Trust Fund below on page 47.

f) OPPAGA could and should have indicated that the $. 89 indirect cost attributable to Commission expenditures is in large part an artifact of the fact that two of the four Commission correctional institutions were in a "ramp-up" mode during the period under consideration. This indirect cost figure already has declined dramatically with the Lake City Correctional institution and the South Bay Correctional institution achieving full operational status.

g) OPPAGA failed to follow the mandate of §957.07, which requires recognition of various taxes contract vendors pay. Had all local, state, and federal tax payments been taken into account in accordance with generally accepted accounting principles and the statutory requirement, additional cost savings would have been proven.

OPPAGA Comment: The executive director criticizes OPPAGA for failing "to follow the mandate of s. 957.07, F. S., which requires recognition of various taxes contract vendors pay." In fact, we requested this data from the commission, which oversees vendor operations, and discovered that the commission had not required either vendor to maintain records of taxes paid. Both vendors submitted gross estimates of sales taxes paid, which included items that do not require sales tax payments, such as food, and items for which the state also pays sales taxes, such as travel. Based on the documentation provided by the vendors through the commission, we made a fair estimate of taxes paid to the state of Florida.

More Specific Points of Disagreement

The OPPAGA effort yields a host of additional items of a more specific nature that the Commission believes to be incorrect. Illustrations include the following:

1. OPPAGA did not take into consideration the statutory and contractual fact that the performance obligations imposed on Commission correctional institutions are more demanding than are the obligations that must be met by correctional institutions operated by the Department.

2. OPPAGA contends that the Commission did not structure its contracts to ensure that the required level of cost savings would be achieved and then maintained. We flatly reject this unsupported conclusion. In fact, we believe the opposite conclusion is easily proven. The average annual increase in the operating cost per diem at these two Commission correctional institutions has been less than the average annual increase in appropriations for the major institutions of the Department.

OPPAGA Comment: The vendor contracts included 4% annual increases for the second and third years of operation, which means that the 1996-97 contract was 8.2% higher than the 1994-95 rate. Increases in the Department of Corrections costs have typically varied from year to year. From Fiscal Year 1992-93, which the Auditor General used as a basis for certifying the cost of public prisons, until Fiscal Year 1996-97, the overall public prison inmate per diem rate increased by 12%, or slightly less than 3% per year.

3. OPPAGA repeatedly points to differences in inmate welfare fund uses that exist between public and Commission correctional institutions and asserts that the Commission "has allowed the vendors to retain possession of the earnings of the inmate canteens and telephone commissions."

Subsequently the Department was given a "credit" for the inmate welfare funds it expended with the effect that the Commission was penalized for being frugal in this area. We believe the approach taken by OPPAGA is wrong-headed. To begin with, contract vendors may not expend these funds for any purpose without the expressed authorization of the Commission. All funds are thus under the firm control of the Commission at all times. Further, unlike the Department, the Commission, pursuant to §957.04( 1)( f), must provide programs designed to reduce recidivism in its correctional institutions. Effective programs clearly would yield consequential financial benefits. The Commission continues to work with its contract vendors with the objective of designing and implementing innovative programs and has made it clear that it will authorize inmate welfare funds to be expended for such programs. Further still, many if not most welfare fund expenditures in public correctional institutions involve services which the Commission requires its Contract Vendors to provide within the contracted per diem. The Commission has never requested any supplemental funding for these services. Yet the Department of Corrections was given a $1.41 "credit" and the Commission, in effect, was penalized for its more cost-effective and far-sighted policies.

OPPAGA Comment: The executive director states that we have penalized the commission for "being frugal in this area," and that the approach taken by OPPAGA is "wrong-headed." He also contends that these funds are "under the firm control of the commission."

The basis for our approach is to determine the cost to the Florida taxpayer of public and private prisons. Just as we have given credit to the vendors for documented tax payments to the state, we gave credit to any prison that returned its canteen profits and telephone commissions to the state treasury. Expenditures from the Inmate Welfare Trust Fund are included in the reported costs from Lawtey; therefore, we have credited the income received by the state from Lawtey. The Inmate Welfare Trust Fund has been used differently, however, for the private prisons. Expenditures from the fund are not included in the costs of the private prison, and the receipts from the fund have not been returned to the state.

As to the issue of controlling the funds, CCA's financial audit disclosed that CCA was not depositing all telephone commissions into the Inmate Welfare Trust Fund, and that CCA owed the fund over $100,000. Although the commission requirement for a financial audit succeeded in detecting and correcting this procedure, we question the executive director's assertion that these funds have been in the "firm control of the commission at all times."

4. OPPAGA attempts to critique the difference between the marginal cost rate at Bay and Moore Haven and either implies or concludes that this has had an altogether negative effect on Commission cost savings efforts. The OPPAGA conclusions would have been different had they examined actual expenditures. In point of fact, the total expenditure during the fiscal year at Bay was lower than the total expenditure at Moore Haven even though Bay accounted for a larger number of prisoner mandays than did Moore Haven. Specifically, during the time period under consideration Bay received operating cost payments of $11,425.243.11 and accounted for 258,302 prisoner mandays. Moore Haven received operating cost payments of $11,700,600.98 and accounted for 257,739 prisoner mandays. The reason, of course, is that the $44.37 base cost at Bay for the first 675 prisoners is $2.77 below the base cost at Moore Haven. The OPPAGA conclusion that "The Commission did not structure the contracts to ensure that the projected level of cost savings was achieved" because in part of an unfavorable structured contract with regard to the Bay marginal per 54 54 Page 55 56 48 diem rate, is simply inaccurate because OPPAGA ignored the lower base cost of the Bay Contract. In fact if the contract had been adjusted the way OPPAGA has indicated the State of Florida would have paid more money rather than less money.

OPPAGA Comment: The executive director defends the high marginal per diem rate for the CCA contract by arguing that the overall operating cost for Bay is slightly lower than the overall operating cost for Moore Haven. While this is true, the state paid $5.6 million more for construction of the Bay facility than for the Moore Haven facility. Given the high cost of constructing Bay, we would expect CCA to provide the state with substantial operating cost savings.

The executive director notes that the guaranteed rate for Bay is lower than the guaranteed rate for Moore Haven. If both prisons were maintained at 90% capacity, the unadjusted cost for Bay would be almost $600,000 less than Moore Haven. However, the savings achieved at that level of occupancy is then progressively given back to CCA as the inmate population increases. OPPAGA believes that it is the responsibility of the commission to achieve the best possible deal for the state in negotiating contracts with private prison vendors. It does not represent good economy for the state to pay over $41 per inmate when the same inmate can be housed elsewhere for less than $9.

Flawed Estimates of Construction Cost Comparisons OPPAGA makes an effort to estimate construction cost savings and concludes that building costs for Bay and Moore Haven were higher than the three public correctional institutions it chose for comparative purposes. We believe the effort was misguided in its logic and flawed in its conclusions. The basic reasons for our criticism are summarized below:

1. The selection of the public correctional institutions appears to have been based largely or entirely on the fact that they opened at approximately the same time as did Bay and Moore Haven. This ignores the fact that the public and private correctional institutions have major design and capacity differences and that these differences are highly correlated with building costs.

2. Commission building costs are all inclusive. Per bed building costs for Bay and Moore Haven, for example, include all furnishings, fixtures, and equipment. The Commission has never sought, for example, as much as a dollar in additional appropriations for major maintenance or repair or for such commonly encountered Department requests as "security enhancements."

3. Each Commission construction contract guarantees that each private correctional institution will have sufficient space, furnishings, and equipment to meet its statutory obligation to provide sophisticated programs designed to reduce recidivism. The Department has no comparable statutory duty and few if any such programs. The effect of this difference on construction costs should be obvious.

4. Each Commission construction contract guarantees that each private correctional institution be designed and constructed in such a way as to control long-term maintenance and repair costs. We 55 55 Page 56 57 49 do not know whether the Department is similarly concerned about long-term costs. We do know that requests by the Department for major maintenance and repair costs are rather routine.

5. Each Commission construction contract guarantees that each private correctional institution will have "state-of-the-art" security equipment so that the public safety mandate of the Commission can be discharged in a highly effective manner and the number of full-time employees required to staff Commission correctional institutions can be minimized without causing any compromise in the caliber of services delivered within the correctional institutions. We have reason to believe that Department construction costs do not provide for the purchase and installation of such equipment. Again, the effect of this difference on construction costs should be obvious.

OPPAGA Comment: We agree with the executive director that numerous differences between the private and public prisons that were opened in 1995 limit the conclusions that can be reached in comparing construction costs. However, it was clearly not the Legislature's intent to build expensive private prisons that do not translate into cost savings for the state. The Legislature expected the commission approval of private prisons to provide the state with at least a 7% cost savings. The fact that Bay was 33% more expensive than the other private prison built in response to the same RFP is perhaps the most compelling evidence that the construction of Bay did not provide the state with any cost savings.

None of these building cost-related factors were acknowledged by OPPAGA in the cost comparison. Defects in the OPPAGA Performance Assessment The OPPAGA Report appears to be contradictory with regard to the performance assessment. In one spot OPPAGA indicates that the "performance outputs and outcomes for the Corrections Corporation of America and the Wackenhut Corrections Corporation compare favorably with the Department", in another spot in the Report OPPAGA indicates that Bay and Moore Haven are said to be performing at a level "generally consistent with performance reported by public prisons". The Commission is distressed by the cursory and we believe understated conclusions OPPAGA draws regarding performance. Independent audits of Bay and Moore Haven yield superior performance assessments. The independent accreditation audits by the American Correctional Association yield superior design, construction, and performance assessments. The public record regarding escapes, disturbances, homicides, suicides, major disciplinary reports, and assaults on staff yields superior results. The public record on the percentage of prisoners involved in and completing educational programs, substance abuse programs, and vocational programs yields superior results. Notwithstanding mounds of contrary evidence Bay and Moore Haven are said to be performing at a level "generally consistent with performance reported by public prisons." It would be accurate to conclude that the Commission flatly rejects this demonstrably invalid conclusion. It is unseemly that the Commission's record of solid accomplishment is trivialized in this manner. As an example of this trivial manner, on Exhibit E-3, OPPAGA lists the number of GED program completers. While the two private correctional institutions represent a very small percentage of the entire State of Florida inmate population, the data listed on Exhibit E-3 indicates that these two private correctional institutions have provided around 22% of the overall GED program completers. 56 56 Page 57 58 50 Areas of Agreement The Commission believes OPPAGA made a good faith effort to provide as comprehensive a report as it was able to prepare. Equally importantly, there are several general areas of agreement with the Report that the Commission would like to identify for the record which we have noted below:

1. We believe OPPAGA is correct in its conclusion that the assignment policies of the Department are contrary to the statute which mandates the transfer by the Department of "a cross section of the general inmate population, based on the grade of custody or the offense of conviction, at the most comparable correctional institution operated by the Department" (§ 957.08). Judged by any prevailing national professional standard, Bay and Moore Haven would provide the most cost-effective benefits to Florida were they used primarily or exclusively for the housing of prisoners with medium security classifications.

2. We believe OPPAGA is correct in its conclusion that negotiations for the contract renewal on Bay can include an adjustment in the marginal cost which is significantly below the existing marginal cost (i. e., the per diem allowable for prisoners in excess of 90 percent of the design capacity of Bay). However, as we have already indicated such an adjustment over the time period covered in the Report would have cost the State of Florida more money, rather than produced a cost savings.

3. We believe OPPAGA is correct in its conclusion that an effort should be made to further reduce construction costs associated with new Commission correctional institutions and any expansions of existing Commission correctional institutions via reliance on inmate labor if, of course, such reliance would have cost benefits. 4. We believe OPPAGA is correct in its conclusion that "additional cost savings could be achieved by increasing the size of the private prisons." If so authorized by the Legislature, the Commission is prepared to move forward with immediate expansions of Bay and Moore Haven.

Overall Response to the Report The Commission appreciates the effort expended by OPPAGA and the cooperative manner in which OPPAGA attempted to meet its statutory duties. OPPAGA has acknowledged that there are no public correctional institutions which are substantially comparable in design, location, size, and mission. We concur. We had hoped for a more sophisticated analysis. Perhaps limitations of time and data made that impossible. However, the unfortunate result is that the Report provides little by way of its analytical approach or empirical findings which can guide and direct the Commission or the Legislature. The additional result is that still more confusion and disagreement will exist regarding the considerable contributions privatization has made in Florida. The on balance conclusion of the Commission is that a fair and reasonable reading of what OPPAGA attempted coupled with the comments we have made here will provide more than persuasive evidence that Bay and Moore Haven are providing services superior to what one finds in broadly comparable public prisons in Florida at a cost below that of typical Department correctional institutions. Those achievements standing alone are both positive and noteworthy. Although the Report rather clearly underestimates actual cost savings, and although our separate and on-going analysis persuades us that the 7 percent cost savings goal is being achieved, we, like OPPAGA, cannot precisely quantify the actual cost savings. There simply is no public correctional institution that is sufficiently similar to either Bay or Moore Haven. Consequently, future efforts to ignore this unavoidable reality can only result in confusion, disagreement, and a pointless expenditure of both time and appropriated funds. 57 57 Page 58 59 51 OPPAGA Comment: We are interested by the executive director's assertion that the 7% cost savings is being achieved, and that therefore there is no further need to waste time attempting to quantify that savings. The commission cannot have it both ways: either the savings can or cannot be quantified. If the savings cannot be quantified, then the commission cannot fulfill the duties of s. 957.07, F. S., that require that the commission not enter into a contract unless it determines that the contract will result in a 7% savings over the public provision of a similar facility.

Should you have any questions or require additional information, please do not hesitate to call me at 921-4034.

Sincerely, /s/ C. Mark Hodges Executive Director

CMH: kls cc: Joel J. Freedman, Chairperson 58 58 Page 59 60 52

Florida Department of Corrections
2601 Blair Stone Road
Tallahassee, Florida 32399-2500
(850) 488-5021

March 20, 1998
Mr. John W. Turcotte, Director
Office of Program Policy Analysis and Government Accountability
111 West Madison Street, Room 312
Claude Pepper Building
Tallahassee, Florida 32301

Dear Mr. Turcotte:

Staff of the Office of Program Policy Analysis and Government Accountability (OPPAGA) are to be commended for undertaking and completing the very difficult task of comparing the costs and benefits of privately contracted prisons with those built and operated by the Department of Corrections. Although we do not agree with all of the conclusions and with some of the assumptions made for developing the costing methodologies, we certainly recognize the complexity of the issues involved and appreciate OPPAGA's efforts to maintain as much objectivity as possible in an otherwise highly subjective exercise.

Our primary area of disagreement is with OPPAGA's decision to use Lawtey Correctional Institution as the singular public facility to use in evaluating the operating cost efficiency of the two contracted facilities. Although similar in size and custody level, Lawtey has many characteristics that make it significantly different from the two private facilities. OPPAGA was thorough in its attempt to identify, measure, and adjust for many of these differences.

However, we believe that limiting the comparison to one public facility distorts the magnitude of the quantitative differences. By using a group of similar public facilities and adjusting for measurable differences, we believe that the unavoidable distortions would be diluted and a more realistic and meaningful conclusion would result. That is why the department chose to compare the cost of the two private facilities to the average of nine department facilities, one of which is Lawtey Correction Institution. Although the remaining eight comparable departmental facilities are all larger than either Lawtey or the contracted facilities, they are more representative of what typical departmental facilities are like. As you have accurately pointed out on page 14, these larger facilities are much more cost efficient that either Lawtey or the contracted facilities. It is important that the readers of this report understand that if the inmates at Bay or Moore Haven had been assigned to a departmental facility, the vast majority would be housed in one of the larger institutions at considerable cost savings to the taxpayers (DC analysis indicates that cost savings would range from 11 percent to 14 percent). It is therefore misleading to state that one of the contracts (Moore Haven) provides a 4 percent savings.

OPPAGA Comment: Using the nine public prisons suggested by the department would require adjustments for size, work camps, and close custody inmates. While other public prisons may be more representative of typical department prisons, they are not as similar to Bay and Moore Haven as Lawtey.

Other areas of concern are summarized as follows:

1. Including Everglades C. I. as a comparable facility for construction costs causes considerable distortion to the information presented in Appendix C. The selection of the Everglades site was based on the critical need to provide more prison beds in Dade County and not on maximizing cost effectiveness. Located in that urban area, construction costs were significantly greater than department norms. On the other hand, both Taylor and Washington represent sites where the department had considerably more discretion to economize on construction costs. As indicated in Appendix C, both of these department-constructed institutions cost much less (cost per bed) than either of the contracted facilities.

2. The report did not elaborate on the reasons why the department's site preparation costs appeared to be higher. The department's long range planning invariably includes a determination whether acquired sites can be utilized for future expansion to include a work camp and/ or an annex. Both Taylor and Washington sites were master planned to include both a work camp and an annex. Part of the cost of the original site preparation contract included work related to future expansion (approximately $750,000 for Taylor and $1,250,000 for Washington). These costs benefit future facilities and should not be included in the comparison with the private prisons. Both Taylor and Washington have their own wastewater treatment and water plants, valued at $1,750,000 per site.

The private facilities are both connected to local municipal plants. This investment in institutional infrastructure will result in lower operating costs over the life of the facility and should not be shown as an added cost when comparing construction costs with the privates.

3. The schedule on page 30 clearly demonstrates that Bay and Moore Haven are not comparable to Everglades, Taylor, or Washington in regards to the custody level of inmates housed or the numbers of medically and psychologically restricted inmates. Unlike the private institutions, all three public facilities house close-custody inmates, including high-risk close management inmates, and Everglades and Washington have psychological impairment workloads of 30 percent and 39 percent, respectively. It is therefore misleading to compare these institutions on the basis of selected custody and control measures, such as escapes, assaults, and disciplinary reports (Exhibit E-1), or on the frequency of deficiencies cited by the Correctional Medical Authority (Exhibit E-2).

We certainly concur with the report's findings and recommendations. To summarize: · The Bay and Moore Haven contracts have not provided the required cost savings to the state. 60 60 Page 61 62 54

· These contracts should be renegotiated to provide a minimal 7 percent savings. We further believe that these renegotiations should take place prior to the upcoming contract renewal period (4 th year).

· Bay and Moore Haven could be made more cost efficient if their capacities were increased. However, it should be emphasized that any expansion of these facilities should also meet the 7 percent cost savings test.

· Contract costs need to be continually evaluated to ensure that the required level of savings are maintained.

· The most meaningful method of comparing operating costs is to focus on direct costs and avoid over-analyzing and over allocating indirect costs to either public or private institutions.

· A more effective use of the housing capacity at Bay and Moore Haven would be to place higher custody inmates into their more secure single-cell units.

· The department should reduce the inmate population at Bay Correctional Facility to the 90 percent guaranteed minimal level until their excessive marginal rate is renegotiated.

· The department should adjust the inmate mix at private facilities to assign more medium-custody inmates and fewer minimum-custody inmates.

· Canteen profits and telephone commissions should be deposited with the state to be used to offset the costs of the private prisons.

· Each contract provider should be required to maintain and report data on selective performance measures, similar to these required of the department.

We appreciate the opportunity to respond to this report. OPPAGA staff have conducted their review in a very professional manner and have remained open and responsive to the department's input on these critical issues.

Sincerely, /s/ Harry K. Singletary, Jr.

Secretary

HKSJr/ JNB/ mt

CORRECTIONS CORPORATION OF AMERICA
10 Burton Hills Boulevard
Nashville, Tennessee 37215
Phone: 615-263-3000
Fax: 615-263-3140

March 19, 1998
Mr. John Turcotte
Director
Office of Program Policy Analysis and Government Accountability
The Florida Legislature
111 W. Madison St., Room 312
Tallahassee, FL 32301

Dear Mr. Turcotte:

In accordance with the provisions of s. 11.45( 7)( d), F. S., I am responding on behalf of Corrections Corporation of America (CCA) to the Office of Program Policy Analysis and Government Accountability's (OPPAGA) preliminary review findings on the costs and benefits of two private prisons operated under contract with the State of Florida through the Correctional Privatization Commission (CPC).

While CCA certainly appreciates and in fact encourages OPPAGA's review of these two contracts, we strongly disagree with many of the conclusions contained in the preliminary draft of Report No. 97-68. In short, we believe that these conclusions are based on the review's fundamentally flawed methodology--comparing the costs and operations of two privately operated facilities with a "similarly Department of Corrections-operated facility. Notwithstanding what I am sure was OPPAGA's dedicated and professional attempt to meet the Legislature's mandate for a study to determine whether privately operated prisons achieved "at least a 7% cost savings over the public provision of a similar correctional institution," one simple fact remains. Neither the two privately operated prisons nor the publicly operated facility to which they are compared are similar in any substantial way. Therefore, I believe that it is impossible and, indeed, even reckless to attempt to make any comparisons between or among the three facilities, let alone make any conclusions based on these comparisons. In fact, it is a matter of record that both the Department of Corrections and the CPC have acknowledged that there are no public correctional institutions in Florida which are sufficiently comparable to either CCA's Bay Correctional Facility (" the Bay facility") or Wackenhut's Moore Haven Correctional Institution (" the Moore Haven facility").

Left with no substantially similar facilities to compare, it appears as though OPPAGA's preliminary findings are completely dependent upon assumptions made or not made. These assumptions led to a series of vague and tortuous cost adjustments on both the public and private side of the ledger which resulted in comparisons and conclusions that are virtually meaningless. To any interested observer, it is these cost adjustments which are perhaps the most damning evidence of the fundamental flaws in OPPAGA's review.

If the facilities being examined were sufficiently comparable, this tortuous cost adjustment exercise would not have been necessary.

Before responding in detail to the flawed conclusions in OPPAGA's preliminary review, l would be remiss if I did not state for the record one significant comparison which was completely ignored in OPPAGA's review--the state's actual expenditures on the two privately operated facilities. If OPPAGA had investigated the state's actual expenditures further, it would have realized something very telling. During the fiscal year in question, the state's total expenditure at CCA's Bay facility was lower than the total expenditure at Wackenhut's Moore Haven facility, even though the Bay facility accounted for a larger number of prisoner mandays than did the Moore Haven facility. The Bay facility received operating cost payments of $11,425,243.11 and accounted for 258,302 prisoner mandays. The Moore Haven facility received operating cost payments of $11,700,600.98 and accounted for only 257,739 prisoner mandays.

Based on these simple and crucial figures, it absolutely defies logic for OPPAGA to make the claim, as it does in the preliminary review, that the state's contract with CCA was more expensive than the one with Wackenhut. There is not a mathematics class in the country where OPPAGA's claim could pass muster, and it is a prime example of the fundamental flaws which render this preliminary review meaningless in its present form.

OPPAGA Comment: CCA points out that the total operating costs of Bay were approximately $275,000 less than for Moore Haven during the 1996-97 fiscal year. At that rate, it will take more than 21 years for the state to recover the $5.9 million additional construction costs expended for Bay in comparison to Moore Haven.

If one considers only operating costs, one would expect Bay to be less expensive than Moore Haven due to its location. Because we considered the location of the facilities, as required by law, we estimated that the state would spend about $500,000 more in higher officer salaries for a prison located in Glades County compared to a prison located in Bay County. This adjustment for location costs results in a more favorable analysis of Moore Haven operating costs.

As mentioned above, CCA strongly disagrees with many of the preliminary review's comparisons and conclusions. The most glaring deficiency of the review is its premise that the state-operated Lawtey Correctional Institution (" the Lawtey facility") is substantially similar to both the Bay facility and the Moore Haven facility and that this similarity is sufficient foundation from which to make comparisons for government cost accounting purposes. In point of fact, the Lawtey facility and the privately operated facilities are fundamentally dissimilar in nearly every way--from the size and design of the facilities to the type of inmates held in each facility:

1) First, the review wrongly implies that the inmate population of the Lawtey facility (733 inmates as compared to 708 at the Bay facility and 706 at the Moore Haven facility) makes it sufficiently comparable to the privately operated facilities. OPPAGA failed to account for the design capacities of the three facilities--555 at Lawtey, and 750 at both the Bay facility and the Moore Haven facility. It is common knowledge that major economies of scale are created at a facility which is operating above design capacity as compared to a facility which is operating below its design capacity. Because OPPAGA failed to account for this factor, the resulting operational and cost comparisons are irrevocably skewed.

OPPAGA Comment: During the 1996-97 fiscal year, Bay did not house significantly more medium custody inmates than Lawtey. (See Table F-1, page 42.)

2) All three facilities vary widely in design and, therefore, construction costs. OPPAGA acknowledges this fact on Page 7 of the preliminary review, stating that "Cells are generally more expensive to construct than dormitory housing."

All inmates at the Bay facility are housed in two-person cells, whereas the Moore Haven facility is a combination of dormitory units and cells. It is our understanding that the Lawtey facility is comprised primarily of dormitory units.

3) The Lawtey facility houses a substantially higher percentage of inmates with minimum security classifications than do the privately operated facilities. As all correctional professionals know (and as OPPAGA also states), it is less costly to house minimum security inmates than medium security inmates. I was very shocked to learn that OPPAGA did not account for this discrepancy while it was making all of the other cost adjustments.

Having established at length the reasons why these facilities are not comparable, I can partially understand the basis for OPPAGA's reliance on assumptions and cost adjustments to complete this review and meet its legislative mandate. Although well-intentioned, this exercise failed in nearly every respect. For the sake of brevity, I will not attempt to critique every assumption and cost adjustment. However, some of them are so egregious that they beg further comment:

1) OPPAGA factored $3.41 out of the state's costs for operating the Lawtey facility due to its "higher medical grade inmates." Notwithstanding the fact that this adjustment confirms our contention that the facilities are not comparable, OPPAGA failed to grant CCA and Wackenhut similar adjustments for housing significantly more medium security inmates than the Lawtey facility.

2) Each privately operated facility was given an adjustment to account for the differences in inmate population. CCA was given an adjusted per diem rate of $41.60 for these additional inmates, while Wackenhut was given an adjusted per diem rate of $8.87. Although this difference can be attributed to language which is specific to CCA's contract with the state, OPPAGA fails to recognize that CCA's overall per diem rate is lower than the other vendor's.

3) OPPAGA factors $0.98 into CCA's cost in the form of Department of Corrections oversight of the CCA contract. Given the fact that ample oversight of the contract is already provided by the CPC, CCA believes that this cost adjustment is not justified and is, in fact, a redundancy that could be translated as a cost savings for the Department of Corrections.

4) The state was given a $1.41 reduction for spending its inmate welfare trust fund money to lower inmate costs. CCA and Wackenhut received no similar credit from OPPAGA, which failed to account for the fact that the vendors are prohibited from expending these funds without the express authorization of the CPC. Further, it should be noted that CCA is working closely with the CPC to utilize these funds in implementing cost-effective and innovative programs aimed at fostering rehabilitation and decreasing recidivism over the long-term.

In effect, the review penalizes CCA for not expending these trust funds in favor of more cost-effective policies.

5) Lastly, OPPAGA failed to properly account for the various taxes that the privately operated facilities are required to pay. As you will note in the attached cost comparison, CCA believes it should be properly credited for both Sales Tax payments and Franchise & Excise Tax payments. Had these payments been accounted in accordance with generally accepted accounting principles, additional cost savings to the state would be evident.

I was particularly disturbed by OPPAGA's conclusion that the Bay facility was performing at a level "generally consistent with performance reported by public prisons."

The record in this regard clearly shows that the Bay facility has achieved a much greater degree of success. Independent audits of the Bay facility by the American Correctional Association (ACA) have yielded superior design, construction and performance assessments. Further, as I am sure the CPC will concur, the Bay facility's educational, substance abuse treatment, and vocational programs have also yielded superior results.

In fact, although both the Bay facility and the Moore Haven facility comprise an extremely small percentage of the Department of Corrections' inmate population, the CPC estimates that the inmates completing GED programs in the two facilities represent approximately 22% of the state's total GED completion percentage. Further, some mention should also be made of the widely respected study on recidivism rates recently done by the University of Florida. Although the study's findings are preliminary, it indicates that offenders who have been incarcerated in privately operated facilities in Florida have much lower recidivism rates than offenders who have been incarcerated in publicly operated facilities in Florida.

Although we have many serious concerns with the preliminary review, it should be noted that CCA does agree with some of the recommendations OPPAGA details in Chapter 4 (" Conclusions and Recommendations"). First and foremost, we agree with the review's recommendation that "the Legislature authorize the Correctional Privatization Commission to renew its contracts with the Corrections Corporation of America for the operation of Bay Correctional Facility and with Wackenhut Corrections Corporation for the operation of the Moore Haven Correctional Institution..." based on certain conditions.

Specifically, we agree that the facilities would provide the most cost-effective benefits to the state were they expanded to 1,000 beds and were used primarily to house medium security inmates. The cost of expansion construction would be significantly lower than the initial per bed construction cost and the per diem rate of a facility expanded to 1,000 beds would be substantially lower than the current per diem being paid by the state to the private operators.

CCA appreciates OPPAGA's effort to meet its statutory duty with this review. And we acknowledge the factors that prohibit a meaningful analysis of cost and performance. To correct these deficiencies, CCA strongly recommends that the State of Florida follow the example set by the State of Louisiana. Louisiana's experiment, which is perhaps the best-devised instrument to date for comparing the costs and performance of public and private prisons, established three identical medium security prisons--one operated by the state corrections department, one operated by CCA, and one operated by Wackenhut. Over the course of seven years, the costs and performances of these three facilities were closely monitored by independent experts. At the end of the monitoring period, the resulting assessments found that the privately operated facilities outperformed the state operated facility in every regard.

In the absence of this type of experiment, it is perhaps impossible to accurately quantify cost and performance factors. However, CCA has provided a revised cost comparison that we believe more accurately depicts the true costs of the privately operated facilities versus the costs of the Lawtey facility. We stand filmy behind our contention that we have provided the state a savings of at least 10%, and we would welcome your close review of this comparison.

OPPAGA Comment: CCA modifies our cost analysis in three basic ways:

· CCA deleted the $3.41 adjustment that we made to account for differences in medical services. CCA, where less than 10% of its inmates have medical restrictions, purports to compare the services it provides to the services provided at Lawtey where over 27% of the inmates have medical restrictions. The CCA did not change any of the similar adjustments for education or substance abuse programs that we made that favored CCA.

· CCA gave itself a credit for canteen profits and telephone commissions even though those funds are still in vendor accounts and have not been used to offset the cost of private prisons to Florida taxpayers.

· CCA made cost adjustments for its estimated sales taxes and for its franchise and excise tax (corporate income tax) payment to Florida. The income tax credit claimed by CCA is not only for Bay, but also includes income from all of its Florida facilities. CCA did not provide documentation to us that would indicate what portion of that tax was based on income from the Bay facility.

In closing, let me state for the record that I strongly believe the State of Florida, through the CPC, has an acceptable means of continuing the superior performance of the privately operated facilities. We continue to be impressed with the CPC's oversight and monitoring of the Bay contract, and we look forward to working with the CPC and the Department of Corrections to strengthen our partnership and provide Florida taxpayers with quality corrections at a lower cost. Although OPPAGA made no attempt to assess the effect of private competition on the public operation of correctional facilities, it seems clear that the real winner in this debate are the Florida taxpayers. I think you will agree that a monopoly is rarely as cost effective as a competitive environment.

Sincerely, /s/ John D. Rees Vice President -Business Development

JR/ jw Attachment

cc: The Honorable Allen Trovillion All House Corrections Committee The Honorable Ron Silver The Honorable Al Gutman The Honorable Alex Villalobos Mark Hodges 67 67 Page 68 69 61 ATTACHMENT PROVIDED WITH RESPONSE FROM THE CORRECTIONS CORPORATION OF AMERICA REVISED APPENDIX D Operating Coats Comparison of Bay and Moore Haven Private Prisons With Department's Reported Costs for HYPOTHETICALLY Similar Lawtey State Prison Fiscal Year 1996-97 Private Prisons State Prison Bay Moore Haven Lawtey

Total Due the Vendors Based on Contracted Per Diem Rates 11,723,543 12,122,767 Less: Deduction for On-site Monitor (74,543) (74,406) Less: Deduction for Property taxes included in Contract But not Paid to Counties (223,757) (347,699) Less: Medical Co-Payments Collected from Inmates Retained by the Vendors (2,975) (1,471)

Total Payments to Vendors $ 11,422,268 $ 11,699,191 $ 12,232,407 1996-97 Per Diem: Operating 36.65 36.40 36.73 Health Services 5.46 6.63 8.55 Education Programs 2.11 2.36 0.45

Total 1996-97 Unadjusted Per Diem $ 44.22 $ 46.39 $ 45.73 ADJUSTMENTS Adjust Private Prisons to Lawtey's Population of 733 Bay (708): Add 25 inmates @ $41.60/ day (0.14) Moore Haven (706): Add 27 inmates @ $8.87/ day (1.41) On-site DC Classification Staff at Private Prisons 0.30 0.32 Hospital Cost Limit (>$ 7,500) for Private Prisons 0.21 0.30 Sales Tax Adjustment (0.35) (0.35) Florida Franchise & Excise Tax Adjustment (0.18) (0.18) Payments to Maintenance Reserve Fund for Private Prisons (0.28) (0.28)

Credit for Cost of Supervising Public Work Squads (0.62) Credit for Inmate Welfare Trust Fund Net Revenue (1.10) (1.10) (1.41) Adjust Medical Costs for Higher Medical Grade Inmates -Enhanced Education Programs 1.95

Enhanced Substance Abuse Treatment 0.53 RECIDIVISM IMPACT ? ? Net Total of Adjustments (1.54) (2.70) 0.45 Indirect Cost Benefit From: Department of Corrections 0.98 0.89 3.33

Correctional Privatization Commission 0.89 0.89 Other State Agencies 0.05 0.04 0.24 Total Indirect Benefits 1.92 1.82 3.57 Total Adjusted Per Diem $ 44.60 $ 44.51 $ 49.75 Estimated Cost Savings (Per Diem) $ 5.15 $ 5.24 Percentage Cost Savings Based on State Cost 10.4% 10.5%

TRUE COST COMPARISON

Unadjusted Per Diem $ 44.22 $ 45.39 $ 45.73 Indirect Costs excluding DOC from private costs $ 0.94 $ 0.93 $ 3.57 Total Adjusted Per Diem $ 45.16 $ 46.32 $ 49.30 Estimated Cost Savings (Per Diem) $ 4.14 $ 2.98 Percentage Cost Savings Based on State Cost 8.3% 6.0%

1996-1997 Payments to Private Vendors Total Payments $ 11,425,243 $ 11,700,661 Mandays 258,302 257,739

Wackenhut Corrections Corporation Executive Offices:
4200 Wackenhut Drive #100
Palm Beach Gardens, Florida 33410-4243
March 17,1998

Mr. John W. Turcotte, Director
Office of Program Policy Analysis and Government Accountability (OPPAGA)
Post Office Box 1735
111 West Madison Street, Room 312
Claude Pepper Building
Tallahassee, Florida 32302

Dear Mr. Turcotte:

Thank you for extending to Wackenhut Corrections Corporation the opportunity to comment on the OPPAGA Draft Report to the Legislative Audit Committee on the Moore Haven and Bay Correctional facilities. As you are aware, Wackenhut Corrections designed, constructed, and operates the Moore Haven facility while Corrections Corporation of America performed similar tasks for the Bay County facility.

Wackenhut Corrections is very proud of the two correctional facilities it operates in Florida (the Moore Haven facility in Glades County and the South Bay facility in Palm Beach County) and their contribution to addressing the fiscal and operational challenges facing the State of Florida. The success of these facilities in delivering a high quality, cost effective correctional product to the State is a testimony to the privatization strategy endorsed by the Legislature several years ago and executed by the Privatization Commission. Our corporate objective clearly was to deliver these needed facilities on time and on cost, and to operate the facilities in such a manner that they would set an example of excellence. The achievement of these objectives in our opinion would foster a competitive environment whereby privatization is viewed as an alternative approach to "business as usual". These challenges were not inconsequential and we stand on our record of performance.

The task assigned OPPAGA was also not inconsequential and was, to say the least, daunting in its scope. The key challenge confronting the OPPAGA team was to develop a methodology for assessing comparability of costs of the two privatized correctional facility against the pool of publicly run, geographically diversified State facilities with substantial differences in sizes, complexity of operation, commitment to programming activities, and security missions. In addition to the direct cost comparisons, the whole arena of indirect or overhead costs application was also addressed. The complexity of these assessments required a high degree of judgmental evaluation by OPPAGA in order to bridge these complex differences, with each and every judgment being critical to the overall evaluation.

One factor, however, recognized by OPPAGA as a significant determinant of the overall cost equation is the relationship of size of the facility to the cost per inmate.

We have in our operations throughout the U. S. seen a marked increase in cost efficiency produced by economies of scale in facilities of 1000 beds or more in size. This key factor puts the current 750 bed facilities at a significant disadvantage when compared to the majority of public run facilities that are over 1000 beds in size.

As you are aware, we designed the Moore Haven facility to meet the original specification for a 750 bed; however, we anticipated an evolving need and have already prepared the site for an additional 250 bed housing unit. Wackenhut strongly endorses the OPPAGA recommendation on expanding the present size of the Moore Haven Facility in order to capture the economies of scale a large size facility would produce.

The primary ingredient of the direct cost base in either the public or private sector is labor costs. The OPPAGA report utilized the Annual Report of the Department of Corrections as the baseline for assessing the staffing of the publicly run facilities. In addition, OPPAGA determined that the most comparable publicly run facility was the Lawtey facility. In our fact finding we discovered a major discrepancy between the actual staffing of the Lawtey facility as contrasted with staffing numbers in the Annual Report. We believe the magnitude of this difference is approximately forty staff which, conservatively estimated, should have a value of approximately $1.5 million. This issue has a significant impact on the overall cost evaluation and should be explored further by the OPPAGA Audit team.

OPPAGA Comment: Wackenhut's reference to "a major discrepancy between the actual staffing level of the Lawtey facility as contrasted with staffing numbers in the Annual Report" of the Department of Corrections is unfounded. Wackenhut raised questions upon review of the draft report regarding the number of staff positions at Lawtey and whether all of these positions had been accounted for in our cost calculations. We verified that all positions had been included. Therefore, there is no "$ 1.5 million discrepancy in staffing."

Costs are not the sole determinant of the value equation. Quality must be elevated since it defines any successful operation. Wackenhut believes the quality of services delivered at the Moore Haven Facility is outstanding. This is evidenced by several factors to include the achievement of American Correctional Association Accreditation with a score of 99.5% within one year of facility reaching full capacity. In addition, during the period of the OPPAGA analysis, the facility awarded to inmates:

185 GEDS
206 Vocational Awards
677 Substance Abuse Awards
697 Life Skills Course Completion Certificates

OPPAGA Comment: These numbers do not agree with the numbers provided to OPPAGA by the department through its automated database, or by the commission after it reviewed a draft of our report. We have included a footnote to Table E-3 based upon documentation provided to us by the vendors through the commission.

This demonstrates both our corporate commitment and ability to deliver high quality programmatic service to the inmate population -a key component to our contractual responsibility to the State of Florida.

In the OPPAGA Report there was no assessment of either content nor quality of programs, of the outstanding number of GED or completed programs, or of the type of programs. The evaluators looked solely at the numbers of inmates in programs to do their comparisons. The Report does recognize that the programs at Moore Haven are extensive and rich, however it fails to factor in the relative value of these programs by focusing only on program numbers.

Competition is healthy and focuses on the energies of both the public and private sectors in most efficiently using the fiscal resources available to the State. As outlined in the OPPAGA Report, the cost of construction of the Moore Haven Facility is comparable with three similar public built facilities, even though the public built facilities had access to inmate labor with a corresponding 16% reduction in labor costs. In addition, the OPPAGA Report clearly shows at the Moore Haven Facility a minimum savings to the State of just under $500,000. A credit for the funds held in a trust from inmate telephone revenues adds significantly to this savings.

State law regarding the use of inmate welfare telephone revenue funds at public facilities was rewritten during the period after the issuance of the RFP and preparation of this report. Prior to the RFP, State facilities used the fund money to offset the cost of running their facilities. With the change, all funds collected are returned to the State as revenues. The private vendors have collected the money and Wackenhut has not spent more than a nominal amount since opening. If a credit was given to the Moore Haven facility for the funds available from inmate telephone revenue, an additional $3560,000 in cost saving would be possible thus achieving the 7% savings target.

Wackenhut has recommended to the Privatization Commission, however, that these funds be utilized to construct an industry building at Moore Haven which we would utilize for inmate participation in the Federal Prison Industry Enhancement (PIE) Program.

With proper statutory authorization and based upon our experience in a similar size facility in another State, we believe this investment can be returned to the State coffers through taxes and collections for cost of housing within three years with significant benefits accruing to the inmate population through direct access to challenging jobs at prevailing wages.

In summary, Wackenhut Corrections Corporation appreciates the efforts of the OPPAGA team in attempting to compare the costs of operations of public versus private facilities; however, we would recommend a more detailed cost audit of the Lawtey facility to establish a credible cost baseline for comparability. A $1.5 million discrepancy in staffing undermines the conclusions and recommendations of this current report.

Sincerely,
/s/ John G. O'Rourke
Chief Financial Officer

The Florida Legislature
Office of Program Policy Analysis and Government Accountability

OPPAGA provides objective, independent, professional analyses of state policies and services to assist the Florida Legislature in decision-making, to ensure government accountability, and to recommend the best use of public resources.

Copies of this report in print or alternate accessible format may be obtained by telephone (850/ 488-0021 or 800/ 531-2477), by FAX (850/ 487-3804), in person (Claude Pepper Building, Room 312, 111 W. Madison St.), or by mail (OPPAGA Report Production, P. O. Box 1735, Tallahassee, FL 32302).

Web site: http:// www. oppaga. state. fl. us

Project supervised by: Byron Brown (850/ 487-9215)

Project conducted by: Lee Cobb (850/ 487-9273) and Charles Barrett (850/ 487-9275)

ANNOUNCEMENT

The Office of Program Policy Analysis and Government Accountability announces the availability of its newest reporting service. The Florida Government Accountability Report (FGAR), an electronic publication specifically designed for the World Wide Web, is now up and operating for your use.

FGAR provides Florida legislators, their staff, and other concerned citizens with approximately 400 reports on all programs provided by the state of Florida. Reports include a description of the program and who is served, funding and personnel authorized for the program, evaluative comments by OPPAGA analysts, and other sources of information about the program.

Please visit FGAR at http:// www. oppaga. state. fl. us/ government. Your comments and suggestions about improving our services are always welcome.

Gena Wade, FGAR Coordinator (850) 487-9245 74