Chapter 5

Private Prisons in South Africa

Prisons in South Africa were constructed in an era which did not embrace the rehabilitation ideal but were intended instead to warehouse prisoners as labour commodities. The shortage of prison space is severe, and many of the existing prisons require significant repair even without the added strain of overcrowding. The African National Congress (ANC) government, however, has adopted an economic strategy which emphasizes reduced government spending. The Minister of Finance is encouraged by foreign governments and international lending agencies which often attach conditions of government deficit reduction to loans or aid packages. Thus, private prisons were first considered as an option in South Africa in 1994, although the first contract was not signed until more than five years later.

Prison policy and the development of private prisons

The average prison in South Africa is forced to accommodate up to twice its intended capacity. In the larger prisons, many of the cells intended for 18 are crowded with 50 to 60 inmates. Often, there are not enough beds or blankets and new arrivals must soon learn to align themselves with powerful gang members in order to obtain the basic necessities.

The food provided is also substandard. Usually, a prisoner in South Africa is given breakfast of mielie pap and then dinner and lunch are combined into an afternoon meal at 15h00. This meal usually consists of ground fish meal and several slices of bread. Meat, fruit, and vegetables are scarce, and the food that does arrive at the prison is often smuggled and stolen so that it can be sold to the highest bidder. Most prisoners are poor and so the medium of exchange becomes sexual favours. This results in victimisation, if not outright assaults, between prisoners and the situation is only made worse by increasing overcrowding and decreasing availability of basic provisions.

The Department of Correctional Services is aware of the problems and challenges faced by the prison system. The department has recognized that the rehabilitation of offenders should be a primary function, but also remains aware that this goal is difficult, if not impossible, to attain given the current situation of overcrowding. The department has adopted several strategies to attempt to deal with overcrowding, including the construction of more prisons. However, the government is also aware that it is impossible to build itself out of the overcrowding problem. In addition to providing more and better facilities, DCS is exploring options such as electronic monitoring and early release of those sentenced for petty offences.

Given the enormous demand for public services, most notably health and education, it is understandable that the Department of Correctional Services is not likely to receive the necessary infusion of capital to reform and refurbish its operations and facilities. For these reasons, DCS began to explore the option of involving the private sector. In 1997, then Minister of Correctional Services, Dr. Sipho Mzimela visited the United States and the United Kingdom in order to learn more about their prison systems. After his trip, Mzimela concluded that "wherever the private sector got involved, they have delivered a better service, and have done it at less cost to the taxpayer."1

The criminal justice budget has grown in proportion to overall government expenditure, from just over 9% in 1994 to 13% in 2001. The proportion of the criminal justice budget allocated to the Department of Correctional Services has increased slightly over the last five years, and currently stands at about 23%.

Particularly in the case of DCS spending, the most significant growth has been in capital expenditure, as opposed to personnel. A great deal of this increased spending on capital projects is due to prison construction. The Department of Correctional Services’ expenditure priorities listed in the 2001 Budget Review are the financing of operational costs and the provision of more prison space over the medium-term. These priorities are reflected in the budget allocations. Over the next three years, budgeted amounts for prison construction are set to grow at nearly 27% each year. This increase does not include the budgets for private prison construction, which is planned to be almost R500 million in 2003/04.2 DCS’ planned spending seems to be designed specifically for additional contracting with the private sector, as the budget aims to pay operating costs (fees per prisoner place) and aims for additional space in the medium term (shorter construction times). Fees per prisoner place and shorter construction times are the exclusive characteristics of private prisons.

The Correctional Services Act of 1998

The Correctional Services Act of 1998 (CSA 1998) contained a section which specifically authorized the government to contract out prison services to the private sector. Chapter XIV, "Joint Venture Prisons", Section 103(1) provides that:

"The Minister may, subject to any law governing the award of contracts by the State, with the concurrence of the Minister of Finance and the Minister of Public Works, enter into a contract with any party to design, construct, finance and operate any prison or part of a prison established or to be established..."

The CSA 1998 goes on to list very specific conditions and requirements for private prison contracts. The legislation states that:

Contracts cannot exceed 25 years.



The contractor must "contribute to maintaining and protecting a just, peaceful, and safe society".



The contractor is responsible for enforcing the sentences of the courts, detaining prisoners in safe custody, ensuring the prisoners’ human dignity, and promoting the human development of all prisoners.



The contractor is explicitly prohibited from taking disciplinary action against prisoners or from involvement in determining the computation of sentences, deciding at which prison any prisoner will be detained, deciding on the placement or release of a prisoner, or grant temporary leave.

The latter safeguards are important to include in legislation, and not just in a prison contract, because this adds an authority and independence which assists in enforcement and monitoring. Also, the CSA 1998 does a great deal to ensure that the responsibility for punishment rests with the state and that only the services are delegated to the contractor.

One interesting component of the CSA 1998 is that it explicitly forbids private contractors from becoming "involved in the implementation of community corrections."3 Community-based corrections refers to an effort to build or rebuild social ties and a connection to the community in order to prevent future violations. This usually includes obtaining employment or education and assisting the offender with adapting to the routine functioning of society. In the United States, prison researchers contend that the criminal justice system needs to incorporate more prison alternatives offered by community corrections, such as halfway houses, parole, probation, and work-release programs.

The private sector has been a significant part of community corrections in other countries, so it is unclear why this area which is most likely to provide innovation and reform benefits from the private sector would be specifically labelled off-limits in South Africa.

Various monitoring programmes have been tried in the countries which have introduced prison privatisation, and none of them have proven to be immune to the phenomenon of capture. However, the monitoring mechanisms which have been most successful at protecting the public interest are those which are designated by statute rather than by a section in a contract.

The CSA 1998 states that a controller must be appointed by the Commissioner of Correctional Services for each private prison, and then goes on to list the duties of the controller. However, the controller still reports to the Commissioner, rather than to a separate regulatory body. It should be noted that the CSA 1998 also requires all prisons, public or private, to be monitored by the Judicial Inspectorate which is an independent agency. The prison visitor from the Judicial Inspectorate for the private prisons will perform his or her normal monitoring duties, in addition to that provided by the appointed controller. Thus, it seems probable that appropriate monitoring, at least as it is designed in the legislation, will be in place for private prisons in South Africa.

Ikhwezi Bloemfontein Correctional Contracts

Once the enabling legislation was passed, DCS was able to begin negotiations for the first private prison contracts. The office which handled the contracts was the Asset Procurement and Operating Partnership System (APOPS). This government agency was set up in 1996 to oversee public-private partnerships.

The first private prison opened in Bloemfontein in July 2001, and the second will open in Louis Trichardt in early 2002. The first contract, with Ikhwezi Bloemfontein Correctional Contracts (BCC) is to design, finance, build, and manage a 3,024 bed facility in Grootvlei, Bloemfontein. The second contract is with South African Custodial Services (SACS), a consortium which includes Wackenhut Corrections Corporation. The sites were chosen because DCS already owned sufficient land in these two areas, as both are adjacent to existing prisons.

Because only the BCC prison was nearing completion at the time of the writing of this monograph, it is the one which will be examined in depth. BCC is a consortium made up of Fikile Projects cc, Ten Alliance Holdings, Group 4 Correctional Services, Murray and Roberts Construction and Bloemfontein Community Trust. Fikile Projects, Ten Alliance, and Bloemfontein Community Trust are black empowerment companies. Each company owns 20% of BCC, which is providing 10% of the funding required for the project. The remaining 90% is financed by Investec and ABSA banks.

Costs

The total cost of the BCC contract is R1,764,644,196 and the expected cost per ‘bed’ is R66.04 per day.4 The government claims that the BCC costs represent a saving of 5% on the DCS cost per prisoner per day. In truth, these cost comparisons are meaningless. Firstly, DCS determines its cost per prison per day by taking its annual budget and dividing it by the number of prisoners in custody. The annual DCS budget does not contain all the costs associated with running a prison. Most significantly, the cost of the facilities is included in the budget of the Department of Public Works. Also, the prison service is increasingly charged with incarceration of awaiting trial prisoners, who would more appropriately be included as responsibilities of the Department of Justice. Finally, the great majority of costs incurred by the DCS is for fixed overhead both in terms of personnel salaries and general sunk costs. These are costs which do not vary with the size of the prison population, and as such should be treated differently when making cost comparisons.

For example, the head of a prison will be paid a salary regardless of whether his prison has the designated 1,000 prisoners it was intended to accommodate or if it is at 200% capacity and crowded with 2,000 prisoners. The cost per prisoner for the head of prison’s salary will be cut in half when there are 2,000 prisoners as compared to when there are 1,000 prisoners. This does not represent an actual cost savings. The same holds true with all fixed costs in the DCS budget. They will be considerably less when divided by the number of prisoners because they are not increased by an increase in the prison population.

A better means of comparison can be found when one compares the variable costs. A variable cost is one which changes with volume. For example, the cost of issuing each prisoner a blanket will increase as more prisoners enter the prison system. If a blanket costs R10, then DCS will have to spend R10,000 on blankets at a prison with 1,000 prisoners. If the prison finds itself with 2,000 prisoners, then it will need 2,000 blankets and the costs increase to R20,000. This of course assumes that the prison is actually able to pay for and obtain sufficient blankets for the additional prisoners, but the concept remains true.

Assuming the total costs for providing correctional services could be captured, and not just the DCS annual budget, and that these costs could be divided into categories of fixed and variable costs, it is still unlikely that an appropriate cost comparison could be made. The BCC prison is designed for 3,024 prisoners. Westville Medium B has approximately 2,700 prisoners. The prisoners at BCC will be in cells designed for either two or four prisoners. The prisoners at Westville Medium B have an average of 48 prisoners per cell, most of which were designed for 18 inmates. At BCC, the prisoners will be unlocked from 7h30 each morning until 19h30 each night and will be fed three times a day. At Westville, the prisoners will be unlocked from 7h30 until 15h30, and are fed twice a day. Clearly, standard of care is not comparable between the two facilities and so cost comparisons become difficult.

It is impossible to ascertain whether private prisons in South Africa are cheaper than publicly run prisons because the standard of care offered by private prisons is entirely unmatched in the public sector. Perhaps the greatest contribution the BCC prison will make to correctional services in South Africa is that for the first time the government will learn exactly how much it costs to provide conditions of humane detention for prisoners.

Quality

The presence of a single employee at BCC makes South Africa’s first private prison an instant improvement on the public sector prisons: the Director of Inmate Development. Most of the prisons in South Africa have a staff of social workers and psychologists on hand to attend to the needs of inmates. And some prisons have explored constructive relationships with the National Institute for Crime Prevention and the Rehabilitation of Offenders (NICRO) so that rehabilitation and development programs can be introduced. But only the BCC prison has a full time management position created expressly for the purpose of inmate development. The inmate development program at BCC consists of four components:

Psychological/mental well being: services to be provided include counselling and support groups.



Social services: the services will include social workers who have been trained in therapy skills.



Religious care: the program has already recognized the need for 31 different denominations and will link and organise with other religious leaders in the community.



Education: the services to be provided at BCC make it truly exceptional, as it is registered as an adult education centre with a private adult training services provider. High school courses will be offered through to matric, and other courses offered will range from basic skills to financial management and entrepreneurial training. The vocational skills programs are all geared towards a goal of encouraging self-sufficiency so that upon release the former prisoner will not need to search for a job but will have the knowledge and ability to start his own enterprise.

Private prisons will necessarily be an improvement on public prisons because it would be almost impossible to perform any worse. It should be pointed out that many of the officials and employees of DCS are hardworking and devoted civil servants who deserve to be commended for their efforts in strenuous and stressful conditions. The impact of backlogs in the Department of Justice and a lack of resources have significantly contributed to the current state of the public prison system and these are factors which are beyond the control of even the most senior DCS official. That being said, the fact remains that conditions in South African prisons are deplorable and the private sector cannot help but be an improvement.

There is a risk that because the public sector cannot, at the moment, compete with the private in terms of provision of correctional services, the government will become dependent on private prison companies. This scenario is still a long way off, however, because private prisons are just beginning in South Africa. The possibility exists that eventually the number of private prisons will outnumber the public ones, and DCS will find itself at the mercy of private companies operating in facilities which the state neither owns nor can afford to purchase.

Monitoring

The CSA 1998 provides that a controller must be appointed and enumerates some of the general duties of the controller. The contract lists the specific performance standards required of BCC in a section entitled "Schedule D". The performance standards in Schedule D assist the controller with his monitoring duties but they are also used by an audit review team which will periodically visit the prison. Schedule D was adapted from the contracts used in the United Kingdom with moderate variation. It includes seven basic goals which the prison is expected to attain:

Keep prisoners in custody.

Maintain order, control, discipline, and a safe environment.

Provide decent conditions and meet prisoner needs.

Provide a structured day programme.

Prepare prisoners for their return to the community.

Delivery of prison services.

Community involvement.

It is difficult to ascertain whether these goals will be achieved and whether the monitoring mechanisms will be effective because the prison has only recently opened. However, the specificity of the contract is encouraging. Under each of the goals, as many as 30 separate points were listed as necessary activities. Some examples include the requirement that the contractor will provide postage for one letter per week for each prisoner, each prisoner will receive a free haircut every three weeks, and that there should be one newspaper per 15 prisoners. Some of these seem too specific to be effectively monitored, but their inclusion is encouraging proof that the government and the contractor are serious about taking the provision of correctional services to a level unseen in South Africa.

The security at the BCC prison is very specifically outlined in the contract, and is also necessarily improved from the public sector given the modern design of the prison itself. All the gates can be opened electronically, whereas most public prisons operate entirely with keys. The design of smaller cells with few prisoners in each one allows for much more effective monitoring of harmful situations, including bullying. Each cell at the BCC prison has a panic button which the prisoner can press if he is in need of assistance from the guard. This button turns on a red light which cannot be switched off without the guard actually venturing over to the cell to check out the problem. A computer automatically records the date and time when the button is pushed, and so incidents can be more easily verified. The BCC prison is designed and organized in such a way that it holds much promise for being considerably safer and more secure than the public prisons in South Africa.

The number one complaint amongst prisoners in South Africa is that the food is terrible. Not just the lack of nutritional content, but also the timing of meals is unsatisfactory as prisoners are fed only twice each day. In one prison, the management claimed that the prisoners are fed three meals, but two are served simultaneously.

In the BCC contract, the contractor is specifically required to serve three meals each day, one of which must be hot, and the time between each meal is not to be less than four and a half hours or more than six hours. The time between dinner and breakfast is not to exceed 14 hours, and 30 minutes will be provided to eat breakfast while 60 minutes will be provided for lunch and dinner. Food is never to be withheld as a disciplinary measure, and appropriate utensils and condiments must also be provided. This is an obvious improvement on the food service standards in public sector prisons, although the existence of the standards on paper is not as convincing as observing such standards in practice. Still, the fact that these standards are written down and signed by both parties in a binding legal document is extremely encouraging.

The enforcement of these performance standards will depend on the monitoring environment. If the controller appointed by DCS becomes co-opted into the prison management structure and total capture occurs then it is possible that none of the laudable standards in the contract will be implemented. Many of the factors which lead to capture are present:

The controller who has been appointed for the BCC prison was hired from the DCS, as was the director and his deputies.



The controller is given an office and a small support staff on the actual premises of the BCC prison. This may seem like a good way to ensure consistent compliance, however, it is also likely to lead to an informal and overly sympathetic relationship with BCC management and staff.



The controller will be in daily contact with BCC and will become extremely sensitive to the interests and operational challenges faced by the prison.

It remains to be seen, however, if the controller will place the interests of BCC management above his duty to ensure appropriate contract compliance and performance.

The contract

The contract between BCC and the South African government has not been made accessible to the public. The reasons given by government officials ranged from a claim that it was necessary to protect the commercially sensitive nature of the contract to the need to protect security information. One government official explained that it may become public information eventually but that would be decided in a committee meeting which had not yet taken place. Lengthy discussions and informal examination of Schedule D was permitted, but the remainder of the contract has still been kept secret. The South African Constitution, Article 32 (1) guarantees access to any information held by the state which is required for the protection of rights. The Promotion of Access to Information Act was passed in 2000 in order to give effect to this constitutional right. In the preamble, the Act recognises that:

"the system of government in South Africa before 27 April 1994, amongst others, resulted in a secretive and unresponsive culture in public and private bodies which often led to an abuse of power and human rights violations."5

The preamble concludes with a statement regarding the intentions of the Act, which includes a desire to "foster a culture of transparency and accountability in public and private bodies."6 In light of the constitutional rights of South African citizens and the intentions of the recently passed Act, the contracts with private prison companies should be released to the public without further delay.

Financing

The BCC contract is a lease-purchase arrangement, wherein the government pays rent for the use of the facilities. Once all the payments are completed, title for the facility will be turned over to the Department of Public Works. Most ‘design, construction, finance andmanagement’ (DCFM) projects involve a separate financing arrangement which extends for anywhere from 20 to 30 years. The management contract however is usually limited to five years. This is intended to give the government the often praised ‘flexibility’ of having the option to change contractors or discontinue private sector involvement entirely. However, this alleged benefit is entirely negated by the presence of long-term financing arrangements which involve the government almost inextricably with the private prison company. The contract with BCC does not even attempt the facade of flexibility as both the financing and management services are contracted for a full 25 years–the maximum allowed by the CSA 1998.

The problem with contracts which stretch out for this length of time is the loss of flexibility and leverage. The introduction of competition is intended to keep the private contractor on its toes and striving to innovate and provide better services at a better value. However, if the company is already guaranteed the customer for 25 years, there is no looming date that the contractor must compete with for renewal or re-bidding for the contract. Termination of the contract is also difficult, which reduces the leverage of the government should the contractor fail dismally to perform its requirements as outlined in the contract. If the government terminates the management portion of the contact, it will also terminate the financing. The full capital cost would then come due immediately, which clearly the DCS does not have the resources to meet. The only recourse open to DCS would be to take all its prisoners out of the facility and walk away, losing the costs already sunk into lease payments and stuck trying to find prison beds for 3,024 inmates.

Mr. T. E. Manchide, Director of APOPS at the Department of Public Works, pointed out that the private prisons are in fact public-private partnerships and not a privatisation of the prison service. "The government is not abandoning responsibility in terms of executing correctional services. We are not selling off government assets either," Manchide explained, "the contracts do not exist in perpetuity". This is true, but 25 years is sufficiently long that it carries with it many of the disadvantages of a full scale handover of the prison to the private sector. It is entirely up to the consortium whether it decides to meet its stated performance goals, as a failure to do so can only result in very limited government action. In effect, BCC can hold the Department of Correctional Services ransom for the full capital cost of the new prison in Bloemfontein. Since the South African government does not have sufficient funds to pay that ransom, the government–and the South African public–is at the mercy of BCC.

The public interest

There is no reason to suspect that BCC does not have the public interest in mind in carrying out its duties and fulfilling its contractual obligations. The BCC prison is already having a positive impact on the surrounding community through the employment opportunities it has created. The company was inundated with 26,000 applications for 400 job openings. The contract required that at least 5%, and eventually 10%, of employees must be recruited from historically disadvantaged groups. BCC not only met but exceeded this requirement as 94% of staff and 60% of management were drawn from these designated populations. The staff were deliberately not recruited from the prison service although a significant portion of management positions were filled by former DCS employees.

The management approach of BCC is very similar to that of Wolds in the UK. The philosophy is that each prisoner should be treated with respect and dignity, and this is considered the most effective means of maintaining an environment of respect and security. As each prisoner arrives at the prison, he is greeted individually in a reception room which resembles a doctor’s office more than a penitentiary. While his paperwork is completed, he is invited to be seated on a couch and offered coffee or tea. If a prisoner does not behave appropriately in this environment then he is detained in a solitary confinement cell.

One of the primary objectives of this entirely different approach is to build better morale amongst both prisoners and staff. This is another reason why the staff have not been recruited from DCS, because it is believed that it will be more difficult to make this attitude change among those who have already developed an attitude towards this kind of work. It could be argued that management is also hoping to hire staff who are not already affiliated with the Police and Prison Officers Civil Rights Union (POPCRU). However, both management and staff at BCC affirmed that any employee is permitted to join POPCRU or whatever union they may choose as this is their constitutional right.

Public health

The private prison companies who are pioneering the South African market are all from wealthy nations and will have to learn a great deal to adapt to the realities of South Africa. Group 4 has operated in many countries, including the UK, Canada, Australia, as well as several European countries. Although Group 4 is a publicly traded multinational enterprise, there are several aspects about doing business in South Africa which are unlike anything the company has faced before. The company appears to have anticipated and/or adapted well to many of the variations which are specific to the South African business environment. For example, BCC has successfully met or exceeded empowerment clauses and affirmative action requirements. But there remains one particular aspect of South African life, particularly prison life, which will pose special challenges to the BCC prison. This is the issue of public health in the prison system.

Public health is prison health, and this maxim is particularly true in South Africa. There are three illnesses which are rampant in South African prisons: tuberculosis (TB), syphilis, and HIV. Approximately 98% of prisoners will return to the community at some point, and anywhere from one third to one half of them will suffer from one of the illnesses listed above.

Any efforts to train and educate prisoners for their return to the community will be in vain if that prisoner is not diagnosed and treated for TB, and is not educated on the health risks of high risk sexual behaviour. TB is highly contagious and both cheaply and easily treated. The contract does not contain appropriate provisions for public health issues, save for one section which requires that communicable diseases are monitored and that HIV tests are only performed when ordered by a doctor. The prison environment affords an unparalleled intervention opportunity to reach those sections of the population which are both at highest risk and the most difficult to access. The costs of releasing prisoners who are sicker and at higher risk of transmitting HIV and/or TB are not borne by the consortium but by the impoverished communities to which prisoners return. Thus, it is not likely that a private prison company will emphasize public health issues unless specifically required to do so in the contract.