Prison Service, private sector, Contractual Management of Custodial Services, Privatizing Correctional Services, contractual management, United Kingdom, Correctional Services, performance, Wolds, prisons, public sector, private prisons, population, Home Office, Buckley Hall, Doncaster, London, Immigration Service, the contract, John Learmont, operational, medium security prison, Chief Inspector, contractual arrangements, operational costs, Sir John, maximum security prisons, Prison & Court Escort Services Ltd., Fazakerley Prison Services Ltd., Serco Ltd., HM Prison, HM Prison Service, Premier Prison Services Ltd., joint venture company, prison, the consortium, prison population, security prisons, Prison & Court Services Ltd., custodial services, the prison
Contractual Management of Custodial Services in the United Kingdom TIM WILSON This chapter reviews the provision of custodial services by the private sector
under the Conservative governments in power in the United Kingdom from 1979 to 1997. This account of government policy in the United Kingdom, since it was written in July 1996, could not anticipate government policy after the election of a Labour Government in May 1997; an indication of the approach to prison management that the new Government may take is given in appendix A. The development of contractual management--a term more appropriate than "privatization" for the approach taken in the United Kingdom--occurred in three phases: (1) the debate and passage of legislation (19871992); (2) the introduction of contractually managed prisons and court escort services (19921995); Notes will be found on page 90. 63
64 Privatizing Correctional Services (3) the letting of contracts under the Government's Private Finance Initiative (PFI) for the private design, construction, management, and financing of prisons (DCMF) (1995 1997). Background Her Majesty's Prison Service is the only penal service within England and Wales entrusted with criminal and the small number of civil prisoners. It is also responsible for some immigration detainees (others are held by contractors1 appointed directly by the Immigration Service or that service itself). There are separate prison services responsible for Scotland and Northern Ireland
but, while these have contracted some functional services such as catering and building maintenance, at the time of writing they have not awarded any contracts for the management of prisons. Population The population of England and Wales is 51.6 million. The prison population, after a period of rapid growth, exceeded 54,600 by July 1996. Of equal significance are the changes within the overall prisons population. Over the last 10 years, the proportion of sentenced prisoners convicted of violent or drugs-related offences has increased from 39 percent to 59 percent; the number of adults serving determinate sentences of four years or more has doubled from about 6,500 to 13,000; and the number of "lifers" has gone up by about a third to some 3,000. There are some 130 prisons and young-offender institutions, ranging in size from 60 places or so to well over 1000 places. These prisons and institutions are located in a variety of buildings, including former grand houses, military camps and, more commonly, Victorian gaols as well as the 22 new medium and high security prisons built since 1979. In the 1996/97 financial year, the Prison Service's estimated outturn for running costs is Ј1.4 billion (approximately $3 billion in Canadian currency). Capital expenditure will account for an additional Ј117 million (CDN$246 million). This will be funded from an overall budget for the criminal justice system
of Ј10.1 billion (CDN$21.2 billion). The average annual cost per prisoner was Ј24,266 (approximately CDN$55,800).
Contractual Management of Custodial Services 65 Characteristics of the Prison Service Nearly all medium-security, and all higher-security prisoners are held in single or double cells, which are designed to provide a considerable degree of privacy. Dormitories are being phased out even at low-security and open prisons. Staff are unarmed, relying instead upon a combination of physical barriers, operational procedures, and a high level of direct contact with prisoners. A modern English prison is surrounded by a concrete wall and an inner fence. It has only one point in the perimeter for entrances and exits. The cells are envelopes of concrete and steel, within larger secure envelopes, the houseblock or wings. Prisons are organized on managerial/supervisory lines: although not allowed by law to strike, staff are highly unionized and there is no semi-military rank structure or ethos. With a massive modernization and new construction program, "slopping out" (the use of a bucket instead of a lavatory and wash basin in the cell) ended in older prisons in 1995, but there remains a major legacy of older buildings that require major refurbishment or are expensive to run. A great deal of emphasis is placed on the time prisoners are unlocked--12 hours a day or more for 37.5 percent of the population in 1995/96--and upon purposeful activity (work, education, and training) amounting to 25.2 hours on average a week. Wages are always paid for work and participation in education courses. The rate is on average Ј7.00 per week (about CDN$15), but there are pilot schemes paying Ј23 or more per week. Whenever possible, prisoners are held close to home. We encourage frequent visitation; most take place as "contact visits"; i.e. visits where there is no barrier preventing physical contact between prisoner and visitor. We seek to counter the risk of drug abuse resulting from frequent contacts by increasing surveillance via closed-circuit television and the mandatory testing of inmates for drugs. At many establishments, prisoners--particularly those who are qualified for standard or higher privileges-- wear their own clothes. Rapid change in the Prison Service No longer a department of the Home Office, the Prison Service is now an agency with more powers devolved upon it and its director general now has a more public role. The Prison Service
66 Privatizing Correctional Services was formerly organized as a London-based headquarters dominated by administrators and directing "outstations" but its the ethos has now shifted to management through the operational line, with a major devolution of additional (non-custodial) management responsibilities to the governors of prisons. While 1995 and 1996 saw a sharply increased focus on security following two high-profile escapes from maximum security prisons, in fact during the last three years, escapes have been cut by 78 percent. Measures such as mandatory testing and more structured incentives schemes to encourage good behaviour are helping to create more drug-free, structured, and controlled regimes. Involvement of the private sector In 1987, following a visit to the United States, a House of Commons Select Committee recommended that, as an experiment, private-sector companies should tender for the construction and management of prisons, particularly remand centres for unsentenced prisoners. This was followed by a Green Paper (1988) intended to encourage informed discussion coupled with work by consultants on the option of private-sector involvement. The objective was: to improve the cost-effectiveness of the remand system by . . . making additional remand accommodation available more quickly and flexibly than would otherwise be possible . . . thus accelerating the . . . elimination of overcrowding and improving conditions for remand prisoners. (UK, Home Office 1988: 9, para. 51) This passage captures the essential link between greater cost-effectiveness and improved conditions during the development of contractual management. Having considered the responses to the Green Paper and the results of feasibility work by consultants, the Government announced on March 1, 1989 that contracting the construction and operation of remand centres was feasible in practice and acceptable in principle. It was also announced that the contracting-out of court-escort (i.e., transfer) duties should go ahead. This reflected earlier studies that identified a wasteful duplication of effort and inefficiency within the police and prison services. Further, because of the priority that had to be given to the courts, court-escort and custodial duties had to be undertaken frequently at the expense of maintaining the quality of prison regimes.
Contractual Management of Custodial Services 67 The Legislation The resulting legislation, the Criminal Justice Act 1991, provided the necessary legislative framework of safeguards, controls, and accountability. In addition to the public scrutiny that applies also to directly managed prisons--detailed external audit by Her Majesty's Chief Inspector of Prisons; regular visits by the local community watchdog body, the Board of Visitors; direct and unfettered access for prisoners to senior Prison Service management, the Prison Ombudsman, the courts (via a state-funded legal aid scheme) and to members of Parliament--there is a government official at a contractually managed prison 365 days a year. This official, the Controller (Prison Service governor grade) together with his deputy controller monitor contract compliance. Also, under statute, they alone, not the contractor's staff (including the Director who manages the prison), are responsible for: · authorizing temporary release from custody · conducting disciplinary adjudications against prisoners · authorizing the removal of prisoners from association in the interests of good order or self-protection and the use of restraints · investigating allegations against staff. The arrangements for court escort services are slightly different as they provide monitors, not controllers, as well as lay observers who have a role in respect of those in police custody similar to that of a Board of Visitors in the prisons. Custodial duties and powers at contracted prisons are entrusted to prisoner custody officers, who are vetted prior to employment by the Prison Service as well as by their employers, and trained to an approved syllabus. In 1994, the legislation was amended so that prisoner custody officers and contractually managed prisons can be fully integrated in the Prison Service's mutual aid arrangements to deal with serious incidents. Thus, assistance will come from the nearest prison irrespective of who manages it and of what prison requires assistance. The director of a contractually managed prison (equivalent of the prison governor or warden) must be a Prisoner Custody Officer (PCO) specially approved by the Home Secretary. Finally, the Prison Service is empowered, should it appear to the Home Secretary
68 Privatizing Correctional Services that the director has lost, or is likely to lose, effective control of the prison or any part of it, to take over the facility in order to protect life or to prevent serious damage to property. The last point to be made about this legislation is that, when originally presented to Parliament, contractualization was restricted to escort services and the custody of remand prisoners. A power was added to the Bill to extend contractualization to the management of new and existing prisons holding sentenced prisoners. Current contracts The award of contracts has been determined largely by the pace of new construction and an assessment of bids in terms of three criteria, price, deliverability, and quality. The three most recently constructed secure prisons were reserved for contractual management and no in-house bids were allowed, with the following results: (1) The Wolds Remand Prison (designed normally to hold 320 remand prisoners) was awarded to Group 4 Prison and Court Services Ltd., a subsidiary of Group 4 International Corrections BV, itself a wholly owned division of Group 4 Securities (International) BV. Wolds Remand Prison opened in April 1992. Group 4 had never run a prison but have a 90year record in the private security industry and already held important government contracts. In addition to recruiting some senior staff from the Prison Service, they employed Carter Goble Associates to assist them on prison planning and operational programs. (2) Blakenhurst Prison (designed for a mixed population of 649 male prisoners) opened in May 1993 and is managed by United Kingdom Detention Services (UKDS), a consortium owned initially by two British construction companies, Sir Robert McAlpine & Sons and John Mowlems & Co., and the Corrections Corporation of America (Mowlems are no longer members of this consortium). (3) Doncaster Prison (mixed population of 771 male prisoners) opened in June 1994, run by Premier Prison Services Ltd., which is jointly owned by Serco Ltd. and Wackenhut Corrections Corporation.
Contractual Management of Custodial Services 69 The Prison Service competed against contractors, however, for Manchester Prison--virtually rebuilt after a major riot in 1990-- and the Prison Service's first "fast-build" prison, Buckley Hall-- constructed on the site of a former young offender institution. (1) Manchester Prison was awarded to an in-house bidding team and its service-level agreement took effect in March 1994, with a 6.8 percent saving in running costs and with major improvements in the quality of service. For example, the present daily time out of cell of 11 hours is almost equivalent to the weekly hours out of cell before the riot. (2) Buckley Hall Prison (a 350-place medium-security training prison) was awarded to Group 4 and opened initially with inmates in temporary accommodation for a maximum of 90 inmates. The site became fully operational in July 1995. Against the background of this steady increase in contractually managed capacity (4 percent2 of the total population in May 1996) Ministers and the Prison Service took two key decisions that have shaped a partnership between the Prison Service and private operators. When the Prison Service became an executive agency in April 1993, it was decided that the Contracts and Competition Group (under an earlier name), which had previously reported directly to the Permanent Secretary (the most senior official) in the Home office and not to the Director General, should become part of the Prison Service. Eventually this allowed the contractual prisons to become integrated with the Prison Service's operational line so that, after the procurement phase, the Area Manager (who is responsible for Prison Service establishments in the prison's vicinity), rather than the Contracts and Competition Group, became the contractor's client. This has undoubtedly contributed greatly to acceptance at a working level within the Prison Service of the need to co-operate on a daily basis with contractual prisons. A common identity within an area has encouraged recognition of, and cooperation with, directors and prisoner custody officers as fellow professionals. At the same time, a target for the number of contractually managed prisons in the initial phase of development was set. In September 1993, the Home Secretary announced that some 10 percent of establishments (13 in all) would be so managed.
70 Privatizing Correctional Services This figure was to include the three new prisons I have already mentioned, six newly constructed private DCMF prisons and a remaining four, which were to be contractualized if, after market testing, the in-house bid team lost. In the event, Lowdham Grange was added to the DCMF sites and, after the Manchester and Buckley Hall market tests, the market-testing program was temporarily halted. Initially this was because of a procedural challenge by the Prison Officers' trade union, the Prison Officer's Association (POA), but, as the director general explained to the Home Affairs Select Committee on June 5, 1996, against a rapidly rising population, the future of the program is being reconsidered because of the likely need to reduce numbers temporarily for decanting when an establishment is transferred to contractual management. The 10 percent target was regarded by some as too small to ensure a competitive market, with each participant having a sufficiently large market share to be truly efficient and to provide career opportunities for their staff. In the event, the Government had by then announced in its White Paper, "Protecting the Public" (April 1996), that 12 additional DCMF prisons would be required by 2011/12. After allowing for the existing contracted prisons and the outstanding DCMF schemes, this will increase to 15 percent at least the contractually managed share of the estate. In parallel with the prison program, court-escort work has been contracted out. The first escort service began in April 1993 in the East Midlands and in Humberside. The contract was awarded to Group 4, who also commenced management of the Wolds Prison, located in the same area, in that month. Since then, the following contracts have come into operation: · London, June 1994 (Securicor Custodial Services Ltd., a part of the Securicor Group) · East Anglia, together with Manchester, Merseyside and North Wales, July/August 1995 (Group 4 Prison and Court Services Ltd.) · South West England and South Wales
, April 1996 (Reliance Custodial Services Ltd., a part of the Reliance Group) · South and North East England, together with West Midlands and Central Wales, April 1996 (Premier Prison Services Ltd.)
Contractual Management of Custodial Services 71 What kinds of success? Results of privatization of prisons Her Majesty's Prison Service was able to report in its Corporate Plan for 1996-99: The Prison Service has achieved a great deal to improve efficiency since it became an agency. The work of escorting prisoners to and from court has now been contracted out to the private sector in most of the country (7 of the 8 court escort areas), at an estimated cost saving in 199699 of around Ј13 million, improving performance and reducing disruption to the operation of prisons. There are now 4 prisons run by the private sector with plans for the private sector to construct, finance and run more over the next three years . . . Work carried out to date suggests that, on 1993/94 data, private sector establishments operate at a cost of 15 percent25 percent lower than comparable public sector establishments, while maintaining equal or better levels of performance . . . the evidence currently available suggests that it is possible to run prisons safely and effectively at a substantially lower cost than the public sector now does. What independent evidence is there to substantiate those statements? And do they represent the full picture? It has sometimes been suggested by critics of contractually managed prisons that, unlike directly managed prisons, they are deliberately not overcrowded and the population is carefully selected to make management easy. Both are incorrect. Until 1995, Prison Service policy was not to overcrowd any new accommodation. In the face of a rapidly rising number of prisoners, this was relaxed with the exception of training prisons--that is prisons, such as Buckley Hall, for sentenced prisoners only, where currently there is no overcrowding, irrespective of date of construction. As a result, during 1996 Wolds was overcrowded by 6 percent, Blakenhurst by 10 percent, and Doncaster had a population 13 percent in excess of its designed capacity. If we consider the population mix in 1994/95, it is clear that none of the contracted prisons has had an easy ride. The Wolds opened as the only prison solely for remand prisoners. The availability to contractual staff only of weak sanctions to ensure cooperation and work coupled with the scant provision for purpose-
72 Privatizing Correctional Services ful activity in the original design was criticized by HM Chief Inspector. The Prison Service subsequently provided the capital investment to create workshop places and changed the role of the Wolds prison so that it also took sentenced prisoners. Blakenhurst operated with a significantly lower proportion (50 percent) of more stable inmates (i.e., sentenced adults) compared with an average of 63 percent at comparable prisons of the same design. Moreover, there was an even greater difference in the proportion of the population who belonged to the medium security group (category C prisoners): 15 percent at Wolds compared with 23 percent at similar prisons. Doncaster's population was composed of only 19 percent sentenced adults (comparator group average is 63 percent) but 37 percent unsentenced young offenders (comparator average is nil). There had been a major disturbance when the Prison Service had earlier opened Moorland Prison nearby with a similar young offender population. During 1994/95, there were about twice as many recorded assaults at contractually managed prisons than at comparable directly managed prisons. This is a problem at all newly opened prisons, although the population profiles described above had an impact and so may have the above-average number of hours out of cell required under the contracts. In all cases, the numbers of assaults have reduced significantly since the first year of operation. HM Chief Inspector of Prisons inspected the Wolds 13 months after it opened (May 1993). He was concerned about the lack of constructive activity, the use of drugs, and the need to integrate Wolds into the mainstream Prison Service area management. These concerns were all addressed in 1993/94. For the most part, however, his report was highly positive, acknowledging good relationships between staff and prisoners, the generous time out of cell, civilized mealtimes compared with the rest of the estate, good-quality food and better opportunities for visits. In March 1994, the National Audit Office, which reports directly to Parliament, examined the procurement process and contract compliance for Wolds. It concluded that overall there had "been significant successes in the placing and operation of the contract" (UK, National Audit Office 1994: 2). But it called for better comparisons with performance, results, and costs in publicsector prisons, suggesting that the poor statistics could be attributed, in part, to the remand-only population in Wolds and its
Contractual Management of Custodial Services 73 separation from the Prison Service's area-management structure. Blakenhurst too was inspected barely a year after it opened (May 1994). While the Chief Inspector identified areas needing further development--particularly, systems for controlling inmate behaviour and more work to tackle re-offending--the overall conclusion was that prisoners were held securely in good modern conditions and were treated humanely. The enthusiasm and potential of staff was seen as impressive; the education and training offered to prisoners, reception into custody, and visitation arrangements were all singled out for praise. More recently, in March 1996, Doncaster was inspected by a newly appointed Chief Inspector (UK, Home Office 1996). In the light of what he saw and aware that some parts of the media had given intense coverage to Doncaster's initial problems, the Chief Inspector set aside usual procedure and called a press conference
to put this into context. The national press reported his description of Doncaster as "showing an example which the rest of the Prison Service could follow." There are also now the results of an academic research project at the Wolds (Bottomley, James, Clare and Liebling 1997). The findings relate primarily to the first two years of the prison's operation (the main field work was carried out from November 1992 to October 1993, with follow-up visits until March 1994) when the Wolds catered solely for remand prisoners and before its integration within the Prison Service area structure. Two important issues were highlighted in the report; both have been encountered not just at the Wolds and the second, not just in contracted prisons. (1) Wolds opened inspired by the American model of direct supervision. It was successful in certain key respects, most notably in the quality of relationships between staff and prisoners, but this also led to difficulties because of differences in British and American prison design, regime, and legislative framework: · the lack of effective sanctions for contractor's staff for dealing with disruptive prisoners · an inability to operate each living block autonomously (e.g., there were no separate exercise and educational facilities for each block).
74 Privatizing Correctional Services (2) Most staff were recruited locally. This had a tremendous advantage in the creation of a new ethos for better relations between staff and prisoners and greatly improved conditions for remand prisoners, but the lack of experience created a major burden for management because of the degree and amount of initial support and guidance needed by basic-grade staff. The inmates, of course, are well seasoned. There is a general question for penal administrators about how to achieve the right balance of control between staff and inmate when opening a new prison, particularly when the emphasis is on more time out of cell and greater respect for prisoners as individuals. The Chief Inspector commented on this in a report on a directly managed prison of a similar vintage to the contracted prisons (UK, Home Office 1995a). It had 635 prisoners (a significant proportion of whom were sentenced), had been built to the same design as Blakenhurst and had a regime with 101/2 hours out of cell. The Chief Inspector wrote: "inevitably, with little to occupy them during that time, prisoners had taken advantage of the largely inexperienced staff to shift the balance of control in their favour. The result had been a dirty prison with a good deal of bullying and other unacceptable activities" (18). He concluded "inmates had been allowed too liberal a regime at first, but staff were now regaining control" (25). Thus, contractors and the Prison Service itself are experiencing and overcoming similar problems. This point is emphasized because media coverage of prisons tends to concentrate on the opening phase. The overall research findings
of Bottomley et al. (1997) were that · the contractor had run this new prison for its first two years with relatively few major incidents, none of which involved loss of control of the prison · while the only sample was small and over a short period of time, nearly 80 percent of prisoners at Wolds regarded it as better than other prisons that they had experienced and, although a small minority felt unprotected and vulnerable, the majority supported the contracting-out of more prisons, insofar as this would lead to improved conditions for prisoners. The researchers question the extent to which general conclusions could be drawn from their study. They consider that com-
Contractual Management of Custodial Services 75 parable achievements were to be found in some public-sector local prisons and believed that there is no evidence, therefore, that Wolds's achievements were necessarily related to its contracted-out status. Moreover, they suggest that although staffing levels did not lead directly to any serious compromises in the security of the prison, "there were situations in which higher levels of staffing might have improved the maintenance of order and helped to reduce staff concerns about their potential vulnerability" (Bottomley et al. 1997: 56). This study was concerned exclusively with the Wolds during the phase when it housed only remand prisoners. The change in population saw the introduction in parallel across the Prison Service of an emphasis on greater control, coupled at the Wolds with a more structured regime, the installation of zonal fencing to restrict movement outside houseblocks, the creation of workshops (previously there were none), and an increase in staff. The researchers also acknowledge that the achievements at the Wolds had taken place in the face of considerable hostility from many different quarters, not least from the media. In the meantime, Doncaster's security procedures were audited in June 1995 on the same basis as directly managed prisons and it was then one of only two local prisons to be given a superior rating. The final independent testimony comes from General Sir John Learmont's report, Review of Prison Service Security in England and Wales (UK, Home Office 1995b). His investigation followed the escape of three prisoners from the maximum-security prison at Parkhurst in January 1995. Having visited three contractually managed prisons, he and his team, which included Prison Service governors and senior policemen, were clearly impressed: "conditions were clean, the chain of command in each was taut . . . The management obviously saw its role as to manage; and the value of the front line operational staff was recognized through higher pay for those who worked in daily contact with the inmates" (para. 3.138). The key was seen to be clear and effective management
accountability and compliance audit: The Directors of the contracted out prisons are accountable to their parent companies. Delivery of performance, measured against the specification in the contract, is monitored by a Controller, who is an experienced Governor, with a small support staff provided by the Prison Service and located at
76 Privatizing Correctional Services the prison. The Controller is also required to carry out all adjudications on prisoners and to advise the Area Manager on all matters regarding the running of the prison. The Area Manager is expected to ensure compliance with the contract, to identify any variations necessary and generally to be satisfied with the way the prison is working. Such arrangements raise the question as to whether the problems at Whitemoor and Parkhurst3 would ever have arisen if the Prison Service had the same monitoring and audit arrangements which are compulsory for the private sector. Through this mechanism, performance standards in private prisons are kept under review, with the result they are maintained at a high level. In comparison with public sector prisons, private prisons demonstrate the advantages of a tauter management chain, greater freedom for Governors to manage and continuous audit on site. (UK, Home Office 1995b: para 3.139) Sir John Learmont had one major reservation: that minimum statutory staffing levels were not included in any of the contracts. The recommendation in favour of minimum staffing levels was considered carefully. On the one hand, as Sir John recognized, "potential involvement of the private sector and the prospect of competition promotes internal re-examination. It stimulates the adoption of modern best practice and business methods. Experience has also shown that competition is likely to cut the cost of the services and increase their efficiency" (UK, Home Office 1995b: para 3.137). Nevertheless, Sir John commented that with the wrong contractor or contractual arrangements there is the risk "that pursuit of cost control
, either to maximize profit or to win a market test, could lead to ill-judged economies that threaten physical or dynamic security" (para 3.137). Having weighed these issues, the Prison Service decided not to set minimum staffing levels. Instead it was decided to continue, for bid evaluation purposes, to analyze the staffing levels in great detail, including a 15 minute breakdown throughout the time that prisoners are unlocked and the distribution of staff in cell blocks. In addition, indicative staffing levels are included in the contract as an audit aide for those who will monitor the delivery of the service. While clearly impressed by what he saw at contractually managed prisons, Sir John Learmont concluded by recommending
Contractual Management of Custodial Services 77 that a substantial public-sector penal service should be retained: "it provides a safeguard against private sector bankruptcies, withdrawals or monopolies. However, there were lessons to be learned from the private prisons which should be applied across the Service" (UK, Home Office 1995b: para 3.141). The next item of research to be published, (13 June 1996), was a cost analysis for 1994/95 by the consultants, Coopers and Lybrand (1996). The results, based on a more vigorous analysis than had been achieved before, confirmed that there are significant savings on operating costs, in the order of 13 percent to 22 percent relative to the average costs of a group of comparable prisons. In addition, on average, contractually managed prisons perform well or better on escapes or visiting entitlement, the number of hours during which prisoners are unlocked is higher and, when allowance is made for remand prisoners, who do not have to engage in work, the average weekly hours of purposeful activity are also higher. On the debit side, the study indicated that assaults were also higher than at comparable public prisons. Results of privatization of escort services The escorts contracts also contribute significantly to an understanding of what can be achieved through private sector involvement. In addition to savings that will rise to approximately Ј90 million (CDN$189 million) by 1999/2000, when this work is fully contracted in England and Wales. The original escort contractor (Group 4 in the Humberside and East Midlands escort area) has cut escapes from custody over three years by 82 percent compared with the service provided jointly by HM Prison Service and the police. In addition, a customer-satisfaction survey seeking the views of police, court, and Prison Service staff was undertaken in this area in 1994. It showed that 86 per cent of those who responded were either satisfied or very satisfied with the service provided by contracted-out escort services. A further survey was conducted in both East Midlands and London in 1995. This showed an improvement in the level of satisfaction with Group 4's performance to 95 per cent, a level that was also achieved by Securicor in London. Above all, this policy means that highly trained police officers
are spending more time on patrol, responding to calls for assistance, and seeking to clear up crime, rather than on ferrying prisoners from one location to another.
78 Privatizing Correctional Services The evidence reviewed so far indicates why HM Prison Service indicated in its Corporate Plan for 1996 to 1999 that contractualization is seen as a spur to better performance and reorganization within directly managed prisons. Within our own custodial facilities "the emphasis has to be on reducing costs; by finding more flexible ways of working, with reductions in service as the last resort. The Service is also developing imaginative new ways of working with the private sector, as part of the Private Finance Initiative" (UK, HM Prisons Service 1995: para 4.6) The nature of future contracts for private prisons The program to design, construct, manage and finance prisons by private means (DCMF) is a new chapter in the development of contracted management, undertaken as part of the Government's wider Private Finance Initiative (PFI). The initial phase consists of: · Bridgend, an 800-place, high and medium security prison for remand and sentenced prisoners, including young offenders for which a contract has been awarded to a consortium of Securicor, Sieferts and W.S. Atkins4 with Costains and Skanska as construction partners · Fazakerley, a 600-place, high and medium security prison for remand and sentenced prisoners including young offenders for which a contract has been awarded to a consortium of Group 4 and Tarmac5 · Lowdham Grange, a 500-place, high security prison for sentenced adult prisoners for which a contract is expected to be let in the autumn of 1997. All three should be operating at the maximum capacity for which they were designed by early in 1998/99. The third project, Lowdham Grange, will achieve this because, first, by learning from the Bridgend and Fazakerley competition we are able to use bankable contractual terms and, second, while not limiting innovation in design, construction, and operation, we have adopted a procurement strategy driven by fast-track and lower cost construction techniques developed within the Prison Service. The duration of each contract will be 25 years from the time that each prison becomes operational; at the end of this term, the property will revert to the Crown.
Contractual Management of Custodial Services 79 In accordance with the principles of PFI, the procurement of the custodial services was based mainly upon an output specification, setting out what needed to be delivered, but not how. (The main exceptions to the purist PFI approach were mandatory security processes.) Through a combination of payment mechanisms and specific contract conditions, there is substantial transfer of risk to the private sector, both through the design, planning, and construction phases and during operation, the degree of risk being based on which party is best able to bear it and at a price that represents value for money. The contracts are placed with special project companies, a procedure that ensures a similar transfer of risk within that consortium to the member best able to bear it. For example, the building contractor may work to a guaranteed maximum price for the design and built contract and also take completion risks. The operator similarly must ensure that performance meets the contractual requirements and does not result in deductions from the revenue stream. The project company and its equity takes an overall risk on the assumptions about revenue and the placement of risks within the consortium. The main features of the contracts negotiated are the transfer of key risks through the structure of the payment streams. Availability The contractor is paid a daily rate for prison places made available, hence no payment is made until the prison is up and running. Performance Building on our experience of management contracts, performance is monitored by a range of indicators for effective prison management, including escapes and provision of out-of-cell activities for prisoners. Payments to the prison operator will depend on performance; persistent poor performance can result in termination. Volume Volume risk lies with the public sector, not the private sector. There are volume-related payments if the prison operators are required to take in prisoners beyond the contracted numbers. This payment reflects the extra costs involved and was fixed in the competitive bidding process. Other major areas of risk transferred under the contracts are:
80 Privatizing Correctional Services (1) Responsibility for obtaining detailed planning permission. (2) The risk of construction-cost and time overruns must be controlled by the contractor because of their effects upon, respectively, the payment and the revenue streams. The previous program for the construction of new public-sector prisons was marked by an average 18 percent cost overruns at the post-tender stage and significant delays. No allowance is made in the price the Prison Service pays for building-cost overruns, and there is no payment until the service is available. Additionally, delayed opening results in liquidated damages that reflect the shortage of prison accommodation and the high cost of keeping prisoners in police cells. (3) Responsibility for maintenance backed up by regular dilapidation surveys throughout the contract and prior to contract end. (4) Responsibility for changes in general legislation, rather than that specific to custodial services. Due to the contract length and high proportion of labour costs relative to the contract value, there is a five-yearly price review mechanism to cover external factors that have significantly increased or decreased costs for all prison operators. The contract also encourages continued technical innovation, with the sharing of financial benefits between the contractor and the Prison Service. If the contractor materially defaults on its obligations, or becomes insolvent, the Prison Service may terminate the contract. Where there is outstanding debt, however, the consortium's lenders have the right to provide an alternative operator. If the consortium's lenders fail to do so, the Prison Service has the right to step in to support the continued safe operation of the prison itself at cost plus 5 percent or terminate and pay off the lower of outstanding debt and the net present value of remaining contract. After five years, the contract may be terminated voluntarily, by paying the consortium its outstanding debt and fair market value for its equity. But the overall thrust in the contract is, first, one of remedy or replacement of a faulty element rather than termination and, second, flexibility in providing a framework for a contractual service over 25 years by managing the introduction of changes over that period.
Contractual Management of Custodial Services 81 It was estimated that the aggregate net present values of the custodial services represent a saving in excess of 10 percent over the contract life, compared to risk-adjusted public sector comparators. As with previous procurement exercises, the bids for Bridgend and Fazakerley were assessed on the basis of deliverability, price, and quality, but with the addition of extra credit for innovation. A weighting was applied to both the quality and innovation scoring: keeping prisoners in custody (20 percent), maintaining order and control (20 percent), providing decent conditions and meeting prisoners' needs (15 percent), providing adequate management systems (15 percent), providing a positive regime to address prisoners' offences (12 percent), helping prisoners prepare for their return to community (12 percent), and demonstrating sensitivity to the local environment (6 percent). This criteria was made available to bidders in the invitation to tender (ITT) to ensure that they were aware of how the Prison Service would be assessing their bids. The pricing mechanism is as follows. Once the prison has been opened, a place must be available for a prisoner before there is an obligation to pay for it, regardless of whether or not it is filled. A place is a series of functions, not just a cell; for example, there must be adequate staffing levels, health care
and food. To be available at all, a place must satisfy certain requirements set out in the contract. Also, the full price per place is subject to reductions. · If the amount of doubling (two prisoners sharing a cell) exceeds a permitted level for more than two days, the price for each doubled place will be reduced by up to 5 percent. The permitted level of doubling allows the consortium to manage the cells for maintenance but discourages undesirable levels of overcrowding. · If the regime in the prison fails to meet certain standards, the amount payable will also be reduced by up to a maximum of 5 percent. There is also a mechanism that will increase prices. Two of the component parts of the price payable will be indexed; the third--notionally for project costs--is not. The rates of indexation are the Retail Price Index (RPI) and RPI plus 2 percent;
82 Privatizing Correctional Services these rates were set in competition. The RPI plus 2 percent relates to the labour element of the operational costs and RPI relates to the rest. Irrespective of indexation and performance, there are five ways in which the price payable can be varied: · if a change in the way the prison is built or operated is required; · if legislation is introduced that affects prisons directly and has an effect on costs; · if legislation comes before Parliament relating to Health and Safety issues that would have a cost impact on the prison, the consortium can raise this as a reason for an increase in price, but the Prison Service then has the opportunity to seek off-setting cost reductions on other grounds; · where there is a change in the manner in which security and control are maintained using new technology, the Prison Service is entitled to a share in any savings; (For the points above the parties must attempt to agree any variation, but if they fail, the matter goes to binding arbitration.) · where there has been a change in the consortium's costs of more than 5 percent and beyond its control, or a change in costs generally affecting prisons that is not accurately reflected in the indexation provisions, the consortium may seek on the fifth anniversary of the opening date and on every fifth year thereafter a change in price such that it is able to obtain the level of return expected in its original finance plan. In the latter case, if the parties cannot agree upon a variation, the matter is taken to an expert. The consortium must accept the expert's opinion if HM Prison Service does, otherwise the contract is terminated for default. If, however, HM Prison Service rejects the decision and the consortium accepts it, the contract will terminate and HM Prison Service must pay the consortium a higher level of compensation based on the net present value. The final form of contract will reflect more general commercial issues and some aspects will change as the market develops. Due to the undeveloped insurance market for prison operation and as a result of the contract negotiations, it was decided that
Contractual Management of Custodial Services 83 the Prison Service would insure Bridgend and Fazakerley prisons in respect of malicious risks, if such insurance was not generally available at a commercial price in the United Kingdom. In these circumstances, the Prison Service would benefit from a fee reduction. This arrangement strikes a balance that gives valid protection to the private sector operator in circumstances where the general market cover for malicious damage risk disappears, while preserving the principles of risk allocation. If, for example, a particular operator is unable to obtain insurance due to doubts over its own performance, then the risk in these circumstances must be borne by the project sponsors and the Government will not provide cover. In the ITT for Lowdham Grange, however, the Prison Service did not offer to provide such cover. As the market develops, issues of this kind need to be retested in competition. The strength of interest commercially is indicated by the fact that, when the financing of one of those prisons was recently syndicated, three British banks were joined by three French, two German and two Dutch banks. As mentioned above, in accordance with the principles of PFI, the procurement of the custodial services was based mainly upon an output specification, setting out what needed to be delivered, but not how. While it is imperative that the statement of requirements is output orientated in order to capture innovation, where there are clear operational or design constraints, these need to be made clear. These may be overriding policy requirements (e.g., 24-hour access to sanitation) or essential operational requirements (e.g., solid perimeter barrier for high security prisoners and not a chain link fence). The key issue for the procurement organization is to decide what is a core requirement or essential process and what is not before the tender stage. Equally, there is a danger that output specifications can be too broad, providing no indication of a minimum threshold for service delivery. From the foregoing, you will appreciate the complexity of the DCMF contracts; such contracts take time to procure. The average time taken within the public sector to design and construct new prisons has been four years and, before 1991/92, the average time was seven years. Bridgend and Fazakerley prisons will come into operation within three and a half years from the date of the initial ITT. For Lowdham Grange, the whole process--procurement, design and, construction--will take just over two years.
84 Privatizing Correctional Services The principal advantages of the DCMF approach First, prisoner places will be delivered on time and within parameters set by the pricing mechanism, as the contract is all-embracing and thus more binding than a traditional procurement. Second, higher quality, more efficient service will be delivered by harnessing the synergy achieved through operator-lED Design
. Third, innovative design and operational features, particularly those based on new technology, can be assessed with a view to its introduction in directly managed prisons earlier than would otherwise be possible and at private-sector risk. As with management contracts, there will also be benefits from innovation that may be transferred across HM Prison Service as a whole. The aim is that contractors should remain under pressure to improve their effectiveness in order to continue to compete against an increasingly effective public sector. General lessons Are there any general lessons from the British experience? It is necessary to be cautious about applying the practices of one jurisdiction to others but, in the course of this review, a template for contractual management has emerged. It includes: · a clear and comprehensive statutory framework with a precise and politically acceptable allocation of duties in terms of compliance monitoring, disciplinary adjudication, and emergency intervention · clarity about services to be purchased and the criteria against which bids will be assessed · a willingness to look abroad while ensuring that contractors bringing in foreign expertise take due account of the ingrained penal culture that will determine inmate behaviour and expectations · an ability to listen to the market as, ultimately, commercial terms can be no better than what the market at any time will bear; if they are not likely to be acceptable to the contracting authority it should not, after initial feasibility work, run a procurement competition · the development of a contractual sector at a pace that allows both the penal authority and contractors to learn from ex-
Contractual Management of Custodial Services 85 perience, ideally beginning with contracts of shorter duration and covering only the less complex task of managing prisons and only later moving to longer, more multidimensional DCMF-type contracts, particularly if capital is to be sought from the private sector · a willingness to seek useful innovation and the ability to recognize it when it is offered6 · above all, a state of mind that seeks to work in partnership with private contractors and to use what is learned from private contractors to improve the public sector's own operation. This is essential if contractual management is to work to the full benefit of taxpayer, state penal service and prisoner. Appendix A Extract from House of Commons Official Report (Hansard) for June 19, 1997 Prison Service (Private Sector Involvement) Mr. Patrick Hall: To ask the Secretary of State for the Home Department [Rt Hon Jack Straw MP] what conclusions he has reached about future private sector involvement in the prison service. Mr. Straw: The Government have expressed reservations about the principle of contracting out the management of prisons. It is generally accepted that responsibility for the incarceration of offenders must remain with the state. The issue is whether that responsibility should, as a matter of principle, be discharged through direct management in the public sector or whether it can properly and effectively be discharged under a regulatory framework. This was addressed in the recent report [UK, House of Commons 1996] on prison management by the Home Affairs Select Committee; and the Government now wish to give further and careful consideration to the arguments deployed in that report before settling their overall approach to this issue.
86 Privatizing Correctional Services Existing management contracts will be honoured, as I have already made clear. But we are considering ways in which the current regulatory framework might be strengthened. At present, all disciplinary hearings in contractually managed prisons are conducted by a state official--the Controller, a Prison Service governor, who also monitors the performance of the contractors on behalf of the responsible Prison Service Area Manager. This is an important principle from which we will not depart; and we are currently reviewing the scope for extending the Controller's powers in respect of prisoners' requests and complaints, sentence calculation, scrutiny of security classifications and frequency of security audits. On the letting of new contracts and the renewal of existing contracts, our immediate approach is necessarily determined by current expenditure plans, to which we are committed, and by operational requirements. We have inherited a prison population rising sharply (well above projections), with expenditure provision for new prisons limited to establishments to be designed, constructed, managed and financed by the private sector. Even with the new prison ship, Her Majesty's Prison Weare, there is huge pressure on capacity. We have therefore decided to proceed with existing procurement plans to provide additional new places on this basis by 1999/ 2000. This is an urgent operational requirement to overcome projected shortfalls in accommodation and to avoid dangerous levels of overcrowding and the use of police cells. we will shortly be launching competitions to provide an 800 place Category B local prison at Agecroft, Salford, and a 400 place Young Offenders Institution at Pucklechurch, near Bristol. I have also authorised the renewal of the contract for the operation of Her Majesty's Prison Blakenhurst with the private sector operator, United Kingdom Detention Services. The existing contract expires on 25 May 1998; and we are required to give the operator one year's notice of a decision to renew the contract. The prison has been performing well against the current contract requirements and in comparison with comparable prisons in the public sector; and I am satisfied that the terms that have been negotiated for renewal of the contract offer value for money. The establishment could not be returned to public sector control
Contractual Management of Custodial Services 87 without incurring additional expenditure, which would require offsetting savings elsewhere, and in the light of the current budgetary pressures on the Prison Service and pending our consideration of the Select Committee report, I am persuaded that no other course of action could be justified in current circumstances. However, in order to ensure that future options are kept as open as possible while our overall policy is under review, I have decided to renew the contract for 3 years only, the minimum practical--and contractually permissible--period. Moreover, before further decisions are made on prison procurement beyond the extra places within existing plans or on the renewal of any other existing contracts, I have asked the Prison Service to: (a) explore all possible methods of using private finance on terms which offer value for money including design, build and maintain new prisons, with the public sector providing custodial services; and (b) to seek to demonstrate that the public sector Prison Service has the capacity to match the performance of the private sector in comparable prisons, and that realistic plans can be developed to establish a case for returning contractually managed prisons to the public sector on value for money grounds when consideration is given to the issue on the expiry of contracts. I have asked senior Prison Service manages to engage in consultations with the Prison Service unions to this end. Proposals for private sector involvement in other areas of the Prison Service, which do not raise the same issues of principle, will be treated on their merits. I have authorised the extension of the contract with Group 4 to operate the court escort and custody services in Area 7. This is essentially a specialist service, distinguishable from the management of prisons; and I am satisfied both with the quality of the service being provided and with the terms negotiated for extension of the contract.
88 Privatizing Correctional Services ? Appendix B Private adult correctional facilities in the United Kingdom
Group 4 Prison & Court Escort Services Ltd.
Group 4 Prison & Court Services Ltd.
Premier Prison Services Ltd. Corp. (a Wackenhut Correc- tions Corp. joint venture company with Serco Ltd.)
Name of facility
HM Prison Wolds HM Prison Buckley Hall
HM Prison Doncaster
Primary source of prisoners
HM Prison Service HM Prison Service HM Prison Service
Type of prison
category B local
cat. C (sentenced category B local (alprisoners only) so young offenders)
Designed number 320
Av. num. of prisoners 326 (April 1996)
all (except maximum)
341 medium public
841 all public
Prisoners received April 1992
December 1994 June 1994
new (public sector) new (public sector) new (public sector)
Expansion plans? none
Note: All prisoners in the United Kingdom are classified as to the level of security risk, category A being the maximum, and category D the lowest.
Contractual Management of Custodial Services 89
UK Detention Ser-
vices Ltd. (a Corrections Corp. of America joint venture with Sir Robert McAlpine & Sons
Bridgend Custodial Services Ltd. (Group 4 and Tarmac Consortium)
Fazakerley Prison Services Ltd. (Securicor, Seifert & At- kins consortium with joint venture partners, Costain and Skanska)
Premier Prison Services Ltd. Corp. (a Wackenhut Correc- tions Corp. joint venture company with Serco Ltd.)
HM Prison Blakenhurst
HM Prison Parc
HM Prison Fazakerley
HM Prison Lowdham Grange
HM Prison Service HM Prison Service HM Prison Service HM Prison Service
category B local 649
category B local (al- category B local (al- cat. B (sentenced so young offenders) so young offenders) prisoners only)
all (except maximum) public April 1993
private (reverts to private (reverts to private (reverts to
public sector after public sector after public sector after
(November 1997) (December 1997) (February 1998)
new (public sector) new (private sector) new (private sector) new (private sector)
90 Privatizing Correctional Services Notes 1 At the time of writing, 84 percent of places in the Immigration Service estate (524 places in total) were contractually managed. 2 With overcrowding at three of these prisons, contractual management accounted for 2,274 out of 54,640 places in use on 4 June 1996. These figures exclude immigration detainees not in prison custody but held directly by, or on behalf of, the Immigration Service. 3 Two maximum-security prisons managed by the Prison Service where there were serious escapes during 1995/96. 4 Siefert's is a leading British architectural practice and W.S. Atkins a major civil and mechanical engineering group. 5 Tarmac is one of the Britain's largest construction companies. 6 Unquestionably the real potential for this and risk transfer can only come if contractors take responsibility for designing and building prisons as well as running them. In this process, it is the future operator, not the architect or construction contractor, that needs to be in the driving seat. References Bottomley, A. Keith, Adrian James, Emma Clare and Alison Liebling (1997). Monitoring and Evaluation of Wolds Remand Prison. Centre for Criminology and Criminal Justice, University of Hull, and the Institute of Criminology, University of Cambridge. [Obtainable from the Home Office, UK.] Coopers and Lybrand (1966). Review of Comparative Costs and Performance of Privately and Publicly Operated Prisons for HM Prison Service. Prison Service Report No.1 (June). Deloite, Haskins, and Sells (1989). A Report on the Practicality of Private Sector Involvement in the Remand Systems. Livingstone, Stephen, and Tim Owen (1993). Prison Law: Text and Materials. Oxford: Oxford University Press. United Kingdom, HM Prison Service (1994). The Escape from Whitemoor Prison on Friday 9 September 1994. London: HMSO ------ (1995). Annual Report
and Accounts. London: HMSO United Kingdom, Home Office (1988). Private Sector Involvement in the Remand System. Green paper, CM434 (July). London: HMSO. ------ (1991). Custody, Care and Justice: The Way Ahead for the Prison Ser- vice in England and Wales. White paper. London: HMSO
Contractual Management of Custodial Services 91 ------ (1993). Wolds Remand Prison: Report by HM Chief Inspector of Prisons (August). London: Home Office Publications. ------ (1995a). Report of a Full Inspection by HM Chief Inspector of Prisons: HM Prisons Ballingdon and Oxford. London: Home Office Publications. ------ (1995b). Review by Sir John Learmont KCB CBE of Prison Service Security in England and Wales and the Escape from Parkhurst Prison on Tuesday 3rd January 1995. CM3020 (October). London: Home Office Publications. ------ (1996). Report of a Full Inspection by HM Chief Inspector of Prisons: HMP Doncaster. London: Home Office Publications. United Kingdom, House of Commons (1987). Fourth Report of the Home Affairs Committee: The Contract Provision of Prisons. HC291 (March). London: HMSO. ------ (1996). Home Affairs Committee Second Report: The Management of the Prison Service (Public and Private) TSO, 1997. HC57, sess. 1996/97. London: HMSO. United Kingdom, National Audit Office (1994). Report by the Comptroller and Auditor General. (30 March) London: HMSO. United Kingdom, White Paper (1996). Protecting the Public. (CM 3190, April). London: HMSO. Woolf, Harry, and Stephen Tumim (1991). Prison disturbances April 1990: Report of an Inquiry by the Rt. Hon. Lord Justice Woolf and His Hon. Judge Stephen Tumim (February). [Woolf Report]. London: HMSO.