Safety and Cost Effectiveness of Private Prisons

Private prisons accommodate around one third of the state’s male prison population. The safe, secure and cost effective operation of these prisons is essential for the effective functioning of Victoria’s corrections system and for community safety. Like the broader prison system, private prisons face significant challenges. The male prison population has increased by approximately 50 per cent over the last seven years, primarily driven by an increase in remand prisoners.

This audit examines whether two of Victoria’s private prisons—Port Phillip Prison and Fulham Correctional Centre—are safe and cost effective. We looked at:

• private prisons’ management of critical safety and security risks 
• private prisons’ performance against key service delivery measures, costs and risk transfer expectations under the original contracts 
• the process for the recent contract extensions and whether they delivered value for money.

This is the first audit in which we used our ‘follow-the-dollar’ powers, directly engaging the private prison operators and requesting information from them.

We made six recommendations for DJR, and two for the Department of Treasury and Finance related to its role in the private prison contract extension process.

Back to top

Ordered to be published

VICTORIAN GOVERNMENT PRINTER March 2018

PP No 384, Session 2014–18

Back to top

ACI	Australasian Correctional Investment Ltd
CMB	Contract management branch
CCTV	Closed‑circuit television
CV	Corrections Victoria
CVIU	Corrections Victoria Intelligence Unit
DJR	Department of Justice and Regulation (and its predecessor, Department of Justice)
DPC	Department of Premier and Cabinet
DTF	Department of Treasury and Finance
GEO	The GEO Group Australia Pty Ltd
G4S	G4S Custodial Services Pty Ltd
IMR	Internal management review
JARO	Justice Assurance Review Office (formerly the Office of Correctional Services Review)
KPI	Key performance indicators
MAP	Melbourne Assessment Prison
MRC	Metropolitan Remand Centre
OHS	Occupational health and safety
OCSR	Office of Correctional Services Review
OPCAT	Optional Protocol to the Convention against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment
OV	Occupational violence
PIMS	Prisoner Information Management System
PIU	Prison intelligence unit
PPP	Public–private partnership
PSC	Public Sector Comparator
PPRF	Private Prison Reporting Framework
SDO	Service delivery outcome
SSU	Security Standards Unit
VAGO	Victorian Auditor-General's Office

Back to top

Victoria's prison system faces significant challenges and risks:

• Male prisoner numbers increased by approximately 50 per cent over the last seven years, including significant growth in remand (unsentenced) prisoners.
• The prisoner population is increasingly complex, with mental health conditions, drug and alcohol issues, and chronic illnesses.
• Across the system, 43.6 per cent of prisoners returned to prison within two years in 2016–17.
• Young prisoners, prisoners with disabilities and prisoners of Aboriginal and Torres Strait Islander background are over‑represented.

Private prisons are part of the broader prison system in Victoria and, as such, are not immune to the challenges posed by a growing and increasingly complex prisoner population.

Victoria's privately operated prisons accommodated around one third of the state's male prisoners in December 2017. The safe, secure and cost-efficient operation of these private prisons is essential, both for the effective functioning of Victoria's corrections system and for community safety.

Corrections Victoria (CV), a division of the Department of Justice and Regulation (DJR), is responsible for managing the contracts with the private prison operators to ensure the safe custody and welfare of prisoners.

In this audit, we examined two of the three privately operated prisons in Victoria:

• Port Phillip Prison (Port Phillip) is a maximum-security men's prison in Melbourne's west. G4S Correctional Services (Australia) Pty Ltd, and its successor G4S Custodial Services Pty Ltd (G4S), has operated Port Phillip since 1997. G4S recently commenced a new contract term of up to 20 years. Port Phillip has more prisoners than any other prison in Victoria and, at December 2017, nearly half of these were remand prisoners. Port Phillip is also unique in that it provides statewide prison medical services and specialist services for intellectually disabled prisoners. It has more prisoner movements, in and out, than any other prison in the corrections system.
• Fulham Correctional Centre (Fulham) is a medium-security men's prison near Sale. The state has contracted Australasian Correctional Investment Ltd (ACI) to operate this prison since 1995. ACI is a special-purpose company established by The GEO Group, with The GEO Group Australia Pty Ltd (GEO) subcontracted as the prison operator. GEO commenced a new contract term of up to 19.25 years in July 2016. Fulham first accepted remand prisoners in September 2015. In December 2017, remand prisoners accounted for around 25 per cent of Fulham's population.

In this audit, we focused on whether these private prisons are safe and cost effective. In particular, we examined how well the private prisons are managing safety and security risks and whether they met the state's expectations for service delivery, cost and risk transfer during the original contract terms. We also assessed how well DJR and the Department of Treasury and Finance (DTF) managed negotiations for the new contracts and whether they achieved value for money.

We did not examine Victoria's third private prison, Ravenhall Correctional Centre (Ravenhall), as it commenced operation in November 2017, after this audit began.

Conclusion

G4S and GEO deliver cost-efficient services for the state that have largely met the contracts' service and performance requirements. Port Phillip and Fulham cost up to 20 per cent less to run than the average for publicly operated prisons of the same security rating.

CV demonstrates sound contract management and genuine engagement with the private operators of Port Phillip and Fulham. DJR successfully negotiated new contracts for the operation of these prisons that address key weaknesses in the initial contractual arrangements, at a cost consistent with government expectations.

However, the prison operators are not always meeting the state's service and performance requirements to run safe and secure prisons, particularly in relation to assaults at both prisons and drug use at Port Phillip. This is consistent with system-wide performance.

Serious incidents at both Port Phillip and Fulham have, in some instances, exposed weaknesses in how G4S and GEO manage safety and security risks, and neither operator is investigating serious incidents using methods that effectively identify root causes. The operators and CV need to improve how they address the risks inherent in the prison environment to better discharge their shared duty of care to prisoners, prison staff and the community.

Prisoner-on-prisoner assaults is increasing across the prison system, as prisoner numbers and complexity increases. Private operators and CV have violence‑reduction strategies in place, but there is a need to better evaluate these and share the lessons learnt.

Findings

Private prison performance

Performance results

CV measures prison performance against service delivery outcomes (SDO), which cover prisoner safety, security, health, welfare, activities and programs. Applicable SDOs and performance thresholds vary for each prison, depending on the prison's security level, prisoner profile and past performance. DJR's target is to meet 90 per cent of SDO thresholds.

Fulham's performance against its SDOs has been largely positive since 2010−11, apart from a decline in performance in 2012–13, partly related to a single incident involving several staff injuries.

Port Phillip has failed to meet a greater number of performance thresholds than Fulham relating to prisoner-on-prisoner assault, assaults against staff and positive random drug tests.

The new contracts include key performance indicators (KPI), with 18 that apply to Port Phillip and 16 to Fulham. These cover health services, facility and asset management, performance and incident reporting, prisoner reintegration and prisoner management. The KPIs significantly increase the requirements for the private operators to maintain their facilities and, importantly, require private prisons to turn around poor performance quickly and provide accurate performance data. These KPIs do not apply to public prisons. Under the new contracts, CV continues to measure the performance of private prisons monthly, and calculates performance-linked payments quarterly rather than annually. This increases pressure on private prisons to consistently meet performance requirements.

Monitoring private prisons

CV scrutinises private prison performance more than public prisons, including validating the operators' self-reported performance data.

This is important, as achieving an appropriate transfer of risk to the private operators was a key objective of the initial contracts. Those contracts allocated the financing risks for construction of the new prison facilities to the private operators. The operators also took on the operational risks associated with accommodating prisoners and delivering correctional services to agreed service standards.

DJR's management of the private prison contracts has significantly improved since our 2010 audit Management of Prison Accommodation Using Public Private Partnerships. CV has an almost constant presence at both prisons and routinely checks contractual compliance to support performance and maintain the allocation of risks agreed in the initial contracts.

CV has responded appropriately to poor performance including significant safety and security breakdowns, requiring the operators to develop and implement plans to address the underlying issues and causes.

CV uses a variety of graduated responses to address operators' underperformance, and engages and supports the operators to manage risks. This reflects the mature contractual relationship developed between the parties. Since 2012, CV has appropriately issued five contract default notices in response to serious performance concerns. To date, the operators have complied with the requirements of the default notices to implement corrective action plans, known as 'cure' plans.

CV's increased oversight and scrutiny of private prisons is positive, however, it is not possible to determine whether it has resulted in improved performance. CV and the prison operators monitor and regularly discuss these corrective action plans, but they do not routinely evaluate the impact of their improvement activities.

CV collects a significant amount of data from all prisons, but its legacy IT systems are not integrated and this hinders its ability to analyse data effectively. CV cannot use this data to fully understand the trends and issues that may affect prisons' performance, safety, security and costs across the corrections system.

Despite the large amount of information CV holds on prison performance, little data on whether Victoria's prisons are safe, secure and meeting performance requirements is publicly available. CV currently publishes information about drug testing and contraband seizures in prisons, but does not publish prison‑specific performance results relating to other safety and security matters such as assaults, deaths and escapes. This reduces the transparency of the Victorian corrections system.

Costs

The privately operated prisons cost the state up to 20 per cent less to run than the average for publicly operated prisons of the same security rating. This is consistent with DJR's advice to government in recent years that the two privately operated prisons are cheaper to run than public prisons, largely due to more efficient staff shift patterns.

Safety and security of private prisons

Prisoner-on-prisoner assaults are increasing at Port Phillip and Fulham, particularly Level 1 assaults, which involve an injury that does not require admission to hospital. This is consistent with the broader prison system as it deals with capacity pressures, increasing remand population, increased prisoner movement through the system and an increasingly complex prisoner profile. Assaults on staff also increased in 2015–16 and then again in the first half of 2016–17 across the corrections system, and are more common at maximum‑security prisons including Port Phillip.

Fulham and Port Phillip have both developed violence-reduction strategies to address this risk. Port Phillip's strategy is comprehensive and involves a broad range of activities targeting violence, which reflects the higher risks at that prison. It involves dedicated resources and has recently broadened to include a specific focus on occupational violence (OV). Fulham's strategy also includes a range of activities to minimise the risk of assault on prisoners and staff, such as staff communication and training, prisoner employment and programs. Both prisons have a strong focus on reducing violence and improving performance in this area. They closely monitor incidents of violence and SDO performance, however, neither prison periodically evaluates its strategies to ensure they are working and based on current evidence.

CV has a system-wide OV strategy, but this does not specifically address prisoner-on-prisoner assaults. There is no planned evaluation of the strategy and no coordinated process for capturing and sharing the lessons of the various prison strategies.

Serious safety and security incidents have occurred at both private prisons since 2010–11, including escapes, a riot, unnatural deaths and serious assaults. CV and the private operators have worked well together to respond to these incidents, but the investigation process lacks a contemporary methodology to analyse the root cause. This creates the risk that corrective actions will not successfully prevent further incidents.

Negotiating new private prison contracts

While negotiating new contracts, DJR successfully navigated significant challenges to ensure service continuity at Fulham and Port Phillip. The new contracts addressed key weaknesses in the initial contractual arrangements and broadly met the cost limits approved by government. Advice to government throughout the process was sound, with some minor exceptions that did not invalidate the outcomes achieved.

These transactions were unique given the context within which they occurred. The processes used by DJR did not fully align with DTF's Partnerships Victoria requirements for public–private partnership (PPP) projects. However, DTF accepted these departures from the usual requirements, and the close involvement from senior DJR officers, DTF and the Department of Premier and Cabinet (DPC) meant there was adequate oversight and review of the options analysis and negotiation process.

DJR analysed the most viable procurement options early enough to protect the state's negotiating position. This analysis found that the state was likely to incur significant additional costs if it competitively tendered the operation of the prisons or decided to operate them itself.

The expected additional costs related largely to the need to buy out leases held by GEO and G4S over the prison sites, which extended well beyond the 2017 expiration dates of their contracts. These long‑term land leases were part of the initial contract arrangements entered into in the 1990s.

The original Crown leases were to expire in 2035 for Fulham and 2046 for Port Phillip. This misalignment of the lease terms with the service contract terms was a key reason for the state negotiating extended contracts with the incumbent operators, and ultimately served to reduce competitive tension. The revised lease terms now align with the new service contract terms of up to 20 years.

DJR developed and implemented a robust strategy to negotiate the new contracts with the operators. Cost was the main criteria assessed and highlighted in advice to government on whether to enter into new contracts. DJR took reasonable steps to gain assurance about the value for money of the proposals put forward by the operators, including targeted examination of the operators' actual operating costs.

DJR only examined service performance at a cursory level, and we saw no detailed assessment of whether the incumbent operators were capable and high-performing providers. DJR's concerns about G4S's performance during 2014 received little coverage in advice to government on whether to negotiate with the incumbent operator.

A commercial adviser working for DJR and one of the operators during the negotiations for the new contract did not adequately disclose conflicts of interest and these were not adequately assessed, or acknowledged in advice to government.

The new contracts negotiated with the operators addressed substantial weaknesses in the previous contracts. They will place the state in a much stronger position at the end of the new service terms to consider and pursue all available options, including competitively tendering the contracts, unencumbered by substantial legacy issues or costs.

DJR needs to manage these contracts effectively to achieve value for money in both performance and cost in the future. This is a significant challenge, but DJR has established robust contract administration guidance and processes to support this task.

Recommendations

We recommend that the Department of Justice and Regulation:

1. update relevant approval processes for changes to system-wide operational requirements and standards so that contractual implications are adequately identified, considered and addressed before changes are implemented (see Section 2.2)

2. address known issues with legacy IT and data systems by integrating offender management systems, to improve data management and analytical capability (see Section 2.3)

3. improve the transparency of the prison system by increasing public reporting on the performance of individual prisons and the system as a whole, against applicable service delivery outcomes and key performance indicators (see Section 2.4)

4. monitor and enforce compliance with the contract requirement that serious incident investigations in privately operated prisons follow contemporary root-cause analysis methodology (see Section 3.2)

5. evaluate the effectiveness and impact of violence-reduction efforts across the system, share the findings for continuous improvement and lead the development of a system-wide violence-reduction strategy that includes occupational violence and prisoner-on-prisoner violence (see Section 3.5)

6. resolve system issues so private prisons have access to the corrections intelligence system that is equivalent to public prisons' access (see Section 3.6).

We recommend that the Department of Treasury and Finance:

7. ensure that its advice to government, and associated public information on Partnerships Victoria and other major projects, should wherever practicable present costs and benefits in nominal and present value terms, with the discount rate (nominal and/or real rate) and other key assumptions explicitly stated and justified (see Sections 5.2 and 5.3)

8. update relevant guidance to require probity reports and sign-off letters for major procurement transactions to disclose any material probity issues that arose during the relevant project, even where the issues were managed to the satisfaction of the probity practitioner and project governance group (see Section 5.3).

Responses to recommendations

We have consulted with DJR, DTF, G4S and GEO, and we considered their views when reaching our audit conclusions. As required by section 16(3) of the Audit Act 1994, we gave a draft copy of this report to those agencies and asked for their submissions or comments. We also provided a copy of the report to DPC.

The following is a summary of those responses. The full responses from each agency are included in Appendix A.

DJR acknowledged the insights the audit has provided and accepted all of the recommendations. DTF also accepted the recommendations, and both agencies have provided an action plan detailing how they will address the recommendations.

While the recommendations were not directed to GEO or G4S, both private prison operators supported the audit findings and the recommendations made.

Back to top

The Victorian adult corrections system includes more than 50 community correctional facilities, 14 prisons and a transition centre. Figure 1A shows the prisons and transition centre, along with their operational capacity, at December 2017.

Figure 1A
Victoria's prison system

(a) Ravenhall was partially operational as at December 2017. Its full operational capacity is 1 000.

Source: VAGO based on operational capacity data provided by CV.

Each prison has unique characteristics and its own complexities and challenges, including:

• prisoner profile—sentenced or remand prisoners, sex offenders, young adult offenders, prisoners with intellectually disabilities and those requiring protection or separation due to particular risks or vulnerabilities
• the type of service provided—reception prison, preparation for release, hospital services.

Victoria's privately operated prisons

The initial contracts required accommodation services for around 600 prisoners in fit-for-purpose buildings and facilities, with the majority of prisoners in single cells, and a safe and secure environment, including a secure physical prison perimeter.

Since 1997, Victoria has contracted three privately operated 'full-service' men's prisons under PPP arrangements—Fulham, Port Phillip and Ravenhall. We did not examine Ravenhall in this audit as it only commenced operations in November 2017. The Dame Phyllis Frost Centre, previously the Metropolitan Women's Correctional Centre is a maximum‑security women's prison, which was contracted to a private operator in 1996, but returned to state control in 2000 due to concerns with the management of the prison.

Under their contracts, the operators of Fulham and Port Phillip provide accommodation services (suitable facilities for prisoner containment) and correctional services (safe custody and welfare of prisoners in their care). Correctional and accommodation services are required to comply with relevant legislation and state policies, as well as contractual obligations.

Contracted correctional services included: safe, secure, humane and just containment and supervision of prisoners opportunities for rehabilitation work for prisoners to support their return to the community health and other services for prisoners.

The private operators, sub‑contractors or CV also provide prisoner education and employment programs and health services.

The initial private prison contracts specified SDOs and performance thresholds, and enabled performance-based payments to the operators. The contracts addressed the state's access rights, insurance requirements and required facility maintenance.

GEO operates Fulham, a medium-security men's prison located near Sale, in eastern Victoria. Fulham accommodates medium-security mainstream and protection prisoners in cellblocks and cottages, as well as minimum-security prisoners in units. Fulham also has a minimum-security annexe, Nalu, a unit initially intended for young offenders aged 18–25 that now houses all age groups.

G4S operates Port Phillip, a maximum-security men's prison located in Truganina, to the west of Melbourne. Port Phillip accommodates prisoners with diverse needs, including mainstream prisoners and those requiring protection or separation in high-security units. It has a psychosocial unit, a youth unit for young adult offenders, and a special care unit for vulnerable and intellectually disabled prisoners. Port Phillip has a 20-bed in-patient facility within its walls, providing primary and secondary health care to prisoners from across the corrections system. Port Phillip is also responsible for managing the security of all prisoners in the secure or non-secure wards at St Vincent's Hospital.

In comparison to other Australian states, Victoria has the largest percentage of its total number of male prisoners in private prisons. With Ravenhall now in operation, Victoria has more privately operated prisons than any other state in Australia. Figure 1B shows key metrics for Fulham, Port Phillip and Ravenhall.

Figure 1B
Key metrics for Victoria's full-service private prisons at December 2017

Metric	Private prison
	Fulham	Port Phillip	Ravenhall
Operational capacity	893	1 087	700(a)
Percentage of state bed capacity in men's prisons	12%	14%	9%
New contract commencement	1 July 2016(b)	10 September 2017	Build phase—15 September 2014 Operational phase—November 2017
Contract term	11 plus 8.25 years	10 plus 10 years	25 years (operational phase)
Contract expiry	October 2035	September 2037	October 2042

Note: Operational capacity percentages are rounded figures.

(a) Ravenhall has not reached its full operational capacity of 1 000. Ravenhall also includes an option to increase capacity to 1 300.

(b) The initial Fulham contract was due to expire in April 2017. The state and GEO signed the new contract in April 2015 and agreed that it would commence in July 2016.

Source: CV.

Terms of the Fulham and Port Phillip contracts

The initial contracts for Fulham and Port Phillip required the design, finance, construction, operation and maintenance of the prisons for up to 20 years. The contracts included an option to negotiate an extension. The initial contracts were due to expire in 2017.

Significantly, the initial arrangements also provided the operators with long‑term leases over the land on which the prisons were constructed. These Crown leases were to expire in 2035 for Fulham and 2046 for Port Phillip. The misalignment of the lease terms with the service contract term was one of the reasons for the state negotiating extended contracts with the incumbent operators.

DJR finalised negotiations during 2015 for new contracts with both operators. The new contracts have maximum terms of up to 20 years, subject to the operators meeting performance requirements, and are forecast to cost the state approximately $4.5 billion in nominal terms if they run their full term. The revised lease terms now align with the 20-year service contract terms.

Systemic issues and challenges

Prison performance is influenced by many factors, and the Victorian system faces ongoing challenges due to a growing and increasingly complex prisoner population.

Increase in prisoner numbers

In December 2017, Victoria had 7 131 prisoners, 93 per cent of whom were male. As shown in Figure 1C, prisoner numbers have fluctuated but overall increased from an annual average of 4 272 in 2010–11 to 6 383 in 2016–17. During this period, Port Phillip's prisoner numbers increased by an average of 5 per cent annually and Fulham's numbers fluctuated but on average increased by 2 per cent annually.

Figure 1C
Annual average number of male prisoners, 2010–11 to 2016–17

Source: VAGO based on the daily and monthly averages for each prison provided by CV.

Prison utilisation rates

The operational capacity of a prison is the total number of beds available for prisoners. In December 2017, the men's prison system had an average daily operating capacity of 7 694 places for 6 638 prisoners. Operational capacity includes any additional beds in cells over and above their design capacity—for example, single cells that now accommodate bunk beds. Excluded from operational capacity are cells dedicated to short-term management of prisoners (often after an incident), some specialist beds such as hospital beds and observation cells, and cells closed for an extended period of maintenance.

CV regularly monitors each prison's utilisation rate based on its operational capacity. DJR's output performance measures published in the 2016–17 State Budget included a utilisation target of 90 to 95 per cent and it achieved an outcome of 94 per cent in 2016–17. CV advised that this target reflects the need to have sufficient flexibility in the system to deal with maintenance, prisoner movement and placement needs, and to respond to incidents.

Figure 1D and Figure 1E show that utilisation rates have fluctuated in all men's prisons since 2011–12. Port Phillip has consistently higher utilisation rates than the average of the other maximum-security prisons and is consistently at or above the 95 per cent utilisation target except in quarter four 2014–15. CV advised that Port Phillip's utilisation rate can reach above 100 per cent due to the use of special-category prison beds, such as hospital beds, that are not counted in operational capacity.

Fulham's utilisation has fluctuated, and between 2014–15 and 2015–16 dropped below the average of other medium-security prisons. CV attributes this to the opening of the Middleton facility at Loddon prison and expansion at other prisons.

The riot at the Metropolitan Remand Centre (MRC) in June 2015 also resulted in an increase in utilisation rates at both Port Phillip and Fulham in 2015–16, after a period of lower utilisation in 2014–15.

Figure 1D
Daily average utilisation at Port Phillip compared to the average at other maximum-security prisons (excluding Port Phillip), 2011–12 to 2016–17

Note: Port Phillip's average daily utilisation reached above 100 per cent due to the use of special-category prison beds (such as hospital beds).

Source: VAGO based on data provided by CV.

Figure 1E
Daily average utilisation at Fulham compared to the average at other medium-security prisons (excluding Fulham), 2011–12 to 2016–17

Source: VAGO based on data provided by CV.

Increase in remand prisoners

A remand prisoner is a person against whom a charge has been laid but not proven in court. The prisoner has not been released on bail and is in prison awaiting trial.

The profile of prisoners has changed over the last decade, with an increase in the remand population from 17 per cent of the total men's prison population in the first quarter of 2010–11 to 30 per cent at the end of 2017, as shown in Figure 1F. This increase in remand prisoners is linked to reforms of the parole and bail systems, as well as a greater number of police detecting more offences. The MRC riot in June 2015 also resulted in remandees being moved to other maximum- and medium-security prisons.

In December 2017, 45 per cent of Port Phillip's prisoners were remand prisoners, down from an annual average of 51 per cent in 2016–17. Fulham started to receive remand prisoners in 2015 and, in December 2017, remand prisoners accounted for 25 per cent of Fulham's population.

Figure 1F
Percentage of remand prisoners in all male prisons, July 2010 to December 2017

Source: VAGO based on data provided by CV.

A November 2016 report by DJR's Office of Correctional Services Review (OCSR)—now the Justice Assurance Review Office (JARO)—found that remand prisoners may face particular challenges. The report found that, for the period reviewed, male remand prisoners were potentially more vulnerable than sentenced prisoners due to the effects of drug and alcohol withdrawal, young age, potential of undiagnosed medical illness and the risk of being subject to standover behaviour from other prisoners.

In addition, remand prisoners have comparatively short and uncertain periods in custody and contribute to the increasing number of prisoner movements, as they have more court appearances than a sentenced prisoner. Based on the presumption of innocence, remand prisoners are entitled to less restrictive conditions and have different rights to sentenced prisoners. They are not mandated to participate in programs or work. While remand prisoners should have separate accommodation units to sentenced prisoners, this is not always achievable due to system-wide capacity issues.

Prisoner movement

There are many prisoner movements to, between and from Victoria's prisons, including:

• reception into prison
• discharge from prison
• progression or regression within a prisoner's sentence or classification
• court appearances
• medical needs and hospital treatment
• short prison sentences
• prison transfers for safety needs or to access particular leave permits.

Reception prisons are the first point of contact for prisoners entering the prison system. By their nature, reception prisons have a high volume of prisoner movements. In Victoria, men's reception prisons include MRC, Melbourne Assessment Prison (MAP) and, since November 2017, Ravenhall. Port Phillip can act as a reception prison, but the state has not used this capability frequently in recent years.

The transient nature of prison populations results in a constantly changing prison culture and dynamic. This poses a risk for prison operators in ensuring the safety and security of prisoners and prison staff. The number of prisoner movements varies greatly between prisons depending on the particular services a prison provides. For example, Port Phillip is the largest male prison and is a central hub for the prison system—prisoners may transfer to Port Phillip from across the system to be near Melbourne's courts, to receive medical treatment and to prepare for discharge. Remand prisoners also experience more movements due to court appearances.

Port Phillip has the greatest number of prisoner movements of any Victorian prison, because of its large population and the services it provides. Figure 1G shows that the number of prisoner movements at Port Phillip is consistently higher than MRC and the average of other maximum-security prisons. Movements increased at Port Phillip and MRC in 2014–15, with court-related movements and prisoner discharges contributing the most to this increase. The difference between Port Phillip and MRC prisoner movements increased over the last two years, potentially due to the impact of the MRC riot and increased proportion of remand prisoners. In comparison, the average of all other maximum-security prisons is lower and this is primarily due to Barwon Prison having a low number of movements.

When movements are calculated as the number of movements per prisoner per year, Port Phillip and MRC are the same, with an annual average of 19 moves per prisoner between 2010–11 and 2016–17.

Figure 1G
Prisoner movements at Port Phillip and MRC compared to the average at maximum-security prisons (excluding Port Phillip), 2010–11 to 2016–17

Note: Movements include all transfers in and out of the prison, receptions and discharges. Court transfers and movements for leave permits (related to health, justice administration, community work, rehabilitation and transition and other approved reasons) are counted as movements.

Source: VAGO based on movement data provided by CV.

Figure 1H shows that Fulham has consistently more prisoner movements than Marngoneet Correctional Centre (Marngoneet), a comparable public medium‑security prison. This difference is partly due to higher prisoner numbers. When compared to the average of all other medium-security prisons, Fulham's average annual number of movements is around double the average for other medium-security prisons.

However, when we take into account the number of prisoners at a prison, Fulham's rate of movement per prisoner per year is similar to the average of other medium-security prisons. On average, Fulham's annual average rate of movements between 2010–11 and 2016–17 is 9.6 movements per prisoner and the average of other medium security prisons is 9.5 movements per prisoner.

Figure 1H
Prisoner movements at Fulham and Marngoneet compared to the average at medium-security prisons (excluding Fulham), 2010–11 to
2016–17

Source: VAGO based on movement data provided by CV.

Increase in assaults

Victoria's prison system is experiencing an increase in assaults. Figure 1I shows the rate of assaults against prisoners and staff in Victoria's prisons over the last four years in men's prisons. In July 2013, there were 63 assaults and 5 022 male prisoners, and in June 2017 there were 133 assaults and 6 622 male prisoners. While the rate of assaults per 100 prisoners fluctuates monthly, on average it increased by 2 per cent per month across all men's prisons in this period. Many prisons, public and private, are not meeting performance thresholds for prisoner‑on‑prisoner assaults.

CV and prison operators attribute the increase in assaults to rising prisoner numbers and movements, the increasing proportion of remand prisoners and a more complex prisoner profile including a higher rate of prisoners with mental illness or intellectual disabilities, or suffering drug withdrawal.

Figure 1I
Monthly rate of prisoner and staff assaults per 100 prisoners in men's prisons, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

1.1 Relevant legislation and regulation

Corrections Act 1986

The Corrections Act 1986 (the Act) and the Corrections Regulations 2009 (the Regulations) provide the legislative basis for adult correctional services in Victoria. Relevant parts of the Act:

• provide for the establishment, management and security of prisons and the welfare of prisoners
• allow the state to contract out the provision of correctional services
• assign clear responsibilities to the Secretary of DJR and, specifically, the Commissioner of CV (the Commissioner) to monitor the performance of all correctional services to achieve the safe custody and welfare of prisoners.

This means that, while the Act allows the state to contract out the operation of prisons and ancillary services, the state retains a duty of care to all prisoners and is accountable to the community for the operation and cost of all prisons.

Victorian Correctional Management Standards for Men's Prisons

The Victorian Correctional Management Standards for Men's Prisons (the Standards) reflect the requirements of the Act and the Regulations. The Standards focus on the desired outcomes and outputs for public and private prisons, and provide a basis for prison operating procedures. They also provide the framework against which CV monitors prison services.

The Standards cross-reference the national Standard Guidelines for Corrections in Australia, to which CV is a signatory. These guidelines act as outcomes or goals to be achieved by correctional services, rather than a set of absolute standards or laws to be enforced.

Commissioner's Requirements

The Commissioner issues Commissioner's Requirements to all prisons to promote consistency and continuity of operational practice across the whole prison system. CV consults with prison management to develop the Commissioner's Requirements.

There are 66 Commissioner's Requirements related to:

• security and control
• prisoner management
• programs and industry
• prisoner services
• prisoner health.

The private prisons are required to document their operating procedures in an operating manual, which must comply with the Commissioner's Requirements. The operating manual includes a range of operating instructions, which the Commissioner endorses.

1.2 Roles of agencies and associated entities

Corrections Victoria

CV, a division of DJR, is responsible for the establishment, management and security of all public and private prisons in line with the relevant legislative requirements. CV's key functions include:

• the statewide provision of correctional operations
• setting standards and monitoring performance
• developing and delivering correctional strategy, policy and programs.

These functions aim to achieve the safe custody and welfare of prisoners and community-based offenders.

Justice Assurance Review Office

JARO operates as an internal review and assurance function to advise the Secretary of DJR on the performance of Victoria's youth justice and corrections systems. JARO is separate from the department's Youth Justice Division and CV, and acts as an additional line of defence against emerging and enduring risk within both systems. JARO reviews all deaths in custody and some serious incidents, as well as conducting thematic reviews of systemic issues.

Private prison operators

GEO operates Fulham under a contract with ACI, which has the contract with the state. ACI is a special‑purpose company established by The GEO Group, which is an American multinational company providing corrective and detention services worldwide.

Port Phillip is operated by G4S—part of a large multinational company operating in over 100 countries.

Monitoring prisons

CV's various divisions support and monitor the private prison operators, as shown in Figure 1J.

Figure 1J
CV divisions and their roles in service delivery and overseeing private prisons

Business services	Manages contracts for management of private prisons Monitors contractual compliance Provides an onsite presence at prisons
Strategic policy and planning	Validates prison SDOs Analyses performance and trends Develops strategy and policy for all prisons
Operations	Acts as the operational interface between CV and prisons Provides post-incident management and response Develops and updates Commissioner's Requirements Collects and calculates prison performance data
Offender management	Provides and oversees offending behaviour programs
Sentence management	Classifies prisoners by security and risk Allocates prisoners to a prison Manages major offenders and high-risk prisoners
Security and intelligence	Manages CV intelligence Oversees and assists prison intelligence units (PIU) Operates the emergency response team Sets security standards

Note: This list does not show all functions provided by CV divisions. It includes only the functions that relate to this audit.

Source: VAGO based on information from CV.

Performance framework

All prisons must report on their performance against specified SDOs, which cover prisoner safety, security, health, welfare, activities and programs. The performance thresholds for SDOs vary for each prison based on the security level of the prison, prisoner profile and past performance. CV monitors SDO performance and provides monthly and quarterly performance reports to the Commissioner.

The new contracts for Fulham and Port Phillip, signed in 2015, retained the existing SDOs and included 18 additional KPIs for Port Phillip and 16 for Fulham. The contracts also redesigned the performance framework to:

• focus more strongly on facility maintenance
• introduce a quarterly performance outcome, instead of annual
• restructure the thresholds and associated performance payments so they are scaled according to the prison's level of performance.

The new KPIs applied to Fulham for the 2016–17 year and Port Phillip from September 2017. They cover a range of performance areas including health care, facility maintenance, reintegration programs and recidivism, although not all the KPIs apply at both prisons. The KPIs do not apply at Victoria's public prisons, and CV has no plans to implement them due to the administrative burden this would place on the prisons.

The eight SDOs and two KPIs relating to prison safety and security are:

• SDO 1—escapes
• SDO 2—assaults on staff by prisoners
• SDO 3—out-of-cell hours
• SDO 4—number of unnatural deaths
• SDO 5—self harm
• SDO 6—assaults on prisoners by other prisoners
• SDO 7—assaults on prisoners by staff
• SDO 8—random general urinalysis
• KPI 6—releases on the correct date
• KPI 11—incident reporting.

The safety and security SDOs and KPIs in the new contracts use various measures:

• zero tolerance—requires 100 per cent compliance or nil incidents
• a percentage result—for example, positive drug tests
• number of incident points per quarter—adjusted to account for prisoner numbers.

Excluding the zero tolerance indicators, each of the SDOs and KPIs above has a threshold that the prisons are required to meet. See Appendix B for full details on SDOs and KPIs.

Unlike public prisons, private prisons have a performance regime with associated payments outlined in their contracts. In the private prisons, performance thresholds form part of the commercial negotiations between the state and the operator, and there are financial implications when private prisons do not meet the thresholds. In the public system, performance outcomes are not linked to financial penalties. CV considers that this provides it with more flexibility to alter the thresholds for public prisons and potentially implement stretch goals.

Our May 1999 audit Victoria's Prison System: Community protection and prisoner welfare found that the initial contracts with the operators of Fulham and Port Phillip did not encourage service excellence. The SDO thresholds in these contracts were not based on average performance but on the lowest standard achieved by comparable public prisons during the previous three-year period. The audit also found that DJR applied less scrutiny to public prison operations and performance than to private prisons.

The initial contracts allowed CV to amend SDO thresholds over time based on the state's requirements and commercial negotiations with the operators. These negotiations and updates to SDO thresholds occurred as part of the five-yearly service term renewal process under the contracts.

1.3 Previous audits

Our September 2010 report Management of Prison Accommodation Using Public Private Partnerships found that the former Department of Justice faced significant challenges and problems managing these long-term contracts. This was partially due to the contracts not adequately defining quality standards, but was also the result of ineffective administration of the contracts, lack of adequate performance monitoring, complex governance arrangements and inadequate risk management.

Our October 2013 report Prevention and Management of Drug Use in Prisons found that the processes for identifying prisoners who use drugs were generally effective. However, weaknesses in performance reporting and evaluation meant that the former Department of Justice could not determine whether initiatives to manage drug use in prisons were effective.

Other recent audits of prisons include Prison Capacity Planning (November 2012) and Prisoner Transportation (June 2014).

1.4 Why this audit is important

In recent years, there has been growing public concern about safety risks at Victoria's prisons, and their implications for prisoners, employees and the public. Recent incidents have contributed to the public's concerns—for example, prisoners escaping over the wall at Fulham (2016), a prisoner escaping from Port Phillip's custody at St Vincent's Hospital (2017), prisoners rioting at MRC (2015) and numerous other violent incidents at Victorian and interstate prisons and youth justice facilities.

The operation of Victoria's prisons is a considerable expense for the state. Understanding the cost of operating private prisons compared to public prisons is important for an accountable and transparent prison system. Equally important is the ability to evaluate whether the recent contract extensions for Port Phillip and Fulham provide value for money.

1.5 What this audit examined and how

The objective of this audit was to determine whether Victoria's private prisons are safe and cost effective.

We examined whether:

• private prisons are appropriately managing critical safety and security risks
• private prisons met the state's key service delivery, cost and risk transfer expectations under the old contracts
• the recent contract extensions achieved value for money.

We limited our comparison of the private prisons' performance and cost to other male prisons:

• We compared Port Phillip's performance outcomes to MRC, and we compared Fulham's to Marngoneet, to reflect similarities such as prison profile, size and volume of prisoner movements.
• We compared Port Phillip's financial performance to MRC and Barwon, and we compared Fulham's to Loddon and the average of other medium‑security prisons (which includes Loddon). These financial comparisons are consistent with the value-for-money comparisons the state undertook when negotiating the new contracts.

We excluded the two women's prisons due to operational differences and complexities.

Our analysis of prison performance covered the period July 2010 to June 2017. We examined prison incident data for four years, from July 2013 to June 2017.

We conducted our audit in accordance with section 15 of the Audit Act 1994 and ASAE 3500 Performance Engagements. We complied with the independence and other relevant ethical requirements related to assurance engagements. In this audit, we used our 'follow-the-dollar' powers, directly engaging the private prison operators and requesting information from them. The private operators cooperated with all our requests.

The cost of this audit was $1 020 000.

1.6 Report structure

This report is structured as follows:

• Part 2 examines private prison contract management and performance
• Part 3 examines the safety and security of private prisons
• Part 4 examines prison costs
• Part 5 examines the negotiation of the new private prison contracts.

Back to top

The Act requires DJR to monitor the performance of all correctional services to achieve the safe custody and welfare of prisoners. DJR is accountable to the community for the operation and cost of Fulham and Port Phillip. Contracting private companies to operate prisons does not lessen the state's duty of care to prisoners.

The contracts for Fulham and Port Phillip transfer significant operating and financial risks to the private operators and require them to meet specified service standards and performance expectations. These contracts provide the state with broad monitoring and access rights, as well as penalty and intervention options if the operators fail to meet expectations.

This part of the report examines DJR's approach to performance monitoring and the actual performance results for Fulham and Port Phillip. The performance of the private operators needs to be considered in the broader context of the challenges facing the entire corrections system, outlined in Parts 1 and 3.

2.1 Conclusion

GEO and G4S largely met the state's service delivery expectations for correctional and accommodation services over the life of the original contracts for Fulham and Port Phillip.

CV rigorously monitors the performance and contractual compliance of the private operators. It works collaboratively with the private operators to improve their performance and applies contractual penalties in response to serious performance concerns.

CV collects a large amount of data on prison performance, however outdated information systems limit its ability to produce meaningful analysis. CV's limited public reporting on prison performance restricts the transparency and public scrutiny of the prison system.

2.2 CV's scrutiny of operator performance

CV has multiple ways of gaining insight into and scrutinising the operational practices and performance of the private operators—see Figure 1J in Part 1. Reliable and accurate performance data and effective contract management is crucial for CV to meet its performance monitoring obligations and to hold private prison operators to account.

Contract management branch

CV's contract management branch (CMB) is responsible for managing the contract for each private prison. CMB's key functions include:

• assessing the operators' service delivery against requirements
• managing the state's oversight of the prisons' operations manuals and ensuring effective review of operating instructions
• managing the process for authorising private prison staff under the Act
• managing and monitoring prisons' reporting requirements and preparing reports for the Commissioner on prison performance
• validating and verifying performance data reported by the prisons
• managing contract variations, including analysis of operators' proposals to ensure a value-for-money outcome
• managing payments to private operators under the contracts, including payments linked to performance.

CV is efficiently administering its contract management responsibilities. Since our 2010 audit Management of Prison Accommodation Using Public Private Partnerships, CV has increased resources in CMB including a dedicated contract manager for each private prison. It also implemented a new information management system for contract administration. CMB works collaboratively with the private operators and has a strong presence at both private prisons.

Operating instructions in private prisons

Operating instructions guide day-to-day operations in the private prisons. Prison safety and security rely on their effective implementation.

The Commissioner reviews and endorses operating instructions submitted by the private operators. In 2012, CMB began supporting the Commissioner's reviews by checking operating instructions to ensure their consistency with contractual, legal and other requirements and to gain assurance about private prisons' operations. Prior to this, CV conducted an annual high-level review of the prisons' operating manuals.

We examined endorsement processes for 114 operating instructions and found that the review process is rigorous. There were some minor record-keeping issues identified in 12 of the processes.

CMB undertakes regular audits to test the implementation of operating instructions. The audit schedule and methodology is risk based and, importantly, these are not desktop audits—they involve CMB staff talking to staff and prisoners, observing day-to-day prison operations, and checking source documents and registers. Prison management must immediately rectify any critical noncompliance. There is no equivalent audit program in public prisons.

Contractual responses to poor performance

The initial contracts enabled the state to address poor operator performance by:

• reducing the fees operators received for accommodation and/or correctional services if any services were not satisfactory
• reducing the annual performance-linked fee, a payment based on the state's assessment of the operator's performance against the SDO performance thresholds
• issuing default notices
• ending the contract and appointing a new operator if an operator was no longer capable of operating the prison satisfactorily, or if it became insolvent.

Imposing contractual sanctions is not necessarily an appropriate first response to poor performance by the operators. This is particularly so when system-wide challenges and risks, outside the control of individual operators, contribute to performance failures. CV's approach of applying graduated contract responses provides notice and time for the operators to improve their performance. For example, in early 2017, CV responded to a number of serious assaults at Port Phillip with a peer review to assist the operator to improve safety and security. Section 2.5 includes more information on the peer review process and results.

Figure 2A shows the default notices for correctional services issued to the operators of Fulham and Port Phillip, between 1997 and 2017. While public prisons are not subject to contractual defaults or similar processes, it is important to note that similar serious incidents also occur in public prisons.

Figure 2A
Contractual defaults at Port Phillip and Fulham

Period	Incident
Port Phillip
1997–2006	Six defaults over this period including for: death in custody serious staff misconduct systemic noncompliance with operating procedures
April 2016	Breach of firearms licensing process
July 2016	Death of a prisoner after ingestion of drugs
March 2017	Prisoner escape from St Vincent's Hospital
Fulham
April 2016	Prisoner escape
August 2016	Infant cannabis seedlings found on prison grounds

Source: VAGO based on information from CV.

Default notices trigger a requirement for operators to develop 'cure' or improvement plans within specified time periods. CV reviews the adequacy of these plans and is rigorous in tracking operator actions to implement the agreed improvement plans through onsite validation and monitoring at the quarterly performance meetings.

Determining performance against SDOs

The private prison operators self-report performance against SDOs to CV, and CV then determines the performance outcomes. The operators' initial reports can change over time based on the outcome of other processes and investigations, including:

• data adjustments initiated by operators following their own review processes, investigation of particular incidents and finalisation of pending drug test results
• validation reviews of operator performance data by DJR
• coronial investigations into deaths in prisons which can extend over years
• determinations issued by the Commissioner based on detailed examination of particular incidents.

This means that there is often a delay between the reporting of an incident and finalisation of the performance outcome for a particular SDO. As an example, the SDO relating to assaults on prisoners by prison staff requires the prison operator to refer the incident to Victoria Police and investigate the alleged assault to either prove or dismiss the allegation. Port Phillip reported no assaults by staff on prisoners in 2010–11. CV later assessed Port Phillip as failing this SDO following an OCSR review of an incident.

The new contracts address the lengthy delays in determining contractual outcomes for prisoner deaths by allowing the Commissioner to issue a financial penalty, known as a 'charge event', when the Commissioner has sufficient information to establish that the operator's failure to meet contractual requirements contributed to the death. The Commissioner does not need to wait for a coroner's finding or other investigations to issue a charge event for a prisoner death.

The Commissioner and Deputy Commissioner also receive requests from both public and private prisons for special consideration when they assess final performance against particular SDOs. This is often the case for SDO 3 (out-of-cell hours). Consistent with the SDO guidelines, prison operators regularly request retrospective approval from the Deputy Commissioner for instances where they have restricted prisoners to their cells or units, known as a 'lockdown'. The Deputy Commissioner must decide whether the request is reasonable—for example, after a serious incident to maintain safety—and, if the request is approved, CV excludes the lockdown from SDO calculations.

CV gave both operators time to phase in the use of a new method for measuring performance against SDO 23 (case management) in 2016–17. This is because CV had imposed a new measurement tool for use across the system without properly considering the contractual implications for the private operators—see Appendix C. This situation demonstrates how important it is for CV to consider the contractual implications of any planned substantive changes in system-wide standards, initiatives or performance measures.

CV's oversight of occupational health and safety performance

Victoria's Occupational Health and Safety Act 2004 and associated regulations provide the framework for ensuring a safe workplace. In the context of prisons, the obligation for ensuring the health and safety of workers, prisoners and others resides with the employer, in this case G4S and GEO. However, the contracting of correctional services to private operators does not completely remove the state's obligations under the Occupational Health and Safety Act 2004.

Although CV has an SDO measuring occupational health and safety (OHS) performance, it does not apply it to private prisons. We found that CV had minimal oversight of the private operator's OHS performance. In addition, CV did not collect key information on how the private operators manage their OHS obligations, for example, the number of notices issued by WorkSafe Victoria under the Occupational Health and Safety Act 2004.

We raised these issues with CV during the audit and it has now increased the OHS reporting requirements for private prisons. This is a positive step to improve CV's oversight and better understand whether private operators are meeting their OHS obligations.

2.3 Prison performance data

Accurate and reliable performance data is important for transparent reporting on prison operations and costs. We reviewed CV's processes to obtain assurance on the accuracy and completeness of this data, with a focus on safety and security performance data.

Validation of incident data

The Prisoner Information Management System (PIMS) is CV's system for recording incidents in Victorian prisons. In 2016–17, there were 24 335 incidents recorded in PIMS for the men's prison system, a rate of 3.8 incidents per prisoner. This has increased from 10 400 incidents recorded in 2010–11, a rate of 2.4 per prisoner. Of the 2016–17 incidents, 5 per cent were notifiable, that is, the most serious incidents, described further in Part 3.

Every incident entered into PIMS, across both public and private prisons, is classified according to the nature and severity of the incident. CV staff manually check the details of each incident, which provides assurance that incidents have been accurately classified and recorded. It is important for the private operators to classify incidents correctly because many of the incidents recorded in PIMS are relevant for assessing the operator's performance which, in turn, has financial implications for the state and the operators.

Between December 2016 and May 2017, Port Phillip had 272 incidents which required reclassification (9.1 per cent of its incidents), and Fulham had 133 incidents (8.1 per cent of its incidents). System-wide in men's prisons, 8.7 per cent of incidents required reclassification. Reclassification of PIMS incidents does not always mean there is an error. Rather, the classification may change if further information becomes available, for example, a prisoner requiring hospital admission after an incident.

SDO data validation

CV regularly validates the SDO performance data self-reported by the private prison operators to ensure the accuracy of the results.

To do this, CV visits each prison and checks a sample of reported SDO data against source records. These records include the daily logbooks held in every prison accommodation unit, training and work attendance registers, and lab reports supporting urinalysis results.

CV is introducing a new approach to its validation of SDO data for public and private prisons in response to an external review that it commissioned in 2016. The review found that:

• there was too much focus on data verification, rather than data and trend analysis
• the length of time taken to collect, report and verify data prevented real‑time responses to issues
• prisons should assume more responsibility for data accuracy
• there was inconsistency in validation efforts across public and private prisons
• a streamlined data collection method was required because of the different data systems in use
• there was a need for a new risk-based validation method.

A CV reform project intends to establish a new risk-based methodology for validating self-reported data, including the frequency of validation required. The new contracts also include KPIs related to data accuracy, which shifts responsibility to the private operators.

Initial contracts

Under the initial contracts, the former OCSR validated SDO data annually. CV took over responsibility for data validation in 2015–16 and now performs monthly validation checks.

CV's validation of Port Phillip's SDO data for 2015–16 resulted in changes to reported performance data for five SDOs. The validation of Fulham's SDO data for 2015–16 resulted in changes to reported performance data for three SDOs. The changes did not affect the overall pass or fail rate of either prison.

CV's SDO calculator

CV's SDO calculator is a complex macro-driven Excel spreadsheet developed as the primary analytical tool to generate SDO and KPI performance reporting across all prisons. The spreadsheet performs both data storage and calculation. We reviewed the 2015–16 SDO calculator and found that it accurately computed the performance outcomes based on the input data.

However, there are numerous procedures required to manually input, verify and modify the source data. We are also concerned that the SDO calculator is highly customised, with complex design and bespoke back-end scripting. While there are basic usage guides in place and staff who are familiar with its operation, the primary capability and knowledge of the design and coding behind the calculator is limited to a single CV employee.

This dependence on one staff member potentially creates a single point of failure for the entire SDO performance reporting process. In addition to this, there are risks of untracked data changes to the data and issues with the functionality of the data.

Data analysis

Easy access to and effective use of relevant and reliable data is critical for CV to make informed decisions about current risks, emerging trends, operational requirements and strategic planning decisions in the corrections systems. A 2016 review of CV's SDO validation and review process found that CV has limited time to undertake meaningful analysis for the quarterly performance meetings.

CV regularly collects a significant volume of data on the operation and performance of both public and private prisons. However, the various information systems do not easily facilitate data analysis and the multiple systems do not readily interface, making it difficult to integrate data and produce meaningful analyses.

E*Justice holds a range of prisoner information and overlaps with PIMS. It records prisoner risk ratings, daily counts of prisoners (musters) and prisoner movements among other important information. Victoria Police uses E*Justice data via a different interface.

Consistent with our previous audits Managing Community Corrections Orders (2017) and Administration of Parole(2016),we found that CV's data is managed across several divisions and is contained in disparate legacy systems. Data is stored in CV's two main systems—PIMS and E*Justice—as well as over 200 locally managed 'shadow systems' such as Excel spreadsheets. Vendor technical support has expired for both PIMS and E*Justice.

There are clear opportunities to improve CV's data analytics capability, to better understand the performance challenges facing the prison system. For example, using data analysis to identify trends and potential correlations between prisoner profiles, the proportion of prisoners with psychiatric conditions, and incident trends would help CV to develop strategies for managing risks, particularly the higher instances of violence across the system. Another example is CV's current inability to analyse and use data to produce evidence on the impact of 'double-bunking' on prison incidents and risks. Double-bunking is when a cell designed for single occupancy is changed to accommodate two prisoners on a set of bunks.

We have seen isolated evidence of recent proactive data analysis by CV. For example, in August 2017, CMB analysed data at Port Phillip and found a potential correlation between 'shivs' (improvised knives) and the number of positive urinalysis results and tobacco-related incidents in the same location. As a result, CMB issued a contract administration note requiring Port Phillip to develop and implement mitigation strategies. Port Phillip provided an action plan to monitor and mitigate these issues, and CMB is monitoring its ongoing implementation.

CV recognises the weaknesses of using outdated legacy systems. However, it has been unsuccessful in securing funding for its preferred replacement—an integrated offender management system. In lieu of an integrated system, CV has developed an Excel-based incident-analysis tool that integrates some of the data between PIMS and E*Justice. The private prison operators find this tool valuable as it allows them to interrogate incident data in detail.

2.4 Prison performance reporting

Performance analysis and discussions

CV meets quarterly with the private prison operators to consider a wide range of performance information and data. These meetings are open and collaborative. They identify key issues and trends, and allow CV to hold the operators to account for their performance. CV also provides the Commissioner with a monthly report highlighting key performance data and trends.

In 2014, CMB implemented the Private Prison Reporting Framework (PPRF) to capture a range of data not directly related to SDOs or KPIs. The PPRF provides CV with greater oversight of prison operations and performance, including the following information:

• workforce data—sick leave, staff turnover, staff disciplinary actions
• prisoner movement data
• prison searches for contraband and drugs
• prison visits
• prisoner disciplinary processes.

CV also contributes information to the PPRF, including the results of operating instruction and other compliance audits. The PPRF allows CV and the prisons to consider a broader range of information when monitoring and analysing trends.

Public reporting on private prisons

Transparent public reporting on the performance of the prison system is important for making DJR and private prison operators accountable and for building public trust in the system.

The limited information reported in DJR's annual report and Budget Papers provides little real insight into the operation of the prison system and no detailed information on the relative performance of individual prisons, including private prisons. The exception to this is the detailed performance information that DJR publishes on urinalysis drug results in individual prisons.

There is limited Victorian public reporting on private prison contracts, operations and performance. Other jurisdictions provide more detailed information, for example, the Department of Corrective Services in Western Australia publishes an annual report on the performance of each private prison, including detailed information about its service and financial performance.

Aside from ad hoc reviews or investigations by other integrity bodies such as the Victorian Ombudsman, there is very limited information in the public domain about the performance of private prison operators. JARO reports are not publicly available because they provide internal assurance rather than external oversight. Western Australia and New South Wales have independent inspectorates that monitor prisons and report publicly on this work. There is limited public transparency about the performance of Victoria's prisons.

In 2014, DJR and DTF advised government that they would work together to identify more suitable Budget Paper 3 measures to improve the transparency of the corrections system. To date, there have been no significant changes to those measures. CV advises that it is currently working on options to improve performance measures but that their implementation is dependent on a new integrated offender management system.

Australia's ratification of the Optional Protocol to the Convention against Torture and other Cruel, Inhuman or Degrading Treatment or Punishment (OPCAT) in December 2017 will increase the scrutiny of all places of detention in Victoria, including prisons. OPCAT is an international human rights treaty that aims to prevent abuse of people in detention by opening these places up to independent inspections by United Nations experts and independent local inspection bodies. Victoria is yet to finalise its implementation of OPCAT, but the November 2017 Victorian Ombudsman report Implementing OPCAT in Victoria: Report and inspection of the Dame Phyllis Frost Centre demonstrates the increased public scrutiny that all prisons should expect.

2.5 Performance results

GEO and G4S largely met the state's service delivery expectations for correctional and accommodation services over the life of the original contracts for Fulham and Port Phillip.

Accommodation services

The private operators met the state's requirements for accommodation services over the life of the initial contracts. However, these contracts included no SDOs directly related to facilities maintenance and specified only high-level obligations for repair and maintenance. The contracts required the operators to keep the facilities 'in good and substantial repair and condition' but did not define what this meant.

DJR advised government during negotiations for the new contracts that the assets and facilities at Port Phillip and Fulham were structurally sound and fit for purpose. DJR provided this advice based on its detailed asset condition surveys that also identified many defects in the facilities. The defects were rectified prior to commencement of the new contracts at the operators' expense, or under the new contracts as part of the asset life cycle management regime funded by the state.

Port Phillip's SDO results and performance payments

G4S has consistently underperformed at Port Phillip against the state's expectations between 2010–11 and 2016–17 in three key measures:

• SDO 2—prisoner assaults on staff or other persons
• SDO 6—prisoner-on-prisoner assaults
• SDO 8—positive urinalysis results.

See Part 3 for more information on these performance results.

Figure 2B shows G4S's performance in meeting SDO thresholds since 2010–11 and the performance payment outcomes for each year.

Ten of Port Phillip's SDOs in the old contract require 100 per cent compliance and failure can result from a single instance of noncompliance, regardless of the overall performance rate.

Figure 2B
Port Phillip SDO performance

Year	SDO thresholds achieved	Failed SDOs	Performance payment outcome
2010–11	17 of 19 89.5 per cent	Assaults on staff or other persons Assaults on prisoners by staff	Reduced by 3 per cent
2011–12	15 of 19 78.9 per cent	Assaults on staff or other persons Assaults on prisoners by other prisoners Assaults on prisoners by staff Timely prisoner disciplinary hearings	Reduced by 13 per cent
2012–13	18 of 21 85.7 per cent	Assaults on staff or other persons Prisoner participation in pre‐release programs Provision of chronic health care plans	Full performance payment
2013–14	16 of 22 72.7 per cent	Assaults on staff or other persons Assaults on prisoners by other prisoners Assaults on prisoners by staff Random general urinalysis Prisoner participation in pre‐release programs Timely prisoner disciplinary hearings	Reduced by 22.5 per cent
2014–15	19 of 22 86.4 per cent	Assaults on prisoners by other prisoners Random general urinalysis Provision of chronic health care plans	Reduced by 3 per cent
2015–16	18 of 22 81.8 per cent	Assaults on prisoners by other prisoners Random general urinalysis Prisoner medical screening within 24 hours Timely completion of prisoner health care plans	Reduced by 17 per cent
2016–17	17 of 21 80.9 per cent	Escapes Assaults on prisoners by other prisoners Random general urinalysis 'At risk' assessments within two hours	Reduced by 20 per cent

Source: VAGO based on information from CV.

G4S has only received the full performance payment once in the last seven years. There are a number of examples where CV has worked cooperatively with G4S to address its performance.

CV regularly meets with G4S to discuss performance, both formally and informally. Along with these structured and ad hoc conversations, there have been two occasions since mid-2014 when the Commissioner has formally 'called in' Port Phillip management to discuss particular concerns about its operational performance.

G4S responded to CV's concerns, whether raised formally or informally, by developing strategies and plans. CV rigorously monitored the implementation of agreed actions.

CV's increased oversight and scrutiny of private prisons is positive, however it is not possible to determine whether it has resulted in improved performance, as CV and the prison operators do not routinely evaluate the outcomes and impacts of their improvement strategies. Section 3.5 discusses an example of this relating to violence-reduction strategies.

In May 2017, after an escalation in serious incidents, G4S agreed to a peer review process at Port Phillip initiated by CV, designed to review the current operations and identify areas for improvement. This collaborative process sought to identify issues and share lessons learnt from the broader prison system. Through the process, CV's subject-matter experts and management from other prisons reviewed a range of different areas including:

• the spread and effectiveness of middle management
• barrier and perimeter security
• prisoner cohort management, including placement and vacancy management
• management of external escorts
• culture, including staff and prisoner interactions
• training, including for new recruits, refresher training, de-escalation, use of force and incident management
• compliance management
• implementation of the action plan to improve Port Phillip's PIU.

The reviews completed to date identified some areas of positive performance, including the management of visits, external escorts and the relationship between the prison and CV's sentence management function. They also identified about 40 areas for improvement, which included:

• communication between management and general duties staff
• development of supervisors, availability of supervisors in the unit and visibility of senior management
• balancing compliance and prisoner management activities
• use of resources in the PIU, particularly dog handlers and staff escorting prisoners outside of the prison
• development and succession planning for staff.

It is positive to see Port Phillip and CV working together to identify issues and share the lessons learnt across the system.

Fulham's SDO results and performance payments

Fulham has generally performed well in meeting the state's expectations over the life of the initial contract, apart from some issues in 2012–13. There were also issues in 2016–17, the first year of the new contract. The prison has accepted remand prisoners since September 2015, and in December 2017 they made up around 25 per cent of all prisoners at Fulham.

Figure 2C shows GEO's performance against the SDO thresholds since 2010–11 at Fulham and the performance payment outcomes for each year.

Figure 2C
Fulham SDO performance

Year	SDO thresholds achieved	Failed SDOs and KPIs	Performance payment outcome
2010–11	16 of 17 94.1 per cent	Unnatural death	Reduced by 15 per cent
2011–12	16 of 17 94.1 per cent	Assaults on staff or other persons	Reduced by 10 per cent
2012–13	13 of 17 76.5 per cent	Assaults on staff or other persons Assaults on prisoners by other prisoners Completion of education and training programs Prisoner participation in pre‐release programs	Reduced by 15 per cent
2013–14	All 18 SDOs 100 per cent	Nil	Full performance payment
2014–15	All 18 SDOs 100 per cent	Nil	Full performance payment
2015–16	16 of 18 88.9 per cent	Prisoner escapes Timely completion of prisoner health care plans	Reduced by 15 per cent
2016–17 (new contract which includes KPIs)	17 of 18 SDOs 94.4 per cent	Assaults on prisoners by other prisoners	Reduced by 3.7 per cent
	13 of 16 KPIs 81.3 per cent	KPI—Provision of timely primary health services KPI—Performance data KPI—Providing plans and reports

Source: VAGO based on information from CV.

GEO received the full performance payment twice in the last seven years. It made a submission to CV for reconsideration of the 2012–13 outcome, based on the fact that one incident in June 2013 resulted in 13 assaults on staff. GEO asserted that this one incident disproportionately affected otherwise good performance, but CV did not alter the performance outcome. GEO did not receive the full payment for 2016–17, the first year of operation under the new contract.

Back to top

Prisons are complex and high-risk environments with constant threats to safety and security. Port Phillip and Fulham are integral to the safe and secure operation of Victoria's prison system—in December 2017, they housed around 26 per cent of the state's male prisoners. The private operators and CV both have responsibilities for maintaining and improving the safety and security of these prisons.

Private operators are responsible for the day-to-day running of the prison and complying with legal, contractual and other requirements. This includes a suite of performance measures on safety and security risks.

CV is ultimately responsible for the safe and secure management of all prisoners and for monitoring and managing the prisons' performance.

In this part of the report, we examine whether the private prison operators and CV are effectively managing their risks and performance relating to key safety and security measures.

3.1 Conclusion

Private prisons are not always meeting the state's service and performance requirements to run safe and secure prisons, particularly in relation to assaults at both prisons and drug use at Port Phillip. This is consistent with system-wide performance.

There have been serious incidents at both Port Phillip and Fulham which, in some instances, has exposed weaknesses in how they manage safety and security risks. Neither private prison operator is investigating serious incidents using methods that effectively identify the root causes.

Prisoners assaulting other prisoners is increasing across the prison system, as prisoner numbers and their complexity increases. Private operators and CV have violence-reduction strategies, but there is a need to better evaluate these and share the lessons learnt.

3.2 Serious incidents

Serious safety and security incidents such as deaths, riots and escapes attract considerable public attention and threaten the safety of prisoners, staff and the community. While these incidents are not always avoidable, it is crucial that:

• there are effective risk management and compliance programs to reduce the risk of serious incidents
• responses to incidents are effective
• investigations identify the root causes of incidents, and preventative actions are identified and implemented.

Serious incidents in private prisons

CV classifies incidents into two categories—notifiable incidents and reportable incidents. Notifiable incidents are the most serious, including deaths, escapes, fires, riots and serious assaults. Prisons must report notifiable incidents to CV within 30 minutes. Reportable incidents are less significant and prisons must formally report these to CV within 24 hours.

Figure 3A compares the number of notifiable incidents per 100 prisoners (therefore standardised for prisoner growth) at Port Phillip and Fulham to the average of other prisons with the same security classification. It shows that Port Phillip's rate of notifiable incidents is similar to the average of other maximum‑security prisons from 2011–12 to 2013–14. In 2014–15, Port Phillip's rates decreased and subsequently increased in 2015–16 and 2016−17, potentially related to the increase in remand prisoners after the MRC riot.

The increase in the average rate of notifiable incidents at other maximum‑security prisons in 2014–15 is partly due to a spike in notifiable incidents at MRC due to the riot on 30 June 2015. MAP's rate of notifiable incidents also increased between 2014–15 and 2016–17, on average 199 per cent higher than Port Phillip's rate of incidents. The high rate of notifiable incidents at MAP is not unexpected, as it provides assessment and orientation services for all new male prisoners and provides statewide psychiatric services.

Fulham's rate of notifiable incidents fluctuated from 2010–11 and has been lower or the same as the average of medium-security prisons in all years, excluding 2011–12. The spike in 2011–12 is attributable to a riot in January 2012 involving multiple prisoners. The increased average rate for medium-security prisons from 2014–15 to 2016–17 is partly due to the high rates at the newly opened Middleton and Karreenga facilities. In 2016–17, Fulham's rate of notifiable incidents reached the highest level since 2010–11 at 10.5 per cent.

Figure 3A
Notifiable incidents per 100 prisoners at Port Phillip and Fulham compared to maximum- and medium-security prisons, 2010–11 to
2016–17

Source: VAGO based on data provided by CV.

The most common notifiable incidents are consistent across private and public men's prisons. These are:

• 'good order' incidents, which cover a wide range of events that may risk the good order of the prison but don't fall into another category—for example, seizure of contraband such as mobile phones or weapons, procedural failures such as keys left unattended and use of force
• medical incidents where a person is hospitalised
• assaults on staff—CV classifies all assaults on staff where an injury occurs as notifiable incidents, but prisoner-on-prisoner assaults are only classified as notifiable if admission to hospital is required
• incidents relating to drugs, alcohol or other substances—such as possession or selling—but not including positive drug tests.

Preventing serious incidents

Preventing serious incidents requires effective risk management and compliance. Both private prisons have adequate risk management and compliance frameworks, policies and processes. These policies and processes are consistent with contractual requirements and standards. While the private operators are not required to comply with government guidelines on risk management, their processes follow the same principles.

Both private prisons are accredited for a range of Australian and international standards related to risk management and compliance, which provide a level of assurance to the state that the prisons have effective management systems.

Despite both private prisons having processes in place, there are examples in each where compliance failures have contributed to serious incidents. Prison operators have taken appropriate action to address the weaknesses identified by incident investigations.

Port Phillip

In 2014, G4S developed an improvement plan in response to CV's concerns about its performance at Port Phillip. Part of this plan involved addressing deficiencies in risk management and compliance. Remedial actions implemented included:

• locating the G4S national risk and compliance manager at Port Phillip to review the compliance processes, with direct reporting to the general manager
• increasing senior management's oversight of compliance and reviewing the compliance model with a focus on violence reduction
• restructuring weekly compliance meetings to focus on key risks
• improving the analysis of violent incidents and trends.

CV validated the implementation of these actions, and we found that this has improved risk management and compliance at Port Phillip. In particular, the weekly compliance meetings are an effective forum for overseeing risks, with a strong focus on reducing violence.

It is difficult to attribute performance outcomes relating to assaults to particular actions or interventions when there are so many contributing factors. However, the SDO performance data in Section 3.4 demonstrates some improvement in prisoner assaults on staff in the second half of 2014–15 that could be linked to the improvement plan. Improvement in prisoner‑on‑prisoner assaults in 2014–15 was not sustained and performance declined over the following two years.

In 2016 and 2017, repeated incidents of noncompliance with firearms procedures and licensing highlighted gaps in Port Phillip's risk and compliance management. Port Phillip developed and implemented corrective action plans, and CV continues to monitor compliance with these actions. We observed rigorous checking of firearms licencing at Port Phillip and an appropriate focus on this risk.

Fulham

In 2016, there were two serious incidents that highlighted gaps in Fulham's compliance activities. An incident involving prisoners growing cannabis seedlings on site exposed weaknesses in Fulham's search procedures and compliance checks. Similarly, a failure to carry out control room procedures contributed to an escape in April 2016.

As a result of these incidents and the contractual default notices issued by CV, Fulham developed and implemented action plans to address the identified gaps, including those in compliance processes. CV monitors and validates the implementation of these action plans.

Investigating serious incidents

CV's operations division and the prisons work collaboratively to respond to and investigate serious incidents. Figure 3B provides a summary of this process.

Figure 3B
Summary of notification and investigation processes after a serious incident

Note: PIMS is the system used to record incidents in prisons.

Source: VAGO.

Internal management reviews

Internal management reviews (IMR) are investigations undertaken at a prison following the most serious incidents like deaths, serious assaults and escapes. They can also be used to investigate workforce-related incidents or issues.

The private prisons have their own policies and processes for initiating and conducting IMRs, which differ to public prisons. In 2016–17, Port Phillip conducted six IMRs and Fulham conducted eight. Prisons submit IMRs to CV for review and monitoring of any recommendations and corrective actions.

Root-cause analysis is a process used to identify the underlying causes of system failures. It provides the information needed to solve problems and address failures.

We reviewed 12 IMRs at Port Phillip and 15 at Fulham relating to incidents between 2010 and 2017, and found that none used an investigation methodology specifically focused on identifying root causes. The terms of reference in only two IMRs explicitly stated that the review should attempt to establish the root cause of the incident, but the investigations did not reflect this requirement. This creates the risk that investigators will not identify the root cause of the incident, and that preventative actions will not be appropriately targeted.

IMRs often focus primarily on whether staff involved in incidents complied with operating procedures and policies. While it is important to identify whether procedural noncompliance was a contributing factor, investigations must analyse the reasons for the error or noncompliance. We found IMRs often attribute an incident to complacency and look no further into the underlying reasons for this. Prisons could improve their investigations and ensure that investigators understand the human factors that contribute to incidents and the cause of human errors.

The standard of IMRs from Port Phillip varies, and CV recently identified that two IMRs lacked detail and missed key factors that contributed to the incident. CV requested Port Phillip address this concern, but did not specify that Port Phillip adopt a root-cause-analysis methodology.

The new contracts require the operators to use contemporary root-cause analysis processes to investigate serious adverse incidents. The operators need to ensure they have the processes and capability to comply with this requirement, and CV should monitor the operators' performance. Fulham has not reviewed its IMR process since the new contract commenced to ensure it complies with this requirement. Port Phillip has acknowledged that its investigation process needs improvement but has not made any progress with this.

IMRs in public prisons Level 1—completed by prison staff. Level 2—completed by senior prison staff and senior staff from another prison. Level 3—completed by CV's Systems Performance Branch, senior staff from another prison and a subject-matter expert.

In the public system, there is a more clearly defined process for completing IMRs. The seriousness of the incident determines who conducts the IMR and what process is used.

We reviewed four IMRs from public prisons—one Level 1 IMR and three Level 2 IMRs—and found that, like the private prisons, the IMRs did not use a defined root-cause-analysis methodology. CV does not view the IMR as a process for conducting a root-cause analysis. However, there are no other processes or guidelines that effectively document how to conduct serious incident investigations to ensure that root causes are identified. As with the private prisons, this creates a risk that preventative actions will not be effective.

IMRs are not the only post-incident process, as shown in Figure 3B. We found that CV has extensive involvement in post-incident support and review. Briefs provided to the Minister for Corrections or Commissioner regarding the incident and the implications were thorough and effectively highlighted key issues. This allows CV to properly consider any potential system‑wide safety and security risks, and make informed decisions on any contractual implications for private operators and actions required to prevent future incidents.

JARO's role in reviewing serious incidents is the same in private and public prisons. In 2016, JARO conducted 72 reviews—56 reviews of deaths, seven reviews of serious incidents, and nine thematic reviews. More recently, JARO has changed its focus from reviewing serious incidents to conducting thematic reviews of the system in an attempt to respond to risks proactively. JARO's reviews objectively assess serious incidents and require CV to respond and develop an action plan.

3.3 Security performance

Physical security refers to the buildings, external walls, internal fences and other infrastructure that prevents prisoners from escaping and minimises potential for harm. It includes other technology and systems, like barrier control and closed‑circuit television (CCTV). It differs from dynamic security, which is the interaction staff have with prisoners, the relationships they develop and their awareness of what is going on in the prison.

The riot at MRC in 2015 exposed significant weaknesses in the prison's physical security. CV's Security Standards Unit (SSU) has subsequently completed physical security risk assessments at all prisons, including Fulham and Port Phillip.

CV established the SSU in 2013 within the Security and Intelligence Branch to:

• set physical security standards for public prisons
• provide expert advice on security
• conduct security risk assessments
• manage locksmiths for public prisons and other security-related functions.

The SSU was not initially involved in setting standards and monitoring performance in the private prisons—this was a conscious decision by CV to manage risk allocation under the contracts. SSU is becoming more involved in supporting private prisons, including conducting detailed risk assessments at both prisons in 2016. It is important that SSU maintain some oversight of physical security at the private prisons, to ensure consistency and to share lessons learnt and information across the corrections system. It is also important that CV has assurance that the private prison operators are appropriately managing risks identified by SSU.

Port Phillip's physical security risks

SSU's January 2016 risk assessment at Port Phillip identified 49 risks. It rated seven risks as very high or high, with three requiring immediate corrective action.

These risks are on the Port Phillip security risk register developed by SSU, which documents ratings and accountability for each risk. Many of the security improvements required significant investment and were part of the new contract negotiations. For example, during the course of this audit, there have been major improvements to the quality and effectiveness of the CCTV systems at the prison and upgrades to the control room.

G4S and CV have adequately identified physical security risks and are in the process of addressing them. G4S conducted a review of its physical security risk register in October 2017.

Fulham's physical security risks

SSU's January 2016 risk assessment at Fulham identified 48 risks. It rated 13 as high or very high, with six requiring immediate corrective action.

As with Port Phillip, Fulham has a security risk register developed by SSU. There is a program of work underway to address these risks. GEO has integrated the physical security risks into its prison risk register, however there has not been a review of the risk assessment and register due to a lack of clarity on whether the prison or SSU is responsible for this. CV has since clarified that GEO is required to review and update the risk assessment and register, with support from SSU if requested, and that CMB will monitor progress on this.

CV definitions of escape severity levels Level 1—a prisoner absents him- or herself from supervised leave, for example, a work group, or from an open prison perimeter or fails to return from unescorted leave. Level 2—a prisoner, while being escorted by an escort officer, absents him- or herself from the view of that officer for any period of time without lawful authority. Level 3—a prisoner escapes from within the secure perimeter or while under escort from the Security and Emergency Services Group (SESG), or absents him- or herself for any period of time without lawful authority while under supervision of SESG.

Escapes

Keeping prisoners in secure custody is the core objective of the corrections system and is a fundamental measure of a prison's security. Since 2010, there have been 25 escapes from Victorian men's prisons. Escapes are categorised according to the seriousness of the security breach—of the 25 escapes, 21 were Level 1 escapes, which are the least serious. While escapes occur at both public and private prisons, 12 of the escapes occurred at a single minimum-security public prison.

Fulham and Port Phillip have each experienced a serious escape in the last two years, highlighting key security failures.

The SDO threshold for escapes at Fulham and Port Phillip did not change during the term of the original contract, however the new contracts impose a tougher threshold to minimise escapes.

Port Phillip

Since 2010, there have been two escapes from maximum-security men's prisons in Victoria, one of which was from Port Phillip. They both occurred during prisoner escorts outside of prison walls.

The Level 2 escape at Port Phillip involved a prisoner escaping during a supervised hospital visit in January 2017. The investigation into this incident found procedural failures in prisoner supervision and restraint, and as a result CV issued a contractual penalty to G4S. This triggered the development of an improvement plan, and G4S is currently implementing corrective actions.

Fulham

The 2016 escape of two prisoners from Fulham was a serious safety and security failure, and resulted in Fulham failing to meet its SDO threshold. It was the only Level 3 'over the wall' escape in the Victorian system in the last seven years. CV and GEO investigated the incident and uncovered several contributing factors, including human error and procedural failures, as well as infrastructure and environmental factors.

As a result, CV enforced a financial penalty and required GEO to implement an agreed plan to address the risks. CMB monitors the implementation of this plan and regularly updates the Commissioner on progress. GEO has implemented all the short-term actions, however it failed to provide a timely response to CV's request for upgrades to the perimeter. This resulted in CV issuing a minor service failure notice.

3.4 Safety performance

We examined the safety performance of the privately operated prisons against CV's performance expectations and compared their performance to that of similar public prisons where appropriate. The private prisons mostly met the safety incident thresholds set by the state. However, consistent with system‑wide trends, the private operators are experiencing an increase in assaults, which often exceed the performance thresholds.

In this section, we analyse performance of the private operators for assaults, deaths, self-harm and drug use in prison. We reviewed data on the actual number of incidents and the severity of the incidents, as well as SDO performance, which is recorded in 'incident points'. Incident points are calculated according to the severity of the incident. The more severe an incident, the more incident points a prison will incur—for example, for SDO 6 (prisoner-on-prisoner assaults):

• a Level 2 assault will incur 12 incident points
• a Level 1 assault will incur 4 incident points
• a Level 0 assault will incur 1 incident point—inclusion of Level 0 assaults in incident point calculations commenced in 2014–15.

CV introduced incident points for SDO 2 in 2012–13 for the private prisons. Prior to this, SDO 2 reporting did not capture the severity of incidents. For this reason, we have only analysed SDO 2 results from 2012–13 to 2016–17.

It is important to note that CV sets different performance thresholds for each prison—these consider the prison's profile, any commercial negotiations (for private prisons) and potential 'stretch' targets (for public prisons). Therefore, we have only compared the private prisons' performance against their own thresholds.

Assaults in prison

There are inherent risks in the prison system that increase the likelihood of assaults. Prisoners often have complex needs. Many have drug and alcohol problems, mental health conditions, disabilities, gang affiliations and histories of disadvantage and trauma, or may be violent.

We found that, like the broader prison system, private prisons are not consistently meeting the thresholds set by the state for prisoner-on-prisoner assaults and, to a lesser extent, prisoner-on-staff assaults. This is particularly the case for Port Phillip and most maximum-security prisons.

Prisoners assaulting staff

The majority of assaults against staff occur at maximum-security prisons. Figure 3C shows that rates of assaults on staff have fluctuated at maximum‑security prisons and spiked in the first quarter of 2015–16 after the tobacco ban and MRC riot of 30 June 2015. These events resulted in the dispersion of the remand population and disruption throughout the corrections system, which contributed to the increased assaults. A second spike occurred in the second quarter of 2016–17 due to increased assaults at MAP and Barwon. CV does not attribute this to any particular factor. Medium- and minimum‑security prisons have small numbers of assaults, so identifying trends is more difficult.

Figure 3C
Assaults on staff per 100 prisoners, by prison security classification, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Figure 3D shows that the least serious Level 0 assaults occur at a greater rate than more serious assaults across Victoria's male prisons. While all violence directed towards staff must be treated very seriously, it is reassuring that the majority of incidents do not involve an injury. Reporting minor incidents where there is no injury can be an indicator of a positive reporting culture, which allows prison management to intervene and prevent a more serious incident.

Figure 3D
Assaults on staff per 100 prisoners across all men's prisons, by assault level, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Level 0 assault on staff—occurs where there is an assault but no injury. Level 1 assault on staff—results in an injury but does not require the victim to be admitted to hospital. Level 2 assault on staff—results in the victim being hospitalised.

Port Phillip

Assaults on staff occur at a greater rate in maximum-security prisons than medium or minimum-security prisons. The number of assaults on staff at Port Phillip per 100 prisoners fluctuated between 2013–14 and 2016–17, as shown in Figure 3E. MRC and the average rate of other maximum‑security prisons also fluctuated during this period, more so than Port Phillip in the last two years.

Figure 3E
Assaults on staff per 100 prisoners at Port Phillip and MRC compared to the average of other maximum‑security prisons (excluding Port Phillip), 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Between July 2013 and June 2017, Level 0 assaults made up 62 per cent of all assaults on staff at Port Phillip. Level 1 assaults made up 37 per cent and Level 2 assaults were less than 1 per cent.

Port Phillip met its threshold for SDO 2 (assaults on staff) in 55 per cent of quarters over the period 2012–13 to 2016–17. Port Phillip's performance threshold for this SDO did not change during the period we examined, and there was no adjustment in the new contract. Figure 3F shows that Port Phillip's incident points experienced several spikes in 2012–13 and 2013–14, and from 2012–13 to 2016–17 Port Phillip's incident points were higher than MRC's in 70 per cent of quarters.

Figure 3F
Performance results for SDO 2 (assaults on staff) at Port Phillip and MRC compared to the average of other maximum-security prisons (excluding Port Phillip), 2012–13 to 2016–17

Source: VAGO based on data provided by CV.

Fulham

Fulham has a low number of staff assaults per month, which makes it difficult to identify trends and, therefore, we have not included analysis of this data. For example, in quarter four of 2016–17 there were four staff assaults and an average prisoner population of 862.

Between 2013 and 2017, across all medium-security prisons:

• 74 per cent of staff assaults were Level 0 assaults (causing no injury)
• 26 per cent of staff assaults were Level 1 assaults (causing injury but not requiring admission to hospital).

There were no Level 2 assaults (requiring admission to hospital) in any of the medium-security prisons during this time. Fulham experienced an increase in the number of assaults on staff in 2016–17, which was due to an increase in Level 0 assaults. The increased reporting of less serious assaults may indicate a positive reporting culture.

Figure 3G shows Fulham's incident points, which are used to calculate SDO performance for staff assaults and reflect the severity of incidents. Compared to Marngoneet—a similar medium-security prison—Fulham's performance is worse in 65 per cent of the quarters from 2012–13 to 2016–17. The spike at Fulham in quarters three and four of 2012–13 was caused by a serious assault in which multiple staff were injured.

Figure 3G
Performance results for SDO 2 (assaults on staff) at Fulham and Marngoneet compared to the average of other medium-security prisons (excluding Fulham), 2012–13 to 2016–17

Source: VAGO based on data provided by CV.

Fulham met its threshold for SDO 2 (assaults on staff) for 80 per cent of quarters between 2012–13 to 2016–17. Fulham's threshold for this SDO did not change from 2012–13 to 2015–16, but it is more difficult to achieve in the new contract.

Prisoners assaulting other prisoners

There were 1 503 prisoner-on-prisoner assaults in Victorian men's prisons in 2016–17:

• 53 serious assaults where a prisoner was admitted to hospital (Level 2 assault)
• 1 162 involving a less serious injury (Level 1 assault)
• 288 assaults with no injury (Level 0 assault).

Across the prison system, the rate of prisoner-on-prisoner assault is increasing, as shown in Figure 3H. From July 2013 to June 2015, the average statewide rate of assault was 1.17 per 100 prisoners. From July 2015 to June 2017, the average rate of assault per 100 prisoners increased to 1.81.

Figure 3H
Prisoner-on-prisoner assault per 100 prisoners in men's prisons, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Maximum and medium-security prisons have both experienced increases from 2015–16, as shown in Figure 3I. From July 2013 to June 2015, maximum-security prisons had an average rate of assault of 2.0 per 100 prisoners, which increased to an average rate of 2.9 between July 2015 to June 2017. During the same period, medium-security prisons had an average rate of assault of 0.6 per 100 prisoners, increasing to 1.15.

Figure 3I
Prisoner-on-prisoner assaults per 100 prisoners, by prison security classification, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

As shown in Figure 3J, the rate of Level 1 assaults have increased more than the rate of Level 2 and Level 0 assaults. In July 2013, the rate of Level 1 assaults was 0.76 per 100 prisoners and in June 2017 it was 1.48. On average, the rate of Level 1 assaults increased by 3 per cent each month.

Figure 3J
Prisoner-on-prisoner assaults per 100 prisoners in all men's prisons, by assault level, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Port Phillip

The complexity of the prisoner cohort at Port Phillip is a significant consideration when assessing the number of assaults in the prison.

Figure 3K shows that the rate of Level 1 prisoner-on-prisoner assaults at Port Phillip increased after the MRC riot on 30 June 2015. Following the riot, there was an increase in Port Phillip's remand population up until the third quarter of 2016–17. Level 1 assaults are the most frequent type of assault at Port Phillip. Between July 2013 to June 2015, the average rate of assault was 1.74 per 100 prisoners, increasing to an average of 2.61 from July 2015 to June 2017.

Figure 3K
Prisoner-on-prisoner assaults per 100 prisoners at Port Phillip, by assault level, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

The comparison of maximum-security prisons in Figure 3L shows that Port Phillip's rate of prisoner-on-prisoner assaults is similar to MRC, excluding the 12 months after the MRC riot. Barwon, which has different infrastructure, a different profile of prisoners and less prisoner movements, has a lower rate of prisoner-on-prisoner assaults.

Figure 3L
Prisoner-on-prisoner assaults per 100 prisoners at maximum-security prisons, 2013–14 to 2016–17

Note: The dip in assault numbers at MRC in July 2015 is the result of the transfer of a substantial number of prisoners out of MRC after a riot on 30 June 2015.

Source: VAGO based on data provided by CV.

Figure 3M shows that Port Phillip's performance for SDO 6 (prisoner-on-prisoner assaults), as measured by incident points, is consistently worse than the average of other maximum-security prisons. Prior to the MRC riot, Port Phillip's incident points were higher than MRC's, aside from three quarters. From quarter one 2015–16 to quarter one 2016–17, Port Phillip's incident points increased on average 15 per cent each quarter, going from 284 incident points to 499 incident points. While the incident points do reduce in quarters two, three and four of 2016−17, they do not return to previous levels.

Port Phillip met its performance threshold for SDO 6 in 39 per cent of quarters from 2010–11 to 2016–17. There was no change in the performance threshold for this SDO at Port Phillip during the period we examined or as part of the new contract negotiations.

Figure 3M
Performance results for SDO 6 (prisoner-on-prisoner assaults) at Port Phillip and MRC compared to the average of other maximum-security prisons (excluding Port Phillip), 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

Fulham

Consistent with the system-wide trend, Level 1 assaults are the most common assault type at Fulham, however, the rate of Level 1 assaults fluctuates, as shown in Figure 3N.

Figure 3N
Prisoner-on-prisoner assaults per 100 prisoners at Fulham, by assault level, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Figure 3O shows that Fulham's rate of prisoner-on-prisoner assault is consistent with the average of other medium-security prisons, which experienced an increase in assaults in 2015–16.

Fulham agreed to take remand prisoners for the first time in 2015. Since then, both the proportion of remand prisoners and number of assaults have increased. The average rate of assault was 0.6 per 100 prisoners between July 2013 and June 2015. From July 2015 to June 2017, the average rate of assault increased to 1.1.

Figure 3O
Prisoner-on-prisoner assaults per 100 prisoners at Fulham and Marngoneet compared to the average of medium-security prisons (excluding Fulham), 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Figure 3P shows that Fulham's incident points for SDO 6 (prisoner-on-prisoner assaults) have fluctuated and increased in 2016–17. Fulham's SDO performance was worse than Marngoneet in 22 of the 28 quarters from 2010–11 to 2016–17. Fulham met its contractual performance threshold for prisoner-on-prisoner assaults in 67 per cent of quarters from 2010–11 to 2016–17.

Fulham's threshold for prisoner-on-prisoner assaults did not change under the new contracts, and it failed to meet the threshold in 2016–17 (the first year of the new contract) resulting in a reduction of its performance payment.

Figure 3P
Performance results for SDO 6 (prisoner-on-prisoner assaults) at Fulham and Marngoneet compared to the average of medium‑security prisons (excluding Fulham), 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

Assaults and remand prisoners

In 2016–17, the majority of prisoner-on-prisoner assaults at Port Phillip involved remand prisoners. Remand prisoners accounted for approximately 50 per cent of Port Phillip's population but were involved in 63 per cent of assaults (up from 26 per cent in 2013–14). Due to data limitations, we could not determine whether the remand prisoners were the victims or perpetrators of the assaults.

This over-representation of remand prisoners in assaults is potentially due to remand prisoners being new to the prison, the effects of drug and alcohol withdrawal, undiagnosed medical illness and other factors described in Part 1.

Fulham has a smaller remand population. In December 2017, they represented 25 per cent of its total prisoner population. Unlike Port Phillip, sentenced prisoners are involved in the majority of Fulham's assaults.

Assaults on prisoners by staff

CV takes a zero-tolerance approach to staff assaulting prisoners. Any proven assault by staff will result in the prison failing this SDO. Prison operators must refer any allegations to Victoria Police for investigation, as well as following their own investigations and procedures for employee misconduct. Performance against this SDO does not require a criminal conviction to prove the assault. The Commissioner determines whether an assault occurred, based on the facts and any investigations by Victoria Police or the prison, or reviews by JARO.

Between July 2010 and June 2017, there were 299 allegations of staff assaulting prisoners across all Victorian men's prisons. Prisons record all allegations but these do not affect SDO performance until proven. In addition, since July 2010, there have been seven proven incidents of staff assault on prisoners.

At Port Phillip, there were 92 allegations of staff assaulting prisoners between July 2010 and June 2017, and five proven assaults—three in 2013, one in 2012 and another in 2011. As a result, Port Phillip failed this SDO measure in 2010−11, 2011−12 and 2013−14, and received reduced performance payments.

From July 2010 to June 2017, there were 12 allegations of staff assaulting prisoners at Fulham, but none were proven. Fulham passed the SDO for assault on prisoners for 2010–11 to 2016–17.

The remaining two proven assaults during this period occurred in public maximum-security prisons.

An unnatural death is subject to the coroner's finding that the death was caused by a homicide, suicide, an accidental cause, or a drug overdose, and the coroner finds that a person or provider contributed to the death of that prisoner. A death will be treated as unknown while the coroner determines a cause of death and/or whether the prison contributed to the death. A coroner may also make an open finding when the cause of death or circumstances cannot be established.

Deaths in prison

Deaths in prison are categorised as natural, unnatural or of unknown causes. Most deaths in prison result from natural causes relating to age or medical conditions. The majority of natural deaths in Victoria's prison system occur at Port Phillip, as it provides medical services, including palliative care, for the whole system.

Unnatural prisoner deaths are not common. In some instances, they demonstrate a failure to protect a prisoner's welfare, which is a fundamental goal of prisons. All prisons have a zero target for this SDO, and any unnatural death where the coroner finds that a person or provider contributed to the death of the prisoner will result in the operator failing this SDO for the year. In addition to operators failing the SDO, the new contracts can impose a charge event for an unnatural death, resulting in financial penalties for the operator. This can occur prior to a coronial finding, where there is sufficient information to prove that noncompliance with contractual requirements contributed to the death. If CV issues a charge event, a subsequent failure of the SDO will not result in a further financial penalty.

The coroner investigates all deaths in prisons and determines whether the death was unnatural or natural. JARO also conducts reviews into deaths in custody. JARO's detailed reports include a review of prisoner management in custody and the prison's response to the incident, and may inform the coronial inquiry.

Since 2010, there have been 12 unnatural deaths across the corrections system, including two at Port Phillip and one at Fulham. There were also eight deaths throughout the system—including four at Port Phillip—where investigations by the coroner have not established a cause.

Port Phillip

In February 2013, an unnatural death occurred at Port Phillip, when a prisoner died from self-harm. A subsequent review by the former OCSR found that the operator's management of the prisoner prior to his death and management of the incident was appropriate. OCSR made one recommendation but, even if implemented, it would not have changed the outcome in this instance. The coroner's report did not attribute any blame to G4S for this death, so this did not result in an SDO failure.

The second unnatural death at Port Phillip occurred in 2015, where a prisoner went into cardiac arrest after ingesting illicit drugs. The coroner's report highlighted that G4S and CV have been able to learn from this death to reduce contraband in prisons during visits. The coroner's report did not attribute any blame to G4S for this death, so this did not result in an SDO failure.

There have been four further deaths of unknown causes at Port Phillip between 2014 and 30 June 2017. These deaths are currently under examination by the coroner, and therefore SDO outcomes have not been finalised. As noted in Figure 2A, the Minister for Corrections issued a default notice to G4S in regard to one of these unknown deaths, as CV found evidence of systematic failings.

S1 prisoners are at immediate risk of self‑harm and require observation every 15 minutes. S2 prisoners are at significant risk of self‑harm and require observation every 30 minutes. S3 prisoners are a potential risk of self-harm and require observation hourly. S4 prisoners have a history of self-harm but are not currently at risk of self-harm. There is no observation of S4 prisoners.

Fulham

In April 2011, a prisoner at Fulham died after an assault. Correctional staff did not detect his injuries for a number of hours. Investigations found that governance failures, staff complacency during and after the incident, and noncompliance with operating procedures contributed to this death. GEO failed this SDO in 2010–11 and received a reduction in its performance payment, but CV did not impose a contractual default for this unnatural death. CV required GEO to develop and implement an action plan to prevent similar incidents at Fulham, and CV monitored the delivery of the agreed actions.

Prisoner self-harm

CV assesses all prisoners when they first enter prison to determine if they are at risk of self-harm. At-risk prisoners are assigned a self-harm risk rating ranging from S1 (immediate risk) to S4 (history of self-harm but not currently at risk). This risk rating determines the level of observation required and is determined by a multi-disciplinary risk review team, which includes a mental health professional.

In July 2017, 1 513 of Victoria's male prisoners (23 per cent) had a self-harm risk rating. Port Phillip had the second-highest number of prisoners with a self-harm rating—336 prisoners. This represented 31 per cent of prisoners in Port Phillip in that month. In the same month, Fulham had 154 prisoners with a self-harm rating, representing 18 per cent of its total population.

Figure 3Q
Self-harm incidents per 100 prisoners in men's prisons, by security classification, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Level 1 self-mutilation —visible injury or injury confirmed by medical practitioner not requiring admission to hospital. Level 2 self-mutilation—injury requiring admission to hospital. Level 2A attempted suicide—not requiring admission to hospital. Level 2B attempted suicide—admission to hospital is required.

Figure 3Q shows a slight increase in self-harm rates in medium-security prisons, particularly in 2015–16 and 2016–17. Rates of self-harm at maximum-security prisons fluctuate, and there were a few large spikes in 2015–16.

There are four classifications of self-harm incidents, ranging in severity. The most prevalent is self-mutilation where there is a visible injury but hospitalisation is not required (Level 1). We analysed the incident data for Level 1 self-mutilation at Port Phillip and Fulham, and found that 59 per cent of the prisoners had an S4 risk rating. Commissioner's Requirement 2.3.1 Management of At Risk Prisoners states that there is no minimum review period to assess the risk of S4 prisoners. Instead, it is the responsibility of correctional staff and appropriate psychological services to report changes in the behaviour or mental state of S4 prisoners, if detected. This is consistent with the operating instructions at Fulham and Port Phillip. The S4 segment of the population is over-represented in self-harm incidents, potentially due to a lack of ongoing assessment of their risk levels.

Port Phillip

Figure 3R shows that Port Phillip's rate of self-harm incidents has fluctuated and reached its highest rate of 1.9 incidents per 100 prisoners in March 2017. On average, Port Phillip's monthly rate of self-harm between 2013–14 and 2016–17 was marginally higher than MRC. From July 2013 to June 2017, MAP's rate of self-harm was, on average, 171 per cent higher than Port Phillip's. MAP has a much higher rate of self-harm incidents because it is a reception prison and cares for prisoners with mental illnesses. The remaining maximum‑security prison, Barwon, has a lower rate of self-harm in comparison.

Figure 3R
Self-harm incidents per 100 prisoners at Port Phillip, MAP, MRC and Barwon, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Port Phillip consistently met its threshold since 2010–11. CV imposed a tougher threshold on Port Phillip in 2012–13 as part of end‑of‑service‑term negotiations and again in the new contracts.

Fulham

Figure 3S shows the low rate of self-harm incidents at medium-security prisons, making it difficult to identify trends. Fulham does not accommodate prisoners with an S1 risk rating and has very low numbers of prisoners with S2 and S3 risk ratings. Fulham consistently met its SDO performance threshold for self‑harm since 2010–11, excluding quarter two of 2015–16. The new contracts introduced a tougher threshold for Fulham, which the prison continued to meet in 2016–17.

Figure 3S
Self-harm incidents per 100 prisoners at Fulham and Marngoneet compared to the average of medium‑security prisons (excluding Fulham), 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

Drug use in prisons

CV records incidents of drug possession as well as usage. Figure 3T shows a slight increase in drug-related incidents in male prisons across the system. These incidents include possession of drugs and positive test results. The majority of these incidents (54 per cent) occurred at maximum-security prisons, while 40 per cent occurred in medium-security prisons.

Figure 3T
Drug incidents per 100 prisoners across all male prisons, 2013–14 to 2016–17

Source: VAGO based on data provided by CV.

CV monitors drug use in prisons through SDO 8 (random general urinalysis). PIMS randomly selects a number of prisoners for testing each week. For SDO 8, overall at men's prisons, there is a system-wide increase in positive drug tests, from approximately 3 per cent in 2010–11 to 5 per cent in 2016–17.

Port Phillip

Port Phillip's positive random urinalysis results have fluctuated since 2010–11. As shown in Figure 3U, Port Phillip's performance is worse than MRC and the average of other maximum-security prisons in nearly all quarters since 2013. Port Phillip experienced a sharp increase in positive tests in 2012–13, which has not returned to previous levels. Between quarter one of 2010–11 and quarter one of 2012–13, the percentage of positive urinalysis at Port Phillip was, on average, 6.59 per cent. From quarter three of 2012−13 to quarter four of 2016−17, the average percentage of positive urinalysis results nearly doubled to 12.1 per cent.

Figure 3U
Percentage of positive random urinalysis results at Port Phillip and MRC compared to the average of maximum‑security prisons, 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

In the last quarter of 2016–17, Port Phillip experienced a spike in positive test results, to its highest level since quarter four of 2013–14. This coincided with an increase in prisoners refusing a urinalysis test, which counts as a positive result. The majority of these prisoners were located in the same unit, and prisoners were advised that refusing a test results in a positive finding.

CV successfully negotiated a tougher threshold for Port Phillip in 2013–14, and the new contract maintains this threshold. This change to the threshold has not resulted in any improved performance, as shown in Figure 3U. Port Phillip failed to meet its threshold for positive urinalysis results in 12 of the 16 quarters.

Prison visitors present a significant risk to controlling drugs and contraband. Port Phillip has visits scheduled six days a week and, in 2016–17, managed nearly 20 000 visits. Fulham had approximately 10 500 visits in that period. Recent analysis by the CV Intelligence Unit (CVIU) stated that Port Phillip accounts for 20 per cent of the strip searches undertaken across the prison system, and 21 per cent of visitor bans.

Fulham

The percentage of random positive urinalysis tests varies at Fulham, with a noticeable spike in 2012–13. As shown in Figure 3V, from the quarter three of 2014–15, Fulham's results have been better that Marngoneet.

Figure 3V
Percentage of positive random urinalysis results at Fulham and Marngoneet compared to the average of medium-security prisons (excluding Fulham), 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

Fulham performed well against its performance thresholds for urinalysis between 2010–11 and 2016–17, meeting the threshold in all but one quarter. The threshold at Fulham did not change from 2010–11 to the end of the initial contract, but CV lowered it in the new contract and Fulham met this tougher threshold in 2016–17. The threshold reduction in 2016–17 was partly due to combining the reporting results for Fulham and the minimum-security annexe, Nalu. A further decrease in the performance threshold was negotiated through the contract extension process.

Tobacco in prisons

Victoria implemented a smoking ban in all prisons on 1 July 2015. This created a new type of contraband for the prisons to manage and presented a unique challenge, as tobacco is a legal product.

Tobacco is easy for visitors to purchase and is difficult for prisons to detect. CV has not conducted any system-wide analysis of this issue but, anecdotally, prisons report a link between violence and the tobacco trade in prisons.

Figure 3W shows the rate of smoking incidents, or findings of tobacco per 100 prisoners since May 2015. The prisons with the highest number of tobacco incidents are the Judy Lazarus Transition Centre, Beechworth, Dhurringile, Fulham and Port Phillip, as shown in Figure 3X. While Port Phillip and Fulham do not have the highest rate of tobacco incidents in the state, they do have the highest rate of tobacco incidents for their respective security classifications.

Figure 3W
Tobacco incidents in men's prisons per 100 prisoners

Note: The spike in July 2015 marks the introduction of the tobacco ban. Most incidents relate to the possession of tobacco rather than the smoking of tobacco.

Source: VAGO based on data provided by CV.

Figure 3X
Ranking of men's prisons' monthly rate of tobacco incidents per 100 prisoners, July 2015 to July 2017

Prison		Prison security classification	Average monthly tobacco incidents per 100 prisoners
1	Judy Lazarus Transition Centre	Minimum	3.7
2	Beechworth	Minimum	2.9
3	Dhurringile	Minimum	2.7
4	Fulham	Medium	2.1
5	Port Phillip	Maximum	1.7
6	Middleton (annexe)	Medium	1.1
7	MAP	Maximum	1
8	Loddon	Medium	0.9
9	MRC	Maximum	0.8
10	Marngoneet	Medium	0.7
11	Barwon	Maximum	0.6
12	Hopkins	Medium	0.3
13	Langi Kal Kal	Minimum	0.2
14	Karreenga (annexe)	Medium	0.1

Source: VAGO based on data provided by CV.

Recent CV analysis of several serious assaults at Port Phillip, demonstrated a potential correlation between the presence of tobacco, illicit drugs and shivs, and the occurrence of assaults in individual prison units. The analysis covered January 2017 to June 2017.

3.5 Violence-reduction strategies

Implementing effective strategies to decrease violence in prisons is challenging, particularly because prisoners move frequently between prisons and their profile is constantly changing. Violence-reduction strategies must be tailored to the unique profile of each prison and need thorough research, including detailed analysis of data to gain an understanding of the causes of the violence. Violence-reduction strategies must be subject to continuous review and updating, as well as periodic structured evaluation to ensure they are working.

Port Phillip

As Port Phillip is Victoria's largest maximum-security prison, it is not surprising that it experiences the most assaults. G4S led the way in developing and implementing a violence-reduction strategy in August 2013, which clearly states the desired outcomes and covers a broad range of initiatives and activities that could reduce violence. G4S continually reviews incidents of violence at Port Phillip through its daily management meetings and weekly compliance meetings. In these forums, the activities targeting violence are discussed and adjusted as necessary.

However, the strategy lacks detail on how the prison will measure its effectiveness, and a review of the strategy is overdue. While G4S's SDO measures are closely monitored and they provide a high-level picture of whether assaults are increasing or decreasing, they do not allow G4S to assess which violence-reduction activities are effective. G4S should review its strategy as a priority to ensure it is based on current evidence and adequately resourced, and that its impact can be evaluated and measured. The lessons from this can be shared across the prison system.

G4S has had several iterations of the violence-reduction strategy, and the current version involves several approaches:

• Prisoners who are perpetrators of violence are identified and allocated a violence-reduction 'VR' rating based on the severity of their behaviour. Their rating may result in sanctions and the removal of privileges, among other management strategies.
• The operator has a violence-reduction database that records all incidents of violence and details of the prisoners with a VR rating who are subject to monitoring and sanctions.
• A dedicated violence-reduction coordinator is responsible for monitoring and reporting on the strategy, with the clear objective to reduce violence.
• There are a range of violence-reduction activities, including training prisoners and staff, engaging prisoners at risk of being violent in developing strategies, and utilising Port Phillip's Incentives and Earned Privileges Scheme.

G4S also engaged a health and safety consultant to help develop and conduct a self-assessment of occupational violence (OV) risks at Port Phillip. This identified a range of areas for improvement, including OV policy development, staff training, environmental design in the medical area, and improving procedures to consider OV risks. G4S has developed an action plan to improve its management of OV at Port Phillip, which is monitored by senior management.

G4S's OV assessment identified that staff are particularly at risk when intervening in prisoner-on-prisoner assaults or disputes. This highlights that G4S should integrate OV strategies into broader violence-reduction strategies and not consider OV in an isolated way.

Our analysis of prisoner-on-prisoner assault data at Port Phillip in Figures 3K and 3L, along with the SDO performance data in Figure 3M, does not show any sustained improvement after the introduction or review of violence-reduction strategies. Figure 3E shows that Port Phillip's rate of prisoners assaulting staff improved in the second half of 2014–15 but this improvement was not maintained. Figure 3F shows a decrease in prisoner-on-staff assault incident points in 2014–15. This improvement was partially maintained despite the overall rate of assaults not improving, which is consistent with the system-wide trend. The increased reporting of the least serious assaults may be a result of the strategy developed in August 2013.

Fulham

GEO first developed a violence-reduction strategy—known as an assault‑reduction strategy—in 2013 by analysing incident data and consulting staff to identify the key factors that contribute to prisoner‑on‑prisoner assaults. It identified a large range of possible contributing factors, including drugs, staff interaction with prisoners and broader prisoner management.

The strategy documented current and new activities that may reduce assaults. GEO closely monitors all assaults at Fulham, and regularly discusses the activities aimed at reducing assaults. There is a strong focus on SDO performance as a measure of the strategy's success.

GEO reviewed its strategy in 2017 by again consulting with staff and identifying potential causes of assaults. It also engaged an external consultant to review its approach in November 2017. The consultant's report details the activities and initiatives GEO has in place at Fulham, but does not offer an objective evaluation of which activities are working.

It is positive to see that GEO consults with staff to identify causes of assaults and used some data analysis to inform its activities. However, it has not conducted a more structured evaluation of the assault-reduction strategy to make sure it is based on current evidence and clearly identifies what parts of the strategy are effective.

Public prisons

In response to increasing violence, CV developed a violence-reduction strategy in 2015. This strategy used the experiences of Port Phillip but focused more on reducing OV due to concerns about assaults on staff. It does not specifically focus on the escalating number of prisoner-on-prisoner assaults.

The strategy requires each prison to have its own violence-reduction plan and sets out five objectives, including:

• to ensure policies, frameworks and procedures are in place to mitigate the risk of OV and direct staff in the use of safety measures, and emergency management processes
• to equip staff to manage difficult situations, to protect their own health and safety
• to ensure staff are trained and competent in assisting prisoners who are affected by mental health issues, drugs and alcohol
• to provide technology and equipment solutions that support staff in frontline roles and protect their safety
• to ensure zero tolerance messages are made clear to prisoners and appropriate action is taken against those who choose to breach standards.

There are some supporting guidelines and resources for prisons, and CV is currently implementing an OV training program at all prisons. Like the private prisons, CV needs to evaluate the impact of this strategy on reducing violence and develop a system-wide strategy for reducing violence in prisons, both towards staff and prisoners.

New contract requirements for managing safety

The new contracts give significantly more detail on risk management. Some of the risk management requirements, such as alignment with the Standards, are already in place. The new contracts explicitly require the operators to develop a just safety culture that:

• encourages and supports staff to report incidents and near misses
• balances a 'no blame' approach for human error with personal responsibility for following policies and procedures
• promotes organisational learning from incidents.

These inclusions in the contract are a proactive approach to managing safety and security risks, and CV should monitor both private prison operators' activities to ensure they are progressing with this requirement. At 31 December 2017, neither operator had any plans or activities that build on existing processes to ensure they comply with this new requirement.

3.6 The role of intelligence in safety and security

The aim of intelligence is to proactively analyse information and take action against threats before they occur. The CVIU manages system-wide intelligence functions, and each prison has its own PIU.

CV Intelligence Unit

'A strong correctional intelligence system has the potential to prevent serious security breaches in prisons, including assaults, trafficking contraband, deaths and escapes. Every staff member can be a source of or receive intelligence information and has an obligation to ensure this information is provided to the Corrections Victoria Intelligence Unit.' —CV's Sentence Management Manual

The CVIU plays a significant role in the management of safety and security risks across the corrections system by:

• gathering, assessing, evaluating and disseminating intelligence across the entire corrections system, including prisons and offenders on community correction orders
• managing intelligence information flows within and external to CV
• providing intelligence training to staff at public and private prisons
• reviewing intelligence units in individual prisons
• producing reports on intelligence activities and ensuring risk information and assessments are shared appropriately between prisons.

There has been substantial reform of CV's intelligence functions and systems following several significant reviews and investigations over the last 10 years. In 2008, CV commissioned a review of its intelligence functions (the Comrie review) which identified serious risks with CV's intelligence function and recommended significant reform and investment in intelligence systems and resources. The Victorian Ombudsman's April 2012 report The death of Mr Carl Williams at HM Barwon Prison: Investigation into Corrections Victoria also found some ongoing failings in the management of corrections intelligence and delays in implementing the recommendations of the Comrie review.

Given this, our audit reviewed whether CV and the private prison operators are monitoring and responding to intelligence risks and issues, but did not extend to a detailed analysis of the effectiveness of intelligence activities.

CVIU provides equal support to private and public prison staff, including training and onsite support to PIUs. CVIU actively monitors intelligence activities at the private prisons, including collecting data on the number of intelligence reports lodged, and recording this in Centurion, CV's new central intelligence database.

Centurion was a critical factor in CV's intelligence reform. Staff working with the new database all reported that it is has significantly improved the capacity for prison intelligence officers and CVIU staff to review and analyse prisoner information. We observed positive examples where staff used Centurion to help analyse patterns of violence and target responses. Centurion has provided improved capacity to bring together a range of prisoner information from CV and external systems, to analyse relationships and prisoner associations.

However, in contrast to public prisons, 'general duties' correctional staff working directly with prisoners in the private prisons do not have direct access to Centurion. They submit information reports to their PIU by email or verbally, and PIU then enters this information into Centurion. This potentially reduces the volume and timeliness of intelligence information received from private prisons, although both private prisons have workarounds to minimise this risk. The number of reports made by general duties staff at Port Phillip and Fulham increased in 2016–17.

The lack of access to Centurion in prison units also restricts unit supervisors' ability to review and analyse information to understand risks and issues in their unit.

Prison intelligence units

PIUs are integral to the safe and secure management of prisons. They collect and analyse information reports, review and approve prisoner mixing and visit requests, determine search operation locations, and conduct random and targeted drug testing.

The demands on PIUs have increased with shorter prisoner stays, increasing movement through the system and the challenges of gang associations and terrorism.

We observed that both private prison operators rely heavily on their PIUs to inform and support their efforts to reduce violence and drugs. This was evident in weekly compliance meetings at Port Phillip and the 'Safe in Fulham' weekly meetings, which consider detailed prisoner-specific intelligence reports. We saw examples at both prisons of intelligence activities leading to significant seizures of drugs and other contraband.

In January 2017, G4S invited CVIU to formally review the PIU at Port Phillip. This report highlighted significant gaps in the PIU, including staff capability and practices, communication within the prison, upgrades required for facilities and cultural issues. G4S accepted all recommendations and has commenced a program of work to improve the PIU function.

Some of the issues identified in the CVIU review related to the increase in the overall number and proportion of remand prisoners, increasing movements, and a changing prisoner profile. In particular, it related some of the gaps in Port Phillip's PIU function to the increasing proportion of prisoners that have to be separated from other prisoners. For these reasons, after a negotiation process with G4S, CV has agreed to partially fund some of the improvements required, including additional resources, with G4S funding the proposed redesign and refurbishment of PIU accommodation.

GEO has not asked CVIU to conduct a review of its PIU at Fulham. CV should continue to monitor Fulham's PIU function through existing mechanisms, such as auditing operational instructions, to ensure it is compliant with all relevant requirements.

Back to top

Achieving sustainable cost savings is one reason governments choose to outsource the operation of prisons to private companies. Demonstrating that prisons can operate at a lower cost while maintaining quality standards can support efforts to improve efficiency in the public prison system.

We have used CV's data on prison costs, including operating costs directly attributable to individual prisons. For public prisons, this includes labour, locally incurred supplies and consumables expenses. For private prisons, this includes contract payments.

We have also included the costs CV incurs centrally for managing the system and allocated these across individual prisons. We based the allocation on each prison's share of the total average prison population for the year, adjusted where necessary to better reflect cost attribution. For example, we allocated CV's centrally incurred contract management costs to the privately operated prisons and the three public prisons with significant facilities provision and management contracts—Hopkins, MRC and Marngoneet. Broader DJR corporate overhead costs and the capital assets charge applied by DTF as part of the output budgeting process are not included in these costs.

The cost information provided in this section excludes the state's two women's prisons—the Dame Phyllis Frost Centre and Tarrengower Prison.

4.1 Conclusion

The private prisons cost the state up to 20 per cent less than the average for publicly operated prisons with the same security rating and system-wide average costs. More efficient staff shift patterns largely account for these savings.

4.2 Total system costs

Total system costs have increased steadily since 2010–11 due to ongoing growth in prisoner numbers driven by sentencing and parole reforms, and an increasingly complex prisoner population. Figure 4A shows that the total system costs for male prisons increased by 90 per cent from 2010–11 to 2016–17, rising from $425.9 million to $811.2 million. Prisoner numbers increased by 49.5 per cent during this period.

Prison direct costs—include labour, catering, utilities, cleaning, asset depreciation and maintenance costs for public prisons, and contractual payments and asset depreciation for private prisons. Other costs—include centrally managed system-wide support functions such as operational support, offender management, sentence management, security and intelligence, health services costs, business support and contract management costs. Note: JARO (OCSR) costs are not included in prisons' direct or other costs.

Figure 4A
System-wide costs for male prisons, 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

Prison direct operating costs increased by 86 per cent over the period. Other costs increased by 105 per cent, with the largest increases relating to IT system costs, sentence management, and security and intelligence.

The average number of male prisoners in the system increased from 4 272 in 2010–11 to 6 383 in 2016–17. Figure 4B shows the average annual costs per prisoner, for both total system costs and direct prison operating costs. These costs grew by 27 and 24 per cent respectively during this period. The average total cost per prisoner per year increased from $99 703 in 2010–11 to $127 092 in 2016–17.

Figure 4B
Average annual system-wide costs per male prisoner, 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

4.3 Prison cost comparisons

We used CV information on system-wide and individual prison costs, from 2010–11 to 2016–17, to compare the costs of private prisons against system‑wide averages and public prisons with the same security ratings and similar characteristics.

While these comparisons yield insights, they should be treated with caution. Differences in the characteristics of individual prisons including location, age, size, physical layout, security classification, prisoner profile and their role or function within the overall system influence their operating costs. These differences make it difficult to compare costs on a 'level playing field'.

Despite this, our comparisons show that the private prisons cost the state less than the average for publicly operated prisons with the same security rating. DJR's advice to government in recent years, and our review of available benchmarking, indicates that the lower costs are largely attributed to more efficient staff shift patterns.

Average cost per prisoner

Figure 4C shows that the average annual total costs per prisoner at Port Phillip and Fulham were lower than those for public prisons with the same security classifications.

Figure 4C
Prison costs—average annual cost per prisoner, 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

Port Phillip

Figure 4D indicates that Port Phillip has lower average operating costs than publicly operated maximum-security prisons.

Figure 4D
Port Phillip—average annual per prisoner cost comparison, 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

The value-for-money assessment undertaken by CV when negotiating the new contract for Port Phillip concluded that the cost bid from G4S was significantly lower than the operating costs at Barwon and similar to, but still lower than, the costs at MRC. CV considered MRC the best cost comparison for Port Phillip due to its similar prisoner profile and level of activity.

Fulham

Figure 4E shows that Fulham has been consistently below the average per‑prisoner costs across all publicly operated medium-security prisons—Hopkins, Loddon, Middleton and Marngoneet—since 2012–13. The average annual per-prisoner costs across all publicly operated medium-security prisons for 2016–17 include costs for Karreenga, a medium-security facility opened in September 2016 adjacent to Marngoneet Correctional Centre.

Figure 4E
Fulham—average annual per-prisoner cost comparison, 2010–11 to 2016–17

Source: VAGO based on data provided by CV.

The value-for-money assessment undertaken by CV when negotiating the new contract for Fulham concluded that GEO's pricing was lower than the benchmarked costs.

4.4 Why do private prisons cost the state less to operate?

DJR's advice to government in 2014 and 2015 recommending the negotiation of contract extensions for Fulham and Port Phillip consistently indicated that these prisons were more than 20 per cent cheaper to operate than publicly operated prisons.

A number of reviews found that private prisons:

• are less costly for the state than publicly operated prisons
• appear to operate at lower operational staff-to-prisoner ratios, particularly during daylight hours
• have more flexible staff scheduling to better accommodate the day's structure, with shifts of various lengths between 7.60 hours and 12.40 hours and shorter shift-overlap periods.

In September 2014, the government responded to these reviews by asking CV to investigate opportunities for workforce and rostering efficiencies within public prisons. CV advised us that it continues to examine and progress these opportunities.

Under the new contracts, the average annual per-prisoner costs are broadly consistent with the past few years. See Part 5 for discussion of the value‑for‑money assessments that informed the new contracts.

4.5 Comparing prison cost elements

We examined and compared private prison costs, as agreed in the new contracts with public prison costs for 2016–17.

The private operators have lower labour costs, but their utility costs are mixed.

Labour

We compared the agreed base labour costs in the new contracts for Fulham and Port Phillip with labour costs for broadly similar public prisons. These comparisons covered all labour costs including custodial, management and administration roles and show significantly lower labour costs in the privately operated prisons. Under the new contract:

• Fulham's expected labour costs per prisoner are around 33 per cent lower than comparable labour costs at Loddon
• Port Phillip's expected labour costs per prisoner are around 45 per cent lower than comparable costs at Barwon and 25 per cent lower than MRC.

We did not find significant differences in the staff pay rates between the public and private prisons for the most common roles, such as correctional officers.

Utility costs

We compared the agreed base utility costs in the new contracts for Fulham and Port Phillip with utility costs for broadly similar public prisons, covering water, electricity and gas. Our comparison shows significant variation across the public and privately operated prisons.

Under the new contracts, Fulham has a higher average utility cost per prisoner than Port Phillip, primarily due to higher expected water and sewerage charges for its location, which also contributed to Barwon's high utility costs.

4.6 Actual costs under the new contracts

As part of the contract negotiations, both operators submitted detailed financial bids using cost templates specified by CV. CV incorporated the final agreed cost templates into the new contracts as an attachment to the payment schedule, and these costs are key 'base amounts' underpinning payments made to the operators.

We examined the operator's claimed and actual costs under the new contracts using actual staffing rosters for labour costs and invoices for subcontracted services, local government charges and utility costs passed on to the state. We found no significant variances.

4.7 State-funded capital works

The state has funded multiple capital projects at Fulham and Port Phillip since 1997. These projects usually involve adding new bed capacity or security upgrades.

The private operators are responsible for the design, costing and procurement processes for state-funded capital works on their sites. CV actively reviews key steps, agrees outcomes and then pays for the works. The state provides funding for capital costs based on an up-front agreement of both the capital cost and the operational cost impact.

The state enters into a contract with the prison operator for each capital works project and, in turn, the operator enters into a 'back to back' contract with a construction firm. CV advised that the private operators bear the risk of cost and time over-runs on capital works.

CV performs due diligence and other processes, including reviewing pricing and costs and, in some cases, engaging quantity surveyors to support decisions on new state-funded capital works. We reviewed CV's due diligence for two state‑funded capital projects—one at Fulham and one at Port Phillip—and found CV's processes for assessing and overseeing both projects were rigorous.

Back to top

The original service contracts for Fulham and Port Phillip were due to expire in 2017. DJR recommended negotiating long-term contract extensions with the incumbent operators, partly because other options would involve significant costs to 'buy out' the site leases held by the operators which extended for many years beyond 2017. DJR maintained alternative options, such as re-tendering the contracts, as a fallback in case the negotiations did not achieve a value‑for‑money outcome.

Government accepted this recommendation in 2014, and new contracts were finalised during 2015. These contracts have performance-based terms of up to 20 years that align with the revised site lease terms. DJR forecast the new contracts to cost the state around $4.5 billion in nominal terms, or $2.4 billion in net present cost terms, if they run for the full term.

In this part of the report, we examine whether DJR:

• thoroughly analysed the procurement options
• developed and implemented a comprehensive negotiation strategy to manage the risks arising from the lack of competitive tension and delivered value for money
• provided robust advice to key decision-makers, including government, on options and outcomes
• used the new contracts to adequately address key weaknesses in the previous contracts.

5.1 Conclusion

DJR successfully navigated significant challenges to negotiate new contracts with the private operators of Fulham and Port Phillip that addressed key weaknesses in the previous contracts and broadly met the government's approved cost limits. DJR's advice to government on whether to enter into the new contracts was sound, with some minor exceptions that did not invalidate the outcomes achieved.

5.2 Options analysis

The state needed Fulham and Port Phillip to continue operating beyond the expiry of the initial contracts to accommodate rising prisoner numbers.

Options to achieve this included:

• negotiating short- or long-term extensions with the incumbent operators
• directly operating the prisons
• competitively tendering for the operation of the prisons.

DJR tried to negotiate surrender of the leases as part of the end-of-service term reviews for both prison contracts in 2011 and 2012. This would have provided the state with unencumbered access to the prison facilities and discretion to competitively tender future service delivery. However, the complex issues around the leases meant DJR did not pursue surrender of the leases at that time.

Legacy issues with the initial contractual arrangements meant the state could not consider each of these options on a 'level playing field'. The key issue was that the operators held leases over the prison sites that extended many years beyond the service contract terms. The lease granted for Fulham in 1995 extended 18 years beyond the service term to 2035, and the Port Phillip lease, granted in 1996, extended 29 years beyond the service term to 2046.

The misalignment between the contracts' service terms and lease terms created significant uncertainty about the state's ability to secure access to the prison facilities in 2017 if it decided to let the operators' service contracts expire and either engage other operators or directly operate the prisons.

While DJR considered compulsory acquisition in its options analysis in 2013 and 2014, the state's lease acquisition rights were seen as complex and would have involved considerable risk and cost if pursued. Given this, DJR appropriately treated the lease term misalignment as a serious constraint on the state's options.

The government decided to negotiate new contracts with the incumbent operators based partly on advice that pursuing other options would involve significant additional costs.

Timeliness of options analysis

DJR analysed the most viable procurement options sufficiently early to protect the state's negotiating position.

The initial contracts for Fulham and Port Phillip included an option for the state to negotiate extensions beyond the 20-year service terms ending in 2017. The state had to give the operators three years' notice if it wanted to negotiate extensions—meaning by April 2014 for Fulham and September 2014 for Port Phillip.

These time lines meant the state needed to consider all options well before the expiry of the contracts to optimise its leverage in any negotiations with the operators. Positively, DJR and DTF considered an appropriate range of available options as early as 2011, years in advance of contract expiry.

Adequacy of options assessment

DJR's options analysis and advice to government was adequate.

DJR provided the Minister for Corrections with high-level advice in May 2013 on the misalignment of the lease terms and the options available to keep the two prisons operating. The minister approved:

• more detailed options analysis to inform recommendations to government by mid-2014 on whether to negotiate contract extensions
• initial contact with the two operators to advise them that the state was willing to discuss possible contract extensions beyond 2017.

Following the ministerial approval, DJR commenced the detailed options analysis and identified the following alternative options and constraints:

• allowing the contracts to lapse and moving prisoners elsewhere—not considered feasible given capacity pressures across the prison system and the lead time required to expand or build additional prison capacity
• allowing the contracts to lapse and the state resuming the leases and operating the prisons itself—considered feasible subject to potential cost and risk associated with the lease issue, but considered likely to result in higher operating costs given the benchmarked costs of state-run prisons
• negotiating extended contracts with the incumbents—the sole-source procurement risks were noted along with the potential upside of resolving the lease issues
• re-tendering for the operation and maintenance of each prison—constrained by the lease issue and the relatively small number of private prison operators.

The analysis also considered hybrid options including separately tendering the provision of correctional services and accommodation maintenance services.

The final advice to government in July 2014 justified the recommendation to renegotiate with the incumbent operators on the basis that:

• the operators had lower operating costs than publicly run prisons
• the operators' service delivery performance was adequate
• the terms for the new contracts could be aligned with the lease terms
• an open tender process was unlikely to deliver an outcome that bettered existing costs and recovered the transaction costs, which potentially included lease buy-out costs of tens of millions of dollars for each prison.

This advice to government was adequate overall, aside from the following issues:

• The summary advice to government did not adequately explain that the estimated lease buy-out costs were preliminary notional amounts that were indicative of potential ambit claims from the operators. DJR sourced these estimates from commercial advice obtained in 2013. Subsequent formal valuation advice to DJR indicated that the approach used in this commercial advice was reasonable, but the underlying estimate of the PortPhillip land value was too high. DJR based subsequent advice to government on the potential cost of buying out the Port Phillip lease on more specific information reflecting the likely profits or margin that G4S would forego if it surrendered the lease.
• The advice did not present detailed information on the operators' service delivery performance—for example, DJR's concerns about G4S's performance at Port Phillip during 2014 were given little coverage in the July 2014 advice to government. Subsequent advice during 2015 included summary information on the operator's service delivery performance and advice that the Commissioner assessed Port Phillip as performing consistently with the state's other maximum-security prisons over the initial contract period.
• The discount rates used as part of the net present cost calculations were not disclosed in the advice.
• The advice materially underestimated the actual required costs for asset maintenance and life cycle expenditure for Fulham. The business case estimated these costs at $16.7 million based on the information available to the state at the time, but the cost ended up at $78 million following more detailed asset condition assessment and review.

Despite these issues, the advice to government adequately captured the risks, constraints and opportunities facing the state in relation to these contracts. It supported the conclusion that the government's best option was to try and negotiate with the incumbent operators.

DTF and DPC supported DJR's recommendations and, in July 2014, the government approved negotiations with the operators to extend the contracts subject to the:

• costs of the new contracts not exceeding approved negotiating limits that were based on the costs of the existing contracts, with some incremental cost increases to cover changes in services and additional investment in facilities maintenance and replacement
• service terms for any new contracts aligning with the lease terms and sound end-of-term arrangements.

Ongoing assessment of options

Positively, DJR did not treat the results of the initial options analysis as a given after the government approved new contract negotiations with the operators. DJR appropriately treated the contract extension negotiations as 'plan A', but also adequately considered the merits of a re-tender, which was the state's best 'plan B'.

The options analysis was 'live' during the negotiation process. DJR progressively developed the analysis, using negotiation outcomes and updated analysis of the fallback strategy, including an updated assessment of the costs and risks associated with the lease issues.

This was more formal for the negotiations with G4S for the Port Phillip contract because G4S initially proposed significantly higher‑than‑expected operational costs. In contrast, the operational costs proposed by GEO for Fulham were broadly consistent with the approved negotiating limit and payments under the existing contract.

5.3 Management of negotiations

DJR planned and implemented a robust negotiation strategy. DJR's negotiation process and time line placed the state in a strong position because it retained a realistic alternative to providing the incumbent operators with new contracts—the state had time to competitively tender the operation of the prisons if the negotiations with the operators did not produce a reasonable outcome.

Figure 5A shows key events and decisions in the negotiation of the new contracts.

Figure 5A
Key events and decision points

Date	Event
28 March 2014	The Minister for Corrections (with the approval of the Premier and Treasurer) advised GEO of intention to commence negotiations to extend the Fulham contract.
May 2014	DJR set up a project team to negotiate possible contract extensions for Fulham and Port Phillip.
28 July 2014	The government approved negotiation limits for new contracts with GEO and G4S.
22 August 2014	The minister (with the approval of the Premier and Treasurer) advised G4S of intention to commence negotiations to extend the Port Phillip contract.
15 September 2014	The government signed a contract with a GEO-led consortium for Ravenhall.
2 March 2015	The government approved a new contract for Fulham and continuation of negotiations for Port Phillip.
2 April 2015	The government signed a new contract with GEO for Fulham.
19 November 2015	The government approved a new contract for Port Phillip.
17 December 2015	The government signed a new contract with G4S for Port Phillip.
June 2016	Partnerships Victoria project summaries published for both prison contracts.
1 July 2016	New Fulham contract commences (around nine months prior to expiry of initial contract).
10 September 2017	New Port Phillip contract commences (the initial contract expired on 9 September 2017).

Source: VAGO based on information from CV.

DJR's comprehensive procurement and negotiation strategy was ultimately successful in achieving the state's objectives. DJR successfully negotiated for:

• the continued operation of 20-year-old prisons under contractual provisions that were largely consistent with those designed for the new Ravenhall prison facility which had been subject to a competitive tender process
• new contract term lengths of up to around 20 years depending on performance, which provided a long-term incentive for the operators while allowing:
  • the cost of services to be adjusted through indexation and benchmarking provisions
  • flexibility for the state to terminate the arrangements based on performance or breach of contract
• resolution of the lease misalignment issues that had constrained the state's ability to re-tender the initial contracts by requiring the incumbents to 'give something up'—particularly in the case of G4S where the Port Phillip lease extended for some 29 years beyond the expiry of the original contract
• updated contractual terms, and payment and performance mechanisms, aligned with the state's approach for Ravenhall, meaning greater risk for the operators—a higher proportion of the service payments are based on performance—particularly given Fulham's ageing prison infrastructure and Port Phillip's complex prisoner cohorts.

DJR achieved these outcomes at a cost broadly in line with the government's approved negotiating limits and the cost of the original contracts, noting that:

• for Fulham, debt service payments of around $8 million per year payable under the original contract were reallocated to asset maintenance and renewal (debt service payments for Port Phillip expired in around 2012)
• for Port Phillip, the state agreed to contribute around $9 million towards required capital upgrades in the lead-up to the new contract commencing.

The Fulham agreement was signed on time and only a short extension was required to finalise negotiations for Port Phillip.

Governance and oversight

DJR appropriately allocated significant resources to the options analysis and negotiation processes. Project management and reporting processes were consistent with DJR's Project Management Excellence Framework, ensuring regular, robust monitoring and reporting on progress, risks and budget.

A steering committee oversaw the options analysis and negotiation processes. The committee included senior representatives from DJR, DPC and DTF, supported by external legal, commercial and probity advisers. The committee met regularly between November 2013 and April 2016, and there is clear evidence of comprehensive reporting to and engagement by the committee.

Negotiation strategy and implementation

DJR developed and implemented a sophisticated negotiation strategy that adequately mitigated the sole-source procurement risks. Positive features included:

• definition of high-level project objectives and benefits sought in the project business case, benefits plan and submissions to government
• development of a detailed negotiation strategy by August 2014, which included time lines, strategic context and fallback strategies
• identification of 'pillar negotiation parameters' that were important to the state
• ongoing refinement of the fallback strategy so that the state had a realistic alternative to renegotiation with the incumbent operators
• use of signed documents with the operators, including a 'Heads of Agreement', to formalise their commitment to the process and time lines and to define the framework for further negotiations
• undertaking 'health checks' at key points of the process that were linked to further work on the state's fallback strategy
• defining criteria to measure the 'success' and value for money of the negotiation outcomes.

The negotiation team considered the possible impacts of other projects and initiatives such as:

• the Ravenhall procurement process, where both operators bid and GEO was awarded the contract in September 2014
• the prisoner transport procurement process, where G4S was awarded the contract
• a review of Justice Health
• capacity expansion proposals for Fulham and Port Phillip.

The team was also astute in anticipating that the Ravenhall contractual model would be less familiar for G4S and allowing extra time and resources to bring this operator 'up to speed'.

Compliance with Partnerships Victoria and other requirements

The contract negotiations for the private prisons were in a unique position regarding applicable requirements and oversight mechanisms:

• Although Fulham and Port Phillip were clearly PPPs, they were originally procured before Partnerships Victoria came into existence and the initial contracts were not always treated as being subject to Partnerships Victoria requirements.

The Public Sector Comparator is an estimate of the hypothetical, whole-of-life cost of a project if delivered by the public sector. The PSC is based on the required output specification. It provides government with a quantitative reference point to assess the value for money of private sector proposals.

• The contract extensions involved a continuation of service delivery via existing assets with some upgrades, unlike typical Partnerships Victoria projects where a private consortium finances the construction of new assets.
• The lack of a significant 'new-build' asset component meant that the High Value High Risk (HVHR) framework did not automatically apply.
• The renegotiations were provided for in the original contracts and, in part, were driven by the lease misalignment issue built into the original deals.
• The Act partly drove the approval requirements.

The new contracts for Fulham and Port Phillip were clearly treated as Partnerships Victoria transactions—for example, Partnerships Victoria project summaries were published. The May 2013 Partnerships Victoria Requirements, applicable at the time the transactions were planned and managed, included key requirements such as the development of a formal Public Sector Comparator (PSC) and application of the state's gateway review process.

The processes used by DJR did not fully align with these Partnerships Victoria Requirements. Specifically:

• DJR did not develop a PSC for the Fulham transaction—instead, it developed a public sector benchmark that served an equivalent purpose
• DTF did not undertake the required independent review of the PSC developed for the Port Phillip transaction—DJR had access to operating costs for broadly comparable prisons and could do an open-book review of G4S's existing costs
• no gateway reviews were conducted for the project.

DTF supported this flexible approach to applying the Partnerships Victoria Requirements, as these were not typical Partnerships Victoria transactions. DTF advised that it was aware of and satisfied with DJR's approach given the context for the transactions and the other information available to inform the state's value-for-money assessments.

The level of oversight and review applied to the project was adequate. Government approved all key decisions after they had been considered by the inter‑departmental steering committee. In addition, the Treasurer approved the new contracts under the Act.

DTF advised that gateway reviews were not required because it considered the project medium risk. Our assessment of the project indicates that the project warranted a high-risk rating, given it involved a significant portion of the state's male prison population and material cost to the state. Nevertheless, we recognise that the evolving nature of the negotiations meant that there was no obvious time to conduct the gateway reviews.

Discount rates

The nominal discount rate applied for Fulham was materially higher than that used for Port Phillip—7.62 per cent nominal for Fulham at 30 June 2014 versus 4.13 per cent nominal for Port Phillip at 30 June 2015. The choice of discount rate for Fulham is not a significant concern because:

• the 5 per cent real discount rate used for Fulham was broadly in line with relevant guidance published in 2013, and DTF was informed of the rate used
• the use of a relatively higher discount rate did not distort the analysis as:
  • the selected rate was used consistently in advice to government during 2014 and 2015
  • the focus of the value-for-money and affordability analysis was on the nominal annual contract cost, not a discounted value
• the project summaries published in mid-2016 for Fulham and Port Phillip included the cost of the contracts in both nominal and net present cost terms, and clearly disclosed the discount rates used.

However, the higher discount rate applied to the Fulham transaction makes it look 'cheaper' in comparison to Port Phillip when the net present cost of expected service payments under the contracts is disclosed without details of the discount rate. For example, the May 2017 DTF publication Partnerships Victoria: Excellence in public private partnerships lists the net present cost of a number of projects and shows Fulham at $593 million and Port Phillip at $1 831 million, without disclosing the discount rates or dates. This makes Port Phillip look around three times more costly than Fulham for a similar contract term when, in nominal terms, it is closer to double—$1 451 million for Fulham and $3 113 million for Port Phillip.

Probity issues

Robust management of probity issues—including actual, potential and perceived conflicts of interest—protects the integrity, fairness and impartiality of government procurement processes.

DJR and the steering committee understood the need for probity during the negotiations and engaged an experienced probity practitioner to provide advice and assurance, partly because there was the possibility of a tender if it needed the fallback strategy. Probity issues identified during the negotiations for the new contracts were largely well managed. However, a commercial adviser working for DJR and one of the operators did not adequately disclose potential conflicts of interest and these were not adequately assessed or acknowledged in advice to government.

The definition of conflict of interest in the project and probity plans was consistent with the National PPP Policy and Guidelines. Any affiliation or interest that prejudiced, or might be seen to prejudice, a project participant's impartiality would be considered a conflict of interest.

Much of the national guidance material focuses on actual and potential conflicts involving external advisers to government agencies, because it is not uncommon for actual or potential conflicts of interest to arise for these advisers. Such conflicts do not necessarily compromise process integrity, as long as they are transparently disclosed, assessed and properly managed.

A conflict of interest exists if an advisor to the state on a PPP project is also advising one or more of the private-sector bidders on the same or a related transaction. This is because the adviser:

• may have access to confidential state information that could advantage the private client
• is performing a role with a high capacity to influence the state's procurement decisions.

The National PPP Policy and Guidelines state that an advisor cannot act on both sides of a PPP project where there is a conflict of interest, with only rare exceptions.

The conflict of interest declaration signed by state employees and external advisers listed the incumbent operators and their related entities, including overseas-based parent companies. The declaration required disclosure of individual and firm financial interests in these entities and any actual, potential or perceived conflicts.

The conflict of interest declaration register established for the project recorded more than 60 declarations completed by staff of external advisory firms, including lawyers and commercial and other advisers. None of these declarations disclosed any actual, potential or perceived conflicts. However, seven of the 10 probity issues reported to the steering committee over the course of the project related to actual and potential conflicts of interest for external advisers working for both DJR and the operators.

DJR became aware of these conflicts using information from parties other than the advisers. The pattern of external advisers not proactively disclosing conflicts to the state was disappointing given that the probity advisor clearly and repeatedly communicated the importance for external advisers to meet their disclosure obligations.

The most significant issue related to the state's primary commercial adviser on the project. This adviser played a key role in assessing and advising on the state's options for securing ongoing services at the two privately operated prisons and was responsible for preparing the value-for-money assessments of the incumbent operators' bids for new contracts. Given this, the adviser was in a position to materially influence government decision-making on whether to grant new contracts.

Staff of the adviser's firm completed more than 20 conflict of interest declarations between October 2013 and August 2015, and none disclosed any actual, potential or perceived conflicts. However, in September 2014, the operator of Fulham advised DJR that it was using this firm for tax advice on the Ravenhall transaction and for financial advice on the Fulham contract negotiations. Both situations posed a conflict of interest for the adviser and clearly warranted disclosure to the state. This was particularly the case for the second conflict, where DJR had information indicating that the adviser was acting for both the state and the operator of Fulham on the same transaction.

When DJR became aware of the conflicts in September 2014, it sought information from the commercial adviser and GEO, and advice from the probity advisor:

• The commercial adviser provided a summary of how it managed such conflicts, including processes for separating advisory teams and maintaining the confidentiality of state information, but did not provide any information about the extent of its role in advising GEO on the Fulham transaction.
• GEO confirmed that it used the state's key commercial adviser extensively inboth Australia and overseas for tax advice and asserted its intention to continue to use this adviser.
• The probity advisor did not make any detailed inquiries or document a formal probity conflict risk assessment for either conflict, and advised DJR that there were no probity issues within hours of being made aware ofthem.

DJR relied on the probity advisor's advice and did not seek any further information or commitments from the commercial adviser on how it would manage the conflicts. DJR also advised that it did not seek any further information from the commercial adviser or GEO on the extent of financial advice from the adviser to GEO on the Fulham contract negotiations, as it did not consider this necessary.

The potential risks to the state's interests arising from this situation warranted more detailed probity conflict risk assessments than the cursory advice provided by the probity advisor. DJR should have also required the commercial adviser to supplement its high-level advice on conflict management with specific written assurances on the controls in place to manage the two conflict situations that arose.

DJR has recently received advice from both the commercial adviser and GEO indicating that the adviser did not, in fact, provide financial advice to GEO on the Fulham transaction. DJR should have obtained definitive confirmation on this issue in 2014.

We reviewed publicly available information and found that, in addition to providing services to the operator of Fulham, the state's commercial adviser's affiliated firm in the United Kingdom also provided advisory services to the global parent company for the operator of Port Phillip between 2013 and 2015.

This company was listed on the conflict of interest declaration form as a relevant entity, and the probity advisor reminded the commercial adviser in November 2013 that it needed to consider services provided to the operators and their related entities by its overseas affiliates when completing the declaration. Despite this, none of the adviser's staff disclosed any potential conflicts. DJR and the probity advisor indicated that they were not aware of this situation until we raised it.

The probity advisor's final reports on each transaction were unqualified and included no discussion of the conflict of interest issues that arose during the project and how they were managed. This was inconsistent with the probity plan for the project and the probity advisor's contract. While the project steering committee was aware of the conflict of interest issues, the probity advisor's reports also formed part of the assurance material provided to the Minister for Corrections and government. Given this, the reports should have explained how the potential conflicts of interest for DJR's key advisers were assessed and managed.

5.4 Demonstrating value for money

Recommending new long-term contracts after exclusive negotiations with incumbent operators required DJR to demonstrate to government that the recommended outcomes represented value for money. This is because the government could not compare the offers from the incumbents against bids from other potential operators in an open competitive process.

DJR communicated the merits of the negotiated contract outcomes in terms of cost, service delivery and performance requirements, risk transfer, and improvements over the original contracts. DJR provided advice to government recommending that it sign new contracts with GEO for Fulham and G4S for Port Phillip in March and November 2015 respectively.

Cost was the main criteria assessed and highlighted in this advice to government, and DJR took reasonable steps to gain assurance on the value for money of the operators' proposals.

DJR only briefly described the incumbent operators' service performance in the advice to government and did not support this advice with any detailed assessments of whether the operators were capable and high performing. DJR gave little coverage of its concerns regarding G4S's performance in operating Port Phillip during 2014.

The negotiated contract prices appear to provide sustainable value for money to the government and a reasonable return to the providers, if they meet performance standards.

Assurance of value for money

DJR primarily assessed the value for money of the operators' proposals against the negotiation limits for the contract extensions approved by government in July 2014, which it based on the costs of the existing contracts with some variations in the scope of services.

The government's approved negotiation limits allowed a cost increase of 4 to 5 per cent over the operating costs of the initial contracts. DJR and DTF appropriately highlighted differences between the negotiating limit and the final outcomes in their advice to government.

To assess the value for money of the operators' proposals, DJR examined their actual and proposed operating costs and benchmarked them against costs in publicly operated prisons. The results of this work indicated that the final offers from the operators represented value for money.

DJR based its advice to government about the value for money of the negotiated outcomes on analysis and sign-offs from external commercial, technical and legal advisers including:

• value-for-money assessment reports from the state's commercial adviser
• advice on benchmarking Fulham and Port Phillip's operating costs
• technical advice about the reasonableness of the asset and life cycle costs proposed by the operators
• legal sign-off letters on the contracts summarising departures from standard Partnerships Victoria commercial principles, explaining that these were linked to the unique nature of the projects and were not significant, and summarising differences in risk allocation between Fulham and Ravenhall, and between Port Phillip and Fulham.

Review of operator cost proposals

The state knew what it was paying the operators to deliver accommodation and correctional services under the old contracts but only had limited insight into the actual costs incurred by the operators in providing these services. The original contracts included five‑yearly pricing reviews, but these were informed by benchmarking exercises rather than regular detailed examination of the operator's actual costs.

DJR required the operators to complete a financial bid template as part of the negotiations for the new contracts. The financial bid templates required detailed cost information in the following areas:

• labour costs—including information on salaries per full-time-equivalent staff and staffing numbers across each prison on a unit-by-unit basis, showing daily shifts and weekly hours for each staff 'post' in each accommodation unit
• outsourced services—including costs for external education providers, allied health services, chaplaincy services and catering services for Port Phillip
• maintenance services and costs—these services are contracted out by the operators (financial bid template information indicates G4S is paid a margin on these costs whereas GEO is not)
• supplies—including for cleaning, prison industries and other consumables
• utilities—comprising water, electricity and gas charges
• other costs—including for IT systems and equipment leases.

DJR planned to test the price bids from the operators during the negotiations using 'open book' analysis, cost benchmarking against similar prisons and PSCs. Both operators signed probity and process deeds agreeing to provide the state with full open-book access to all necessary documentation for the negotiations.

DJR undertook reasonable steps to gain assurance about the value for money of the proposals put forward by the operators including targeted examination of their actual and proposed operating costs and benchmarking against costs in publicly operated prisons. DJR was also directly involved in negotiating with both operators and their sub-contractors for the scope and pricing of asset life cycle management and maintenance activities to limit the state's exposure to significant cost increases.

The extent of value-for-money analysis was appropriate given the unique circumstances of the Fulham and Port Phillip negotiations and DJR's advice to government on value for money was sound.

Fulham

The future operational costs initially proposed by GEO for Fulham were consistent with the approved negotiating limit and the payments made under the existing contract. They compared favourably to the costs of similar state-run prisons. DJR identified and assessed the value for money of incremental cost increases proposed by GEO and recommended granting the operator a contract extension without undertaking open-book analysis or a formal PSC as Partnerships Victoria requires. Instead, it compared the operational costs against a public sector benchmark. This benchmarking analysis:

• was prepared with a clear understanding of operating model requirements at Fulham and how the state would operate the facility
• confirmed that state delivery would be significantly more expensive.

DJR advised that the outcome of this benchmarking removed the need for a formal PSC. The value-for-money analysis also considered the costs and risks associated with the best alternative option—a tender for the contract—and concluded that a competitive process was unlikely to deliver a better outcome.

DJR did not highlight the lack of an open-book analysis and formal PSC in its summary advice to government in March 2015, in which it recommended signing the new contract. The summary advice referred to an 'appropriate level of due diligence over GEO's existing cost base'. The detailed supporting information included with this advice noted that GEO provided a detailed financial template that enabled DJR to compare cost assumptions against other data.

Port Phillip

When G4S initially proposed significantly higher-than-expected operational costs under the proposed new contract, DJR took additional steps to assess value for money. These included:

• preparing a formal PSC
• developing a 'tipping point' analysis informed by an updated estimate of the cost to buy out the lease
• undertaking open-book analysis of G4S's main operating costs.

The PSC and other benchmarking analysis included in the value-for-money report confirmed that G4S could operate the prison more cheaply than the state. The tipping-point analysis compared the expected costs of a new contract with G4S against the value that an open tender for the contract might achieve.

The sample open-book process undertaken for Port Phillip focused on key assumptions underpinning G4S's employee costs, which represented around 60 per cent of its operating costs. While the scope of this work was limited and the report documenting the work was not formalised beyond a draft, it served a useful purpose in DJR's final negotiations with G4S on costs under the proposed new contract.

DJR's advice to government in November 2015 recommending a new contract with G4S included a robust summary of the value-for-money analysis. The advice accurately described the open-book analysis as a sample-based process focused on G4S's historical costs and pricing assumptions.

5.5 Negotiation outcomes and new contracts

The government signed new contracts for Fulham and Port Phillip in April and December 2015 respectively. The contracts included some important improvements for the state. Figure 5B sets out key information on the new contracts.

Figure 5B
Scope of the new contracts for Fulham and Port Phillip

Measure	Fulham	Port Phillip
Bed capacity	893	1 087
Commencement	1 July 2016	10 September 2017
Initial contract term	11 years	10 years
Further extension term	8.25 years	10 years
Contract expiry (if full term)	October 2035	September 2037
Estimated total payments – full term (nominal)	$1.4 billion	$3.1 billion
Performance bond	$8 million	$12 million

Source: VAGO based on CV information.

Contractual improvements

It is clear that the new contracts negotiated with the operators address substantial weaknesses in the previous contracts. They will place the state in a much stronger position at the end of their service terms to consider and pursue all available options—including competitively tendering the contracts—unencumbered by legacy issues or costs.

The state's objectives in negotiating the new contracts included:

• updating the contractual provisions to be more consistent with current Partnerships Victoria PPP contracts
• addressing key deficiencies in the original contracts, including unclear service specifications and a low proportion of service payments based on performance.

DJR and DTF advice to government in March 2015 and November 2015 included reasonable claims that the new contracts addressed the major weaknesses in the original contracts and that material improvements had been achieved. The new contracts:

• define the required services more clearly, to assist contract management and enforcement, noting that it remains an 'output specification' that relies on commercial incentives to drive performance
• improve provisions for asset condition and life cycle maintenance and replacement, including better definition of the operators' obligations and specifying agreed planned works to be funded from a dedicated life cycle payment account, with any underspend at the end of term to be shared with the state
• include a payment and performance regime that more clearly allocates prison operation risks to the operators—payments are effectively built up from a zero base according to the number of available prisoner places and the quality of services delivered, rather than being deducted from a standard fee
• include a capacity 'tranching' regime which allows the state to decrease or increase the required bed capacity in the prisons in increments of 50 and adjust the service payment accordingly
• improve contractual security for the state, through a parent guarantee and increased performance bonds—the performance bond for Fulham increased from $3 million to $8 million, and for Port Phillip from $8 million to $12 million
• resolve the lease-term misalignment issue, providing the state with improved end-of-term provisions that will allow it to re-tender all or part of the services at the end of term (around 10 or 20 years in each case, depending on whether the operators secure a further extension term).

Appendix D lists the contractual improvements or 'negotiation benefits' described in CV's contract administration manuals for the new contracts and comments on the validity of the claimed improvements and benefits.

Changes to performance-linked payments under the new contracts

The initial contracts for the private prisons provided for an annual performance‑linked payment to each operator based on their achievement of the SDO performance thresholds.

The maximum performance payment payable to each operator was capped, and the fees were not indexed. As a result, they declined proportionally over time and did not provide a sufficient incentive for operators to improve their performance:

• For Port Phillip, the maximum performance payment was fixed at around 3.5 per cent of the total payments from the state in the first year of the contract. It dropped to around 1 per cent of total payments in 2015–16.
• For Fulham, the maximum performance payment was fixed at around 7.4 per cent of the total payments from the state in the first year of the contract, dropping to around 3 per cent by 2015–16.

The Commissioner determined the amount of the performance payment paid to each operator at the end of each year, based on advice from CV about actual performance. Under the contract, the performance payment would only be reduced if the operator failed a category of SDO performance thresholds—for example, SDOs related to safety and security—by more than 20 per cent.

While the new contracts retain the SDOs and include additional KPIs, they also significantly increase the proportion of operators' total contract payments that rely on performance. The new contracts also introduced contractual penalties for specific performance failures, known as charge events. These events include prisoner deaths, escapes, serious professional misconduct by prison staff and significant breakdowns in discipline, such as riots.

Along with better processes for addressing performance issues, the new contracts also introduced additional incentives for operators to address poor performance quickly. CV can reduce the service-linked fee, which includes performance payments, if the operators fail to meet SDO thresholds in successive quarters. Under the initial contracts, this was an annual assessment and, once an operator had failed for the year, there was little incentive for it to improve its performance on failed SDOs. The introduction of a KPI that penalises successive quarterly failures has significantly increased pressure on the operators to perform.

Risk allocation

The key commercial features of the new contracts, including the risk allocation, are explained in the project summary documents for each contract publicly available from DTF's Partnerships Victoria website.

We examined this information on the risk allocation, the contracts and the sign‑off opinion letters prepared by the state's legal adviser supporting DJR's recommendations to government. The risk allocations were accurately described in advice to government and the public.

We note that there are some differences in the detailed commercial regimes included in the two contracts. This reflects the project-specific outcomes from the negotiations and differences in the operational and commercial risks carried by the operators. For example, there are differences in the contractual provisions which determine whether each operator will secure a further extension term after the initial 10- or 11-year term, including differences in the allowable number of adverse events and payment reductions.

Despite these differences, it is clear that the new contracts achieve an overall risk allocation for Port Phillip that is broadly the same as that for Fulham and Ravenhall, representing an improvement over the original contracts.

Back to top

We have consulted with DJR, DTF, G4S and GEO, and we considered their views when reaching our audit conclusions. As required by section 16(3) of the Audit Act 1994, we gave a draft copy of this report, or relevant extracts, to those agencies and asked for their submissions and comments. We also provided a copy of the report to DPC.

Responsibility for the accuracy, fairness and balance of those comments rests solely with the agency head.

Responses were received as follows:

• DJR
• DTF
• G4S
• GEO

RESPONSE provided by the Secretary, DJR

 

RESPONSE provided by the Secretary, DTF

 

RESPONSE provided by the Managing Director, G4S

 

RESPONSE provided by the Managing Director, GEO

 

 

Back to top

SDOs and KPIs are the key measures of prison performance.

SDOs apply to both public and private prisons, although not all SDOs apply to all prisons.

KPIs only apply to the private prisons—18 of the KPIs apply to Port Phillip, and 16 to Fulham.

The SDO and KPI thresholds use various measures:

• zero tolerance—requires 100 per cent compliance or no incidents
• a percentage result—for example, for positive drug tests
• number of incident points—adjusted to account for prisoner numbers.

DJR allocated a threshold for each SDO and KPI which prisons are required to meet. For some measures, prisons must stay below the threshold—for example, SDO 8 (random general urinalysis)—while for other measures, they need to meet or exceed the threshold—for example, SDO 14 (engagement in meaningful activity).

In addition to the SDOs and KPIs, under the new contracts, the private operators are subject to four charge events. These events are deaths, escapes, acts of material indiscipline, such as riots, and professional misconduct.

If a charge event occurs and the operator's actions or failure to act contributed to the event, the operator must pay a specified amount to the state.

Figures B1, B2 and B3 summarise the charge events, SDOs and KPIs in Victoria's prisons.

Figure B1
CV's charge events for private prisons

Event	Intent	Measure
Prisoner escapes	To maintain lawful custody of all prisoners.	Each Level 2 and Level 3 Prisoner Escape.
Deaths	This charge event is intended to emphasise the importance of: a secure facility the safety of prisoners, staff and visitors staff acting professionally good intelligence models healthy interactions between staff and prisoners.	Each death (primarily caused by a failure of the operator).
Serious professional misconduct	As above.	Each incident of serious professional misconduct.
Acts of material indiscipline (riot/hostage)	As above.	Each act of material indiscipline.

Source: VAGO based on data provided by CV.

Figure B2
CV's SDOs

SDO		Intent	Measure	Prisons measured
1	Escapes	To minimise the number of escapes and maintain lawful custody of all prisoners.	Number of incident points per 100 prisoners incurred in the relevant period due to prisoner escapes.	All
2	Assaults on staff or other persons by a prisoner	To minimise the number and severity of assaults on staff or other persons by prisoners.	Number of incident points per 100 prisoners incurred in the relevant period due to assaults on staff or other persons by prisoners.	All
3	Out of cell hours	To maximise prisoners' daily out of cell hours, consistent with the approved regime under which they are managed, while acknowledging that the maintenance of a safe and secure prison environment will at times require the restriction of prisoners to their cells or specified areas during normal unlock hours.	Number of incidents of failure to provide prisoners with the Approved Out of Cell Hours for the relevant period.	All
4	Number of unnatural deaths	To have no unnatural deaths of prisoners.	Number of unnatural deaths that have occurred in the relevant period	All
5	Self-harm	To minimise the number of incidents of self-harm and ensure the effective management of prisoners at risk of self-harm or suicide.	Number of incident points per 100 prisoners incurred in the relevant period due to self-harm incidents.	All
6	Assaults on prisoners by other prisoners	To minimise the number and severity of assaults on prisoners by other prisoners.	Number of incident points per prisoners incurred in the relevant period due to assaults on prisoners by other prisoners.	All
7	Assaults on prisoners by staff	To prevent any assaults on prisoners by staff.	The number of assaults by staff that have occurred in the relevant period.	All
8	Random general urinalysis	To measure and deter the use of illicit drugs within prisons.	Percentage of Positive and Positive Equivalent Random general urinalysis results of an Effective Sample.	All
9	Health assessment within 24 hours	To assure prisoners have their general and mental health needs assessed promptly.	Percentage of prisoners assessed within the required time frame.	All
10	At risk assessments	To ensure prisoners identified as being 'at risk' promptly receive an 'At Risk' assessment.	Percentage of prisoners seen within the required time frame.	All
11	Chronic health care plans	To identify, manage and improve the health and wellbeing of prisoners in custody.	Percentage of required Chronic Health Care Plans provided in the time frame.	All
12	Maintenance of accredited health provider status	To monitor that accredited health provider status is maintained.	Accredited health provider status is maintained for all providers.	Port Phillip only as part of old contract
13	Professional misconduct (health staff)	To monitor any incidents of professional misconduct by a healthcare provider or healthcare professional.	Incidents of professional misconduct.	Port Phillip only as part of old contract
14	Prisoners engaged in purposeful activity	To promote prisoner rehabilitation and their constructive engagement.	Percentage of relevant prisoners engaged for a minimum of 60 hours per fortnight.	All excluding MAP and MRC
15	Vocational education and training participation	To encourage prisoners to increase their participation in education.	Percentage of prisoners actively participating in approved education and training.	All excluding MAP
16	Education competency completions	To maximise the number of enrolments in education and training units of competency/modules that result in a successful completion.	Percentage of closed enrolments that result in a prisoner successfully completing that training.	All
17	Prison related harm reduction	The provision of information to prisoners regarding strategies to minimise harm to themselves and others while in custody.	Percentage of relevant prisoners provided with training in the required time frame.	Only applied at MAP, MRC and DPFC
18A 18B	Offending behaviour and alcohol and other drugs program	18A—To ensure the delivery of Offending behaviour programs (OBP) and alcohol and other drugs (AOD) programs.	Percentage of OBP and AOD scheduled programs delivered.	Fulham and PPP in the new contract
		18B—To encourage prisoner completion.	Percentage of prisoners who commence then complete OBP and AOD programs.	All except MAP and MRC
19	Governor's disciplinary hearings	To ensure disciplinary hearings are promptly conducted.	Percentage of hearings completed.	Excluded public prisons from 2010–11. Applied at Fulham and Port Phillip until the end of the previous contracts.
20A	Training Assistance Programs	To ensure that prisoners receive transitional support prior to release.	Percentage of prisoners discharged that participate in the leaving prison information session. Percentage of referrals made to post-release services for prisoners assessed with particular support needs, for example housing or Centrelink.	All prisons
20B	Remand Release Assistance Program	To ensure remand prisoners are provided with information about post release support.	Percentage of remand prisoners who attend a release program within the time frame.	Excludes Loddon, Karreenga and all minimum-security prisons.
21	Prisoner survey	To measure indicators of a healthy prison.	Prisoner survey results.	All
22	Prison industries	To ensure prison industries meet agreed sales targets.	Percentage of gross sales against sales targets.	Excludes Fulham, Port Phillip and MAP
23	Case management	To drive effective case management of prisoners.	Average case management audit score.	All
24	Occupational health and safety	To monitor OHS within the work environment.	Days lost per 100 full‑time‑equivalent staff.	Excludes Fulham and Port Phillip
25	Disability training	To ensure custodial staff in the Marlborough unit at Port Phillip are adequately trained to work with prisoners with a cognitive impairment.	Number of hours and the number of staff who have successfully completed the training.	Port Phillip only

Source: VAGO based on information provided by CV.

The following KPIs are part of the new contracts with Port Phillip and Fulham and are consistent with the Ravenhall contracts. Not all KPIs apply at all three prisons, this depends on the nature of the services provided.

Figure B3
CV's KPIs for private prisons

KPI		Objective	Measure	Prisons measured
1	Health discharge plans	To ensure prisoners with significant health issues requiring ongoing health care are released with a health discharge plan to promote continuity of care upon release into the community.	Percentage of relevant prisoners with a health discharge plan upon release in the relevant period. Threshold = 100% compliance	All
2	Communicable diseases services and immunisation	To ensure all prisoners initially received into the prison system are offered screening for communicable diseases and immunisation status.	Percentage of prisoners initially received into the prison system who are offered screening for communicable diseases and immunisation status within the relevant period. Threshold = 100% compliance	All
3	Positive drug test verification	To increase the effectiveness of urinalysis testing programs in prisons by ensuring prisoners who receive a positive urinalysis drug test result have their test result verified in the required time frame to determine whether a prescribed medication caused the positive urine drug test result.	Percentage of positive drug tests that were verified in the required time frame in the relevant period. Threshold = 95% compliance	All
4	Provision of timely primary health services	To ensure prisoners are provided with timely primary health services.	Percentage of primary health services provided within the required time frames during the relevant period. Threshold = 100% compliance	All
5	Completion of identified drug user (IDU) reviews	To ensure IDU reviews occur in a timely manner.	Percentage of IDU reviews completed within the required time frame. Threshold = 100% compliance	All
6	Release on the correct date	To ensure prisoners are released on the correct date.	Number of prisoners in the relevant period who are released on the correct date. Threshold = 100% compliance	All
8(a)	Carrying out scheduled facility management services tasks	To drive the importance of a secure facility, and maintaining infrastructure quality standards.	Number of incident points incurred in the relevant period for failures to undertake planned maintenance and other facility management services tasks. Rated according to severity. Threshold = up to 20 incident points	All
9	Rectifying facilities management service failures	To ensure that failure events are rectified in a timely manner.	Number of incident points incurred in the relevant period for failures to rectify a failure event within the rectification time. Rated according to severity. Threshold = up to 120 incident points	All
10	Responding to service failure notices and default notices	To ensure that service failures as notified by the state are responded to in a timely and effective manner.	Number of incident points for failures in the relevant period to comply with a service failure notice or a default notice within the required time frames. Rated according to severity. Threshold = 0 incident points	All
11	Incident reporting	To ensure that all reportable incidents and notifiable incidents are recorded and reported in the required data systems.	Number of failures to report any reportable incident or notifiable incident in the relevant period. Threshold = 100% compliance	All
12	Performance data	To ensure that the requirements with respect to performance data are met, including provision of data, and ensuring the availability, integrity and accuracy of data.	Number of failures to provide relevant performance data (defined in the contract). Threshold = 100% compliance	All
13	Providing plans and reports	To ensure that the requirements with respect to provision of plans and reports are met.	Number of failures in the relevant period to meet the requirements with respect to plans and reports as outlined in Schedule 25 of the contracts. Threshold = 100% compliance	All
14	Trend performance	To drive the early identification of multiple service indicators that are at risk of not being met, and the development and implementation of strategies to respond to and avoid this actually occurring.	Incidents of multiple failures of service indicators either in the given quarter or in two consecutive quarters. Threshold scaled according to severity of the failure.	All
15	Reintegration KPIs measure whether assessments occur of individual prisoner's needs after release, that referrals for support are made and that these supports are maintained. Not all prisoners are eligible or require all supports and are therefore not included in the measure. Education or training—completion of education or training activities following release. Employment—maintenance of stable employment of 20 hours or more per week for two months following release. Housing—maintenance of stable accommodation for two months following release. Alcohol and other drugs—maintenance of alcohol and other drug treatment programs for two months following release. Mental health—maintenance of mental health treatment for two months following release.			Ravenhall only
16	Reducing recidivism	To measure the rate at which prisoners released from the prison return to any prison within a specified period of time, providing an indication of the influence of the contractor upon recidivism.	The percentage difference between the rate of returns to prison for qualifying prisoners who are released from the prison, and the rate of returns to prison for qualifying prisoners who are released from other Victorian prisons, measured two years after release.	Ravenhall only
17	Provision of timely forensic mental health services	To ensure prisoners have access to timely forensic mental health services.	Percentage of forensic mental health services provided within the required time frame during the relevant period. Threshold = 100% compliance	All
18	Percentage of relevant prisoners with a mental health recovery plan	To ensure that prisoners with a mental illness or a significant mental health issue are provided with a mental health recovery plan in a timely manner.	Percentage of relevant prisoners with a mental health recovery plan established within the required time frame. Threshold = 100% compliance	All
19	Review of mental health recovery plans	To ensure effective and appropriate treatment and ongoing management of prisoners with a mental illness or significant mental health issue.	Percentage of mental health recovery plans that were reviewed within the required time frame. Threshold = 100% compliance	All
20	Follow up of prisoners discharged from mental health inpatient care	To ensure prisoners discharged from mental health inpatient care are followed up within the required time frame to ensure continuity of care.	Percentage of relevant prisoners who have been followed up within the required time frame. Threshold = 100% compliance	Ravenhall and Port Phillip
21	Mental health discharge plans upon release	To ensure prisoners with mental illness or significant mental health issues requiring ongoing care are released from prison with a mental health discharge plan to promote continuity of care upon release into the community.	Percentage of relevant prisoners with a mental health discharge plan upon release in the relevant period. Threshold = 100% compliance	Ravenhall and Port Phillip
22	Forensic mental health service assessment within 24 hours	To ensure that prisoners transferred for the purpose of a forensic mental health service assessment are assessed within the required time frame.	Percentage of relevant prisoners where the forensic mental health service assessment was completed within the required time frame. Threshold = 100% compliance	Ravenhall only
23	Management of referrals to Forensicare	To ensure that the contractor adheres to the criteria developed by Forensicare for the referral of prisoners to the forensic mental health services.	Percentage of referrals to the forensic mental health services in the relevant period that adhere to the criteria for referral of prisoners to the forensic mental health services. Threshold = 90% compliance	Ravenhall only
24	Reintegration assessment and referral	To ensure that the contractor prepares individual reintegration plans and, where a need is identified, refers prisoners to its Ravenhall Alliance partners.	Percentage of prisoners released from the prison for whom individual reintegration plans are developed and who are, where a need is identified in the individual reintegration plan, referred to an appropriate Ravenhall Alliance partner. Threshold = 100% compliance	Ravenhall only

(a) There is no KPI 7.

Source: VAGO based on information provided by CV.

Back to top

CV granted concessions to both operators on the measurement of their performance against SDO 23 (prisoner case management) in 2016–17. The circumstances leading to this decision demonstrate the importance for CV to consider any contractual implications for the state and private operators when planning any substantive changes in system-wide standards, initiatives or performance measurement approaches.

Correctional staff are expected to assess the risks and needs of individual prisoners and engage with them to promote behaviour change by motivating them to participate in programs and services, coordinating access and reviewing progress. These activities aim to improve management of prisoners, reduce risk to the community and decrease the number of prisoners returning to prison.

Each prisoner has an individual management file for recording key information, activities and engagement. SDO 23 relates to prisoner case management, with the operators' performance measured based on audits of a randomly selected sample of prisoner files each month.

CV selects the sample files for audit and mandates an audit tool. The audit tool specifies standard criteria and point scores for expected file content, in the categories of 'unacceptable', 'basic' or 'good'. Extra points can be allocated for additional or exceptional work. The audits produce a score for each file which is averaged across the total number of files reviewed to determine an overall result. The audits are not independent, as each prison operator audits its own files. However, CV undertakes monthly checks on the robustness of file audits and its decision is final.

The SDO definition and measurement methodology, including the audit tool, form part of the agreed contractual arrangements between the state and the private operators. The performance of the operators against SDO 23 contributes to the calculation of any performance payments under the contracts.

CV changed the audit tool for SDO 23 in August 2016. It intended for the amended tool to be effective from 1 September 2016 for all prisons, including Fulham and Port Phillip. However, it did not update the contracts with the two private operators to incorporate the new tool.

GEO raised concerns about the impact of the amended tool on its performance at Fulham, and CV agreed that it did not need to apply the new tool until November 2016. Fulham failed SDO 23 in November and December 2016 and argued that the change to the audit tool represented a material change in standard under its new contract. Port Phillip also experienced difficulties in meeting the SDO threshold using the new audit tool.

CV's CMB reviewed the impact of the new audit tool across the state and found that nearly all prisons were failing SDO 23 since the introduction of the new tool. This compared with only two prisons failing the SDO at the same time in 2015–16. The review also identified that

• practices considered 'good' case management using the previous audit tool are considered 'basic' under the amended tool
• the criteria under the new tool were seen as a stretch target to promote improved case management.

Given these circumstances, CV provided GEO with additional support and guidance in this area, and the Commissioner allowed performance for SDO 23 to be measured using the old audit tool. G4S was granted the same relief, and DJR will exclude SDO 23 results when calculating Budget Paper 3 performance for 2016–17.

CV intends to measure performance for SDO 23 using the new audit tool for all prisons from 1 July 2017. It will need to update the contracts for the private operators to reflect this.

Back to top

The table below lists the contractual improvements or 'negotiation benefits' described by DJR in the contract administration manuals for the new contracts, and our comments on the validity of the claimed improvements and benefits.

Figure D1
Improvements in the new contacts

Claimed feature or improvement	Description and VAGO comment
Long-term contractual arrangements	The initial terms are 11 years (Fulham) and 10 years (Port Phillip). The further extension terms are only available if GEO and G4S achieve a set of contractual, commercial and operational thresholds measured over a specified three-year period. The operators have an automatic right to extension if they meet these thresholds; otherwise, the decision to extend reverts to the state. While these contract terms are longer than typical service contracts, this was seen as necessary to resolve the lease misalignment issues. The potential disadvantages to the state of being 'locked-in' to a long-term contract are mitigated through the performance-based mechanism for earning a further extension term and more generally through provisions which allow the cost of the services to be reviewed, benchmarked and/or indexed. Overall, the contract terms, including the mechanism for earning a further term, strike a good balance between providing long-term incentive for the operators and preserving flexibility for the state.
Updated contractual provisions and commercial regimes, including refined risk allocation and improved enforceability through greater clarity of provisions	The clarity and enforceability of the provisions in the new contracts will be tested over time. Despite this, it is clear that the new contracts represent a fundamental rewrite and have been 'modernised' to align more closely with the Partnerships Victoria standard commercial principles. The contracts align closely with the model used for Ravenhall which should support efficient and effective contract management of the private prisons. That said, the commercial mechanisms in the contracts are complex and will require significant resources to manage.
A rigorous asset management and investment regime	The new asset management and investment regime is an improvement over the initial contracts. The contracts include much more specific facilities maintenance requirements, and an asset life cycle schedule and associated funding schedule that were rigorously tested during negotiations. The parties established a 'sinking fund' model to promote strategic management of asset life cycle and maintenance investment. The state funds this activity and will share in any cost savings at the end of the term.
Improved clarity and detail within the contracted services specification	A detailed output specification defines the services to be provided under the new contract, covering both correctional services and facilities maintenance and management. This is similar to Ravenhall and incorporates system-wide requirements. The services specification is an improvement on those in the initial contracts.
Restructured payment, performance and reporting mechanisms, including a strengthened performance incentive regime	This claim is correct and, at a high level, the risk allocation is the same for Fulham and Port Phillip. The risk allocation is evidenced by the overarching obligations on the operators imposed by clause 4 and related clauses including the payment and performance regime. Under the initial contracts, performance was only assessed once a year to determine the performance payment. This meant that there was no incentive for operators to try to address poor performance on an SDO if it had failed to a degree that would impact its performance payment.
The entire service payment is now at risk—through deductions for poor performance and other regimes in the contract	Due to the operation of the 'Available Prisoner Place' concept and the service-linked fee, it is possible that payments to the operators could be close to zero (ignoring 'Pass Through Payments'). Although, this is largely theoretical, as it is unlikely that all the prisoner accommodation would fail the 'Available Prisoner Place' test at once.
An improved modifications regime within the contract	The new contracts improve the modifications regime compared to the initial contracts. They provide the state with flexibility to adapt services and/or facility requirements, with pre-agreed pricing parameters.
Capacity 'tranching' regime, allows the state to lower or increase the required bed capacity in the prisons in increments of 50 and adjust the service payment accordingly	This regime provides flexibility to the state to adjust prisoner numbers in these two prisons at a pre-agreed price to match system-wide requirements. The regime allows the state to turn tranches of 50 prisoner places off or on and decrease or increase the service payment. There are up to six tranches for Port Phillip and four tranches for Fulham. The value of this mechanism is linked to the size of the relative price reduction and the cost structures of the private operators, so is difficult to assess. We understand that the state has flexibility to implement a more permanent change to prisoner numbers through the modifications regime as an alternative to the tranching regime.
Improved contractual security for the state, including a parent guarantee and increased performance bonds	The new contracts provide the state with improved direct security in addition to its ability to reduce payment for poor performance and its step-in and termination rights. This security is in the form of a parent company guarantee and a performance bond. The performance bond for Fulham increased from $3 million to $8 million, and for Port Phillip, from $8 million to $12 million. Parent company guarantees: For Fulham, the parent company guarantee liability of the guarantor (The GEO Group, Inc.) is limited to the liability of the contractor under the State Project Documents, but no upper cap is specified. For Port Phillip, liability is similarly limited but an overall cap is specified. The Guarantor is G4S Regional Management (UK&I) Limited.
High contractual and commercial consistency with the Ravenhall contract	Consistency across the three full-service private prison contracts should provide contract management and operational management efficiencies for the state. Although there are prison-specific differences, it is reasonable to suggest that the similarities will provide management efficiencies for the state. Despite this, the commercial mechanisms in the contracts are complex and will require significant resources to manage.
Alignment of crown lease and contract term, including improved end-of-term arrangements for the state	The negotiations for the new contracts have aligned the lease terms with the terms of the new contracts and improved end-of-term arrangements for the state. The revised lease arrangements provide for the operators to surrender these leases at the end of the service terms under the new contracts. Ownership of the facilities has reverted to the state from the commencement of the new contracts with the exception of furniture, fixtures and equipment at Fulham.

Source: VAGO based on information from CV and the original and new contracts for Fulham and Port Phillip.

Back to top