Government’s use of private finance for infrastructure

The Treasury has not identified which financing models represent value for money for different types of infrastructure assets. A range of financing models are currently in use for delivering public infrastructure including but not limited to: Contracts for Difference; Regulated Asset Base models; and Public Private Partnerships, of which the extensively used Private Finance Initiative (PFI) is an example. To achieve value for money, it is essential that the additional costs of private finance are justified by the benefits offered. The NAO has previously reported 3 that the Treasury did not consider the cost of government borrowing to be relevant when making financing decisions on PFI deals, and that the value for money assessment favoured off-balance sheet solutions, which gave the illusion of lower public borrowing. The Treasury now says that the correct private finance model should be chosen for projects and that financing decisions should not be conditional on achieving off balance sheet classification. However, the Treasury has yet to identify the types of financing model it will support for various project types, such as energy, transport, or communication. Doing so will allow public bodies to be clearer on how they might deliver infrastructure and encourage investor participation, drive competition, and improve value for money. recommendation To maximise the chances of delivering value for money, the Treasury should evaluate the costs and benefits of alternate financing models, including the different costs of borrowing in the public and private sectors, to identify a preferred model for different types of infrastructure.

There is no central record of private finance for infrastructure investment, which limits the Treasury’s ability to spot themes and patterns and deliver value for money. The overall value and number of infrastructure projects delivered using private finance models other than PFI are significant. For example, on the Regulated Asset Base model, around £9 billion was invested – including for Thames Tideway Tunnel and Heathrow Terminal 5. In addition, the government has recently announced that it will invest £14.2 billion in the Sizewell C nuclear power station. In June 2018, the previous Public Accounts Committee recommended that the Treasury and IPA publish data on the benefits of PFI and set out an approach to evaluating the value for money of operational PFI projects. In 2021, the IPA told the Committee that it was unable to conclude its evaluation, noting that this was because it was unable to make any meaningful conclusions from the data. NISTA, which has replaced the IPA, says it is now collecting more data on performance, asset condition, and PFI contract expiry so that it can learn the lessons. Unlike PFI, other financing models are not monitored by the Treasury or NISTA and there is still no information about how the relevant data will be collected and monitored centrally. recommendation The Treasury and NISTA should publish a central database covering private finance for public infrastructure. This should include (subject to commercial sensitivity), but not be limited to: • project synopsis; 4 • lead department and counterparties; • capital value; • finance type, including any government support packages; • contract length; • project business cases, including evaluation plans; and • forecast equity returns to investors.

The government’s ambitions for infrastructure investment are at risk of not being achieved because of sector specific skills shortages. Using private finance to support investment requires public bodies to develop and maintain specialist commercial and financial skills. The NAO has previously reported that contracting authorities often had limited in- house skills available to make critical decisions on complex projects, which can place the public sector at a disadvantage. Building specialist skills takes time and the uncertain infrastructure environment in the UK has led to a weakening of the essential skills needed for successful project delivery. The Treasury recognises these challenges and in response has created functions and centres of expertise to support public bodies. NISTA says it is running a number of training programmes and has an associate pool of 60 experts in various fields providing advice and support to public bodies. The government has also announced that the 10-year infrastructure strategy will support Skills England in its assessment of where skill gaps exist that will need to be addressed to successfully deliver key infrastructure projects. recommendation The Treasury and NISTA should set out plans for building, attracting and retaining the necessary skills to support public sector infrastructure investment. This includes skills within both the public and private sectors.

There is no comprehensive framework for considering risk allocation between the public and private sector when working in partnership. Risks should be borne by those who are best able to manage them and should be priced appropriately. Not all risks can or should be transferred to the private sector because the cost of inappropriate transfer of certain project risks to the private sector could be disproportionately high, as ultimately, the government may have to pick up those risks, including the completion of projects if the private sector partner fails or is unable to deliver to the requirement. The lack of guidance in quantifying risk as seen for PFI contracts helped to fuel some misalignment between the additional costs of PFI, private sector return on investment (which was considered 5 disproportionately high) and the actual level of risk incurred. The Treasury says that the benefit of private finance includes the incentive for more risk management within the private sector. However, we heard that sometimes there is a misplaced belief that risk transfer to the private sector equates to risk management by the private sector, and therefore there is a false assurance that the problem lies elsewhere. This is not always the case, as ultimately the government may need to step in if a major supplier of critical infrastructure were to fail, as was the case for Carillion. recommendation The Treasury should develop a specific infrastructure financing toolkit to support public bodies in quantifying, allocating, monitoring, and managing risks for infrastructure projects. This should include the consideration of contingencies for supplier failure.

Poor contract management is impacting the quality and condition of PFI assets being handed back to the public sector. Public bodies are due to pay £136 billion in unitary charges up until 2052–53 for all 665 ongoing PFI contracts, with half of these contracts set to expire within the next decade. However, ongoing challenges around PFI asset condition—for example in schools—need careful management to ensure only quality assets are handed back. We were pleased to hear about the hard work that NISTA’s PFI centre of excellence is undertaking to manage these issues, including the use of toolkits and experts. We understand that some issues with contract management stem from the assumption that PFI would be self-monitoring, with private sector providers reviewing their own performance and reporting back to the public body on the other side of the contract. This rarely happened in practice and should not preclude public bodies from robust contract management to ensure that private sector providers comply with their contractual obligations. The Treasury and NISTA say that the earlier PFI contracts did not have clear provisions for the condition of assets on their return to the public sector at the end of the contract. To support public bodies in managing these contracts, NISTA has published an asset condition playbook, rolled out technology and other toolkits to support contract managers. The lessons here apply equally to any other infrastructure financing model. recommendation The Treasury should set out how it will improve central support for public bodies in assessing the state of their privately financed assets and enforcing contractual mechanisms to ensure that well-maintained assets are returned to the public sector. 6 1 Government’s infrastructure strategy Introduction

On the basis of a report by the Comptroller and Auditor General, we took evidence from HM Treasury (the Treasury) and the National Infrastructure and Service Transformation Authority (NISTA) on the lessons learned from government’s past use of private finance for infrastructure projects.1 We also took evidence from Darryl Murphy, Managing Director, Infrastructure at Aviva Investors, and Bruce Dalgleish, Partner at P2G LLP.

To improve the transparency of longer-term investment priorities and to support investors with planning, the IPA had published the National Infrastructure and Construction Pipeline annually from 2016 to 2024 – with the exception of 2019, 2020 and 2022. The 2024 publication contains details of the forthcoming procurement and assumptions for the estimated value of the investment that will be delivered via public finance and private finance across a range of infrastructure types.10 There has been no analysis of past performance for projects in the pipeline, as the data is too inconsistent to compare pipelines year on year, and three missing publication years 7 Committee of Public Accounts, Government’s use of private finance for infrastructure 8 Q 2 9 Qq 10 and 12 10 C&AG’s Report, para 2.11 8 have further reduced the value of the pipeline. The 2024 iteration of the pipeline was also a single snapshot in time, reflecting the previous government’s priorities.11

We highlighted to NISTA that the infrastructure pipeline has been too short-term and lacked consistency, which has caused issues with future planning and made longer-term decision-making for investors difficult.12 Written evidence we received also referenced the lack of a central pipeline of infrastructure projects and investment opportunities in the UK, which has resulted in investors relying on “multiple and disparate sources” of information to identify upcoming projects (largely through private sector advisors).13 NISTA indicated that the real challenge with the pipeline is not so much working up the finance but developing “investable models”. NISTA also told us that their aim, therefore, is to develop an investable set of projects that will attract not just domestic capital, but international capital. This would help to drive down prices and achieve a better out-turn, in terms of the number of projects delivered and their value for money.14

We asked NISTA what it was doing to create a pipeline for investors that supports long-term decision making. It told us that it was aware the previous pipelines were construction industry focused, contained gaps, and only gave information on costs and schedules.15 In response to these issues it has committed to publishing a pipeline every six months, which is investor-focused and provides greater granularity on the types of financing models that various projects will be eligible for.16 11 C&AG’s Report, para 2.12 12 Q 25 13 PFI0005 14 Q 30 15 Q 25 16 Qq 25-27 9 2 Evaluating the choice of finance The value for money of different financing models

The government does not have an evidence base upon which to judge which financing models are best suited for particular projects and as such, the government may not be achieving value for money on some infrastructure projects.17 In addition, the absence of an evaluation means that government cannot make the necessary assessments to determine the appropriateness of financing models. As a result, there is a lack of clarity for investors on what projects the government is willing to support and what financing approach they may take in each case.18

Though government attempts to determine the success of a project through measures such as cost-benefit analysis, they do not routinely evaluate completed projects.19 The UK has used a variety of private financing models to deliver new infrastructure, including but not limited to: Contracts for Difference, which guarantee wholesale prices for generators over 15-year periods (e.g. offshore wind); Regulated Asset Base models, which allow a private sector provider to charge users a regulated price for utilities and include an element in bills to fund new investment; and the Private Finance Initiative (PFI)20 used extensively between 1992 and 2018 to design, build, finance, and operate infrastructure projects for public authorities, such as schools and hospitals.21

The National Audit Office’s (NAO) report on evaluating government spending highlighted that much of government’s activity is either not evaluated robustly or not evaluated at all, despite the government’s commitment to evidence-based decision-making.22 In addition, there 17 C&AG’s Report, para 3.26-3.27 18 C&AG’s Report, para 3.27-3.29 19 C&AG’s Report, para 3.26 20 C&AG’s Report, Figure 6 21 C&AG’s Report, para 1.7 22 C&AG’s Report, Evaluating government spending, Session 2021–22, HC 860, December 2021, p 15 10 have been no assessments of the merits of private financing models relative to public procurement. Attempts have been made to remedy this issue, such as in 2021 when the government set up the Evaluation Task Force to support best practice in public policy evaluation across its departments. The following year, the Task Force published the evaluation strategies setting out how departments would undertake and learn from evaluation activity.23

The Treasury told us that the proportion of major projects that have been evaluated has improved since the NAO’s evaluation report. After our evidence session, the Treasury wrote to us and stated that 68% of infrastructure and construction of major projects have a high-quality evaluation plan. Across all major projects, 63% have at least some form of evaluation plan – but only 34% were deemed high quality. This leaves 37% of all major projects with no evaluation plan.24 The Treasury told us that to address some of these issues, the Evaluation Task Force published an action plan to keep improving evaluation across the major projects portfolio.25

Historically, private financing was sometimes selected over public financing because it did not increase short-term government borrowing. This kept public sector debt levels down, as the assets were considered off-balance sheet for national accounts purposes. The Office for Budget Responsibility called this a “fiscal illusion”: where the accounting measure does not reflect the true fiscal implications of a transaction.26 We asked the Treasury whether infrastructure financed through Public Private Partnerships should be on or off balance sheet. The Treasury told us that the Green book stipulates that the balance sheet classification of a project should not be a determining factor of whether a particular financing model is chosen and that the focus should be on choosing the correct structure to ensure the project will be delivered well and provide value for money, whether on or off balance sheet.27 Centrally held data on private finance

The overall value and number of infrastructure projects delivered using private financing models is significant, for example, on the Regulated Asset Base model, a combined £9 billion was invested in Thames Tideway Tunnel and Heathrow Terminal 5.28 In addition, the government announced in June 2025 that it will invest £14.2 billion in the Sizewell C nuclear 23 C&AG’s Report, para 3.29 24 Q 60; Correspondence from HM Treasury, 27 May 2025 25 Q 60; Government Major Projects Evaluation Review, Evaluation Task Force, April 2025 26 Qq 4 and 6; C&AG’s Report, paras 3.21-3.22 27 Qq 24 and 57 28 C&AG’s Report, para 1.8 11 power station.29 With the exception of PFI contracts, private finance contracts are not monitored centrally because responsibility for them rests with departments and sector regulators.30 Written evidence we received from Affinitext highlighted that the absence of complete, digitised and searchable contracts undermined effective contract management, preparation for assets to be handed back to the public sector at the end of a contract, and value for money assurances.31

The Treasury publishes data on the number of active PFI contracts, which includes information on: the capital value of assets; when contracts are due to expire; and the remaining unitary charges.32 As of 31 March 2024, there were 665 PFI contracts with a capital value of £50 billion – 79% of the value relating to social infrastructure and the remaining 21% to economic infrastructure.33 The NAO has previously reported that there was a lack of systems to collect comparable data for similar projects using different procurement routes, and the previous Public Accounts Committee also highlighted that there was a lack of data available to assess the actual efficiency of PFI projects.34 In June 2018, the then Committee recommended that the Treasury and IPA publish data on the benefits of PFI and set out an approach to evaluating the value for money of operational PFI projects.35 Doing so would allow for lessons learned from the evaluation to be applied to other projects, including on construction or delivery, future expiries, and contract management.36

In 2021, the IPA told the previous Committee that it was unable to conclude its evaluation of the benefits of PFI because it was unable to draw any meaningful conclusions from the data.37 The Treasury told us that the IPA was asked to deliver the information on benefits but found that the data was not available for what the private sector had done and what the public sector comparator would have been. In addition, NISTA told us that it was 29 Qq 24, 43 and 59; C&AG’s Report, para 1.8; HM Treasury, Thousands of jobs to be created as government announces multi-billion-pound investment to build Sizewell C, June 2025; Sizewell C, Green Light for Sizewell C – The First Majority British-Owned Nuclear Power Plant in Over 30 Years, June 2025 30 Q 68; C&AG’s Report, para 1.8 31 PFI0014 32 C&AG’s Report, Figure 3 33 C&AG’s Report, para 1.7 34 C&AG’s Report, PFI and PF2, Session 2017–19, HC 718, January 2018, p 5; Committee of Public Accounts, PFI in Housing and Hospitals, Fourteenth Report of Session 2010–11, HC 631, January 2011, p 10 35 C&AG’s Report, para 3.28; Committee of Public Accounts, Private Finance Initiatives, Forty-Sixth Report of Session 2017–19, HC 894, June 2018, p 5 36 Qq 32 and 61 37 C&AG’s Report, 3.28; HM Treasury, Treasury Minutes Progress Report – Update of the Government responses to the Committee of Public Accounts on Sessions 2010–12, 2012–13, 2013–14, 2014–15, 2015–16, 2016–17, 2017- 19, 2019 and 2019–21, CP 424, May 2021, pp 63-64 12 collecting data relating to asset condition and upcoming expiries in order to target and assist local authorities through the contract expiry and asset hand back process.38 38 Q 32 13 3 Contract management and capability Skills and capability

The Public Accounts Committee has, in recent years, frequently drawn attention to how public bodies lack the necessary in-house skills to deliver on plans and programmes.39 We also heard from Mr Dalgleish that the lack of a credible pipeline has meant investment cannot be deployed, resulting in skills and labour going elsewhere, including the international market.40 Using private finance to support investment requires public bodies to develop and maintain specialist skills.41 The government requires access to sufficient specialist expertise at reasonable costs to work across organisational boundaries to support technical, commercial and financial assessments and decision-making. In recognition of these challenges, the government has created specialised delivery organisations and functions with the purpose of consolidating centres of expertise that other public sector organisations can draw from.42

There is a general recognition that sectors, including the construction industry, may lack the appropriate skills and workforce to deliver on the government’s infrastructure strategy. NISTA told us that, in collaboration with the Department for Business and Trade and the Department for Education, it has a role in supplying the information needed to determine the skills requirement. This will in turn link to the pipeline and the 10-year infrastructure strategy that organisations such as Skills England will use to identify what the long-term requirements for skills in the UK are over 10, 15 or 20 years to ensure that infrastructure projects can be delivered.43 39 Committee of Public Accounts, Delivering value from government investment in major projects, Thirty-Second Report of Session 2023–24, HC 456, May 2024; Committee of Public Accounts, HS2: Update following the Northern leg cancellation, Tenth Report of Session 2024–25, HC 357, February 2025 40 Qq 10-12 41 C&AG’s Report, para 2.17 42 C&AG’s Report, para 2.21 43 Q 68 14

Using private financing requires skills and capabilities from public and private sector partners including, financial, commercial, construction, project and contract management abilities.44 The lack of a stable and consistent pipeline has contributed to the private sector not investing as much as it might have, while long contracts make it difficult to retain staff with particular knowledge and expertise for the full duration. Skills also diminish where departments have few projects.45 NISTA reported that part of the desire to improve the availability of expertise and capability within public authorities was to help reduce the number of costly legal disputes. NISTA told us it has been running several programmes to train contract managers within the public sector, that it has issued guidance and toolkits, and published an asset condition playbook to help contract managers manage their underlying contracts. NISTA also told us that its most “influential” programme is the associate pool of 60 experts from multiple fields who provide advice and support to public authorities.46

The Treasury told us that the formation of NISTA, which brought together the strategy of the National Infrastructure Commission and the delivery of the IPA, brings increased focus on contract management.47 In addition, it told us that public financial institutions, such as the National Wealth Fund, will be responsible for large financial transaction projects on the basis that they have the skills to “do them properly”.48 Risk allocation

There are many risks to delivering major infrastructure projects, such as financing risk, construction risk, and demand for the asset being lower than expected.49 These risks should be transferred to those best able to manage them, and when they are transferred to the private sector, it comes with costs: it is fair that the higher the risk, the higher the return.50 Not all risks can or should be transferred to the private sector because the cost of inappropriate transfer of certain project risks to the private sector could be disproportionately high, as ultimately, the government may have to pick up those risks, including completion of projects, if the private sector partner fails or is unable to deliver to the requirement. A lack of 44 Qq 10-11 and 33; C&AG’s Report, para 2.17 and 2.20-2.21 45 Qq 3, 11-13 and 73 46 Qq 32-33 47 Qq 32 and 68-69 48 Q 32 49 C&AG’s Report, para 3.2; Committee of Public Accounts, Private Finance Initiatives, Forty-Sixth Report of Session 2017–19, HC 894, June 2018, p 13 50 C&AG’s Report, para 3.16; C&AG’s Report, Equity investment in privately financed projects, Session 2010–12, HC 1792, February 2012, p.11; Committee of Public Accounts, Private Finance Initiatives, Forty-Sixth Report of Session 2017–19, HC 894, June 2018, p 17 15 guidance in quantifying risk for PFI schemes, for instance, helped to fuel some misalignment between the additional costs of PFI, the private sector’s return on investment (which was considered disproportionately high), and the actual level of risk incurred.51 For some other types of financing models, the government sought to ensure that excessive risk did not deter private investment, such as in the case of the Carbon Capture, Usage and Storage (CCUS) programme. The private sector invested significantly in CCUS after the Department for Energy Security and Net Zero applied lessons learned from previous failed attempts to launch CCUS in the UK and created business models that allocated costs and risks effectivel

We asked the Treasury how it decides on which private finance model to use, and how it assesses the additional cost of private finance against the benefits of risk transfer. The Treasury told us that the benefit of private finance is that the private sector is incentivised to take on and manage risk. It explained that the private sector performs more front-end due diligence on projects because it tends to have access to more skilled and experienced people for delivery of these projects, which can lead to better performance and outcomes. The Treasury added that the returns that the private sector receives is determined by their assessment of the risk in the contract and the difficulty of delivering it, and so the private sector is incentivised to give the best price for both finance and the delivery contractor.53 The Treasury highlighted that when deciding whether to choose private or public finance, the Green Book dictates that the risk that the public sector is retaining should be considered, even if public financing is cheaper.54

We heard from Mr Dalgleish that sometimes, with private finance, government has a misplaced assurance that risk transfer to the private sector equates to risk management by the private sector, which offers a false assurance that the risk lies outside of the public sector.55 Risk management by the public sector is paramount, as past examples have shown that the government will need to step in when suppliers fail, as was the case when Carillion went into liquidation and government had to complete two hospital projects.56

We asked the Treasury how it supports government departments to prepare for and manage potential supplier failure, and what contingency plans it had in place to address supplier failures, particularly for critical services, to avoid over-reliance on Government support. The Treasury 51 C&AG’s Report, para 3.7 52 Q 62; C&AG’s Report, para 3.9 53 Qq 37-38 and 41, 45 54 Qq 37 and 52 55 Q 3 56 Qq 23, 79 and 81; C&AG’s Report, Investigation into the rescue of Carillion’s PFI hospital contracts, Session 2019–20, HC 23, January 2020, pp 5-6 16 highlighted that being ready for failure was one of the lessons learned from Carillion that the NAO and government had pointed to. The Treasury added that there are “living wills” in place within contracts which list all the services that are reliant on the supplier and how those services will continue in the event of insolvency. The Treasury also told us that all departments need contingency planning and to support this, it had published four playbooks by sector, 12 guidance notes, and a sourcing playbook which includes some of the resolution plans. Where necessary, the Government Commercial Function also assists departments with drawing up management plans.57 Contract management

Half of the 665 ongoing PFI contracts are set to expire within the next decade, at which point the assets usually revert to the public sector.58 It is important that public authorities manage these contracts well so that assets in good condition return to the public sector.59 Unitary charges are the fees PFI companies charge public authorities to make the asset, for example a school, available. These unitary charges, which are agreed at the beginning of the contract, run for the duration of the contract term (typically 25 to 30 years).60 As of 31 March 2024, public sector bodies were due to pay £136 billion in unitary charges up until 2052–53 for ongoing PFI contracts.61

The NAO has previously reported that the public sector does not take a strategic or consistent approach to managing PFI contracts as they end, and risks failing to secure value for money during expiry negotiations with the private sector. Around one-quarter of respondents from public bodies to a survey stated that contracts did not contain any information on how and in what condition assets should be returned. Poorly drafted clauses open to interpretation resulted in differing views between authorities and PFI providers.62 We asked the Treasury and NISTA whether they had learned the lessons from this issue.63 The Treasury told us that at the initiation of private finance there was talk about it being self-monitoring, but this had clearly proven not to be the case. The Treasury acknowledged that contract management was required, and as a result has set up a contract 57 Qq 79-80 58 C&AG’s Report, para 4.18 59 Qq 72-73; C&AG’s Report, para 4.22 60 Q 73; C&AG’s Report, Figure 3, Note 2 61 C&AG’s Report, para 4.18; Appendix One, para 15; Figure 3 62 C&AG’s Report, Managing PFI assets and services as contracts end, Session 2019–21, HC 369, 5 June 2020, pp. 7-9 63 Q 54 17 management capability programme.64 NISTA also told us that the early PFI contracts had poorly drafted clauses relating to hand back provisions and that standardisation of contracts addressed this. NISTA added that issues with the PFI contracts coming up for expiry relate to contracts signed prior to the hand back clauses being added.65

In 2020, the IPA set up the PFI Centre of Excellence to provide expert support and advice to contracting authorities, and in the same year established the PFI Contract Management Capability Programme to support contracting authorities managing PFI contracts through the operational and expiry phases.66 In 2022, the IPA further published practical guidance for contracting authorities on managing PFI expiry and service transition. Additionally, in 2023, IPA published A Guide to PFI Expiry Health Checks to support contracting authorities in assessing their readiness for PFI expiry. Further guidance—PFI Asset Condition Playbook—was released in March 2025.67

The NAO noted in its latest report on Private finance for infrastructure that while there has been some progress made, much remains unchanged, with public bodies continuing to show a lack of preparedness for contract expiry – particularly for long-term contracts.68 We asked Mr Dalgleish how effective he thinks the government is at managing PFI or Public Private Partnership contracts in general. He told us that a large amount of good work has been done by what is now NISTA in terms of producing guidance, however, there was now a large amount of data and guidance for public bodies to try to digest, particularly by public bodies who also cannot afford good contract management capability in the first place.69

One of the underlying principles of the PFI model was its design to be self-monitoring, with the project delivery company responsible for reviewing its own performance and reporting back to the contracting authority. But this self-monitoring should not preclude authorities from having appropriate access to data and information to support their own monitoring of projects to ensure that contractual obligations are being met.70 We asked NISTA what it was doing to support public authorities with contract management. NISTA told us about issues with schools in Stoke-on-Trent, where the conditions of assets were not acceptable, but NISTA managed to work with the private delivery partner to secure 64 Q 60 65 Q 54 66 C&AG’s Report, para 4.19 67 Q 32; C&AG’s Report, para 4.21 68 Q 56; C&AG’s Report, para 4.20 69 Qq 3 and 73 70 Qq 60 and 68; C&AG’s Report, para 4.7 18 a £30 million asset improvement fund.71 Additionally, the Treasury and NISTA committed to acting on the lessons learned including, a commitment to ensuring that construction is monitored by a clerk of works and independently certified.72 71 Q 73 72 Q 49 19