Thirty-Ninth Report - Lessons from major projects and programmes

We are concerned about the continued lack of transparency to Parliament about the progress of government projects. This Committee has examined many programmes where issues were emerging for some time that only became public knowledge much later. For example, the Committee was kept in the dark for over a year before being told about the difficulties on the High Speed Two programme, and the programme now reports to Parliament half-yearly as a result of our work. The IPA agrees that creating a culture of openness and transparency is critical to effective project delivery. However, data on individual programmes can be poor, and there can be trade-offs to manage between commercial confidentiality and public transparency. Departmental processes for publishing information, such as in annual reports, may be delayed and/or inconsistent with information presented elsewhere in government. While the IPA states that it has good quarterly reporting in the GMPP and in its annual report, there is a time lag between the reported data and what’s happening on a project. As a result, we question whether this is useful for helping Parliament or the public to understand how major projects are progressing. The NIC told us that GMPP project business cases, outturn costs and schedules and other long-term evaluations will now be published from April 2021, which we welcome. Recommendation: HM Treasury and Cabinet Office must set out what they are doing to make sure that information on the progress of major projects is transparent and reported to Parliament in a more timely manner, particularly 6 Lessons from major projects and programmes for those projects that are nationally important. They should also set out what they are doing to make sure that all Accounting Officers comply with the rules around making assessments of major projects (against the criteria of propriety, regularity, feasibility and VFM) and to publish summaries of this advice on a timely basis. The Committee reminds all AOs

We welcome the IPA’s plans to strengthen how it assures government’s major projects and the recent clarification of it roles and responsibilities, but it remains to be seen whether the IPA will be able to effectively ensure those delivering projects make the level of improvement required. In the past, we have raised concerns that the IPA is not sufficiently independent of government to confidently challenge it, being both a champion of government project delivery as well as being responsible for assuring it. The IPA recognises that while its gateway review process for assuring projects at key decision points is good, it has not always been applied with the rigour needed. Programmes with significant issues have proceeded through Gateways even when they were rated as amber/red. It is now resetting its assurance of the GMPP, to ensure that departments address issues identified at gateway reviews before a programme can proceed, and points to having already made meaningful impact on the Lower Thames Crossing programme. The IPA is working to have a systemic view of what impact it has on programme performance, though it also told us that it was implementing a system to do this when we spoke to them in 2016. We note that government has reinforced IPA’s mandate, alongside clarifying what the responsibilities of departments are in acting on the IPA’s recommendations. Recommendation: Within six months IPA must write to us setting out the changes it has made to its assurance of major programmes in its reset, and how this will improve project delivery. It should also set out what action it took in relation to those projects on the GMPP currently rated as red/red-amber, and what the result has been. The Cabinet Office and HM Treasury should also write to us detailing the actions taken to ensure that Departments act on the IPA’s recommendations.

There is scope for a more mature approach to setting and monitoring cost and schedule estimates by using ranges which narrow over time. The UK government has some of the most novel and complex programmes in the world. The issues that major programmes face are not unique to the public sector, with research indicating that around 75% of major programmes exceed their intended cost and schedule, including those in the private sector. Nonetheless, there are practices in the private sector that the public sector could learn from, such as there being more front-end development of a programme to define its scope, and on getting cost and schedule estimates right before committing to them. The IPA and the National Infrastructure Commission say that they are working to improve the quality of cost estimates across government, and the understanding of what a good estimate looks like. Programmes should make greater use of cost and schedule ranges to reflect the high degree of uncertainty at early stages, with the expectation that these ranges would narrow as the programme is developed further. Lessons from major projects and programmes 7 Recommendation: The IPA should encourage and support departments to use ranges for cost and schedules estimates, rather than single point numbers, and write to us setting out how it is scrutinising and improving cost and schedule estimates on major programmes.

The calibre and number of people in leadership positions in major projects is still not strong enough. Skills and leadership remain a persistent problem in delivering major projects, particularly in getting the appropriate mix of skills across areas such as civil engineering and digital systems. While the tenure of Senior Responsible Owners (SROs) has improved slightly in recent years, SROs still typically manage four or five different projects and so do not get the time to manage them effectively. Matching up an SRO’s skills with the stage a programme is at is also important. The IPA set up the Major Projects Leadership Academy to train SROs and has begun to offer this training to Ministers with responsibility for major programmes. It is looking to build on this work with a Major Projects Academy which is intended to support training and accreditation for the wider project delivery profession across government. This Committee previously recommended that Permanent Secretaries and other civil servants attend the Major Projects Leadership Academy, but it is clear that more work remains to improve the calibre of leadership for major projects. Recommendation: All SROs and people who work in the project delivery profession should attend IPA’s major projects leadership academy. Cabinet Office and HM Treasury should make sure all departments comply, including ensuring that all SROs have sufficient skills and time to be able to undertake their responsibilities. They should write to us to explain how they have assured themselves that this is happening. The IPA should update us in six months about how many people and in which positions, have been accredited and across which departments. It should also explain how it plans to roll out its training to the government project delivery profession as a whole, and whether the government now has, in its opinion, enough SROs to manage the number and scale of projects promised.

It is vital that Parliament’s own major project, the Restoration and Renewal programme, is an exemplar of an open and transparent project which welcomes scrutiny. The restoration and renewal of the Palace of Westminster is a highly complex programme with a number of uncertainties that underpin it, including the condition of the building, interdependencies with other Parliamentary building projects, and the lack of consensus over the scope of the programme and how to relocate staff. Since the programme and the body overseeing it are Parliamentary, the programme does not fall within the GMPP and the IPA’s formal remit, though it has offered advice informally on the programme when invited to. Given the complexity and difficulty of the programme, it would benefit from all available expertise on major programmes and from the additional transparency that could be brought from the IPA reporting on the programme. Recommendation: The government and Parliament should work together to establish a framework for how the IPA will scrutinise and assure the R&R project. 8 Lessons from major projects and programmes 1 Major programme delivery

On the basis of a report by the Comptroller and Auditor General, we took evidence from the Infrastructure and Projects Authority (the IPA) about the delivery and assurance of government major programmes.1

We asked about the capacity of both government and the private sector to deliver large increases in activity efficiently and effectively. The Government’s commitment to reach net zero carbon emissions by 2050 will add further challenges to meeting these commitments.8 The IPA told us that it sees the government’s plans as ambitious but achievable. It has assessed the supply chain and, provided that there are stable long- term plans, it believes industry will invest in the capabilities needed to meet the level of activity planned.9 The National Infrastructure Commission (NIC) agreed in its written submission to us.10

The NIC stated that new economic infrastructure must take a cross-sectoral, evidence- based and long-term perspective.11 It said that this should increase the likelihood that the strategic case will remain relevant and persuasive as time goes on, which is likely to make it easier for delivery bodies to match the scope of the project with its strategic objectives. It told us that in the course of the work on the NIC’s review of Rail Needs Assessment for the North and the Midlands, it has seen a number of examples of the issues the NAO identified occurring with major rail projects, such as significant changes in scope during a project without sufficient review of the objectives and plans, and not considering operations at an early enough stage.12

Within the National Infrastructure Strategy, the government also announced the initiatives that had come out of Project Speed, which it set up over the summer to review the infrastructure project life cycle and identify where improvements could be made in delivering them faster. These initiatives include measures such as reforms to the planning system and environmental regulations, streamline approval processes, and transforming the construction sector through better use of data and modern construction techniques.13 We were concerned that accelerating programmes could hinder the proper scrutiny and development of major projects. The IPA was optimistic about these initiatives, and told us that Project Speed had brought a focus to the duration of certain processes, and was not an excuse to go too fast in delivering programmes. For example, it pointed to variations in the time projects take to achieve their approvals, with some projects achieving approvals in half the time of others, even under existing legislation.14

Government has also made changes to its Green Book guidance on evaluating investment proposals, such as reducing the dominance of benefit-cost ratios (BCRs) in project appraisal and introducing a new public value framework.15 The IPA stated that being fixated on the BCR in the past has tended to lead to investment concentrating in London and the South of England at the expense of other areas, and that if one wishes 6 HM Treasury, Spending Review 2020, November 2020, para 2 7 HM Treasury, National Infrastructure Strategy, November 2020 8 Qq 8–9 9 Qq 12–13 10 National Infrastructure Commission submission, page 2 11 National Infrastructure Commission submission, page 1 12 National Infrastructure Commission, Rail Needs Assessment for the Midlands and the North, December 2020 13 Q 72; HM Treasury, National Infrastructure Strategy, November 2020, p 78–79 14 Q 72 15 Q 82; HM Treasury, National Infrastructure Strategy, November 2020, p 44 10 Lessons from major projects and programmes to achieve levelling up across the country then a broader lens is needed. It stated that the public value framework will be published by government in December, and that this will give clarity and metrics to Departments which they and the IPA can use to test if they are on track against their strategies and targets.16 However, were the published framework to be vague about the expected outcomes and measures of performance, then the IPA would share the Committee’s concerns about the potential for investment decisions to be made without sufficient demonstration of their value.17 Cost and schedule

In discussing the historic performance on managing major programmes and the challenges they have faced, the IPA stated that the UK government has some of the most novel and complex programmes in the world, that are probably more complex than many encountered in the private sector. Programmes in both sectors face challenges—the IPA stated that a number of studies have suggested that around 75% of major programmes in the private and public sectors, overspend or are delivered late.18 Nonetheless, the IPA stated that there were practices in the private sector that the public sector could learn from. This included there being more front-end development of programmes at the beginning, focusing on getting the business case and cost estimates to a level of accuracy that stakeholders can approve with more confidence. The IPA stated that government tends to keep options open for too long and should be more decisive about the outcome and intent that a programme is delivering and make choices on scope as early as practically possible.19

The NIC agreed, stating that when building infrastructure, government needs to spend sufficient time at the start deciding on the exact outcomes it wants and the ways those outcomes can be best achieved. This involves rigorous option appraisal and, critically, meaningful engagement with the public and stakeholders to understand key issues and concerns. Once the best option has been selected, government should commit to a project’s scope, essential if it wants to stick to a project’s cost forecast and delivery timeline.20

The IPA stated that it is working with government to improve the quality of cost estimation and is currently working on a standard for developing guidance on cost estimating. It has also set up a cross-departmental cost estimating community to share best practice.21 The NIC told us that the Government’s National Infrastructure Strategy promises to establish a cost benchmarking tool, and a separate requirement for all major infrastructure projects to publish cost and performance data at three key stages of their lifecycle.22

The IPA and NIC stated that programmes should provide a reasonable range of outcomes for likely costs and schedule that reflects the level of uncertainty in a programme.23 The Comptroller and Auditor General has also recommended the use of 16 Q 82 17 Q 85 18 Q 3 19 Q 5 20 National Infrastructure Commission submission, page 2–3 21 Q 22 22 National Infrastructure Commission submission, page 2 23 Q 21; National Infrastructure Commission submission, page 2 Lessons from major projects and programmes 11 ranges.24 Ranges will necessarily be large at the early strategic outline business case stage, and as a programme goes forward, should narrow as more becomes known. The NIC said that its Rail Needs Assessment uses its latest best practice on using ranges for cost estimates. The IPA stated that as a programme goes towards the outline business case stage it expects that ranges will narrow down to plus or minus 25 to 30% and would like to see cost estimates with an accuracy of plus or minus 10% by the final business case stage.25 Skills and Leadership

Skills and leadership remain a persistent problem in delivering major projects. Our report on specialist skills in the civil service concluded that skills shortages were responsible for delays, inefficiencies and increased costs in government projects, with the lack of skills in the Home Office partly responsible for the 3-year delay to the Emergency Services Network programme.26 The IPA also stated that government has not got the right level of skills and experience to deliver some of its difficult and complex projects.27 The civil service struggles to attract and retain specialise staff, and the turnover of programme Senior Responsible Owners (SROs) has been an issue which has impacted performance.28 The IPA stated that this has improved slightly in recent years, with tenure in role moving from an average three years to over four and a half years, and it is currently examining whether competency-based pay mechanisms could be introduced to reward good SROs and encourage them to stay within their roles.29

However, the IPA stated that many SROs manage more than one programme, and typically have four or five, which means they are “not being afforded the time in role to do the necessary work.”30 The IPA said that government needed to recruit 25 to 35 more senior responsible owners with the requisite expertise and deploy them to critical programmes. Getting the right skills at the right time can also be a challenge; for instance, skills relevant to a civil engineering stage of a programme may be less needed when the programme enters its systems integration phase.31

The IPA told us that, to remedy this issue, in addition to making attendance at the Major Projects Leadership Academy mandatory for SROs, the IPA aims to give them an assessment and an accreditation to ensure that they have the necessary skills for their roles. This is in addition to introducing the Government Projects Academy next year, which is aimed at improving project delivery skills for professionals across government.32 24 Comptroller and Auditor General, Lessons learned from major programmes, paragraph 20, National Audit Office, HC 960 Session 2019–21; Comptroller and Auditor General, Palace of Westminster Restoration and Renewal, paragraph 13, National Audit Office, HC 315, Session 2019–21; Comptroller and Auditor General, High Speed 2: A progress update, paragraph 29, National Audit Office, HC40, Session 2019–20. 25 Q 22 26 Committee of Public Accounts, Specialist Skills in the civil service, HC 686, December 2020, Conclusions and recommendations, para 1 27 Q 49 28 Committee of Public Accounts, Specialist Skills in the civil service, HC 686, December 2020, Conclusions and recommendations, para 2 29 Q 34 30 Qq 34, 49 31 Q 49 32 Qq 14 49 12 Lessons from major projects and programmes 2 Major projects assurance Transparency

This Committee has examined many government programmes where problems had arisen and were known by those within the programme for some time, but the issues only became public knowledge much later. For example, the Committee was kept in the dark for over a year before being told about the difficulties on the High Speed Two programme, and the programme now reports half-yearly as a result of our work.33 Nonetheless it remains challenging to hold government to account for its delivery of programmes - the Comptroller and Auditor General’s report notes that the pressures organisations are under can make them defensive about a programme or allow a “good news culture” to develop.34

The IPA agreed that creating a culture of transparency and openness are the foundations of effective project delivery. It promotes these values within its principles of project success, namely “Prioritise people and behaviour” and “Tell it like it is”.35 It used Crossrail as an example, where people at the working level understood the lack of progress with systems integration, but this information did not get passed upwards in the organisation. Creating trust within organisations requires leadership; leaders must be willing to hear bad news in order to instil the right behaviours in their teams.36

Projects need good data to appropriately track programme performance, but the underlying data which communicates progress can be poor. The Comptroller and Auditor General’s report refers to multiple estimates and numbers being in use within Departments and insufficient or out of date evidence supporting reporting.37 As we noted recently in our report on EU Exit, Departmental processes for publishing information can also be delayed, or inconsistent with other information presented by government.38 There can also be trade-offs to manage between commercial confidentiality and public transparency in reporting cost and schedule performance.39

The IPA stated that releasing all data at the detailed level would cause issues within the supply chain. It argued that it does not release its own assurance reviews so that it can have effective and candid discussions with programme leaders, which would be undermined if the reviews were made public.40 Nonetheless, it agreed that overarching data on programmes should be available and pointed to the GMPP performance data that is released quarterly and annually alongside the IPA’s annual report. The NIC also told us that government plans to publish business cases, outturn costs and schedules and other long-term evaluations of projects on the GMPP from April 2021, which we welcome.41 33 Q 49 34 C&AG’s Report, para 48 35 Q 37 36 Q 55 37 C&AG’s Report, para 46 38 Committee of Public Accounts, Whitehall preparations for EU Exit, HC 682, 2 December 2020, Conclusions and recommendations, para 6 39 Q 39 40 Qq 39, 45 41 National Infrastructure Commission submission, page 2 Lessons from major projects and programmes 13 However, GMPP data is a snapshot at a point in time and we have reported in the past how the publication of this data months in arrears means there is a lag between the reported data and what is actually happening on a project.42 The Infrastructure and Projects Authority’s Assurance

The IPA provides a range of assurance and support services to major programmes, with one of its key activities being gateway reviews of programmes at key decision points, such as each approval stage of the Business Case. Major programmes typically join the GMPP as they approach the strategic outline business case stage and leave the GMPP when they have delivered their benefits and are considered complete.43 In the past, this Committee has raised concerns that the IPA is not sufficiently independent of government to confidently challenge it. The IPA arose from the merger of the Major Projects Authority, which provided independent assurance, and Infrastructure UK, which promoted the development of infrastructure, and we stated that the conflict between these roles risked compromising its effectiveness. At the time, the IPA argued that the merger of the two bodies would have benefits such as bringing new expertise and increasing the level of support and intervention it could provide to projects.44

The IPA considers that while its gateway review process is good, government has not always applied it with the rigour needed.45 This has meant that programmes with significant issues have proceeded through Gateways even when they were given a delivery confidence assessment rating of amber or red. The IPA is now resetting its assurance of the GMPP, to ensure that departments address issues identified at gateway reviews before a programme can proceed. Assurance will now be risk-based assessments, rather than assessments being a judgement-based opinion based on reviewers’ experience. The intention is to ensure that projects not ready to proceed to the next gateway do not receive further funding until they taken action to resolve the issues they face, and for the assurance process to ensure that these actions have been taken.46 The IPA stated that government has reinforced its mandate, alongside a clarification of what the responsibilities of Departments are in acting on the IPA’s recommendations.47

The IPA pointed to the Lower Thames Crossing as a recent example of where it has made meaningful impact on a programme’s delivery. It stated that the programme had an ambitious schedule and was larger than any other scheme that Highways England had taken on before. The IPA refused to give support to the outline business case, advising that its delivery schedule was unachievable and that it needed to revisit its cost, as well as strengthen its governance model. The IPA said that Highways England revisited its plans for the programme and subsequently addressed all the key elements that the IPA had highlighted.48 42 Committee of Public Accounts, Delivering major projects, Thirty-first Report of Session 2015–16, HC 710, March 2016, Conclusions and recommendations, para 2 43 Q 4 44 Committee of Public Accounts, Delivering major projects, Thirty-first Report of Session 2015–16, HC 710, March 2016, Conclusions and recommendations, para 2 45 Qq 5, 10 46 Q 10 47 Letter from Infrastructure and Projects Authority to Committee dated 6 January 2021 48 Q 16 14 Lessons from major projects and programmes

We noted that the number of projects rated as red has increased from four to eleven in the IPA’s most recent annual report and questioned the IPA on how we should measure its success going forward.49 The IPA stated that one would expect to see more issues being raised in a more robust assurance regime, and if a programme was badly developed in the past it is not necessarily possible for the assurance regime to bring it fully back on track. The IPA stated that its performance should be measured more systemically and that there will be a number of elements needed to improve overall system delivery.50

The IPA told us that it is looking to introduce metrics to help it understand whether it is making a material difference to projects passing through stage gates, and what impact it is having on programmes that are underway.51 The IPA also told us it has in place a “get to green” programme to be clear about the fundamental elements needed for teams to give programmes a green rating at the next stage gate.52 The IPA said it wished to make sure that it is impactful on new programmes joining the GMPP, whilst recognising that improvement will be a slower process for those already on it. We noted, however, that when we last spoke to the IPA in 2016, it told us that a system was in development then to measure the impact it was having on programme performance.53 Restoration and Renewal of the Palace of Westminster

Since the IPA’s remit covers the government major programmes portfolio, it does not extend to the restoration and renewal of the Houses of Parliament, or the associated Northern Estates programme.54 The Restoration and Renewal programme is overseen by a Sponsor body and a Delivery authority, both of which are ultimately accountable to Parliament, and the IPA were not given a role in The Parliamentary Buildings (Restoration and Renewal) Act 2019 (the Act) which set out how the Programme would be managed. The programme is highly complex with a number of uncertainties that underpin it, including the condition of the building, its interdependencies with other Parliamentary building projects, and the lack of consensus over the scope of the programme, and how to relocate staff as the programme is underway.55

Whilst the IPA has provided support and advice in an advisory capacity to the programme, it does not currently fall within the IPA’s formal remit. The IPA was engaged in implementing the recommendations of the Joint Committee report on the Palace of Westminster of September 2016, and did support the Cabinet Office in preparing the Act, which set up the Sponsor Board and Delivery Authority, after which the IPA ceased to have formal involvement with the project.56 The IPA has provided further advice when invited to, such as in March 2020, when asked by the Cabinet Office to give advice concerning the status of the project. At this time, the IPA they confirmed that the rationale for the programme is clear, and that since the programme did not have an agreed scope, budget or programme at the strategic level and is therefore subject to considerable challenges 49 Q 65 50 Qq 65, 66 51 Q 66 52 Q 65 53 Committee of Public Accounts, Oral Evidence: Delivering Major Projects in government, HC 710, January 2016, Qq 83–89 54 Q 27 55 Committee of Public Accounts, Restoration and Renewal of the Palace of Westminster, Nineteenth report of Session 2019–21, HC 549, 2 October 2020, Summary 56 Q 17; Letter from Infrastructure and Projects Authority dated 14 December 2020, page 1 Lessons from major projects and programmes 15 to prepare a deliverable Outline Business Case.57 The IPA told us that it approved of the recent pause to review the programme, for the programme to refine the boundaries of how it will be delivered and what the optimum minimum scope will be.58

When asked, the IPA stated that it could not see a reason why it could not provide further assurance over the programme, if asked to by Parliament.59 When we reported on the programme in October 2020, we welcomed the decision of the Sponsor body to engage with the IPA and were keen that it continued to proactively seek the IPA’s advice.60 57 Letter from Infrastructure and Projects Authority dated 14 December 2020, page 1 58 Q 19 59 Qq 54, 55 60 Committee of Public Accounts, Restoration and Renewal of the Palace of Westminster, Nineteenth report of Session 2019–21, HC 549, 2 October 2020, Conclusions and recommendations para 4 16 Lessons from major projects and programmes