Resetting government programmes

Resets could have been avoided with more realistic upfront planning and scoping, including to better reflect the backdrop against which government programmes operate. This is a critically important part of the inception of any project. We heard that there has been a tendency across Government to re-baseline major projects1, particularly in the Ministry of Defence. The baseline should be agreed at Full Business Case and any scope change should be negotiated with the Treasury. We continually see poor planning as a deep-rooted problem across programmes, which can lead to resets. We would hope to see fewer programmes needing to be reset in the future as the IPA strengthens its assurance regime and its focus on setting up programmes for success. Although some resets are avoidable, for example through investing in up-front planning, other resets will be harder to avoid given the long delivery horizon and complex nature of major programmes. Resets may be a natural consequence for long-term programmes because of macro changes in the economy, technology sector or across society. Political factors, such as changes in ministers or the government composition, may also impact programmes. We heard how Universal Credit had been overseen by around nine secretaries of state since the current Senior Responsible Owner started his role in 2014. 1 Q140 6 Resetting government programmes Recommendation 2: a) Alongside the Treasury Minute response to this report, the IPA should share with us how it is embedding its good practice guidance on upfront planning and scoping, and the changes (with timeframes) it expects to see as a result. This should include providing the Committee with confirmation on how it is using its assurance regime to ensure that programmes do not pass major milestones without having followed planning good practice. The Senior Responsible Owner should produce a reconciliation statement comparing the milestones to what has actually happened on the project. b) By June 2024

Not having the right environment to encourage diverse views, transparency and constructive challenge has created problems in identifying and managing resets. A transparent and honest culture helps create an environment which supports constructive challenge and open conversations. This will encourage risks and issues, such as the potential need for a reset, to be discussed early with a full consideration of options, including if a programme should be stopped. We have seen programmes, such as Crossrail, when there has not been sufficient candour, for example on the civil contact, and as a result resets being undertaken later than needed. Having a diverse range of individuals, such as non-executive directors, to provide views contributes to creating an effective environment and leads to better decision- making. There is more progress to be made in improving diversity amongst those in central government as well as across the programme profession. Recommendation 3: The Infrastructure and Projects Authority should encourage and support departments in developing an environment of openness and transparency across programmes. Alongside the Treasury Minute response to this report, it should write to the Committee setting out how it plans to do this and monitor skills across all aspects of the profession against targets. This should include its pool of assessors, non-executives and practitioners.

On too many occasions, programmes have suffered from resets being done too quickly. The time needed to undertake a successful reset will vary case-by-case, influenced by factors such as the programme complexity and breadth of the reset. Resets of government programmes have on occasion taken less than three months and sometimes more than a year. Teams can underestimate how much time a reset can take. There can be pressures to complete resets quickly, with teams not being given the space and time to think about the next steps; if a reset is being undertaken it must be done properly, otherwise it simply increases the risk of further resets. Recommendation 4: By June 2024, HM Treasury and the Infrastructure and Projects Authority should have drawn together and shared with the profession its insights on the factors influencing the time needed to undertake a reset and Resetting government programmes 7 encourage programme teams to be realistic on the time they need. In doing this they should review their own processes to ensure they do not introduce milestones that incentivise rushed resets.

Broader programme-related good practice, such as having the skills, leadership and governance relevant to the programme stage, has not always been applied. Government departments do not always have the specific skills needed to undertake a reset. This includes not always having the right leadership with an appropriate SRO, the importance of which we have regularly highlighted, to oversee the reset or future programme. Although we are told there has been less churn of SROs than there used to be, civil service pay and promotion constraints can impact an SROs incentive to stay in post and provide continuity if needed. The IPA recognises governance as an area where it needs to sharpen its focus. Recommendation 5: The Infrastructure and Projects Authority should set out its progress, and the actions it has and is taking, to: • ensure programme SROs have the required skills and stay in post the expected length of time. More widely, Cabinet Office and HM Treasury should work with departments to ensure they use any available levers where it is best to incentivise continuity of leadership. • accredit programme professionals across individual departments.

Resetting programmes can create new risks that are not always effectively managed. A programme reset is much more than the routine change management expected within any large programme. To help address the root causes of any underlying issues, a reset will necessarily significantly change what or how a programme delivers. This will introduce new risks and opportunities that need to be identified and managed such as creating new commercial relationships or changing existing ones if necessary by contractual changes. In not recognising these challenges, on its Ajax programme, the Ministry of Defence had to undertake a further reset. It is vital that departments fully appreciate the new risk landscape. Recommendation 6: The Infrastructure and Projects Authority should encourage and support departments to understand the full risks when resetting a programme. Alongside the Treasury Minute response to this report, the IPA should write to us setting out how it is considering this as part of its ongoing assurance and review of major programmes. 8 Resetting government programmes 1 Recognising a reset

On the basis of a report by the Comptroller and Auditor General, we took evidence from HM Treasury and the Infrastructure and Projects Authority (IPA) about resetting major programmes.2 We also took evidence from four senior responsible owners (SROs): Matthew Lodge, Government SRO for Crossrail, Department for Transport; Neil Couling, SRO for Universal Credit, Department for Work & Pensions; Dr David Marsh, SRO for the Ajax programme, Ministry of Defence; and Jim Barton, SRO for the Electronic Monitoring programme, Ministry of Justice.

As government does not have a standard definition for a reset, government bodies may not recognise when a reset is required, being carried out, or has been completed.15 There was some confusion over what was termed a reset. The SRO for Universal Credit suggested he had probably reset the programme several times but not used the term. Also, there were questions raised over the number of resets undertaken for the Ajax programme and whether the 2014 decision to move production to the UK should be counted alongside two later resets which firstly recast the commercial agreement and then sought to address quality and delivery schedule challenges.16 Not clearly defining a reset, will make it harder for government to identify lessons that can be carried over into the same or other programmes and situations.17

SROs described how it would be helpful if resets were seen more positively, which HM Treasury suggested government was moving towards.18 They also commented that resets can be seen as a failure and can be a traumatic event for programme staff, after which the confidence of the team would need to be rebuilt.19 Negative perceptions could mean government bodies continue trying to resolve unresolvable issues and lead to wasted effort and costs.20 The SRO for electronic monitoring commented that “sometimes, the easier path is to simply carry on doing the same thing, but it is negligent if, as an SRO, you do not call it.”21 He spoke about how the IPA confirming a ‘red’ delivery confidence status helped validate their own assessments and made it easier to have conversations about a more fundamental reset. It took the Ministry of Justice nine months from starting to think 7 C&AG’s Report, para 8 8 Qq 2, 4, 8 9 C&AG’s Report, para 26 10 Qq 97, introductory remarks before q102, 134 11 Qq introductory remarks before q102, 157, 169 12 Q 135 13 Qq 142, 169 14 Qq 17, 32 15 C&AG’s Report, para 8 16 Qq 4, 20–21, 66–67, 88 17 C&AG’s Report, para 8 18 Qq 97, introductory remarks before q102, 102 19 Qq 4–5, 18–19 20 C&AG’s Report, para 9 21 Q97 10 Resetting government programmes about a reset to confirming the decision.22 The IPA recognised how its response to red programme could provide a trigger if the Department or the project team did not realise the need for a reset.23 It told us it was now in a better place to discuss with departments when things can no-longer be achieved and what should be done about it.24

Although those deciding on a reset, such as the SRO or programme board, need to be aware of what may trigger a reset we consistently heard how the IPA and central government bodies, could help identify or build the case for significant change.25 The IPA told us how, quite often, although programmes would ideally identify the need to change, those involved would not necessarily see the issues without independent analysis and data.26 We have previously recommended, in our report on improving the accounting officer assessment process, how HMT and IPA should use their information to understand programmes specific risks and wider issues.27 Addressing the factors behind a reset

Various factors, both internal and external to a programme, can contribute to the need for a reset.28 We heard how this can include the programme going off track, the need to improve confidence in supplier delivery, or to resolve technical problems and set a realistic schedule.29 Factors can result from deep-rooted issues caused by poor programme planning, the importance of which we have been stressing for some time.30 The IPA told us it is trying to encourage practitioners to set up projects well given they are much harder to fix in-flight. It described how it has a several tools to help with project set-ups, and therefore reduce resets, and that it was working to equip SROs and project directors to have the right skills and knowledge at the start of a programme.31 The SRO for Ajax commented favourably on the IPAs range of products.32

The IPA suggested that its assurance work has become much more rigorous – it has republished the assurance workbooks it uses to undertake reviews; refreshed its pool of available independent reviewers; and started to provide them with training.33 For the 100 programmes it has the capacity to support, the IPA told us it gets more deeply involved as early as it can to help avoid issues that can lead to a programme failing later. As part of this it will look at, for example, whether there is a robust schedule, a stakeholder alignment plan, and a robust cost estimate.34

Although some resets could be avoided, the IPA suggested there may be occasions where multiple resets are necessary, particularly for long-term programmes.35 External factors, including financial pressures, political changes, technology developments or events 22 Q 11 23 Qq 115, 128–129 24 Q 117 25 Qq 11, 17, 97; C&AG’s Report, para 10 26 Qq 115, 105 27 Committee of Public Accounts, Improving the Accounting Officer Assessment process, Twenty-Eighth Report of Session 2022–23, HC 43, 30 November 2022, recommendation 4 28 Q 89; C&AG’s Report, paras 12–13 29 Qq 2–3 30 Q 138; C&AG’s Report, para 12 31 Qq introductory remarks before q102, 135, 167 32 Q 68 33 Q 118 34 Q 167 35 Q 134 Resetting government programmes 11 such as the COVID-19 pandemic, can lead to a reset.36 At present, those programmes in construction are being particularly impacted by inflation.37 The SRO for Universal Credit felt resets were a natural result of major long-term projects. He highlighted how the United Kingdom leaving the European Union, devolution of benefits to Scotland, the COVID-19 pandemic and ongoing cost-of-living challenges had all impacted the programme.38 He also mentioned how, although there had been one Secretary of State for six years, he is now on his ninth Secretary of State since he started as SRO in 2014.39 He felt it was impossible to ‘freeze politics’, which became more of a challenge for projects extending beyond a parliamentary term.40 Creating the right environment

Without the right environment, individuals may be afraid to report problems in a timely and honest way, which makes it harder to identify the need for a reset and the issues a reset should address.41 We heard from the SRO for Universal Credit that creating this openness was the hardest, but most important, thing to achieve in projects.42 Also, the SRO for Crossrail spoke about how the programme started to believe its own ‘on- time, on-budget’ mantra which stopped many people raising issues.43 He also felt that, in hindsight, the governance structures did not provide the Department for Transport, as sponsor, enough transparency over potential issues.44 The SRO for electronic monitoring described a positive experience with the relevant Minister who had a firm belief in the role of electronic monitoring. He created a space and appetite for candid conversations about the programme challenges and how they could be resolved.45 From a Ministry of Defence perspective, the SRO for Ajax mentioned the additional cultural challenge of being an integrated military and civilian organisation and acknowledged that, for Ajax, it was very difficult for individuals to continually have the confidence to speak up. He explained how the MoD was working hard on this and recently commissioned a report on psychological safety in major programmes.46

We asked SROs about the diversity of their programme boards, which could help establish an effective, open and constructive environment. For Universal Credit, the SRO explained how, over time, the Board had included women and men in almost equal number fulfilling different roles, alongside representatives from HM Treasury, the IPA and a local authority. For electronic monitoring, the SRO described the Programme Board as representative of HM Prison & Probation Service but recognised that this did not equate to being representative of society at large.47 We also asked the IPA and HM Treasury about diversity within their respective organisations. IPA said it has quite a diverse team, both in terms of cognitive diversity and background. HM Treasury agreed that it is similarly diverse in terms of background although officials tended to have a more policy experience. 36 C&AG’s Report, para 13 37 Qq 111–114 38 Q 69 39 Q 99; C&AG’s Report, Universal Credit: progress update, Session 2014–15, HC 786, 26 November 2014 40 Qq 73–75, 79, 90 41 C&AG’s Report, para 16 42 Q 17 43 Qq 16 44 Qq 9, 10, 91, 93 45 Qq 76–77 46 Qq 16, 19 47 Qq 70–72 12 Resetting government programmes HM Treasury set out that across its public spending work, around 50%, or maybe even more, were female. We commented, however, that there still seemed to be some way to go on ethnic diversity.48

We were interested to hear about the role of non-executive directors in creating an environment supportive of challenge and diverse thinking. The SRO for Universal Credit explained how it was useful having a non-executive director chairing the Programme Board.49 The SRO for Ajax spoke about how he has reset the Board to expand the number of voices around the table, trying to bring-in more diversity and non-advocate challenge. For electronic monitoring the SRO told us that, when they were discussing a potential reset, they had strengthened the independent voice with a non-executive challenge panel outside formal governance arrangements. This provided a space for ‘incredibly robust challenge’ of the programme.50 The IPA felt it could do more to explore whether departments bringing in appropriate external director skills.51

It is important to have open and honest conversations on how to progress a programme, including whether a reset is the right solution.52 Where a government body identifies a programme is at risk of not achieving its intended outcomes, it should consider a range of options to address this. This should include whether the programme should be stopped, with costs potentially written off.53 HM Treasury emphasised that it is really important programme teams consider stopping a programme, and starting again, rather than just continuing on the previous course.54 The SRO for electronic monitoring described how, following an IPA review in March 2021, they suspended developing the system which gave them three months’ breathing space to work through whether it was right to carry on with the work or to stop. HM Prison & Probation Service decided to terminate an aspect of its contract with Capita and stopped development of the Gemini case management system.55 On the most recent reset for the Ajax programme we were told that the Ministry of Defence looked at various options, including stopping the programme.56 48 Qq 121–124 49 Qq 12–13 50 Q 15 51 Q 168 52 Qq 104–105, 107 53 C&AG’s Report, para 11 54 Q 116 55 Q 3, C&AG’s Report, Electronic monitoring: a progress update, Session 2022–23, HC 62, 8 June 2022 56 Qq 31–33, 36–37 Resetting government programmes 13 2 Undertaking a reset Allowing time for a reset

The time needed to undertake a reset will vary. Across 24 resets, the NAO found that the time spent ranged from less than three months to more than a year. Various factors contribute to the time needed for a reset such as the complexity of the programme and reset; the need to understand the underlying issues and potential opportunities; involving (and potentially rebuilding) a supply chain; engaging stakeholders, including contractors, so they remain supportive; and seeking programme approvals.57 The SRO for Crossrail described how time may be needed to work through commercial issues. On Crossrail the project team spent time re-establishing contracts and creating the right incentives to get the railway open.58 Political involvement can also mean extra time is needed for a reset. The Ajax SRO told us that over the last 18 months, the programme had benefited from very strong hands-on ministerial involvement, with his biggest lesson being that this created decision-making processes that should be planned for.59 The SROs identified how changing ministers could create additional work for programme teams.60

The SROs highlighted to us the need to take appropriate time to ‘do the job properly’. For the Ajax programme, the NAO found the Ministry of Defence had underestimated how long a reset would take. It expected to conclude negotiations to reset the contract in October 2017, but these took 15 months instead of two, with the legally binding agreement signed in December 2018. We heard from SROs how there is an understandable desire, and sometimes political pressure, to finish a reset as soon as possible.61

The C&AG’s report identified value in creating ‘space’ to reset a programme. To do this teams might generate or capitalise on opportunities to pause a programme (rather than resetting ‘in flight’); establish a dedicated reset team to focus on resetting; or not publicise details of a reset until it has been fully developed given how this could affect some stakeholders’ confidence in the programme.62 The SRO for Crossrail reflected that one of the challenges faced in undertaking a reset was the reset team retaining responsibility for the project at the same time.63 During a later reset on Crossrail, the COVID-19 pandemic forced a pause to work which allowed the team time to replan.64 Building the skills, leadership and governance

Governance arrangements need to evolve throughout a programme lifecycle to reflect the risks and requirements at each stage. This includes both during and after a reset, so governance and approvals processes remain proportionate to the programme risks with skills, expertise and leadership brought in as required.65 The SRO for Crossrail told us how, when they reset the project in 2018, they worked to bring sponsors, such as DfT and 57 C&AG’s Report, para 21–22 58 Q 85 59 Qq 77–78 60 Qq 73–79 61 Qq 75, 95–96, 135; C&AG’s Report, The Ajax programme, Session 2021–22, HC 1142, 11 March 2022 62 C&AG’s Report, para 24 63 Q 2 64 Q 96 65 Qq 95, 120; C&AG’s Report, para 25 14 Resetting government programmes TfL, closer to the Crossrail board to create better conversations and share information. He reflected on whether they got the governance structure right, with the Department for Transport currently undertaking an exercise to identify lessons for project governance which they hope to publish in the summer.66 The IPA told us that it needs to sharpen its focus in the governance area.67

Given a reset may involve different activities, or a different focus, to how the programme was being managed it may require specific skills and expertise such as around planning, negotiation, political engagement, or stakeholder management.68 As such, the skills and leadership for long-term major programmes, may need to change to reflect the stage of the programme.69 The SRO for Crossrail reflected on the importance of bringing in the right people with the right skills, even when things may look to be progressing well. Very different skills were needed at the end of the programme to the start.70 The SRO for Universal Credit spoke about considering carefully the value of a dedicated reset team, giving leadership back to the programme as soon as possible, and whether the current SRO should do the reset.71 A dedicated reset team could bring in the necessary skills and help step back from the programme.72 On Universal Credit, the head of the then Major Projects Authority (now incorporated into the Infrastructure and Projects Authority) headed up the first reset.73 The IPA spoke about more recently being deeply involved in providing expertise and short-term resources to some programmes, including for the Ajax programme.74

As with a programme, those leading a reset, such as the SRO, need to have a clear vision shared with stakeholders alongside the skills and experience to critically examine a programme, motivate staff and contractors, and deal with external pressure.75 This was particularly important for resets as the SRO would need to help maintain focus on the decisions needed despite the likely heavy demands for information from those overseeing the programme.76 While SROs felt that continuity of SROs was a good thing, they stressed that it was important to have the right leader for the right moment.77 Although recognising a reset may lead to a change in SRO, we have previously commented on SROs moving off programmes too quickly.78 We were informed that the situation is improving, although heard about the challenges in encouraging SROs to stay in their role given a lack of pay incentives and the inflexibility of promotions meaning they potentially give up career opportunities.79 In terms of incentivising a continuity in SROs, the IPA told us it can now administer dedicated allowances in certain circumstances and, with minister 66 Qq 10, 68, 91 67 Q120 68 C&AG’s Report, para 28 69 Qq 14, 91–92 70 Q 70 71 Qq 96–97, 135 72 C&AG’s Report, para 28 73 Q 2 74 Qq 138–139 75 C&AG’s Report, para 29 76 Q 68 77 Qq 98–99 78 Qq 96, 98; Committee of Public Accounts, Lessons from major projects and programmes, Thirty-Ninth Report of Session 2019–21, HC 694, 29 January 2021 79 Qq 98–101 Resetting government programmes 15 support can agree specific milestones for SROs to stay in position.80 It also felt that an SRO, or a programme director plays a critical role in building the right team around them, particularly in terms of having the digital skills required.81 Impact of resets on programme risks

As a reset is a fundamental or substantial change to what a programme will achieve, or how it is delivered, it cannot be managed through routine change control procedures.82 We heard from the SRO for Ajax that when undertaking resets, it was important to understand the root causes of issues rather than just the symptoms. The IPA and HM Treasury reiterated the importance of this in helping to get resets right and avoid further unnecessary resets.83 Understanding the causes will also make it easier for programmes to set out the reset aims and focus efforts, as well as establish clear measures to assess if the reset is realistic and monitor its success.84

In focusing on the underlying issues, and as a reset significantly changes what or how a programme delivers, it introduces new risks and opportunities that need to be identified and managed.85 The SRO for Ajax told us how the 2018 programme reset aimed to protect the delivery date, but this introduced new risks impacting the scheduling and has meant the programme has had to be reset again.86 He described a core lesson being the need to fundamentally reassess the risk profile and landscape when making significant programme changes, which had not been fully done on Ajax.87 The SRO for Universal Credit also spoke about managing risks, and trade-offs with the timetable.88

The risk profile of a programme may shift with changes to the commercial approach or how this is managed during a reset. For a reset, the government body will need to consider what existing arrangements with delivery partners mean for how to undertake a reset and what it could achieve.89 Programme teams need to work with suppliers and delivery partners to establish an effective relationship for the reset and beyond.90 The SRO for electronic monitoring mentioned the role of the Government Commercial Function. It was very helpful to them as they were working through how to reset their commercial relationship with a provider who was a strategic government supplier.91 We were told that their relationship with the supplier has improved post settlement.92 The SRO for Universal Credit raised the importance of time to have discussions with suppliers to create a win- win situation.93 80 Qq 161–162 81 Qq 136–137 82 C&AG’s Report, para 2 83 Qq 97, 104, 134 84 C&AG’s Report, para 14 85 C&AG’s Report, para 27 86 Qq 21–22 87 Q 65 88 Q 68 89 Q 150; C&AG’s Report, para 18–19 90 C&AG’s Report, para 18 91 Q 17 92 Q 83 93 Q 84 16 Resetting government programmes