Decommissioning Sellafield

The targets set for Sellafield Ltd neither support DESNZ and the NDA in holding it to account for poor performance, nor allow Sellafield Ltd to demonstrate it is making meaningful progress towards addressing the site’s hazards. In recent years, Sellafield Ltd has missed its most important hazard reduction targets more often than it has achieved them, often by considerable margins. While carrying out work safely must come first, safety cannot be used as a ‘get out of jail free card’ to excuse poor performance. We do not believe that the current target–setting process allows DESNZ and the NDA to distinguish between legitimate reasons for missing a target and underperformance on the part of Sellafield Ltd. Additionally, these annual targets do not allow Sellafield Ltd to show it 3 will be able to pick up the pace in future years in order to achieve the ‘key decommissioning milestones’ it has agreed with its regulators (some of which are decades away). These milestones mark a step–change reduction in the hazard buildings pose. To achieve them, Sellafield Ltd needs to fundamentally transform how the site functions – for example, in the case of MSSS, it wants to be removing 24 times as much waste as it did in 2023–24 within a decade. recommendation a. Sellafield Ltd should publish a set of medium–term targets which demonstrate it is on track to complete key activities within its assumed ‘strategic tolerances’. This should contain sufficient detail to give Parliament, and the public, confidence it has a credible plan to develop and sustain the necessary capabilities. b. It should then report annually on its progress against these medium–term targets. c. Alongside these medium–term targets, it should set out what actions will be needed, including clear milestones, to achieve a completely decommissioned site in the next hundred years. It should report progress against this every ten years.

Sellafield’s ability to deliver its largest projects on time and budget is still too variable. While there are some recent signs of improvement, Sellafield continues to repeat old mistakes. Four of the largest projects at Sellafield will collectively cost £1.15 billion more, and be delivered many years later than Sellafield Ltd expected in 2018, the last time this Committee examined Sellafield. Sellafield has a poor track record of delays and cost overruns, but the NDA reassured us in 2018 that cost forecasting is improving while the Department believed that the estimates of those costs have begun to stabilise. Nevertheless, there are some signs that a corner may finally have been turned. Sellafield Ltd is now working with a group of contractors in a different way, which places more emphasis on planning. As a result, most of the projects that have started more recently are being delivered in line with their business cases. However, one of Sellafield’s more recent major projects, the Replacement Analytical Project (RAP) has been managed very poorly indeed. When Sellafield Ltd ‘paused’ RAP last year, it was expected to cost up to £1 billion more than originally approved while new plutonium analysis capabilities would be at least 5 years late. Sellafield Ltd now believes that it can cut this gap to around one year and save money by converting a different building instead. However, £127 million spent on RAP will have been wasted. 4 recommendation As it carries out more detailed planning for future provision of Analytical Services, Sellafield Ltd should write to the Committee at a suitable point (but within six months) to update us on: • when it expects to be able to carry out the plutonium analysis the site will require (and what level of confidence it has in this estimate); • how much it expects to spend on the projects it has decided to carry out instead of the Replacement Analytical Project; and • what assessment it has made of the costs and value for money of mainta

The failure of the Replacement Analytical Project illustrates the need to improve asset management at Sellafield. Sellafield is frequently in a race against time to complete work before assets (such as buildings or equipment) reach the end of their life or risks materialise (particularly as changes to planning assumptions can affect many other facilities on the site). Poor asset management caused Sellafield Ltd’s long–term plan for analysing waste to become incoherent: it did not understand what physical state its laboratories were in, or do the right remedial work to address their deterioration. Although Sellafield has resolved some of the maintenance issues which emerged during the COVID–19 pandemic, there is still a growing backlog of maintenance tasks that need to be performed. There are worrying implications for safety: malfunctioning equipment has limited Sellafield Ltd’s ability to retrieve waste from its oldest facilities, while its safety experts have warned that the deteriorating condition of assets is making the site increasingly unsafe. recommendation Alongside the Treasury Minute response to this report, Sellafield Ltd should write to the Committee to explain how it is addressing the deteriorating condition of its assets. This should include information about the specific actions it is taking to increase the productivity of its maintenance teams, and what impact these actions have had, and are expected to have.

DESNZ is not doing enough to ensure that long–term value for money is achieved from decommissioning the Sellafield site. The cost of decommissioning Sellafield continues to grow – notwithstanding the billions spent on the site every year. The NDA’s tight funding setttlement for 2025–26 will mean more work slips into future years, increasing the long–term cost to the taxpayer. We were disappointed the Department sought to place the 5 onus for achieving value for money entirely on the NDA and Sellafield Ltd: as the NDA’s sponsor, it has an important role to play as well. We are still not convinced that UK Government Investments plays any useful role in the NDA’s governance that the Department could not fulfil itself. We would also have expected the Department to take a more active interest in the relative cost of different options for cleaning up NDA sites so it fully understands the cost implications of its current policy of full site remediation. Meanwhile, the date when the NDA will be able to store waste permanently in a Geological Disposal Facility (GDF) has slipped from 2040 to the late 2050s, even though the programme is still only at an early stage. For every decade of delay, Sellafield could need to construct another storage building (each costing £500 million–£760 million) and would incur higher operating costs. recommendation The Nuclear Decommissioning Authority and Sellafield Ltd should use their existing long–term financial models to produce and publish a range of scenarios, giving more insight into how government can reduce the lifetime decommissioning cost of the NDA estate. This should include an assessment of the cost (on both a discounted and undiscounted basis) if government funding for the NDA is constrained at current levels in all future years.

The NDA and Sellafield Ltd still have more to do to build a culture where all employees feel able to raise concerns and report poor behaviour. The exceptionally hazardous nature of many of Sellafield’s activities means that it is imperative that all employees and contractors on the site feel able to raise any concerns that they have without fear of consequences. While the witnesses told us that they felt the site’s reporting culture was good, we remain to be convinced, since there are other indications of a sub–optimal culture. This Committee and its predecessors have received a number of letters from whistleblowers in recent years making concerning allegations, and we are aware that the NDA paid £377,200 in 2023–24 to settle employment–related claims. It is good that the Department, NDA and Sellafield Ltd recognise the need to do more to improve the site culture, and we acknowledge the improvement in staff survey results over recent years. However, it is concerning that the Department and the NDA did not identify the need to take action earlier–particularly in light of Sellafield Ltd’s chief executive’s observation that there was a disconnect between Sellafield’s workforce and leadership when he took over in July 2023. 6 recommendation The NDA should publish information in its Annual Report about: • the prevalence and perception of bullying and harassment broken down by site and demographic groups, accompanied by suitable benchmarking; • other staff survey results which it believes are particularly important in assessing the health of its corporate culture; and • metrics to measure the effectiveness of its various reporting mechanisms including safety reporting. 7 1 Addressing the most significant hazards Introduction

On the basis of a report by the Comptroller and Auditor General, we took evidence from the Department for Energy Security and Net Zero (the Department), the Nuclear Decommissioning Authority (the NDA) and Sellafield Ltd on the decommissioning of the Sellafield site.1

We heard from the NDA that the Magnox Swarf Storage Silo is “the most hazardous building in the UK”. It has been leaking radioactive water into the ground since 2018–we calculated, at current rates, enough to fill an Olympic–size swimming pool approximately every three years. The NDA accepted that this leak is its “single biggest environmental issue”. However, the witnesses told us that the radioactive particles are “contained” in the soil and do not pose a risk to the public.9 They also told us that they–and the site’s regulators–believe that the best way of 4 C&AG’s report, paras 13, 3.17, Figure 8 5 C&AG’s report, paras 1, 16, 3.7 6 C&AG’s report, Figure 7 7 Committee of Public Accounts, Nuclear Decommissioning Authority: risk reduction at Sellafield, Sixty–Fifth Report of session 2017 –2019, HC 1375, 31 October 2018 8 C&AG’s report, Figure 8 9 Qq 52–55 9 addressing the leak is to focus on emptying the silo.10 Sellafield Ltd expects that the silo will largely be emptied at some point between 2054 and 2059, which is 8–13 years later than it expected in 2018.11

Since 2020, Sellafield Ltd has missed most of its operational targets for retrieving waste, and turning liquid waste into a glass–like substance (both of which reduce the hazard the waste poses).12 It told us that the COVID–19 pandemic had had a big impact – particularly on the facility which turns highly active waste into glass. This facility stopped processing waste for a period of time and maintenance tasks were not carried out, leading to a number of issues with the equipment, some of which have still not been resolved. Some experienced operators decided to retire, and Sellafield Ltd found it difficult to train their replacements as it could not provide ‘on the job’ training in periods when the facility was not operating.13 Sellafield Ltd also told us that it had made less progress with retrieving waste from the ‘Magnox Swarf Storage Silo’ because of problems with getting newly–installed equipment to work effectively in an older building – including issues with how cranes were powered.14 We were, however, pleased to hear that it achieved its 2024–25 target for removing waste from its other silo, the ‘Pile Fuel Cladding Silo’.15

The Department, NDA and Sellafield Ltd have all taken actions to improve the management of the plutonium stored at Sellafield since the previous Committee expressed concerns about this. The previous Committee raised then the lack of a long–term strategy for the stockpile.16 The Department has now decided that a new facility should be built to put the plutonium beyond use – which clarifies what the NDA and Sellafield Ltd will need to do.17 Sellafield Ltd has taken action to resolve some of the most pressing issues, improving how some of the plutonium is packaged and stored. As a result, the Office for Nuclear Regulation has closed two of its highest–priority concerns.18 While the risks from the plutonium are still greater than the level which Sellafield Ltd is content to accept, it believes that it will be able to reduce them to an acceptable level by 2060.19 10 Qq 52, 54–55, 57 11 C&AG’s report, Figure 8; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 8 12 C&AG’s report, Figure 9 13 Q 41 14 Q 55 15 Qq 12–13 16 Committee of Public Accounts, Nuclear Decommissioning Authority: risk reduction at Sellafield, Sixty–Fifth Report of session 2017 –2019, HC 1375, 31 October 2018 17 Q 18 18 Q 14 19 C&AG’s report, para 2.12 10 Sellafield’s approach to target setting

Sellafield Ltd has–in conjunction with the NDA and the Office for Nuclear Regulation–set a ‘strategic tolerance’ for completing key activities. It described this to us as the date “we really want to hit”, to ensure that the site is fully cleaned up by 2125. We explored the consequences of not achieving ‘strategic tolerances’ with particular reference to the ‘Waste Vitrification Plant’ and were told that there is no fundamental engineering reason why the ‘strategic tolerance’ of 2039 could not be extended in this case. However doing so would need careful consideration as it could affect planning assumptions made elsewhere on the site (such as those relating to Sellafield’s future requirement for sample analysis capabilities).20 The NDA recognises that devising measures which allow measurement and reporting of progress in the short term, while also providing confidence that Sellafield Ltd will achieve long–term outcomes, is challenging.21 It also told us that Sellafield Ltd was right to be “super–cautious” because of the extremely hazardous nature of its waste – though it was conscious that “this may sound like a load of excuses”.22 We sought to understand how the NDA would identify underperformance – particularly in light of the Office for Nuclear Regulation’s request that the NDA do more to hold Sellafield Ltd to account for delivery, safety and security.23 We note that Sellafield Ltd has not only missed most of its targets for retrieving and ‘vitrifying’ waste in recent years, it has generally fallen well short of the level of performance it deems ‘Acceptable’.24

Sellafield Ltd has shown that it is technically possible to retrieve waste from all of its oldest ponds and silos.25 The plan to make the site less hazardous assumes that it will be able to significantly accelerate the speed it works at over the next few years – and then sustain that pace until tasks are completed. This was described to us by the NDA as “being able to crawl before we can walk, and then walk before we can run”. Sellafield Ltd told us that it expects the teams removing waste from the ‘Pile Fuel Cladding Silo’ to remove at least three times as much waste in 2025–26 as they did in 2024–25 (and ideally, five times as much).26 It also expects to remove significantly more waste from the ‘Magnox Swarf Storage Silo’ from the early 2030s (potentially retrieving 24 times as much as it did in 2023–24 by the middle of that decade). Sellafield Ltd told us that its focus over the 20 Qq 42–47, 69 21 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, pages 1–2 22 Q 12 23 Q 84 24 C&AG’s report, Figure 9 25 Q 20 26 Qq 14–15 11 next few years will be on developing the capability to do this (including installing additional machinery to retrieve waste and finishing building the ‘Box Encapsulation Plant’ elsewhere on the site).27

We explored with the witnesses what Sellafield Ltd needs to do to reach the point where retrieving waste becomes an efficient, routine activity. As well as installing the right equipment, they emphasised the importance of “constant iteration” – addressing problems such as the availability of cranes as they arise and the importance of developing a site culture focussed on performance. We were also told about the importance of setting “ambitious but achievable” annual targets which motivate the workforce.28 The NDA accepted that it could be more sophisticated in how it sets targets – potentially setting targets that capture improvements in capability rather than relying on relatively simple metrics such as the quantity of waste removed.29 We also expressed our concern that it is not clear enough what the underlying assumptions behind longer–term goals are, and how optimistic they are. The Department agreed to consider how these could be drawn out and linked to annual targets.30 The NDA subsequently wrote to us to provide additional information about how Sellafield Ltd uses modelling to produce a range of scenarios for retrieving waste, and the range of factors that would need to be optimised to achieve the best case outcomes. However there is currently no visibility of what these underlying assumptions are – or of how likely each scenario is to occur.31 27 Q 23; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 8 28 Qq 15, 48, 55, 89–90 29 Q 19 30 Qq 25, 27–28 31 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 8 12 2 Management of projects and programmes to build or maintain assets Delivering major projects and programmes at Sellafield

When it last took evidence on Sellafield, the previous Committee was told that major project delivery was improving.32 The NDA told it that “cost forecasting is improving” while the Department said that “for the first time now, the estimates of those costs have begun to stabilise”.33 Even then though, the nine major projects underway were, between them, delayed by 165 months and expected to cost £913 million more than originally budgeted.34 However costs, and delays, have risen considerably since then for the four projects which were two years or more away from completion. These four projects are now expected to cost £1.15 billion more between them than when the previous Committee reported, while each will be delayed by between 58 and 129 months. Sellafield Ltd accepted that £546 million of this cost increase was caused by poor planning and poor contractor performance. It attributed £191 million to delays resulting from the COVID–19 pandemic and higher–than–forecast inflation.35

DESNZ told us that it nevertheless believes that “there is some tangible evidence on improvement” and that “the overall direction of travel on project delivery is improving at the site”.36 Four of the five major projects which have been approved since 2018 are on track to be completed in line with the project team’s expectations when the project was originally approved.37 Additionally, the Infrastructure and Projects 32 Committee of Public Accounts, Nuclear Decommissioning Authority: risk reduction at Sellafield, Sixty–Fifth Report of session 2017 –2019, HC 1375, 31 October 2018 33 Committee of Public Accounts, Oral evidence: Nuclear Decommissioning Authority, HC 1375, 16 July 2018, Qq 20–21 34 Committee of Public Accounts, Nuclear Decommissioning Authority: risk reduction at Sellafield, Sixty–Fifth Report of session 2017 –2019, HC 1375, 31 October 2018 35 C&AG’s report, para 2.6 36 Q 74 37 C&AG’s report, Figure 6 13 Authority has given two of the projects a ‘Green’ rating two years running.38 This means that “successful delivery of the project on time, budget and quality appears highly likely”; only 11% of government major projects were in this position at March 2024.39

We asked the witnesses to explain what has led to the apparent improvement in how projects are now being delivered. Sellafield Ltd told us that it was now delivering its largest projects in a new way, which it calls ‘Programme and Project Partners’ (PPP). Under this model, it works closely with four suppliers to deliver a portfolio of projects over a 20–year period – giving them incentives to develop the workforce, avoiding unnecessary costs and improving co–ordination between projects. We were pleased to hear that Sellafield Ltd is optimistic that the PPP model will deliver even better outcomes in future – as the next group of projects will have been entirely delivered under these arrangements.40

More generally, the Department and the NDA told us that they believed that there was a much greater focus on ensuring projects were well–planned. ‘Reference class forecasting’ is being used to help ensure that project baselines are realistic, and there are now more frequent ‘check–ins’ with teams during the planning phase. When members of this Committee visited the site, we were shown a ‘4D’ digital model of one of the projects, which was created during the planning stage and is now used to identify potential issues and ensure the project remains on track. The Department and the NDA also believe that there are now more robust assurance arrangements than was the case previously, with reviews carried out by the Infrastructure and Projects Authority and independent ‘red teams’.41

We asked the NDA and Sellafield Ltd what they were doing to ensure that other projects benefitted from these improvements. Sellafield Ltd told us that it was looking to share learning with those projects which were being delivered under its pre–existing arrangements – such as those which were already well underway when the PPP approach started. It also told us that it was now looking to make sure that its medium–sized projects (those expected to cost over £20 million) were also benefitting.42 The NDA told us that it had established a group–wide community of project leaders which could share best practice between sites – and that staff transferring to other group companies were also helping.43 38 C&AG’s report, para 2.8 39 Infrastructure and Projects Authority, Annual Report on Major Projects 2023–24, 17 January 2025, pages 10 and 15 40 Q 63 41 Qq 63, 73–74 42 Q 63 43 Qq 64–65 14

A notable exception to this positive trend is the ‘Replacement Analytical Project’ (RAP) – which Sellafield Ltd now believes should be abandoned.44 This project would have refurbished a laboratory on the site so Sellafield Ltd would be able to continue analysing samples of waste – which is essential to operate the site safely. It would also deliver new capabilities that will be required by another facility, the ‘Sellafield Product and Residue Store Retreatment Plant’ (SRP). Sellafield Ltd will use SRP to repackage plutonium, so it can be stored more safely.45 When the Outline Business Case was approved in 2020, Sellafield Ltd believed that RAP was very likely to cost less than £800 million, with RAP ready to support SRP by October 2027 (it expected to achieve these outcomes in 80% of the scenarios it modelled).46 By late 2023, the estimated cost had risen to £1,500–£1,800 million, while RAP would be unable to support SRP until November 2033 at the earliest (delaying its opening by at least five years).47 The NDA told us that there were several causes of this: an inadequate understanding of what Sellafield and other customers needed from RAP, over–optimism about how much it would cost, too little consideration of the condition of Sellafield’s buildings, as well as poor project management.48

Sellafield Ltd now plans to convert a different building to support SRP, and also refurbish the existing building so it can carry on using it until 2040 (whereas RAP had been expected to remain in use until 2070).49 The NDA told us that it expects this to cost between £420 million and £840 million – which is considerably less than RAP would have cost. It also told us that it currently expects that SRP will only be delayed by around one year – though there is still a lot of uncertainty around how long the gap would be (or whether there would even be a gap at all). There could, however, be further costs if the facility is required to continue operating beyond 2040. 44 Letter from the Group Chief Executive Officer of the Nuclear Decommissioning Authority relating to Public Accounts Committee’s visit to Sellafield, 17 March 2025, page 2 45 Letter from the Group Chief Executive Officer of the Nuclear Decommissioning Authority relating to Public Accounts Committee’s visit to Sellafield, 17 March 2025; C&AG’s report, para 2.17 46 Letter from the Group Chief Executive Officer of the Nuclear Decommissioning Authority relating to Public Accounts Committee’s visit to Sellafield, 17 March 2025, page 2; C&AG’s report, para 2.18 47 C&AG’s report, para 2.20 48 Letter from the Group Chief Executive Officer of the Nuclear Decommissioning Authority relating to Public Accounts Committee’s visit to Sellafield, 17 March 2025, page 2 49 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 3; Q69 15 The NDA told us that it believes it does not make sense to spend money now to enable this due to the uncertainty about what analysis will be required after 2040 and the potential for technological innovation.50

The previous Committee criticised the NDA in 2018 for failing to learn the lessons from projects it had cancelled after belatedly realising that there were more cost–effective ways to reduce risk (having spent £586 million of taxpayers’ money on three projects it had cancelled since 2012).51 The NDA told us that up to 48% of the money already spent on RAP could be written off – meaning £127 million of taxpayers’ money would have been wasted.52 We wanted to understand why Sellafield Ltd had not taken action sooner in light of the amount of money at stake, and the consequences of the delay to SRP. Sellafield Ltd told us that it now has options which it did not have in 2020, as it has completed the clean–up of the building it wants to use to carry out plutonium analysis. It also told us that the review which led it to ‘pause’ spending money on RAP in early 2024 had only taken six weeks.53 The Department also observed that deciding to stop a project as important as RAP was “a tough call to make”.54 Managing the condition of ageing assets

The NDA told us that its plan for decommissioning Sellafield will take around a hundred years, and has three phases. The current phase, focussing on ‘minimising the risk of offsite consequences’ will last until approximately the 2060s; in the next phase most of the waste will be moved offsite for permanent storage, before the final phase sees all the remaining buildings demolished. The Sellafield site is no longer reprocessing fuel – meaning some of the facilities on the site are no longer required.55 In the words of the Department, “a lot of the work at Sellafield is a race against time [ … ] sorting things out within the lifespans of the existing assets”.56 50 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 3 51 Committee of Public Accounts, Nuclear Decommissioning Authority: risk reduction at Sellafield, Sixty–Fifth Report of session 2017 –2019, HC 1375, 31 October 2018 52 Letter from the Group Chief Executive Officer of the Nuclear Decommissioning Authority relating to Public Accounts Committee’s visit to Sellafield, 17 March 2025, page 3 53 Q 66 54 Q 73 55 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, pages 5–6 56 Q 50 16 The ‘Waste Vitrification Plant’ is a good example – Sellafield Ltd would like to complete processing of ‘highly active liquor’ by 2039 as it expects the plant will be less reliable after that point.57

There is an increasing backlog of maintenance tasks at Sellafield which is affecting the condition of Sellafield’s assets. Sellafield Ltd could not carry out some tasks during the COVID–19 pandemic due to restrictions on working practices to prevent transmission of the virus. However even now, fewer maintenance tasks are being carried out than was the case before the pandemic: 12% fewer tasks were carried out in 2023–24 than in 2018–19.58 We also queried whether Sellafield Ltd knows enough about the condition of its assets – given that it does not have information about the condition of 9% of its most important assets. The Department accepted that this was a challenge – as, for example, older assets might not have been built with monitoring equipment installed.59

We explored with the witnesses the implications of assets’ condition for Sellafield Ltd’s plans to address hazards and clean up the site by 2125 and identified four areas of particular concern. Firstly – poor availability of assets has impeded Sellafield Ltd’s ability to address site hazards (such as retrieving waste from the older ponds and silos).60 The poor condition of its existing laboratory also poses ‘acute risks’ to the site’s ability to operate.61 Secondly, the failure of the ‘Replacement Analytical Project’ was in part caused by an inadequate understanding of the condition of the building being refurbished which led to the project’s difficulty being underestimated.62 The long–term strategy for providing analysis capabilities to the site became incoherent as a result.63 Thirdly, the ‘strategic tolerances’ for completing a set of activities inform the planning assumptions made for other facilities.64 For example, Sellafield Ltd told us that its need for certain types of analysis would be much lower once all the ‘Highly Active Liquor’ on the site has been turned into a glass–like substance – and this was a factor in its recent decision–making.65 In recent years, maintenance issues have seriously affected how much ‘liquor’ can be processed in this way.66 57 Qq 41–42 58 C&AG’s report, para 3.15 59 Q 49 60 Qq 12, 55 61 Q 66; C&AG’s report, para 2.17 62 Letter from the Group Chief Executive Officer of the Nuclear Decommissioning Authority relating to Public Accounts Committee’s visit to Sellafield, 17 March 2025, page 2 63 C&AG’s report, para 2.20 64 Q 48 65 Q 69; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 3 66 Q 41; C&AG’s report, Figure 9 17 Fourthly, the deterioration in the overall condition of assets on the site was one of the factors that led Sellafield’s safety assurance team to conclude in 2022 that the site

The forecast cost of decommissioning Sellafield has risen considerably since we last reported. It now stands at £136 billion, an increase of 18.8% since March 2019.68 We challenged the witnesses about the adequacy of the NDA’s budget as we are concerned that unwillingness to spend money in the short term is compromising long–term value for money as, ultimately, all the buildings on the site will need to be taken down.69

The NDA acknowledged that its funding settlement for 2025–26 means “there are difficult decisions to be made about some of the less critical aspects”.70 It subsequently told us that it would ideally “remove risk/ hazard and liability at the earliest opportunity” and that decommissioning can free up space for other uses.71 This matters because the Sellafield site is highly constrained, and Sellafield Ltd believes it may need to build some new buildings on land outside the current boundary.72 Sellafield Ltd told us that its current preferred approach for providing analytical services–which it expects to cost much less than RAP would have – is only possible because it had “drive[n] on a broad front” and cleaned out another building.73 It is even possible that slow progress could mean facilities become obsolete while there is still a need for them, at great cost to the taxpayer.74 We are aware from evidence submitted by Cumberland Council that the adequacy of budgets is a matter of local concern.75 68 C&AG’s report, para 16 69 Qq 95–97 70 Q 91 71 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 6 72 C&AG’s report, para 3.11 73 Q 66 74 Qq 50, 95 75 DS0006 19

We explored how the Department, as sponsor, ensures that the NDA and Sellafield Ltd achieve value for money. It told us that it was for the NDA board and the Sellafield board to decide how to allocate resources, within the funding envelope set by ministers.76 We also sought to understand what work had been done to understand the cost implications of the current policy of returning Sellafield to a ‘greenfield’ condition. The Department told us that it had not done any analysis recently, but that it doubted that any change to the end–state would affect what Sellafield Ltd needs to prioritise over the next five to ten years. It did however acknowledge that this could be more relevant in the short–term for some of the NDA’s other sites.77

The previous Committee questioned whether UK Government Investments (UKGI) ‘is playing any useful role’ – and whether its function could instead be discharged by the Department.78 The Department told us that UKGI did help it hold the NDA to account, and that its teams had particular expertise in corporate governance, while the NDA told us that it was “a helpful adjunct to how we operate”. However, the specific examples the NDA gave (around appointment of non–executive directors and attendance at the NDA Board) were far from convincing.79

The NDA also needs to construct a ‘Geological Disposal Facility’ (‘GDF’) to store waste permanently – including that currently at Sellafield. This project is still at a very early stage: it needs to drill boreholes before it can even start developing a ‘Preliminary Design’ and also needs to conduct a ‘Test of Public Support’ with the local community .80 It has been considering sites in Cumbria and Lincolnshire – though we understand that Lincolnshire County Council has recently announced it is likely to withdraw, citing “the uncertainty and slow pace of this process”.81 Even though the project is still at an early stage, the opening date has slipped considerably, with the NDA now expecting the GDF to open in “the late 2050s”.82 The previous Committee was told in 2015 that the GDF was expected to open in 2040.83 Delays to the GDF have serious consequences for value for money at 76 Qq 94–97 77 Qq 98–103 78 Committee of Public Accounts, Nuclear Decommissioning Authority: risk reduction at Sellafield, Sixty–Fifth Report of Session 2017 –2019, HC 1375, 31 October 2018 79 Qq 80–81 80 Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 11 81 Q112; Lincolnshire County Council, County council set to withdraw from nuclear waste facility group, 18 March 2025 82 Q 104 83 Committee of Public Accounts, Oral Evidence: Sellafield recall, HC 1096, 11 March 2015, Q105 20 Sellafield as extra buildings could have to be constructed for short–term storage of the waste: a delay of a decade could cost Sellafield Ltd £500 million–£760 million.84 Creating the right culture in the NDA group

The previous Committee took evidence on how the civil service responds to whistleblowing, and found that the government needs to do more to signal the value of whistleblowers.85 It is particularly important that everyone working on a site as hazardous as Sellafield feels free to raise any concerns they have. This Committee–and its predecessors–have received a number of letters from whistleblowers in recent years making serious allegations about how they have been treated by Sellafield Ltd.

Sellafield Ltd told us that it believed it does have a robust process for reporting safety issues as this is one of the conditions of its nuclear site licence. It pointed to the “green rating” it had received the last time the Office for Nuclear Regulation (ONR) had reviewed its compliance with this licence condition. It also considered that the sheer number of reports raised each week about potential safety issues reflected a good reporting culture.86 The NDA referred to the different channels people can use to raise concerns – including informal discussions before work starts, as well as the group–wide ‘Safecall’ anonymous phone service.87 It also told us that it has started tracking whether respondents to its group–wide staff survey are confident that they would not face retaliation should they raise concerns.88

The Department also considered that workers’ willingness to report safety issues is correlated with their willingness to report other poor behaviour.89 We asked about Sellafield Ltd’s use of non–disclosure agreements–a previous Public Accounts Committee report noted that these agreements have been used elsewhere in the public sector to cover up failure.90 The NDA told us that “there may be occasional instances” – and subsequently informed us that Sellafield Ltd had signed 16 non–disclosure 84 C&AG’s report, para 3.17 85 Committee of Public Accounts, Investigation into whistleblowing in the civil service, Thirty–sixth Report of Session 2023–24, HC 457, 25 May 2024 86 Q 9 87 Qq 3, 40; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 4 88 Q 85 89 Qq 85–86 90 Committee of Public Accounts, Confidentiality Clauses and Special Severance Payments, Thirty–sixth Report of Session 2013–14, HC 477, 24 January 2014 21 agreements in the last three years.91 Additionally, the NDA made ‘special payments’ of £377,200 in 2023–24 to settle claims relating to employment matters.92 The NDA acknowledged that “there is more work to do across the group”, and committed to “continue to work them through and investigate where appropriate”.93 Both the NDA and Sellafield Ltd insisted that they were “not complacent” when it comes to addressing bullying and harassment.94

The NAO report recognises that there has been an improvement since 2021 in the number of employees at Sellafield who would recommend it as a great place to work.95 Sellafield Ltd informed us that the most recent survey showed a further improvement, with more employees taking part in the survey than in previous years.96 We explored what the NDA and Sellafield Ltd had done to improve the workplace culture on its sites. The NDA told us that it had introduced more consistent approaches to ‘Speak Up’, supported by a network of 45 ‘Ethics Ambassadors’ across the group who can act as a first point of contact. It told us that the early signs are that its new approach is having a positive impact.97 Sellafield Ltd’s chief executive told us that he had deliberately sought to address a ‘disconnect’ with the workforce upon his appointment in July 2023, while also improving the site’s leadership culture.98 The Department told us that it wanted to see further improvements; its annual letters to the NDA’s chair and chief executive instruct them to continue focussing on culture, while it is now reviewing a new ‘people dashboard’ on a monthly basis.99 91 Q4; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 1 92 C&AG’s report, para 1.11 93 Q 84 94 Qq 3, 7; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, page 3 95 C&AG’s report, para 1.13 96 Q 6 97 Q 3; Letter from the Group Chief Executive Officer at the Nuclear Decommissioning Authority relating to the oral evidence session held on 20 March on Decommissioning Sellafield, 4 April 2025, pages 4–5 98 Qq 2, 90 99 Q 85 22