Defence Nuclear Infrastructure Contents

Conclusions and recommendations

1.The Department’s nuclear infrastructure projects have suffered from major cost increases and delays. While we acknowledge the unique complexity and scale of these projects has contributed to costs escalation and this is common to other comparable programmes in other countries, UK cost overruns were caused in large part by avoidable mistakes, such as beginning construction work without mature designs. The three projects will not be delivered to the original plans, will be delayed by between 1.7 and 6.3 years, and will cost a total of £1.35 billion more than originally planned. In particular, MENSA had seen a cost increase of £1.07 billion. Some of these cost increases were the result of poor planning decisions and were avoidable. For example, 48% of the total increased costs and nearly £400 million for MENSA was a result of construction starting before requirements or designs were sufficiently clear, which was time-consuming and costly to subsequently rectify. The Department states that it immensely regrets the amount of taxpayers’ money lost as a consequence, but acknowledges that costs could continue to increase. It notes that similar projects in other countries had also gone over budget, because of the unique complexity, scale and safety standard requirements of nuclear projects. It states that the big lesson from these experiences was the crucial necessity for proper planning before starting to build.

2.We understand the need to make early progress given the interdependencies between the infrastructure programmes and the requirements of the Dreadnought programme, but the Department did not properly anticipate and manage the risks of starting early. The Department told us it has learned from these experiences and has paused another Atomic Weapons Establishment project—Project Pegasus for the construction of a manufacturing facility for enriched uranium—in order to ensure that the design is mature enough and that requirements for the future warhead are clear.

Recommendation: In the future, the Department must more explicitly identify and manage the risks of initiating infrastructure projects without a fully mature programme design, and plan using appropriate checkpoints within contracts to assess progress.

3.The Department’s previous contracts have been poorly designed, which has left the taxpayer to shoulder the burden of cost increases while doing little to incentivise contractors to improve performance. The defence nuclear field is a monopoly environment and very few companies are able to carry out such work; four contractors hold 97% by value of the Enterprise contracts. The Department says that the defence sector has become increasingly concentrated in a small number of companies. The 1958 Mutual Defence Agreement also means that only Rolls Royce has the capacity to design and manufacture nuclear propulsion systems.

4.The contracts for the three infrastructure projects did not allow the Department to share the financial risk. This means the taxpayer has borne the full cost of budget overruns, including those of its sub-contractors. The Department accepts the National Audit Office’s criticism that its contracts were not well designed and says it would not operate in this way in the future. The Department tells us that a number of steps have been taken to ensure it has more appropriate contracting mechanisms and avoid repeating these mistakes. For example, before any contract is let by a primary contractor for Dreadnought work, it is scrutinised by a new contracts permissioning group chaired by the Submarine Delivery Agency Chief Executive. In addition, around £37 billion of contracts—some 252 in total—are now subject to the Single Source Contract Regulations (the Regulations), which the Department says have helped increase transparency over costs. However, so far it has been unable to persuade Rolls Royce to agree to move the Core Production Capability contract to come under the Regulations.

Recommendation: In the 2020 report to Parliament on the Dreadnought programme, the Department should update us on how it is taking full advantage of the Single Source Contract Regulations, making full use of target cost or firm price contracts, and ensuring that it effectively shares risk with site owners when negotiating commercial arrangements.

5.The current funding regime does not work for the Nuclear Enterprise due to its uniquely long project timescales, and given the impact on the stretched overall defence budget. Instead, there is a strong case for ring-fencing the nuclear budget. The nuclear budget is around 18–19% of the UK defence budget, whereas it is 6–7% of the defence budget of the United States. The Department says it currently has the funding it needs, but will negotiate with the Treasury about future funding as part of the Spending Review. Under current arrangements, the nuclear budget remains subject to the same pressures as other expenditure. The Department is keen to place some form of ring-fence around the nuclear budget, given its high priority and importance.

6.The Department finds that the demands of Treasury annularity expectations and managing to a 31 March deadline are hard for long-running major capital programmes. The need to achieve in-year affordability savings has had some terrible consequences, such as work being deferred, which in the case of MENSA has led in the past to increased costs, with AWE receiving additional fees as part of a £97 million cost increase. It has embarked on a huge programme of recapitalisation across the whole Nuclear Enterprise, involving new nuclear submarines, new propulsion systems and new infrastructure, many of which will be in the top ten largest for the Department. As a result, there are peaks of spending in the coming years, at a time when, as the Committee has reported on a regular basis, the overall defence budget is stretched. This emphasises the importance of managing the portfolio of programmes as a whole, trying to smooth spending and managing the overall risk profile.

Recommendation: Given its impact on the overall defence budget, the Department should make a case to the Treasury for ring-fencing the nuclear budget in the course of the discussions in 2020 for the current Integrated Review and the Spending Review.

7.The Department has belatedly learned through experience the importance of strong relationships between it, the nuclear regulators and the site owners. This has been a missed opportunity to improve mutual understanding and tackle challenges, to the detriment of the Nuclear Enterprise. Regulatory arrangements for nuclear sites involve the Department, the site operators and one or both nuclear regulators, depending on whether the Department or industry own the site. The Department must pay for the infrastructure it needs, but it is the site operator that holds the nuclear licence and must satisfy the regulators’ requirements. Because the Department does not operate most of the nuclear-regulated sites, it does not hold a regulatory role. All parties accepted that some ‘gold-plating’ of specifications had taken place, although the Department does not consider that the regulators try to force expensive facilities on it.

8. More recently, the Department says there has been improved dialogue via senior leadership groups which bring together different responsible parties at Devonport and AWE sites. We commend this continuous learning from experience. A Defence Nuclear Organisation team is also now located on site at MENSA and there is a shared cost management system to share schedule and cost data in real time. The Department says that the Office for Nuclear Regulation has been happy to discuss costs and help avoid ‘gold-plating’ of specifications. It is also confident that there are many issues around the design and build that it can continue to discuss with the regulatory community, which do not require any changes to the current arrangements.

Recommendation: To secure performance improvements across infrastructure programmes, the Department must continue the commendable practice of admitting failures early and learning from its mistakes. We expect to see as standard more robust liaison arrangements between the Department, site owners and regulators, including the use of co-location of teams, consistent with practice in the civil sector to accelerate the process of reviewing and learning.

9.It is unacceptable that the Department in other areas has repeated past mistakes, and has failed to learn lessons from elsewhere. This underlines the importance of developing a strong corporate memory and retaining experienced staff wherever possible. Many of the mistakes made on these projects are similar to those on which the Comptroller and Auditor General reported more than 30 years ago, and have also been seen in the civil nuclear sector and in the United States. The Department cannot explain why its leadership has not ensured that it learned from these experiences, started before requirements were sufficiently mature, and had not engaged sufficiently with the regulators in the past. It accepts that the NAO report contains a full set of lessons and that it should not operate in the same way in the future.

10.The failure to learn from the past underlines the importance of the Department maintaining a robust corporate memory, particularly given that specialist staff are in demand and may move to posts elsewhere in government and the private sector. The creation of the Defence Nuclear Organisation was in part designed to consolidate knowledge in one place, and the recently appointed Director-General Nuclear stated that she had specialists with 20–30 years of experience, but there are personnel shortages. The Permanent Secretary states that he wants a mix of internally developed staff and external expertise, and that itis important that the Department is able to generate its own talent. The Department continues to discuss how to improve the position on specialist skills with the Department for Business, Energy and Industrial Strategy and the Department for Education.

Recommendation: Given the specialist nature of this field, it is essential that the Department has in place effective arrangements to maintain corporate memory, and works with industry and other government departments to develop the skills needed to be able to take forward nuclear work in line with best practice. The Department should update us on the progress it is making in this regard in the 2020 report to Parliament on the Dreadnought programme.

11.Ultimately, the Department retains the risk associated with these programmes and must manage them itself, regardless of whether it owns the relevant sites or not. The Department uses eight nuclear-regulated sites. It owns and operates two of them, with contractors owning or operating the others. The Department acknowledges that the scale of the risks of nuclear programmes, civil or military, means they are too large for private companies to carry and the risk of failure or cost overruns rests with the state, regardless of ownership. The Department accepts that contracts it had signed in the past have not reflected this reality as well as they should have done.

12.On whether it is considering taking on the risk directly and then contracting out the facility to a private company, the Department responded that it is considering where ownership of risk should lie for future projects.We are also concerned that when the Department makes a contract it does not at least jointly own the intellectual property. The Department says that it makes a balanced decision about who should own or use the facilities and intellectual property. Sometimes the contractors have the intellectual property that the Department wishes to benefit from, and on other occasions, the Department generates it.

Recommendation: The Department must avoid writing contracts which purport to transfer risk to the private sector when in reality this is illusory. The Department must only write contracts which are explicit about where risks lie and how those risks will be monitored and managed by both the Department and the contractor. The Department should write to us by 31 December 2020 to provide a detailed assessment of whether the current ownership arrangements for nuclear regulated sites are in the best interests of the taxpayer and whether more could be done to exploit the intellectual property arising from developments on the sites in the national interest.

Published: 13 May 2020