Back to the future: lessons from the UK’s build out of nuclear power
It was a 50 year journey from public to private nuclear power in the UK during which, whilst always challenging, most new build success happened when public owners contracted with private firms.
Between 1946, when the Atomic Research Establishment was created under the Ministry of Supply, to 1996, when privatisation of nuclear power took place in the UK, the base policy assumption was that nuclear power generation would be owned by public institutions.
This was manifested firstly via the Atomic Energy Authority (AEA) - created in 1954 to pick up the baton from the Ministry of Supply and behind the role out of Magnox power plants (4.4GWs in total, at an average size of c500MWs per unit), and then by the Central Electricity Generating Board (CEGB) who drove the development of Advanced Gas-cooled Reactors (AGRs, 8.4GWs, at an average size of c600MWs per unit), and the single Pressurised Water Reactor (PWR, 1.2GWs) unit at Sizewell B.
Whilst ownership in this period remained public, the private sector has always been heavily involved construction of nuclear power in the UK, in both the Magnox and AGR programmes. For example, under the AGR programme, the CEGB selected (via tender) from a consortia of private firms to build out power stations. The three main consortia were the Nuclear Power Group (NPG), Atomic Power Construction (APC), Nuclear Design and Construction Consortium (NDCC. Latterly in 1973, following the collapse of APC, the remaining two consortia were combined to form the National Nuclear Corporate (NNC, majority privately controlled).
This model was not without its problems. Tellingly, right from the start, there were pervasive concerns about the costs of developing nuclear power plants, and the impact on the exchequer in incurring those costs. For example, a well known British atomic scientist, Franz Simon, remarked as early as 1953 that “It would be quite impossible to give an estimate of the cost of a [atomic] unit of electricity other than in the vaguest terms.” Whilst the former Treasury Economic Advisor, David Henderson, described the AGR programme as one of the two most costly government sponsored project errors (second only to Concorde).
In truth, lateness and cost overruns were continuous features of the Magnox and AGR eras. For example, Dungeness B faced design problems which led to a 13 year delay to commissioning (from 1970 to 1983) and considerable financing problems, which ultimately led the to failure of one of the main members of the consortia model. There were also additional costs and bottlenecks arising from the facilitation of different approaches to the detailed design and delivery of AGR units, even though they shared the same underpinning design concept. This was all compounded by construction taking place in a fractious labour market. And iterations to address lessons from prior projects had mixed success. Hinkley Point and Hunterston followed Dungeness but were actually commissioned earlier, but Heysham 1 and Hartlepool - which had been specifically designed to reduce complexity and cost - faced unexpected difficulties in construction arising from those changes.
But despite the thread of these issues running through the entire period of public ownership, we should recognise that over 12GWs of nuclear power generation capacity was delivered over a 40 year period. By contrast, no new nuclear power stations have been delivered in the quarter of a century or so since privatisation, and this - despite a degree of political flip flopping - has decidedly not been the policy intention. This is a significant policy failure, of which there can be no doubt.
Most recently of course, this is all being brought into sharp focus by the continuing escalation of costs and repeated announcements of delay at Hinkley Point C, which has prompted the government to recently consider the implementation of a Regulated Asset Base (RAB) model for subsequent new nuclear roll out, a subject of a previous post by this author.
What lessons can we take from an examination of this history? First I want to make clear that I am setting aside the first order question of whether we should develop new nuclear power stations. I see this as a moot issue mainly because both main political parties are committed to new nuclear power stations. For now, efforts are therefore better focussed on not if, but how such a programme could be implemented. Taking this as read, there are some highly valuable lessons that could and should be shaping our current policies:
New nuclear has always been a highly risky financial venture - Building nuclear power stations requires substantial political commitment, and large resource and financial mobilisation. And, very few are built to time or budget. Nuclear projects tend to exceed timescales and cost plans, with earlier projects much more liable to do so, and to a greater extent. Any political case for the support of nuclear power needs to avoid dressing up, or attempting to hide these risks. And the bigger the project, the bigger the political and economic the risks become.
The public sector takes construction risk one way or another - However much one would like to believe the private sector can own nuclear power plants in construction, and take construction risks, in the end government will always come under pressure to insulate and support private capital, and if it does not or cannot do so, there is a high probability of even more costly failure through write offs of sunk investments.
The broader case for nuclear needs to be made to sustain political and public patience - Patience to last the course tends to be present where there has been clear and consistent articulation of the benefits of nuclear deployment across multiple dimensions - energy security, environment/climate change, and even the inter-relationship between a civil nuclear programme and military and defence capabilities. If government isn’t prepared to make the case in such a way, then there is a high likelihood of weak grounds emerging to defend continuation of new nuclear build programmes. Pure financial returns metrics won’t always look favourable - far from it - particularly during difficult and complex construction phases.
If we reflect on these lessons, then it is clear that we will need a different approach to building out new nuclear power stations to those adopted over the last 25 years or so. As I have written about previously, it is unlikely that the shift over to a RAB model will resolve the fundamental challenges the UK faces in seeking large scale deployment of new nuclear power, and if anything may make the political risks too substantial for Ministers to effectively manage.
Instead, there could be merit in exploring the following model:
Public ownership in construction - We could revert back to an approach of public ownership of nuclear power stations in construction. In truth a RAB model as currently proposed will see bill payers and tax payers sharing in construction risks anyway, but with no direct financial stake in the project that allows citizens to benefit from the financial returns once it is de-risked and operational. Sure, the argument runs that consumers will then enjoy low cost, low carbon power, but if citizens and bill payers are de facto sharing in the construction risks under RAB, then are smeared bill benefits enough? A clearer model, which is decidedly more honest, would be for ownership of projects in construction to remain in public hands, with substantial transfer of risks to the private sector under tendered construction contracts. Then publicly owned projects could be sold off under auction to private ownership once commissioned. The sales would allow repayment of sums invested on the public’s behalf, during the construction phase. Notably, Great British Nuclear are already rumoured to be looking at some ownership of project sites, with an investment in a development site at Wylfa in Anglesey a distinct possibility according to the weekend press. But there could also be a role here under a prospective Labour government for GB Energy and/or a National Wealth Fund.
Risk transfer to private sector construction firms via contracts - As part of this model, the government could revisit at least the spirit of the consortia approach, and invite firms to bid for the construction of sites, entering into contracts with the publicly owned project. The focus of effort and intellect for government then would be establishing contractual frameworks that incentivise out-performance, and penalise under performance in construction. There are clearly well established construction contract frameworks that could be applied here in terms of delay damages, defects liability provisions, performance guarantees and credit support clauses. Nuclear interface risks will be highly complex, noting the possible involvement of multiple contractors. Financial sums involved will no doubt create highly fractious commercial negotiations on risk allocation. There might be a meaningful role here for Export Credit Agencies to wrap risks and enable successful conclusion of contracts, in a manner that we have seen work highly successfully in the offshore wind sector. Of course, it will not be easy to establish viable contractual frameworks, but this is not a challenge unique to this proposed model - such frameworks need to be developed anyhow under the current regime, by private sector owners of projects. A public sector owner tendering for consortia, would at least allow for standardisation, create transparency and open book building blocks for iteration, and mean that the UK tax payer is a direct beneficiary of damages or compensation for contractual under performance.
Auctioned sales of nuclear projects, once commissioned - Once construction phases are concluded, and projects commissioned, projects could be auctioned, with the proceeds raised by sale used to repay construction costs incurred by the UK tax payer. Low cost, low carbon energy would then be available, but in addition, a direct recovery of exchequer costs would be enabled by this model. Discount rates used to value operational nuclear projects would - depending on arrangements for managing waste and decommissioning - likely be low. With very long asset lives and low costs of operation, it is likely that operational nuclear power plants would attract interest from private sector buyers. Consideration could be given by government to award of CfDs to auctioned projects to further enhance the likelihood of higher valuations.
Smaller projects, of the same design, for more manageable risk taking - Of course, this proposed model requires substantial capex costs to be incurred by the government during the construction phase, as oppose to this being supported by the private sector under a RAB or CfD approach. But in the absence of any success in delivering new nuclear under either RAB or CfD in the UK context, then arguably these are false counter-factuals. It is highly likely that one way or another either the public sector will have to bare significant risks under RAB, or we won’t procure any new capacity if we continue with a CfD as the bedrock of raising private capital for the construction of nuclear projects. In any event, to ensure that such an approach is manageable to the public purse under a public ownership model in construction, then it would be better to reduce the scale and increase the number of individual projects, albeit adopting the same over-arching design approach. For example, a project like Hinkley Point C (£40bn+ for 3.2GWs) would be incredibly difficult for government to own during construction, but a few projects such as Tees, using Small Modular Reactors (at a cost of £10bn or less, for c1GW) would be much more manageable, particularly if the design and build approach could be homogenised.
In the absence of a new utopia in which new nuclear power stations can be delivered by private actors in the UK context, and noting that no route to the same outcome is likely to be smooth, then we could do worse than looking back to what underpinned (qualified and not always linear) achievements in previous eras.
If we can tweak and adapt those models, whilst finding the political will to support their application on an enduring basis, then it may be possible to create similar momentum in the next phase of nuclear power in Britain, to what we experienced in earlier decades.
In truth, the current government has begun at least tentatively and partially down a pathway such as the one alluded to above with its £2.5bn shareholding in Sizewell C. This makes the UK government the largest shareholder during its pre-construction/development phase, and represents the first public ownership of a nuclear plant in 30 years in the Uk. But this is still short of taking the project through construction to operation under majority public ownership. The current model anticipates that RAB will lead to more conventional private sector funding at a future final investment decision. It remains to be seen if a RAB regime can achieve that feat given the lessons of recent history.
An election and possible change of government provides the opportunity to pause and take stock, and consider building on such approaches. Let us hope, for the sake of not losing further time, that such an opportunity is taken.