The UK National Audit Office (NAO) has released a new report on the Sizewell C NPP which raises questions about the cost effectiveness of the project. The UK government views Sizewell C as a critical pillar of its long-term energy strategy. The project aims to reduce reliance on volatile imported fossil fuels and establish a baseline of reliable “baseload” power. This steady power supply supports intermittent renewable energy sources like wind and solar, helping the UK meet its statutory target of net-zero carbon emissions by 2050.
Sizewell C is a planned 3.2 GWe nuclear power station located on the Suffolk coast in East Anglia. Once operational, it is expected to generate enough power for approximately 6m UK homes for at least 60 years. The station will feature two European Pressurised Reactors (EPRs) supplied by Électricité de France (EDF). This is a direct replica of the Hinkley Point C (HPC) plant currently under construction in Somerset.
In 2022, the UK government granted planning approval and committed its first major direct financial stake to the project. Formal construction work began on-site in 2024. Following negotiations, the Department for Energy Security & Net Zero (DESNZ) announced in 2025 that it had secured private investment from EDF and other investors. NAO notes that this investment decision was over two years later than anticipated, although it allowed for a fully developed and costed timetable to be developed.
Sizewell C is projected to have a baseline cost estimate of £38.2bn ($51bn), with construction being completed by Summer 2039.The current equity share of Sizewell C is as follows: DESNZ 44.9%; EDF 12.5%, La Caisse 20%; Centrica 15%; and Amber Infrastructure 7.6%. Amber has an option to take a further 2.4%, reducing DESNZ’s share.
The 69-page NAO report warns that while the project can potentially deliver up to £18bn in net societal benefits, the underlying economic model shifts an “immediate and substantial” financial risk onto UK taxpayers and consumers. NAO notes that the project may not achieve a financial break-even point for the public until at least 2064.
Unlike previous projects, consumers began paying upfront for construction costs via energy bills in November 2025 under a Regulated Asset Base (RAB) model. Costs will add roughly £4 per household in 2025–26, rising to £17-19 a year by the scheduled completion date in summer 2039.
Private investors (including EDF, Centrica, and La Caisse) stand to gain £4-4.5bn in financial returns. NAO emphasises that these returns reflect a “nuclear premium” that may be disproportionately high relative to the limited risk these private firms shoulder. It remains unclear if the financing deal sufficiently motivates private backers to keep construction below the £47.7bn upper regulatory threshold.
“Sharing risk between the investors and taxpayers and consumers appears to have reduced the cost of financing Sizewell C, but the rewards for investors still appear high,” NAO says. “Investor financial returns will cost consumers over £4 billion but will be justified if they help the project to cut construction costs and speed up delivery times.”
According to NAO, the DESNZ delivery model for Sizewell C places more risks on taxpayers and consumers than other electricity projects, but the Department believes this model has reduced finance costs and will allow the project to be delivered on time and to budget. This novel approach has costs and relies on big assumptions.
NAO notes that, while capitalising on HPC’s finalised design makes Sizewell C advanced in its planning stages, UK consumers will ultimately pay more for Sizewell’s electricity. This stems from sharply elevated borrowing costs and a lack of mechanisms forcing developers to absorb overruns like EDF did at Hinkley following the triggers of early infrastructure phases.
According to NAO, the construction of Sizewell C is frequently compared to HPC, which is currently expected to double its initial projected cost, with a seven-year delay. This has sparked concerns that these problems may be mirrored in Sizewell C, but DESNZ hopes to avoid this by applying the lessons and final designs from HPC.
“As a result, Sizewell C’s plans are already at a much more advanced stage than Hinkley’s were at the equivalent point, NAO says. “However, although Sizewell C should cost less than Hinkley Point C to build, it is likely that consumers will pay more for energy from Sizewell C. This is because the price of Hinkley’s electricity was set before its cost overran (which has been borne by EDF), and the cost of borrowing has also increased since then.”
Investors told the NAO they are most exposed in “tail‑end” scenarios above the higher regulatory threshold, when the Company would have to apply to the Secretary of State to make decisions over the project’s future, which could involve allowing higher consumer costs (with mechanisms that can limit investors’ financial returns), further government funding (diluting private investors’ equity), or discontinuing the project.
DESNZ believes Sizewell C will lower electricity system costs compared with alternative ways of achieving net zero. This is uncertain because it is hard to predict the future costs and likelihood of success of alternative technologies, but DESNZ has reasonably argued that, because Sizewell C uses a proven technology, it reduces risk across its portfolio. Sizewell C still carries major delivery risks: no nuclear power station of its type has so far been built without delays or cost increases.
NAO concludes: “DESNZ hopes that a new delivery and financing model, designed so the taxpayer and consumer take on more of the project risks, together with replicating the design of Hinkley Point C and more mature cost estimates, mean Sizewell C will deliver to time and budget where others did not. But this approach has costs and its justification relies on some big assumptions.”
DESNZ has intentionally limited the government’s control over the project by sharing ownership with other investors, seeking to avoid governance weaknesses that have beset other mega-projects. “The sharing of risk with the taxpayer and consumer appears to have reduced the cost of financing the project, but the rewards for investors still appear high, given their limited exposure to project risk. The extent to which investors will be incentivised to control project costs in the way DESNZ assumes is unclear.”
