Delivering Sizewell C on time and to budget relies on “big assumptions” finds National Audit Office
20 May 2026
To read the NAO press statement and report, click here: https://www.nao.org.uk/reports/sizewell-c/
Stop Sizewell C commented: “The NAO’s report confirms what we already suspected – that “big assumptions” and the “significant uncertainty” of factors underpinning DESNZ’s claimed benefits could easily turn Sizewell C into a financial disaster, with its investors – thanks to RAB – being the only ones who can’t lose. As the NAO confirms, households are relying on those investors to produce significant savings and reduce Sizewell C’s construction time to justify the nuclear tax on our energy bills, but we share the NAO’s questions about whether investors can or have the incentives to do this.
We had asked the NAO to look at Sizewell C before it reached Final Investment Decision and are dismayed it did not do so, but at least some critical information withheld by the government is now in the public domain. We agree with the NAO that DESNZ must provide transparency of forecast cost and schedule for Sizewell C. We call for the government’s promised Sizewell C Strategy and Delivery plan, containing a public, realistic completion date, to be laid before parliament immediately.”
Key revelations
-
The “benefits” of Sizewell C to consumers would not outweigh the cost until at least 2064 and DESNZ’s modelling involves “significant uncertainty”, such as changes in the cost of other sources of electricity for reaching net zero, whether the reactors will reliably operate at 90% load factor and whether Sizewell C exceeds its higher predicted cost of £47.7bn [2024] (pp6, 33). Stop Sizewell C notes that DESNZ’s modelling also shows realistic scenarios in which its predicted benefits worth £3.9 – £18bn [2024] over Sizewell C’s lifetime would be wiped out. (p32) [1]
-
The NAO calculates that the Sizewell C Regulated Asset Base or RAB will cost consumers £4 – £4.5bn in returns to investors compared to using 100% public finance. (pp10, 41) Investors will need to deliver £3.9bn in savings and reduce delays by 24 months in order to offset their additional cost to consumers, but the NAO says it is not clear whether investors will deliver these savings or whether the deal sufficiently incentivises them to do so. (p11) [2]
-
DESNZ estimates Sizewell C requires £66.8 – £83.1bn (real, up to £100bn nominal) of funding during construction and £241 – £270bn (real) during its operational life (Figure 7, pp36, 37). The construction phase figures include £24.7 – £33.5bn (real) in net consumer payments, compared to £8 – £8.1bn (real) in investors’ equity. The project’s “baseline” predicted cost is £38.2bn. [2024] These are the figures reported by the Financial Times in August 2025.
-
The NAO says that “although Sizewell C should cost less than Hinkley Point C to build, consumers may still pay more for Sizewell C’s electricity”. DESNZ estimates an equivalent strike price of £133 – £155/MWh [2024-25 prices] – after the government’s ‘recycling’ of its own profit from RAB, predicted to be £19.2 – £20.9bn [2024] over Sizewell C’s lifetime – compared to Hinkley C’s “fixed” contract price of £129/MWh [2025] (p9) [3]
-
Investors will earn a “nuclear premium” of £9.73bn – £10.74bn net cash [2024-25 prices], “higher than typically seen in other utility sectors”. RAB means that many risks are shared with consumers, so RAB limits what investors can lose. (p11)
-
RAB could cost average households £17-19/year by the end of construction. The government has stated it would cost on average £1/month. (Figure 10, p46)
-
Sizewell C’s construction schedule “project baseline” is 111 months, being fully operational by July 2039. Unless this changed before the report was published, Stop Sizewell C cannot comprehend how such a date can be described as “mid 2030s”, as Sizewell C and Minister Vallance told us recently. [4] The Lower Regulatory Threshold is 114 months, fully operational by October 2039 and Higher Regulatory Threshold is 160 months, fully operational by August 2043. By comparison Hinkley C is expected to take 132 – 144 months (p28).
-
Doubts about the future of the Nuclear RAB: The NAO says “SZC reflects specific circumstances in the UK nuclear programme, and the approach may not be transferable” p14. Stop Sizewell C considers this calls into question the time and legislative work put into it. The NAO reports DESNZ spent nearly £77 million on external advice. (p21) The NAO recommends that “Any department considering a similar approach should be able to explain why it is necessary for taxpayers and consumers to bear so much risk, and show that investor returns are proportionate to the risks investors actually carry.” (p14)
-
The report also describes the delays to FID – in part due to the General Election – and protracted negotiations with EDF (p21) which attempted to renegotiate the Hinkley C strike price. DESNZ was determined they should retain a stake given Sizewell C would be built with EDF’s/Hinkley C’s technology, but ultimately accepted that this would be reduced from the expected 19.99% to 12.5%. (p41)
-
The NAO recommends DESNZ and the Company should provide transparency of forecast cost and schedule through joint annual updates to Parliament, including formal notifications that key milestones and thresholds such as the lower regulatory threshold are being approached, and that DESNZ should use their influence within the project (including with investors), and influence across government, to keep cost and schedule below the lower regulatory threshold. (p14)
Notes:
1. Although DESNZ claims Sizewell C’s net present social value (NPSV) – ie the saving to the energy system over its 60 year lifetime – would be £3.9 – £18bn [2024], (p3) this figure would be negative if Sizewell C operated at 85% instead of 90%*, for 50 years rather than 60** and if renewables had lower capital costs. (p32)
* To compare a 90% load factor with other operational EPRs;
- Olkiluoto 3: 2023 88%, 2024 70%, 2025 82.6%
- Taishan 1 average load factor to end 2024: 55%. Taishan 2 (same period): 76%
** Campaigners have long expressed concern that Sizewell C’s operational life could be curtailed by erosion on Suffolk’s fragile coast.
DESNZ states that other non-nuclear costs have risen, reducing this break-even point to the 2050s in some scenarios.
The NAO lists the limitations of the Dynamic Dispatch Power Sector model (DDM) that was used for modelling for Sizewell C’s Full Business Plan (p62). Stop Sizewell C has long criticised use of the DDM because of these limitations. Elsewhere is confirmation that at lower levels of energy demand, it shows new nuclear power is not necessarily needed if alternatives can be deployed at scale (p31).
2. DESNZ’s insistence of being a minority stakeholder is revealed (p40). However this means the government cannot direct the project which we consider a weakness.
3. Hinkley’s contracted strike price of £92.50/MWh [2012] reduced to £89.50/MWh [2012] is worth £129/MWh today. The NAO says Hinkley’s electricity is relatively cheaper because the price was set before its cost overran (which has been borne by EDF), and the cost of borrowing has also increased since then. Without RAB, Sizewell C’s strike price would be £152 to £178/MWh [2024-25], almost twice the price of the £91/MWh for fixed offshore wind from Allocation Round 7 in February 2026 (p31), but see the bullet about public finance being cheaper than RAB.
Note also that “recycling” profits is a policy decision by the current government that could be rescinded.
4. The completion schedule has moved from the very vague “mid to late 2030s” to the slightly less vague but still non-specific “mid 2030s” according to Sizewell C Main Site Forum 15 April and DESNZ Nuclear NGO Forum with Minister Vallance, 14 May.