How the UK went from closing reactors to building new ones
Britain invented commercial nuclear power in 1956, then spent three decades methodically shutting it all down — and is now scrambling to rebuild what it threw away.
There is a particular kind of embarrassment that comes from rediscovering something you already knew. Britain opened the world’s first commercial nuclear power station at Calder Hall in 1956, beat everyone to the technology, then spent the following decades making a series of expensive, politically convenient, and deeply consequential decisions that gutted its own industry. By 2015, the last surviving Magnox reactor shut down at Wylfa in North Wales. By 2023, several of the Advanced Gas-cooled Reactor units that followed had quietly closed too. For a country that once powered 26% of its electricity from nuclear at its 1997 peak, the retreat was remarkable.
Now the same site at Wylfa is at the center of Britain’s nuclear comeback. In April 2026, Great British Energy Nuclear and Rolls-Royce SMR signed a contract formally beginning technology design work for three small modular reactors on the island of Anglesey. In July 2025, the government made a final investment decision on Sizewell C, committing £38 billion to build two large EPR reactors in Suffolk. And in November 2025, Prime Minister Keir Starmer stood up and admitted, with unusual directness for a politician, that “years of neglect and inertia” had let the country down.
That is not a bad summary of the last thirty years. Understanding how Britain got here requires understanding why it stopped building in the first place, and what has changed enough to make nuclear politically survivable again.
The long retreat: how Britain walked away from an industry it invented
The UK’s nuclear history is, depending on your tolerance for irony, either darkly comic or genuinely sad. According to Wikipedia’s detailed account of UK nuclear power, Britain launched 26 Magnox reactors between 1956 and 1971, followed by 14 Advanced Gas-cooled Reactors (AGRs), and finally a single pressurised water reactor at Sizewell B, which opened in 1995. Then nothing for thirty years. 🏚️
The reasons layer on top of each other like geological strata:
Privatisation in the late 1980s exposed nuclear’s true economics. When auditors scrutinized the books ahead of privatisation, decommissioning liabilities ballooned from £10.4 billion to somewhere between £22 and £36 billion at current prices. Investors recoiled. The nuclear assets ended up in public hands anyway.
The AGR program was a repeated fiasco. A Works in Progress investigation found that Dungeness B, the first AGR ordered, was awarded to a company that hadn’t expected to win the tender and had no real plan to deliver it. Nine years into the program, Britain still had no operational AGRs, and costs had escalated 50% on average. 😬
Lack of standardization meant each reactor was essentially custom-built. France built a fleet of nearly identical PWRs and drove costs down through repetition. Britain built bespoke reactors at different sites by different contractors, each learning the same hard lessons independently.
Political consensus collapsed in the 1990s. Without a clear energy policy, new nuclear was effectively ruled out until a government review reversed that position in 2006. By then, the industrial skills, the supply chain, and the institutional memory had largely evaporated.
The last Magnox reactor — the original Wylfa unit — finally stopped generating electricity in December 2015. It had been running for 44 years, twice its original design life. That’s a remarkable machine, even if the story around it is painful. Britain’s nuclear fleet is now mostly EDF-owned aging AGRs, several of which have already closed, and Sizewell B, the only PWR, which is currently seeking a 20-year life extension to 2055. ⚛️
The Hinkley Point C education, delivered at great expense
Before discussing what Britain is now building, it is worth spending a moment on what it has been building, because the lesson is central to understanding the SMR push.
Hinkley Point C in Somerset is the UK’s first genuinely new nuclear plant in a generation. EDF began construction in 2017 with expectations of completing the first reactor by 2025. The project now targets 2030 at the earliest for Unit 1, with Unit 2 probably in the early 2030s. Initial costs were estimated at £18 billion. Engineering and Technology Magazine reports total costs have risen to approximately £46 billion at today’s prices. That is not a rounding error. That is the GDP of a small country. 💸
EDF blamed British regulations for requiring 7,000 design changes, including 35% more steel and 25% more concrete. The Office for Nuclear Regulation tersely replied that it did not recognize its requirements as “the principal factor” in the cost increases. Both parties have a point, which is one of the depressing things about this story.
A government-commissioned Nuclear Regulatory Review in 2025 was characteristically blunt: the UK had become “the most expensive place in the world to build nuclear projects.” The review listed the causes:
Complex, risk-averse regulation that has accumulated restrictions without ever pruning them
Loss of nuclear skills and supply chain during the 30-year building gap
No standardization, so each project essentially starts from scratch
Brexit and COVID, which hurt labour supply and materials costs for Hinkley specifically
A planning system that treats each nuclear project as novel regardless of precedent
The uncomfortable truth is that Hinkley Point C is simultaneously a disaster and a necessary disaster. Stuart Crooks, Hinkley’s managing director, described it as “relearning nuclear skills, creating a new supply chain, and training a workforce,” which is an honest acknowledgment that Britain had to pay to rebuild knowledge it once had for free. Whether that knowledge transfer justifies £46 billion is a fair question, and I’m not sure anyone has a satisfying answer. 🔬
Sizewell C and the government’s biggest nuclear bet
In July 2025, Energy Secretary Ed Miliband signed the final investment decision for Sizewell C, committing a project estimated at £38 billion, with the UK government holding 44.9% of equity as the largest shareholder. The other investors are La Caisse, the Canadian pension fund (20%), Centrica (15%), and EDF (12.5%). The National Wealth Fund is providing the majority of debt finance. 🏗️
What makes Sizewell C different from yet another announcement about a nuclear project that might eventually happen is the actual money attached. The government has committed £14.2 billion through the Spending Review and taken on the role of lead investor. That is a meaningful change in how the UK is approaching nuclear: not leaving it to the private sector to figure out, but treating it as infrastructure the state needs to own and build.
Sizewell C is essentially a copy of Hinkley Point C. The government claims it will cost around 20% less because of lessons learned from Hinkley, which implies they have learned something, and also implies the optimism that always precedes nuclear construction. The expected operational date is 2039 at the earliest. Things to know:
The two EPR reactors at Sizewell would each generate 1,630 megawatts, together powering the equivalent of six million homes
China was originally an investor in Sizewell C but was bought out due to “security concerns”
10,000 jobs at peak construction, with 1,500 apprenticeships
70% of construction value is targeted to go to UK suppliers
The government has been consistent in calling this a “golden age of nuclear,” which is the kind of phrase that either ages well or doesn’t age at all. 📋
Wylfa, Rolls-Royce, and the SMR wager
If Sizewell C is the big, expensive, established bet, Wylfa is the more interesting one.
The island of Anglesey in North Wales has a nuclear history that mirrors the country’s own: the original Wylfa Magnox station opened in 1971 and finally closed in 2015. Hitachi then spent years planning to build a £20 billion advanced boiling water reactor there under the name Wylfa Newydd, before formally canceling the project in January 2021, citing economics and the absence of a government financing commitment. The UK government bought the site back from Hitachi for £160 million in 2024. 🏝️
In November 2025, Wylfa was designated as the site for Britain’s first small modular reactors. In June 2025, Rolls-Royce SMR had been selected as the preferred technology after a two-year competition against GE-Hitachi, Holtec Britain, and Westinghouse. The plan is for three 470-megawatt units, which together would produce roughly the same output as one large EPR — around 1.4 gigawatts total.
The Rolls-Royce design is a pressurised water reactor drawing on established PWR technology. Whether it qualifies as truly “small” is a matter of some debate — critics point out that at 470 MW, it is larger than many historical reactors and requires full nuclear safety provisions including exclusion zones and aircraft crash protection. Calling it an SMR, some observers suggest, is more brand positioning than strict taxonomy. The nuclear industry has a complicated relationship with its own vocabulary.
The financial structure is worth noting:
£2.5 billion committed by the government via the 2025 Spending Review for the GBE-N SMR programme 💰
£599 million from the National Wealth Fund directly to Rolls-Royce SMR for reactor development
A final investment decision on the Wylfa project is expected in 2029
First grid connection targeted for the mid-2030s
US firm Amentum selected in January 2026 as programme delivery partner
On 13 April 2026, GBE-N and Rolls-Royce SMR signed the contract that formally commenced design work. Engineers are now conducting geoseismic studies at the site, mapping cooling water availability from the sea, and routing grid connections. It is real work, not just announcements.
If you follow UK energy policy, the question worth asking yourself is: does Britain have the political will to stay the course for the decade-plus it takes to complete these projects? That will isn’t always guaranteed. ⚡
The honest assessment: it’s not a done deal
Anyone who has watched British nuclear policy cycle through promise and disappointment is right to be cautious. The ambitions are real. The obstacles are equally real.
The regulatory system is still the same one the 2025 review called “the most expensive in the world.” Reforms have been promised, but nuclear regulation reform takes years to implement even when everyone agrees it is needed — which they don’t entirely. The planning system adds further friction, and the nuclear skills gap hasn’t been closed; it’s been slightly narrowed.
Key things to watch:
Whether Hinkley Point C actually reaches first power in 2030 or slips again, because every delay feeds skepticism about the whole programme
Whether Rolls-Royce’s licensing process (Step 3 of the Generic Design Assessment) completes on schedule in late 2026, enabling first concrete at Wylfa potentially in 2027
Whether supply chain capacity proves adequate for running Hinkley C, Sizewell C, and Wylfa simultaneously, since all are competing for the same nuclear engineers, specialist contractors, and specialist components
The 2029 final investment decision for Wylfa, which is where the financial commitments either get made or don’t 🔍
There is also the honest question of cost. Stephen Thomas, emeritus professor of energy policy at the University of Greenwich, has argued that Rolls-Royce’s 470-MW design will not deliver the economies of scale that make SMRs attractive in theory — it’s large enough to require all the safety and security infrastructure of a big reactor, without being large enough to benefit from full-scale economies. That’s a criticism worth taking seriously rather than dismissing.
What Britain has, at last, is something it lacked for thirty years: a government that has made actual financial commitments rather than just expressions of interest, a site selection process that has produced a decision, and contracts with real money attached. The country that opened the world’s first commercial nuclear power station in 1956 is not starting over — it’s trying to rebuild a capability it let rust. Whether it can do so at the pace and cost the government is projecting is a different question entirely, and Hinkley Point C’s long shadow makes anyone’s projections look at least a little optimistic.
The Rolls-Royce contract at Wylfa is the most concrete signal yet that Britain is serious this time. If those three reactors begin generating power in the mid-2030s roughly on schedule, the whole programme becomes more credible and the export ambitions — starting with the Czech Republic — become realistic. If they don’t, Britain will have paid a very expensive tuition fee for another lesson it already learned once.
What do you think is the bigger risk: that Britain repeats its AGR-era cost disasters with the new SMR programme, or that it ends up with too little nuclear capacity to replace the aging fleet before the gap becomes a genuine grid problem?