Nuclear's Timing Issue

I recently visited Rheinsberg, a small town north of Berlin that is home to a 70 MW nuclear power plant which first came online in 1966. The plant stopped operating in 1990 and decommissioning began shortly afterwards. More than thirty years later, that work is still unfinished with full decommissioning now expected to be around 2040, with costs estimated at over €1bn. Standing there, the nuclear renaissance that many people, particularly in the United States, are now talking about suddenly felt very far away and the plant reminded me of the huge timing issue with nuclear.

On paper, nuclear looks like the perfect solution for a world searching for abundant clean energy. Its greatest advantage is energy density, meaning that a very small amount of fuel can produce an enormous amount of energy. To put it into perspective, 1 kilogram of uranium can produce roughly as much energy as thousands of tonnes of coal. The heat from the nuclear reaction is then used to drive turbines and generate electricity around the clock with very low operational costs.

The problem is that in the Western world we have largely lost the ability to build nuclear power plants quickly or cheaply. Take Hinkley Point C in the United Kingdom, which was originally expected to be producing electricity by now. The final investment decision was taken in 2016, but when the plant eventually comes online sometime beyond 2030 it will likely have cost well over £40bn, making it one of the most expensive power plants ever built, and at least five years behind plan.

Part of the challenge is that almost every large nuclear project is still effectively bespoke. That is why so much hope is now being placed on small modular reactors, or SMRs. The idea is simple enough, standardise smaller reactors and manufacture them in factories rather than build giant one off projects on site. In theory this should reduce both cost and construction time. The issue however is that there are multiple competing technologies, no clear winning design, and as a result it remains uncertain whether SMRs can be commercialised at scale before the mid 2030s.

The deeper issue though is speed. Even in the most capable jurisdictions, nuclear projects consume extraordinary amounts of time, capital, political attention and institutional capacity. Planning, permitting, financing and construction can easily stretch over a decade and in today’s world that matters enormously because the global economy is now evolving on much faster timescales. Artificial intelligence companies need power now. Data centres are being built now. Countries across the Global South need electricity now. At the same time industrialised economies are electrifying transport, heating and industry while simultaneously powering AI, automation and an increasingly digital economy. These systems evolve in months and years not decades. This raises an important question. If nuclear is so slow to bring to market with costs unclear, why are we suddenly talking about a nuclear renaissance?

Part of the answer is that nuclear fits naturally into the strategic and military worldview of the United States. Nuclear aligns with its desire to preserve leadership in a technology that has been for 80 years at the crosspoint between energy, national security and industrial prowess. SMRs and especially nuclear fusion carry the allure of near limitless and cheap clean energy and the possibility of a breakthrough that could fundamentally reshape civilisation itself, not least because fusion is the same process that powers our sun.

There are whole host of other countries such as the UK and France who are looking to expand their nuclear fleets and then there are countries like Poland and Egypt who are looking to build their first plants, but let’s be clear they are all being built for a mixture broader strategic reasons linked to industrial capability, geopolitical positioning, energy security and defence purposes. The decision to go nuclear is not just about producing electricity.

None of this means nuclear generation has no role to play. Existing fleets in countries such as France and Japan continue to provide enormous value, particularly where the capital costs are already sunk and institutional knowhow remains strong. But one of the lessons I took away from Rheinsberg is that nuclear is not simply an energy technology. It is a multi generational commitment. Countries entering nuclear for the first time are not just committing to building reactors, they are committing to managing those systems, their waste and eventually their decommissioning for decades after the electricity has stopped flowing. And the big question is whether this makes any economic sense or not.