The cost of nuclear

The spread itself is the clue. A solar panel costs roughly the same anywhere on earth, with real but modest variation in installation. Nuclear runs from about $2,200 per kilowatt in South Korea, the cheapest in the world, to about $8,000 for Vogtle in Georgia and probably the recently completed Finnish project, a four-fold gap. Lovering’s explanation is that nuclear is no longer really a traded product. There used to be far more international commerce in commercial reactors, and today a plant is built as an infrastructure project, country by country. Each one is substantially unique, closer to a highway, a dam or a bridge than to a car or an airplane, and uniqueness is expensive.

Opening up the cost pie makes that concrete. For an AP1000, the 1.1-gigawatt Westinghouse design built at Vogtle, roughly half the capital cost sits in the plant outside the reactor: the yard, the cooling infrastructure, the groundwork. Lovering’s phrase for that half is that it has nothing to do with nuclear, though it is still heavily regulated concrete and steel. The nuclear island itself, meaning the reactor and the pressure containment structure, is only about 12%. Another 35% or so is engineering, procurement, construction management and owner’s costs such as interest accruing during construction. So when people say nuclear is expensive, they are mostly making a claim about large construction projects and the cost of money, not about atoms. She reinforces the point from the other direction: much of what nuclear costs is not concrete and steel either, but engineering, design and project management, and she thinks a good part of South Korea’s advantage is simply excellent project management, which is hard to export.

That is also what makes nuclear cheap where it is cheap. Building the same design repeatedly produces what she calls economies of multiples, and South Korea’s own history shows the curve: it started nearer $4,000 to $5,000 per kilowatt and came down. France shows learning too, and tellingly the French curve jumps back up every time a new reactor design is introduced, then descends again. The United States stopped building in the 1980s and never accumulated the repetitions, which is why its handful of recent projects sit at the top of the range.

Which sets up the regulation question directly, and this is where Lovering declines the argument everyone else is having. Her evidence for demoting regulation is clean: South Korea’s regulator is modeled substantially on the US Nuclear Regulatory Commission, which is widely treated as the international gold standard and copied elsewhere. If American regulation were the dominant cost driver, a country regulating on a similar model should not be the cheapest builder in the world. She grants that licensing could be modernized for advanced designs with passive safety features, and later grants something stronger, that genuinely factory-fabricated reactors will require a bulk licensing regime the US does not have, since aircraft are not licensed one airframe at a time. But she will not call regulation a silver bullet, and she spreads blame onto the industry itself as well. Her alternative explanation is demand. Nuclear gets cheaper where electricity demand is growing fast, because the plants are actually needed, and the US had exactly that when it built its existing fleet. What the US never had is demand-pull policy for nuclear. Renewables got three decades of production and investment tax credits and state portfolio standards that manufactured demand and induced the innovation that took solar from something like $100,000 per kilowatt in the 1970s to today’s prices. Nuclear got nothing comparable until the Inflation Reduction Act and a recent shift in some states toward clean energy mandates that count nuclear.