[eukgoekoko]

From the link that you have provided: > The US Energy Information Administration has recommended that levelized costs of non-dispatchable sources such as wind or solar may be better compared to the avoided energy cost rather than to the LCOE of dispatchable sources such as fossil fuels or geothermal.

Since nuclear reactors are dispatchable by nature (through the use of control rods), the LCOE cannot be an accurate indicator.

[beat]

Nuclear reactors are dispatchable in theory, but it's not economical in practice. One nice thing about nuclear reactors is that they can run more or less wide open for long periods of time. But the flip side of that is that the operational cost isn't driven by fuel consumption. It costs about the same to operate a reactor whether you run it at 100% capacity or 10% capacity.

So the financial models for reactors are built around selling electricity at a certain price and a certain capacity for several decades, maybe as long as 50 years. Most of the cost is up-front, which makes it very capital-sensitive. And capital is risk-sensitive, so unpredictability in energy prices and the risk of less expensive alternatives appearing (which is exactly what has happened over the past decade, with three different sources coming it at half the price of nuclear) raises the risk, which makes the capital much more expensive, which makes the reactor that much more expensive, which raises the price targets they need to hit, and thus... at a certain point, it makes zero economic sense to build reactors. It doesn't even make sense to keep running reactors we already have, which is why many are shutting down years ahead of their official end-of-life plans (and causing hundreds of millions, maybe billions, in sunk cost, which has an impact on risk...)

Dispatchability means NOTHING. There's a reactor in Iowa that's getting closed because it lost key industrial customers to cheaper wind. It can't be operated profitably at 50% power.