Last I read none of the closed plants in the USA had been completely decommissioned. Which is why I think the estimates for decommissioning are wrong. But perhaps there are cheaper ways to do it in the future... I don't know.

Yes, fuel cost isn't that much... apparently 0.52 ¢/kWh. But that could still impact the bottom line if costs were 3x as much. But with 70 years of uranium left, I'm not sure prices will stay at similar prices over the life time of a new reactor.

That's more than 95% processing, not raw material cost.

> But with 70 years of uranium left, I'm not sure prices will stay at similar prices over the life time of a new reactor.

That's not true at all. The total accessible uranium resources are near inexhaustible. The 70 years you hear bandied about is proven reserves/consumption, which is completely braindead, because proven reserves do not mean what you think they mean.

If the market price of uranium rose by two orders of magnitude, it still wouldn't materially impact the economics of nuclear fuel plants, and it would make thousands of years worth of uranium reserves economically extractable.

The Trojan Nuclear Power Plant[1] in Oregon comes "close" to being decommissioned. E.g. the reactor vessel itself was barged up the Columbia River and dumped into a big pit in the ground.

Of course, the spent nuclear fuel remains on site. But that's a political issue, not a technical one.

[1] https://en.wikipedia.org/wiki/Trojan_Nuclear_Power_Plant

Much, if not a lot of this has to do with the simple fact that after you mothball the plany, it's easiest and cheapest to wait a while for various radioactive isotopes to decay to less hazardous stuff. As long as you mothball it properly, and maintain that, there's no hurry.