There are several factors, but the single biggest reason for nuclear's ever increasing cost is due to construction productivity.
All other forms of productivity have gone up drastically. However construction productivity remains stuck at a constant. And as other areas are more productive, we need to still pay those in construction competitive wages or else they would switch to more productive jobs with higher wages. (Let's just elide the disconnect between wages rising fully with productivity increases, but they do rise some!)
Despite being far more wealthy today than centuries ago, we don't build cathedrals with super intricate stonework, because labor is so much more expensive.
I fear that nuclear is like the gothic cathedral: something that was far easier to do when labor costs were low, but at wealth increases it becomes far more difficult to make economic sense.
This doesn't measure the cost of providing dispatchable electricity though. If I want 1MWh of electricity at night provided by solar, it's going to cost more than solar's LCoE because I will also need to pay for a way to store and dispatch it.
Which again does not capture the cost of a nuclear plant being forced off the market because no one is buying its electricity during the day and they have to amortize the cost over a 40% capacity factor instead of 85% like they target.
And this can be a purely economical factor. Sure a plant may have a 90% capacity factor but if the market clears at $0 50% of the time they still need to recoup all the costs on the remaining 50%, pushing up the costs to what would be a the equivalent to a 42.5% capacity factor when running steady state.
Take Vogtle running at a 40% capacity factor, the electricty now costs 40 cents/kwh or $400 MWh. That is pure insanity. Get Vogtle down to 20%, which is very likely as we already have renewable grids at 75% renewables and it is 80 cents/kWh.
Take a look at Australia for the future of old inflexible "baseload" (which always was an economic construct coming from marginal cost) plants.
Coal plants forced to become peakers or be decommissioned.
You can say that "no one would do that" but it is the end state of the market.
Electricity is fundamentally priced on the margin and if you start forcing nuclear costs on the ratepayers they will build rooftop solar and storage like crazy, leaving you without any takers for the nuclear based electricity.
Solar has this problem to a much greater extent though. If you have a market where solar is >100% of demand during the day then it will be dispatching at or below $0/MWh for almost all of its life.
But of course the marginal market is not the whole story. In reality solar largely receives effectively fixed prices in most markets (via CfDs or PPAs). Nuclear does the same and can also take capacity payments and sell into flexibility markets where those exist.
And some of those data seems sane: the cost of solar panels reduces due to tech improvements: ok
The cost of coal increases a bit, maybe due to geopolicital issues: ok, seems legit
The cost of nuclear increases .. why ? Why the step between 2016 and 2017 ? Does tech "de-improved" ?
More insights would be interesting
There are several factors, but the single biggest reason for nuclear's ever increasing cost is due to construction productivity.
All other forms of productivity have gone up drastically. However construction productivity remains stuck at a constant. And as other areas are more productive, we need to still pay those in construction competitive wages or else they would switch to more productive jobs with higher wages. (Let's just elide the disconnect between wages rising fully with productivity increases, but they do rise some!)
Despite being far more wealthy today than centuries ago, we don't build cathedrals with super intricate stonework, because labor is so much more expensive.
I fear that nuclear is like the gothic cathedral: something that was far easier to do when labor costs were low, but at wealth increases it becomes far more difficult to make economic sense.
Baumol's Cost Disease
https://en.wikipedia.org/wiki/Baumol_effect
This doesn't measure the cost of providing dispatchable electricity though. If I want 1MWh of electricity at night provided by solar, it's going to cost more than solar's LCoE because I will also need to pay for a way to store and dispatch it.
Which again does not capture the cost of a nuclear plant being forced off the market because no one is buying its electricity during the day and they have to amortize the cost over a 40% capacity factor instead of 85% like they target.
And this can be a purely economical factor. Sure a plant may have a 90% capacity factor but if the market clears at $0 50% of the time they still need to recoup all the costs on the remaining 50%, pushing up the costs to what would be a the equivalent to a 42.5% capacity factor when running steady state.
Take Vogtle running at a 40% capacity factor, the electricty now costs 40 cents/kwh or $400 MWh. That is pure insanity. Get Vogtle down to 20%, which is very likely as we already have renewable grids at 75% renewables and it is 80 cents/kWh.
Take a look at Australia for the future of old inflexible "baseload" (which always was an economic construct coming from marginal cost) plants.
Coal plants forced to become peakers or be decommissioned.
https://www.abc.net.au/news/2024-10-13/australian-coal-plant...
You can say that "no one would do that" but it is the end state of the market.
Electricity is fundamentally priced on the margin and if you start forcing nuclear costs on the ratepayers they will build rooftop solar and storage like crazy, leaving you without any takers for the nuclear based electricity.
Solar has this problem to a much greater extent though. If you have a market where solar is >100% of demand during the day then it will be dispatching at or below $0/MWh for almost all of its life.
But of course the marginal market is not the whole story. In reality solar largely receives effectively fixed prices in most markets (via CfDs or PPAs). Nuclear does the same and can also take capacity payments and sell into flexibility markets where those exist.
1 reply →