The rumored 3 digit prices paid for nuclear contracts he shares, while a decent profit over the marginal running costs of an existing nuclear plant, do not cover even the low end of costs to build a brand new plant in the USA. Which, unlike his Australian audience, has an existing nuclear industry.
I wonder if he knows that and it's why he uses the price of building in China as a benchmark?
The latest Lazard gives $32 to run a nuclear plant, $190 average to build one, with a range of 144 to $222.
They came up with a capacity cost by comparing the cost of a supercritical coal plant in Korea with Australia (modulo size differences), along with an assumed lifespan of 30 years. See section 2.5 and appendix table B.10 of the GenCost report; their estimates are between 103 and 167 USD/MWh, so not a million miles away.
The Australian GenCost report basically ruled out nuclear as a sensible option, with an extra big question mark on the much touted Small Modular Reactors.
> Building a large-scale nuclear power plant in Australia would cost at least $8.5 billion, take 15 years to deliver and produce electricity at roughly twice the cost of renewable sources, the country's leading scientific institution has found.
> In a report that has for the first time compared conventional nuclear power with other options in Australia, the CSIRO concluded the technology's costs were broadly similar to gas- and black coal-fired generation with carbon capture and storage.
> But the CSIRO's GenCost report noted that nuclear was still likely to be at least 50 per cent more expensive than large-scale wind and solar power backed by "firming" technologies such as batteries.
> And it warned that the true costs of utility-scale nuclear were likely to be far higher still, given the major risks of cost and time-frame blowouts for a technology that had never been built in Australia before.
> It said a nuclear power plant in Australia would be a "first of a kind" and, as such, "premiums of up to 100 per cent cannot be ruled out".
> basically ruled out nuclear as a sensible option
The report's conclusion has been drawn from a single comparison (USC coal in South Korea/Australia then scaling the APR-1400 cost), and they don't really show their workings. A factor of 2.8 is asserted/accepted, which does make me query the competitiveness of the Australian economy.
A counterfactual would be to compare using Japan. The USC coal plant in Japan and Australia cost about the same, so nuclear would be 3963 USD/kWe (2018); exchanging for AUD (1.35 AUD/USD) and allowing for perhaps 20% inflation over that time we get 6350 AUD/kWe . Scaling up to a 1400MW plant decreases the capacity cost by 10% to 5715 AUD/kWe ( (1377MW/1152MW)^0.6 = 0.9 ).
So I think there is certainly a weakness in the report's methodology; a national level decision should probably not be based on a single comparison. Getting quotes from a few builders (KEPCO, Westinghouse/Bechtel) might be a good idea before remodelling the energy system.
4 comments
[ 3.5 ms ] story [ 21.0 ms ] threadI wonder if he knows that and it's why he uses the price of building in China as a benchmark?
The latest Lazard gives $32 to run a nuclear plant, $190 average to build one, with a range of 144 to $222.
To be fair, the Inflation Reduction Act Investment Tax Credits does give 30% + 10% + 10% off of new nuclear plants, provided certain criteria are met.
https://www.bclplaw.com/en-US/events-insights-news/inflation...
As for Australia, they do intend on getting nuclear powered submarines. Their recent CSIRO GenCost report on LCoE for Australia did include nuclear.
https://www.csiro.au/en/news/All/News/2024/May/CSIRO-release...
They came up with a capacity cost by comparing the cost of a supercritical coal plant in Korea with Australia (modulo size differences), along with an assumed lifespan of 30 years. See section 2.5 and appendix table B.10 of the GenCost report; their estimates are between 103 and 167 USD/MWh, so not a million miles away.
https://www.abc.net.au/news/2024-05-22/nuclear-power-double-...
> Building a large-scale nuclear power plant in Australia would cost at least $8.5 billion, take 15 years to deliver and produce electricity at roughly twice the cost of renewable sources, the country's leading scientific institution has found.
> In a report that has for the first time compared conventional nuclear power with other options in Australia, the CSIRO concluded the technology's costs were broadly similar to gas- and black coal-fired generation with carbon capture and storage.
> But the CSIRO's GenCost report noted that nuclear was still likely to be at least 50 per cent more expensive than large-scale wind and solar power backed by "firming" technologies such as batteries.
> And it warned that the true costs of utility-scale nuclear were likely to be far higher still, given the major risks of cost and time-frame blowouts for a technology that had never been built in Australia before.
> It said a nuclear power plant in Australia would be a "first of a kind" and, as such, "premiums of up to 100 per cent cannot be ruled out".
The report's conclusion has been drawn from a single comparison (USC coal in South Korea/Australia then scaling the APR-1400 cost), and they don't really show their workings. A factor of 2.8 is asserted/accepted, which does make me query the competitiveness of the Australian economy.
A counterfactual would be to compare using Japan. The USC coal plant in Japan and Australia cost about the same, so nuclear would be 3963 USD/kWe (2018); exchanging for AUD (1.35 AUD/USD) and allowing for perhaps 20% inflation over that time we get 6350 AUD/kWe . Scaling up to a 1400MW plant decreases the capacity cost by 10% to 5715 AUD/kWe ( (1377MW/1152MW)^0.6 = 0.9 ).
So I think there is certainly a weakness in the report's methodology; a national level decision should probably not be based on a single comparison. Getting quotes from a few builders (KEPCO, Westinghouse/Bechtel) might be a good idea before remodelling the energy system.