Very very little is made/manufactured on site. You do not make turbines, computers, pipes, or much at all, on a construction site. You form, weld, pour, and assemble. The goal would be a standard design, to cut those giant engineering and custom fabrication costs to a small fraction.
You don't need to deliver a completed factory on a pallet to save costs (although that's possible with distributed small module reactors, as Asia is ramping up for).
But the "negative learning" was exactly because they abandoned the standardized approach.
"Conversely, the gradual erosion of EDF’s determination to
standardize (caving in to proposals of numerous design changes
in the wake of the ‘‘frenchifying’’ of the Westinghouse de-
sign—the P’4 reactor series—and above all to the new N4 reactor
design pushed by the CEA), as well as the abrupt slowdown of the
expansion program after 1981, paved the way towards a gradual
demise of the French success model, as borne out in lengthened
construction times and ever higher cost escalation towards the
end of the program (cf. Section 4 below)."
As a result, France's electricity grid is almost fully decarbonized (and we'd be in a better state if we didn't shut down one reactor for political reasons) while industrialized countries betting only on NREs are still counting the days they don't need to burn coal. So there are good things about it.
It’s a little more complicated as they’re importing and exporting a great deal of electricity. But overall yea it’s been good for the environment, just expensive.
It's not subsidized. In fact, cheap nuclear electricity is used to subsidize other industries.
The entire nuclear industry (construction, operation, support) cost France € 228 billion and produce 11000 TWh (by 2012). That's 2,07 cents/kWh. Not too shabby.
Your numbers are wildly off and not just from ignoring inflation when using poorly sourced numbers from 2012. Quick, how much did they spend on fuel over that timeframe? Well according to that estimate it was 0, so it’s hardly including the operating costs.
To give some perspective: “In March 2023 France's Parliament formally approved the government's nuclear investment plan – by 402 votes in favor and 130 against – which considers the €52 billion construction of six new EPR-2 PWRs at three sites.” That’s not operations that’s just for construction of 6 reactors when they have 56 in operation and that’s interest free unlike US reactor where interest is included with construction costs. One year later that’s already been increased to 67.4 billion euros: https://www.reuters.com/business/energy/french-utility-edf-l....
France operated as a pay as you go system so they didn’t set money aside for the full cost of decommissioning their reactors etc. Last I checked there was some talk in 2017 of them setting aside 27 billion based on some ridiculously optimistic estimates but mostly the plan is just foist the costs onto future taxpayers.
The article explicitly says "overall cost" and "including all expenditures."
"Following the Fukushima nuclear accident in 2011, the French government requested that the Court of Accounts prepare a report on the overall cost of both public and private investment in the French nuclear power industry from its beginning, including all expenditures.[225] The report estimates that the industry has cost around 228 billion euros for a yearly production of roughly 400 TWh, with a cumulative production of approximately 11,000 TWh. Among the expenses, the Court of Accounts differentiates €55 billion spent on research since 1950 (equivalent to approximately a billion dollars annually) and €121 billion spent on construction, which includes €96 billion on the 58 reactors."
Your quoting estimated future prices does not contradict what has already happened. The new reactors both have higher rated output than what is currently installed and higher capacity factors. Operating costs are small compared to initial investment, and fuel costs are a small part of the operating costs.
Let's math the shit out of this!
Assuming the 1.5 GW output for the EPR2 that's quoted in Wikipedia and a conservative 90% capacity factor, each of these reactors will produce 1.5 * 24 * 365 * 0.9 = 11826 GWh of electricity per year.
80 year running life makes that 11826 * 80 = 946080 GWh of electricity over the lifetime of the plant. Or 946 TWh. That's 946 Trillion Wh, or 946 Billion kWh. If I can sell these 948 Billion kWh for 1 cent / kWh, that's 948 billion cents or € 9.48 billion so close to the estimated cost of constructing each of these plants. So let's assume an extremely unlikely ~100% cost overrun and operating costs that are the same as originally estimated construction costs and we have 3 cents / kWh. Everything after that is profit, even with 100% cost overruns for construction.
Which maybe gives you an idea why, though the French government almost certainly does not like the cost overruns, they don't seem to be nearly as perturbed by them as the anti-nuclear activists.
And neither is the UK government. So nobody is going to claim that Hinkley Point C is going well. Nevertheless, the UK is proceeding with Sizewell C, have just selected a site for an additional pair of reactors and have made it policy to quadruple nuclear output.
Just like neither the Poles nor the Ukrainians let the problems at Vogtle-3/4 keep them from selecting the Westinghouse AP-1000 for 4 reactors each, with site-prep work having started in both countries earlier this year.
All nuclear power plants made since around 1980s or so have been made in factories. The US (and other nations) use them all the time for their submarines- defense concerns correctly override environmentalists (and the oil companies backing them financially).
The problem is a solved one. We're rich enough to afford to tilt at windmills instead- for now, anyway.
If they really work in this factory like shown in that video, handling radioactive material in the production process with little personal protection (especially against airborne dust), then at least I for one would not like to work for them, ever. The safety of "final" product notwithstanding (that's fine), the shaping, size sorting and sintering processes create dust and really, you don't want to get that into yourself.
Very very little is made/manufactured on site. You do not make turbines, computers, pipes, or much at all, on a construction site. You form, weld, pour, and assemble. The goal would be a standard design, to cut those giant engineering and custom fabrication costs to a small fraction.
You don't need to deliver a completed factory on a pallet to save costs (although that's possible with distributed small module reactors, as Asia is ramping up for).