The referendum that made nuclear energy illegal in Italy was held right after the accident of Chernobyl; people voted with their gut, not with their mind.
Curiously enough, some Italian town signs still have a small sub-sign reading "denuclearized municipality" (example: https://live.staticflickr.com/1421/712461235_e46a2c7db5_z.jp... )
Since all plants in Italy are closed since the 90s won’t it take decades and billions of tax euros to get them back operational again?
Studies are converging on renewable energy systems being cheaper in climates ranging from Australia to the Nordics:
Some recent ones:
Finland, 2025: nuclear scenarios 71–84% higher system cost; unrestricted least-cost optimization builds zero new nuclear. https://www.sciencedirect.com/science/article/pii/S036054422...
Denmark, 2026: system-level LCOE including storage, flexibility and system integration; renewables substantially cheaper than nuclear. https://www.sciencedirect.com/science/article/pii/S036054422...
Australia, CSIRO + AEMO GenCost 2025–26: whole-system modelling including transmission, storage and firming; renewables + storage remain the lowest-cost net-zero pathway. https://www.csiro.au/en/research/technology-space/energy/Ele...
And, even under ideal conditions, baseload is still quite hard. Especially in Southern Europe (e.g. which has quite a bit less land than Australia, and less wind than the nordic countries).
Fundamentally, I agree that it boils down to cost, and the people writing $XXB checks are generally pretty good at financial modeling. If nuclear makes sense, and the regulations allow it, they will build it. However, it's fighting against a highly variable, zero-marginal-cost renewable market with something that is high capex. Same reason people don't build new coal in the US despite certain politicians trying to popularize it.
My guess... If it does take off, either it will have to be wildly cheap (unlikely), or there are geopolitical concerns about the gas backup that Italy currently relies on so heavily to make the renewables work.
We fix tail risks when we encounter them. Not by envisioning your pet ones to make your solution fit.
You're suggesting to start fixing the issue after the blackout happens?
Australia's weather is different from what we got in Europe. And let's not pretend many of aemos estimations are already obsolete due to increased costs of transmission and hydro storage that are huilt.
Dk still has worse emissions vs france/sweden, just like SA
If you don't like nuclear, you'll have gas peakers, like Germany
Cheaper for how long? And in what conditions?
What happens in a post petrol world? What happens when china stops exporting panels/batteries/inverters (aka a good 50% of the current supply)?
Edit to add: heh, the CSIRO eh? That bastion of truth.
Please do tell what is wrong with the CSIRO GenCost.
Nuclear power has the opposite spread of ruinously expensive CAPEX and acceptable OPEX.
I would say the pandemic shows that you do not need to account for every possibility. Humanity is amazing at adapting itself and finding new solutions to the problem at hand, rather than wasting all our resources trying to prevent the next made up disaster.
This is why we model with 10 year winters, X00 year floods or similar. Fix reasonable problems, leave the tail risk as tail risk. The worst that happens are rolling blackouts until the grid adapts, which is no different than any other outage due to grid maintenance or someone cutting a cable with a digger.
It mostly happened because covid hit at the same time.
In most countries which don't have the advantages of Australia or Finland, you do need a mix as the OP was saying
I only checked the first paper and it doesn't seem talk about what would happen if a 3 weeks bad wind event would happen in winter, and a rough look at the figures makes me think it just wouldn't work at all. I didn't check the other papers.
And the proof is in the pudding anyways, if it's that cheap and feasible to do, well let's see who's doing it.
https://www.ecoshock.org/2024/02/why-renewables-cannot-power...
I think he points to a valid point but completely over plays his hand. Everyone that has argued these kinds of points have brought valid arguments but have completely under estimated the pace of change.
Do keep in mind that Michaux is a fan of being a little contrarian. He likes the attention it being to him.
He might be right, and him pointing to the issues may actually be something that helps us avoid these issues. Generally, knowing about a problem is a much better position than being slammed by something you didn't anticipate.
I tend to think long term we are going to a much lower energy world, that we made the mistake of building a world and our expectations to the paradigm of fossil fuels. But I also think that we are not going back to the stone age and that renewables will do better than the pessimists. Good chance we will hit a reasonable mid-ground even if it means living standards change, possibly a smaller, slower but more intentional world. It may be nice even if it means you can no longer get a slab of Coke for an hours wage.
Which makes the whole "we'll build renewables and storage" sounds even more unrealistic
[1] https://www.sciencedirect.com/science/article/abs/pii/S03605...
[2] https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#...
I guess ultimately the nuclear vs renewables discussion is coloured a lot by feelings nowadays.
Resources like this help a lot to form a more nuanced understanding grounded in reality - but otoh I have to admit that a lot of this goes over my head - like even trying to understand how the economy works around this is mind boggling complex. I’ll need a deeper look to understand it better and build a more informed opinion tbh
https://analysesetdonnees.rte-france.com/en/publications/ene...
But yeah their most modern nuclear power plant was a BWR 4 with Mark II containment, which would probably not be worth modernizing instead of building a new GenIII+ PWR from scratch that lasts a century.
I mean anything could happen but seems like a long shot
The argument is that small-scale nuclear reactors have been built for military and scientific applications so the tech is not especially novel.
What I think is unproven is whether SMRs can be built and run safely while being cost-effective to other forms of power.
I wouldn’t be surprised, for example, if it actually costs less to build and run one full-scale reactor than it does for multiple SMRs of the same output.
That doesn’t mean they won’t have a use.
And I personally have switched from bull to bear on that. I was rooting for NuScale but they're hitting the same thing as everyone else.
Meanwhile, battery storage has been on an insane cost reduction curve.
It is the same argument I make for Starship. Technically, yeah it should work. The economics of of going to Mars, not so sure about that.
But despite SpaceX's huge successes with Falcon, that doesn't prove they'll cross the finish line with Starship. Falcon and specifically the Merlin engine made just about every choice they could in favor of risk reduction rather than the best possible performance. It was a continuation of work that Tom Mueller did while at TRW.
With Starship they're in more uncharted territory, especially the re-entry concept, and the engine has been designed under the opposite philosophy of trying to push limits.
And no you’re not solving it with a buffer (batteries or any kind of other energy storage for that matter)
Italy is extremely energy poor, it even lacks coal. It does have significant hydro storage though
There are several different concepts being collapsed together here.
“Baseload is a myth” is mostly a semantic dodge. Base load is simply the minimum demand on the system. You do not need a special category of plant called a “baseload generator,” but you absolutely need enough firm capacity and energy to meet demand continuously, including when wind and solar are producing very little.
“The grid can handle peak load, therefore it can handle base load” confuses transmission capacity with generation availability. A grid capable of carrying 50 GW does not produce 50 GW. The engineering problem is having enough generation available at the exact hours when it is required.
And batteries do not make that problem disappear. They are excellent for frequency response, peak shaving and moving solar production from noon into the evening. But a battery is an energy buffer, not an energy source. During a prolonged shortage it can discharge only what was previously stored, and then it must recharge. Once the problem lasts several days rather than several hours, the amount of stored energy required becomes enormous.
Germany provides a very useful real-world example.
Germany began its nuclear phase-out under the 2000/2002 agreement, briefly extended reactor lifetimes in 2010, then accelerated the shutdown after Fukushima in 2011. The final three reactors closed in April 2023.
In 2008 German nuclear plants produced roughly 149 TWh of electricity. Germany subsequently built huge amounts of wind and solar, and those renewables have unquestionably displaced a great deal of fossil generation. But Germany eliminated its existing firm, low-carbon nuclear fleet while it was still burning very large quantities of coal and gas.
And then there is the problem Germans have an excellent word for: Dunkelflaute.
This is not some hypothetical edge case invented by nuclear advocates. Germany’s Bundesnetzagentur defines a Dunkelflaute as a period of at least 48 hours during which combined wind and PV generation remains below 15% of installed capacity. Low-wind periods are recurrent; analysis of four decades of German weather data found roughly a five-day period with average wind output below a 10% capacity factor in a typical year, with substantially longer events occurring less frequently.
Germany got a particularly good demonstration in November and December 2024. During two Dunkelflaute episodes, renewable production collapsed while demand remained high. Germany’s own Bundesnetzagentur says that almost the entire controllable power-plant fleet was brought into operation and that substantial electricity imports from neighbouring countries were required. The regulator concluded that Germany urgently needs additional controllable generation capacity for future Dunkelflauten.
That is basically the entire argument in one empirical example. When there is plenty of wind and sun, Germany can have extremely cheap electricity and sometimes negative wholesale prices. When the weather changes, the system suddenly needs the plants that supposedly became obsolete.
And the price swings are spectacular. Germany’s average day-ahead wholesale price in 2024 was about €79/MWh. During the November and December Dunkelflauten it exceeded €300/MWh for dozens of hours and reached approximately €936/MWh at the peak. The Bundesnetzagentur investigated whether those prices resulted from market manipulation and found no such explanation; the underlying condition was very low renewable production, high demand, extensive use of controllable thermal generation and limited remaining dispatchable capacity.
At the retail level Germany also remains one of Europe’s most expensive electricity markets. In the second half of 2025 a typical German household consuming 2,500–5,000 kWh paid €0.3869/kWh, compared with an EU average of €0.2896/kWh—about 34% more. Germany had the second-highest household electricity price in the EU after Ireland.
Those retail prices cannot honestly be attributed entirely to the nuclear phase-out or to renewables: taxes, levies, network costs, fuel prices and the consequences of the European gas crisis all matter. But neither can anyone point to Germany as evidence that a wind-and-solar-heavy system has somehow made firm generation economically irrelevant. Its own regulator is saying the opposite: when Dunkelflaute hits, Germany needs controllable generation and imports, and it says additional controllable capacity is urgently required.
What supplies that capacity today? Coal and gas remain substantial. Germany has zero domestic nuclear generation, yet it still burns coal and natural gas and participates in a European interconnected system containing large quantities of French and other nuclear generation.
So the actual German experiment is considerably less impressive than the slogan. Germany spent decades building an enormous renewable fleet while simultaneously destroying an existing fleet of firm low-carbon reactors. It succeeded in greatly increasing renewable generation and reducing coal use over the long run, but it did not abolish the requirement for firm generation. When wind and solar disappear together, Germany burns dispatchable fuels, imports electricity, draws on reserves and watches wholesale prices rise.
Which brings us back to the original question.
Once you accept that an electrical system requires firm energy during prolonged periods of inadequate intermittent generation, the menu becomes fairly obvious. Hydro and geothermal are excellent where geography permits them. Storage, transmission, demand response and renewable overbuilding all help. Countries fortunate enough to possess enormous reservoirs, geothermal resources or exceptionally strong interconnections can lean heavily on those advantages.
Those countries are not a universally reproducible model. A solution intended to work in most industrial countries has to survive winter demand, multi-day wind lulls, minimal solar production, droughts, transmission constraints and simultaneous weather patterns across neighbouring regions.
Under those conditions, you need some combination of renewables and firm generation. At present, for most countries without exceptional hydro or geothermal resources, large-scale firm generation means principally nuclear or combustion, with storage and interconnection reducing how much of it must run.
So yes: build wind and solar. Build batteries where they make economic sense. Build transmission. Use demand response. Exploit hydro and geothermal wherever available.
But after all of that, somebody still has to answer the boring engineering question: what produces electricity on the fifth cold, dark, windless day after the batteries have discharged?
“Baseload is a myth” does not answer it. “Batteries” does not answer it unless you specify the required energy capacity and how it is recharged. And Germany, of all countries, does not answer it: every serious Dunkelflaute demonstrates why it still maintains dispatchable fossil generation and depends on the wider European grid.
This is a policy decision, not a technical limitation. Palo Verde has been operating in the Arizona heat for forty years: nuclear can handle situations of low/high-temp water.
The system worked as designed.
Won’t it also take decades and cost billions to roll out renewables to the extend that they can even put a dent in total energy usage? So I don’t see how that is even an argument.
US killing the linear no-threshold model is a good thing. LNT model assumes that absolutely no amount of radioactivity is safe, which is provably wrong.
It has nothing to do with safety if the model that required "as low as reasonably possible" is bullshit per se.
The maximum allowed annual radiation dose for US radiation workers is 50mSv/year. It's just 16x the global average background radiation. And that's the average. In reality it's more like 5x - 25x.
For cancer specifically we're seeing marginally increased risk at 100 mSv/year, but it doesn't really become worrying enough until we're at 300-500 mSv/year.
https://maps.eu-risk.eucentre.it/map/european-seismic-risk-i...
Sardinia?
Is the Italian coastline at high risk of tsunamis as well?
I would say near Pisa, on the Arno River.
Judging from the map, Piedmont, Liguria, Trentino, Tuscany and Basilicata have good sites.
https://www.nytimes.com/2026/09/20/world/asia/philippines-ba...
Italy lost WWII, and became a de facto protectorate/colony of the US. I wouldn't be surprised if behind the nuclear referendum in the 80s there way to gentle hand of US services.