David Sterling: A drugstore sign in Rennes. Forty-three degrees Celsius. That's the image that cuts through all the aggregate data for me — one sign, one city in northwest France, a number that shouldn't be physically possible there.
Megan Skiendel: France's hottest day ever recorded. During what is now the third heatwave of the year.
David Sterling: Third. One month after the second. And the grid — look — Golfech is offline because the Garonne River is too warm. EDF shuts the reactor to stay inside regulatory cooling limits. Same moment Ember's tracking fourteen percent demand growth. The math on that is brutal.
Megan Skiendel: Brutal and, honestly, predictable if you know who Hervé Champenois is. Enedis technical director. He goes public on June twenty-third — underground cables at eighty Celsius, network failures a real possibility. Someone cleared that statement. That's not accidental.
David Sterling: What's the implication — they knew cuts were coming?
Megan Skiendel: I'd want to know who authorized him. But yes — you pre-position like that when the decision is already made internally. And while all of this is happening, Lecornu confirms forty people drowned in unsupervised water in five days. Two children dead in a car in Carpentras.
David Sterling: Prices two to three times normal per Ember. That's the transfer mechanism. And it's running directly toward the people with no optionality.
Megan Skiendel: And here's what I want to get precise about — because the headline is 'heatwave breaks grid' and that's not wrong, but it's not the mechanism. The mechanism is a double-hit. Demand up fourteen percent because everyone's running air conditioning simultaneously. Supply down because Golfech goes offline for the Garonne. It's not one or the other. It's both landing at the exact same moment.
David Sterling: Think of it like a highway at rush hour — someone closes two lanes and adds a thousand extra cars at the same time. The jam isn't the lanes. It isn't the cars. It's the combination.
Megan Skiendel: Exactly that. And what's actually new — the Golfech cooling thresholds aren't a 2026 discovery. The Garonne River temperature limits existed for decades.
David Sterling: Right. Copernicus has Europe warming at point-four Celsius per decade since the eighties. Twice the global rate. The thresholds didn't move. The frequency of hitting them did.
Megan Skiendel: That's the signal. Not that it happened — that it's the third time this year. And the underground cable problem, honestly, is the piece that surprised me. Buried cables hit eighty Celsius because the soil heats up and there's nowhere for that heat to go. Overhead lines don't fail the same way.
David Sterling: So Champenois going public — that's the tell. If Enedis knew cuts were already probable, that warning on June twenty-third wasn't a forecast. It was liability management.
David Sterling: Here's the take I want to stress-test. The one circulating is: spot prices hit two to three times normal, Ember documented it, market signal is loud and clear, grid investment should follow. I don't buy the transmission mechanism.
Megan Skiendel: Why not?
David Sterling: Because the spike lasts ten days. The cable replacement, the substation hardening — that's a thirty-year payoff horizon. The person reading the European Power Exchange at three a.m. is not the person signing the capex budget. Those are two completely different decision loops.
Megan Skiendel: Right, and — honestly, the people controlling grid capex are reading regulatory approval cycles and political calendars. Not spot prices. I mean, France, Spain, Germany, Portugal, Switzerland — all simultaneously under elevated blackout risk, and nobody moved. That's not a market failure. That's a coordination failure between individually rational actors who are institutionally stuck.
David Sterling: Patrick Martin at MEDEF says France is running at a slower pace. Eurostar cancels Paris–London services. The economic cost is landing — diffuse, real — but it's landing on businesses, not the grid investors. So the pain and the accountability are completely decoupled.
Megan Skiendel: Which means the market signal argument isn't wrong about the signal. It's wrong about who receives it.
Megan Skiendel: And that's the part I keep getting stuck on. Copernicus confirms 2024 was the hottest year on record in Europe. This is the third heatwave of 2026. At what point does Météo-France putting fifty-four departments under red alert stop being a crisis response and start being — I mean, is this just the operating environment now? Because if it is, every grid investment target, every generation mix decision from the last decade — they were all calculated against a baseline that no longer exists. And nobody in the regulatory machinery has said that out loud officially.
David Sterling: The nuclear piece is the sharpest version of that problem. Golfech isn't an anomaly — it's France's primary low-carbon asset becoming a supply liability during the exact events decarbonisation is supposed to address. That's not a paradox. That's a design assumption failing in real time. And the engineers already know it.
Megan Skiendel: Champenois went public. The market priced it — two to three times normal. So the engineers know, the market knows. The question is whether there's a single regulatory body anywhere in Europe that's formally acknowledged the baseline has shifted — and if there isn't, what exactly is the investment signal supposed to trigger?