Summary – This text is my contribution to the 2026 Journées de l’économie (Jéco) in Lyon, where on Tuesday 3 November I take part in the panel “What transition for energy infrastructure?”. It was first published, in French, in the Mag des Jéco. The energy transition is largely a matter of infrastructure: what must be built to produce and consume low-carbon electricity, and what will have to be closed. The difficulty is to coordinate a supply and a demand that wait for each other, to move forward through options without closing doors, and to share the roles between markets and long-term policy.
The energy transition is largely a matter of infrastructure: the infrastructure needed to produce, transport and consume low-carbon electricity, but also the infrastructure that serves fossil fuels and will have to be closed. Of course, this infrastructure raises a question of money: climate investment in France reached €102 billion in 2024 (3.5% of GDP), and it would need almost to double by 2030 (I4CE Climate Finance Landscape 2025) [1]. Investment in the electricity grid, for instance, should rise from nearly €4 billion a year today to more than €10 billion a year in 2040 [2].
There are many reasons to invest: beyond geopolitical and climate imperatives, it is sometimes simply a matter of renewal, or of reindustrialisation. Many of these projects tick enough boxes to gather varied support, yet consensus is not in the spirit of the times, and every project finds its opponents. In any case, one major difficulty remains: orchestrating a sustained effort that leaves room for agility, bringing an overall vision into dialogue with the specificities of each territory, and reconciling long-term ambition with the pragmatism that lets us move forward day by day.
Coordinating supply and demand illustrates this difficulty well. On the supply side, the French nuclear fleet was built in a very short time: 4 GW a year commissioned between 1980 and 1990. This success is a fine example of a transition, and our historical fleet an undeniable asset, but today we also pay for the dynamics of that period: part of the loss of skills, the “cliff effect” of reactors that will reach the end of their lives together, and uncertainty about our ability to extend the fleet durably. For renewables, we absolutely must avoid a halt like the 2010 solar moratorium [3]. Preparing for these deadlines and maintaining our industrial base require regular investment, despite today’s overcapacity.
But this overcapacity cannot be allowed to settle in, and it should push us to develop demand, because electrification is one of the main workstreams of decarbonisation. The other low-carbon energies rely mostly on biomass, whose available volumes, outside waste, are not commensurate with needs. Electricity, a minority of our consumption today, is set to become its largest share, with efficiency gains as a bonus. And here, even more than before, everything goes through infrastructure: the electricity grid of course, transmission as well as distribution, but also charging points, battery and vehicle plants, electrolysers for hydrogen for industry, for fuels and for synthetic chemistry. Beyond this electricity grid, other infrastructures are linked to one another: heat networks, hydrogen networks, CO₂ capture, transport, use and storage.
Each of these workstreams is linked to the others, and the risk is not equally shared. On the supply side, the State carries it, through long-term contracts for renewables and public financing of new nuclear. On the consumption side, much remains to be put in place, and everyone is waiting for everyone else: electrolysers make little progress because they struggle to find customers; manufacturers who want to electrify their heat first seek to secure part of their electricity supply. On top of this comes the dependence on networks. Grid connection lead times are lengthening in the short term with the arrival of data centres, and the medium term will require major reinforcements, not always easy to carry out, such as the Jonquières–Fos line. Hydrogen, for industry as well as for shipping and aviation, would develop better with transport and storage infrastructure; but that requires centrally planning large investments, at the very moment when policymakers’ ability to anticipate the role of hydrogen in Europe has been seriously called into question.
To move forward with agility, there are many options, in the literal sense: investing without closing doors. In the electrification of industry, this means allowing faster grid connections in exchange for power that is not guaranteed in all circumstances, and managing congestion through gas-electric hybridisation, high-temperature heat storage, process control or batteries. For hydrogen, we will not switch overnight to a system based on electrolysis connected to a “hydrogen backbone”, when hydrogen is produced today from gas, on site, without a transport network: we have to accept stages. One can, for example, hybridise new electrolysers with existing steam reformers, as Air Liquide is doing with Normand’Hy [4], while waiting to know which infrastructure to build and which storage to use.
That leaves the political and economic organisation of these transitions. Should we envy China’s strategy? Its planned economy gives the illusion of a steamroller, but its bet is on overcapacity that cares little about balancing supply and demand. It holds as long as the world buys and Chinese households, who finance it through their savings, accept it. Our agility must be conceived at the European scale, and Europe is not China. We need well-organised markets that, in operating the system, expose prices reflecting costs as far as possible, and that coordinate actors across countries, across energy carriers, along the whole value chain. But we cannot expose every actor to the resulting volatility, especially for their investments, and particularly in the phases when an industry’s first projects are being launched. Long-term commitments must be organised: the State can secure volumes, as it plans to do in Article 24 of the 2027 Finance Bill [5], or encourage private actors to do so. To short- and medium-term markets, “the commissariat”, to borrow Jean-Pierre Dupuy’s word [6]; to a long-term European and French policy, the direction.
Policymakers must still be able to correct their mistakes without waiting too long: REPowerEU’s hydrogen ambitions targeted far too large volumes, for uses, notably in transport, that were better electrified directly [7]. Europe is now reacting on these points. Policy must also manage to bring people together, and to support the losers, because some infrastructure will also have to close. The gas distribution network is the clearest example: as uses electrify, its fixed costs weigh on fewer and fewer customers. Anticipating these stranded assets and supporting those who bear them, workers, local authorities, captive consumers, is the condition for a shared project.
I will debate this on Tuesday 3 November 2026, 2:00-3:30 pm, at the emlyon auditorium (144 avenue Jean Jaurès, Lyon 7th), with François Chaumont (RTE Auvergne-Rhône-Alpes) and Minh Ha-Duong (CNRS, CIRED), moderated by Simon Brigode (Regards croisés sur l’économie). The conference page (in French).
Notes
[1] I4CE, Panorama des financements climat, 2025 edition: €102 billion of climate investment in France in 2024 (−5% year on year); according to the draft third national low-carbon strategy, it must rise by €87 billion above that level by 2030, i.e. 1.85 times the 2024 level. The 3.5% relates these €102 billion to 2024 GDP (≈ €2,920 billion, Insee). The largest gaps to needs are in building renovation and vehicles.
[2] The two figures do not have the same scope. The “nearly €4 billion” is climate investment in electricity grids as counted by I4CE for 2024 (€3.9 billion: reinforcements and connections by RTE and Enedis); across all items, these two operators invested about €8 billion in 2024, and their programmes approved by the regulator (CRE) plan an increase of more than 70% by 2028 (RTE from €2.3 to €5.9 billion, Enedis from €5.3 to €7 billion; I4CE, p. 77). For the transmission grid alone, RTE plans to go from about €3 billion in 2025 to €8 billion a year by 2040, nearly €100 billion over 2025-2039 (CRE, public consultation 2025-08 on RTE’s ten-year network plan, 18 September 2025). Enedis, for its part, announces about €96 billion by 2040.
[3] Decree no. 2010-1510 of 9 December 2010, which suspended for three months the feed-in tariff obligation for new solar projects, before the tariffs were overhauled in March 2011.
[4] Normand’Hy, at Port-Jérôme: a 200 MW PEM electrolyser, up to 28,000 t of hydrogen a year, connected to Air Liquide’s hydrogen pipeline network and operated in parallel with the group’s steam methane reformers, which smooth the intermittency of production and ensure a continuous supply to customers (Air Liquide; Clean Hydrogen Partnership).
[5] French Finance Bill for 2027, Article 24, measure “Creation of a support mechanism for synthetic aviation fuels”: a double auction run by an intermediary, which buys up to 160,000 t a year of synthetic aviation fuel from producers under contracts capped at fifteen years, then resells it to fuel suppliers; the State covers any deficit (prior impact assessments of the 2027 Finance Bill, § 1.4.4 of Article 24, in French).
[6] Jean-Pierre Dupuy, L’Avenir de l’économie. Sortir de l’écomystification, Flammarion, 2012. “L’intendance” is the commissariat, the quartermaster’s job of supplying an army whose course is set by others. The introduction speaks of “seeing politics let itself be humiliated by the economy, power by the commissariat”; the first chapter says the economy succeeds “when it […] goes beyond its condition of commissariat by letting itself be carried by the transcendence that politics offers it”.
[7] The REPowerEU plan (May 2022) aimed for 10 million tonnes of renewable hydrogen produced in the EU in 2030, and as much imported.