Most arguments about Power-to-X are arguments about efficiency. Conversion losses get counted, round-trip percentages get quoted, and the conclusion writes itself: storing electricity as hydrogen is wasteful, so build less of it. A study from Leibniz University Hannover and the Institute for Solar Energy Research Hameln (ISFH) makes the opposite calculation – it prices what happens when the flexibility is not there. The answer, for Germany, is up to 60 billion euros.
The study is not new. Weniger Abregeln durch mehr Flexibilität im Energiesystem (“Less curtailment through more flexibility in the energy system”) by Alexander Mahner, Raphael Niepelt and Rolf Brendel was published in August 2025 and financed by EWE AG. We are returning to it now for a reason: the arithmetic it performs is exactly the arithmetic missing from the Swiss debate, and a year on, nobody has done it for Switzerland.
A third of renewable electricity will not be used when it is produced
The core finding is a single number. In a cost-optimised German energy system in 2050, roughly 35 percent of all electricity from renewable sources has to be stored or converted into hydrogen before it can be used. It is generated at a moment when nobody needs it, in a place where nobody needs it, or both.
That is not a flaw in the system design. It is the direct consequence of building a power supply out of wind and sun. Once the share of variable generation passes a certain point, the question stops being “how much do we generate?” and becomes “what do we do with the surplus?”. There are only three honest answers: curtail it, move it in time, or move it into another energy carrier. The first is pure loss. The second is what batteries do. The third is what electrolysers do.
What too little, too late actually costs
The team modelled a transformation path from today to 2050 against Germany’s current statutory expansion targets, and then compared it with two less comfortable variants: one in which electrolyser deployment lags by five years, and one in which capacities are capped at what the grid operators’ network development plan foresees.
The reference case has installed electrolyser capacity reaching around 10 GW by 2030 and roughly 70 GW by 2050, with battery storage distributed across the country – around 128 GWh in Bavaria and 95 GWh in Baden-Württemberg by mid-century. Fall short of that, and the total cost of the energy transition rises by up to 60 billion euros.
The mechanism behind that figure matters more than the figure itself. As lead author Alexander Mahner puts it: the costs rise “because we need more imports”. Electricity that could have been captured domestically is curtailed instead, and the energy has to be bought in from somewhere else, in another form, at somebody else’s price. Missing flexibility does not simply waste renewable electricity. It converts a domestic asset into a foreign dependency.
Electrolysers north, batteries south – and why the split is not arbitrary
The study’s second contribution is spatial. It does not just ask how much flexibility Germany needs, but where it should stand.
Electrolysers belong predominantly in the north, close to the wind. Wind surpluses are large, sustained and seasonal; they last for days, not hours, and no realistic battery fleet absorbs them. Converting them into hydrogen turns an overload into a product that industry can use and that can be stored for months.
Battery storage belongs across the country but with its centre of gravity in the south, close to the photovoltaics. Solar surpluses are a day-night problem, not a winter problem – a few hours of shifting is enough to make them useful.
The two technologies are not competitors for the same job. They address different timescales, and the study’s optimised scenario needs both. Any policy discussion that treats “storage” as one undifferentiated category will get the mix wrong.
The electrolyser as grid infrastructure
This is where the study connects to a debate we have been following in Switzerland. Markus Friedl’s observation – that Swiss electrolysers are currently driven by grid needs rather than by hydrogen demand – points at the same structural fact from the other end. An electrolyser is not only a device that makes hydrogen. It is a controllable load, and a controllable load in the right place is worth money to the system whatever comes out of the other end.
Value it only by the hydrogen it produces and it looks expensive. Value it by the curtailment it prevents, the grid expansion it defers and the imports it avoids, and the balance sheet changes. The Hannover model does the second calculation. Most public debate still does the first.
What this means for Switzerland
Switzerland is not Germany. We have no offshore wind, no north-south generation divide on that scale, and we have storage hydropower – a flexibility asset most countries would envy. The 60 billion euros do not transfer.
The method does. On 2 September the Federal Council adopted its report on synthetic energy carriers and seasonal storage, concluding that Power-to-X will play only a limited role in winter electricity supply, with conversion losses and costs among the reasons given. Those losses are real. But a round-trip efficiency figure on its own is only one side of a ledger. The other side – what curtailment, deferred grid expansion and avoided imports are worth – is what the Hannover team modelled, and it is precisely what the Swiss assessment does not quantify.
Switzerland is about to write a national storage strategy, is preparing the Energy Perspectives 2060 to replace data from 2020, and is running a Round Table on Energy Storage in which Power-to-X is explicitly in scope. Every one of those processes needs the full ledger, not half of it. The question is not whether conversion has losses – it does. The question is what the alternative costs.
Germany now has a number for that. Switzerland does not.
Source: Alexander Mahner, Raphael Niepelt, Rolf Brendel, “Weniger Abregeln durch mehr Flexibilität im Energiesystem”, Leibniz Universität Hannover / Institut für Solarenergieforschung Hameln, 2025, funded by EWE AG. Available at repo.uni-hannover.de. Reported by IWR and pv magazine.
🤖 AI transparency: This text and its visual were created using AI, reviewed and approved by a human.
Power-to-X Congress Switzerland 2026 – 22 September 2026, Kursaal Bern: “Reality Check with Net Zero”. Programme and registration.