Hydrogen Auctions Keep Awarding Money That Never Builds Anything
A new Oxford Institute for Energy Studies analysis shows why the European Hydrogen Bank's fixed-premium auctions awarded subsidies that were never large enough to close the cost gap, while the UK's contract-for-difference and Germany's H2Global offtake model actually pushed projects toward final investment decisions. We look at what the withdrawal record and the EU's own cost data mean for how Baltics should design hydrogen support.
NEWS
PtXBaltic
8/10/20264 min read
Winning a hydrogen auction used to sound like the hard part was over. Then you look at what actually got built. Of the seven projects that won the European Hydrogen Bank's first round in April 2024, two have formally withdrawn and not one has announced a final investment decision. That is not a rounding error in an otherwise working system. It is the system telling us something, and a new Oxford Institute for Energy Studies analysis by Martin Lambert sets out exactly what.
The cost gap the subsidy never came close to touching
The first European Hydrogen Bank round awarded €720 million in fixed premium subsidies to seven projects, at levels between €0.27 and €0.48 per kilogram. On the same day, the Bank published its own data showing the levelised cost of renewable hydrogen running from roughly €5/kg in the cheapest European geographies to around €12/kg in France and Germany. Grey hydrogen sits near €2/kg. Do that arithmetic once and the outcome stops being surprising.
Some commentary at the time read those low bids as evidence that buyers were ready to pay a premium for renewable hydrogen. The three years since suggest a simpler reading. Developers bid what they needed to bid to win, and then met the financing reality later.
The third round, announced in early May 2026, landed at €0.57–€0.98/kg in the general category. Higher, but not higher by the order of magnitude the cost data implies is needed. Lambert's judgement is blunt: these subsidies may well prove insufficient too. The maritime and aviation category, awarded at around €3.50/kg, is the one part of the picture with a plausible path — and that covers only two 12 MW projects.
Attrition tells the story better than award announcements
The second round is where the pattern becomes impossible to explain away. A €1.2 billion budget, 15 winners announced in May 2025, 2.33 GW of electrolyser capacity on paper. By September 2025, seven of those fifteen had withdrawn — 1.88 GW of the 2.33 GW awarded. The Commission invited ten smaller reserve-list projects to replace them. In January 2026, six projects signed binding agreements, totalling 380 MW.
From 2.33 GW to 380 MW. Roughly one sixth of the headline, and those six still have two and a half years to reach FID before the money is at risk.
Lambert attributes much of this to thin due diligence on project status and feasibility before award, which lets developers bid unrealistically low and walk away when binding commitments arrive. Notably, the Commission's own December 2025 webinar pack framed the outcome not as a failure of the auction but as confirmation that its design is working as intended. That gap — between what the mechanism was built to do and what the pipeline needed — is the part worth sitting with.
Revenue certainty is what moved projects forward
Two alternative designs have performed better, and both share one feature: they give the developer a dependable revenue line rather than a one-off top-up.
The UK's Hydrogen Allocation Round uses a contract for difference. The project bids a strike price and receives the difference against a market reference price, floored at the natural gas price where no hydrogen market exists. HAR1 strike prices averaged around £9.49/kg (roughly €11/kg), with the lowest near £6/kg. All eleven shortlisted projects signed by January 2026, the 15 MW HyMarnham project entered full commercial operation, and West Wales Hydrogen and Northfleet Hydrogen have taken FID.
Germany's H2Global runs a two-sided auction. The state-backed intermediary Hintco signs long-term purchase contracts on the supply side and auctions the product onward, with government covering the spread. The pilot's renewable ammonia lot went to Fertiglobe's Egypt Green Hydrogen project at €1,000 per tonne of ammonia — equivalent to roughly €4.50/kg of hydrogen — behind a 100 MW electrolyser, and it appears to be moving toward FID.
Neither model is free. A government-backed offtake agreement is expensive from a budget perspective, and the UK route delivers at smaller scale and higher unit cost. But both produced signed contracts and operating assets, which is more than the fixed-premium route has managed at several times the nominal capacity. Oman's hydrogen agency Hydrom has adopted the H2Global model for its next auction round, which is a fairly direct verdict from a jurisdiction with options.
The Baltic position in a price-only European auction
Look at the Hydrogen Bank's own levelised cost chart and the strategic problem for this region becomes visible immediately. Lithuania sits at €8.47/kg. Sweden, the cheapest shown, sits at €5.53/kg. Poland tops the range at €13.50/kg. Latvia and Estonia do not appear in the published country set at all.
In an auction where bids compete on price alone across the whole Union, that spread is not a detail — it is the outcome. A Baltic project with sound engineering, a real offtaker and a credible grid connection can still lose to a project in a structurally cheaper geography that has none of those things and simply bids lower. And on the evidence of rounds one and two, the winning bidder may not build anything either.
So the pan-European auction is a useful instrument for Baltic hydrogen ecosystem stakeholders to compete in, but it might be a poor foundation to plan on. The projects that will actually reach FID here are the ones with a revenue mechanism underneath them that does not depend on winning a Brussels price contest.
Design lessons worth carrying into Baltic support
Three things follow for how support should be shaped in this region.
Design for revenue certainty, not for headline capacity. A CfD or a state-backed offtake gives a lender something to underwrite. A fixed premium against an unknown market price does not.
Screen on readiness, not only on price. Lambert goes as far as suggesting bids should be accepted only from projects close to FID-ready — which raises the cost of bidding and may exclude smaller players, but it stops public budget being committed to projects that quietly disappear.
Pair supply support with demand. Only four of twenty-seven member states had transposed RED III into national law by the end of 2025, and over 99 per cent of the roughly 100 million tonnes of hydrogen used globally each year is still made from unabated fossil fuels. Supply subsidies alone will not shift that.
Coming late to hydrogen support design is not only a disadvantage. The Baltic states get to build their instruments knowing which mechanisms produced signed contracts and which produced press releases. That is a genuinely valuable position, provided the lesson gets used rather than admired.
Source: Oxford Institute for Energy Studies — Oxford Energy Forum, Issue 150