Green Hydrogen: Overhyped AND a Real Bet — Where’s the Line?

YouTube Spotify Apple

Key Takeaways

• It is a split verdict, not a hedge. As the fuel of everything — cars, home heating, storing the grid in a tank — hydrogen is overhyped, and thermodynamics decided that before any policy did. As the cleanup of the hydrogen we already make, it is a real bet. Same molecule, two opposite answers.
• The world already runs on ~100 Mt of hydrogen a year — ammonia for fertilizer, desulfurization at the refinery, methanol for plastics. Under 1% of it is clean, and making it accounts for about 2% of global CO₂ emissions. The customers are not hypothetical. They are buying the dirty version today.
• One number decides everything: 50–55 kWh per kilogram. The thermodynamic floor is 39.4 kWh — you approach it from above, and roughly 20% of headroom is all that is left. There is no Moore’s law here. Electricity is 60–70% of the cost, so this is not a bet on hydrogen. It is a bet on absurdly cheap, absurdly available power.
• It does not pencil yet, and the gap is specific. Unsubsidized PEM in the US runs $4.33–6.05/kg against grey at $1.5–2.5/kg. Take the full $3/kg credit and the best case only just reaches the top of grey’s range. For carbon pricing alone to close it you would need about $300/tonne. Today the sector runs on policy, not economics.
• Follow the money and the discipline gap is glaring: announced 2030 capacity fell from 49 to 37 Mt/yr, and only about 9% has reached final investment decision. The bottleneck is almost quaintly old-fashioned — nobody signs the offtake.

Two Questions Wearing One Name

Most hydrogen arguments are unwinnable because the two sides are answering different questions. One side is asking whether hydrogen should power your car and heat your house. The other is asking whether the hydrogen industry that already exists should be cleaned up. Those have different answers, and pretending they are one question is how you get both a bubble and a backlash.

So take them separately. That is the whole method here.

Hydrogen wins where it is the ingredient. It loses where it is just a courier for electrons.

The Molecule We Already Make

Here is the part almost nobody leads with. We do not need to invent demand for hydrogen. The world quietly consumes about 100 million tonnes of it every year — turned into ammonia for fertilizer, used to strip sulfur at refineries, made into methanol that becomes plastics, paints and adhesives. It is one of the most important industrial chemicals on earth and almost nobody outside the industry thinks about it.

Under 1% of that is clean. In 2025 clean hydrogen production was on the order of a million tonnes; the rest came from natural gas and coal — roughly 290 billion cubic metres of gas a year for hydrogen alone. Producing today’s hydrogen accounts for about 2% of global CO₂ emissions.

Which reframes the whole question. The bottleneck was never finding customers. The customers have been there for decades. The question is whether the clean version can take that business away from the dirty version on price.

The Number That Decides Everything

A commercial electrolyzer eats 50–55 kWh of electricity per kilogram of hydrogen. The thermodynamic floor — the physical wall — is 39.4 kWh. You approach it from above and never pass it, which means the best case left on the table is roughly 20% better than today. No exponential curve is coming to rescue this.

It gets harsher. That kilogram carries only about 33 kWh as fuel. Turn it back into electricity through a fuel cell and you keep 30–45% of what you started with — call it 37% on a normal day. A lithium battery hands back 85–95%. Storing electricity as hydrogen is filling a bucket with a hole in it: pour in at the top and less than a quarter reaches the bottom.

That said, some jobs only this container can do — parking summer sunshine until winter, or shipping energy across an ocean. Those are real. They are just not most jobs.

Go one level deeper and you find the metric that actually binds: electricity is 60–70% of the cost of every green kilogram. Every $10/MWh move in power price shifts hydrogen by roughly $0.50/kg. So this was never a bet on a molecule. It is a bet on absurdly cheap, absurdly available electricity — and most of the planet does not have that.

Does It Pencil?

Like for like, US market, PEM, no subsidy: $4.33–6.05 per kilogram, midpoint around $5.25. Its opponent — grey hydrogen from steam methane reforming — costs $1.50–2.50. Same molecule, two to three times the price.

Take the full US credit of $3/kg and green lands at $2.36–4.08. Read that carefully: the best subsidized case only just steps into the top of grey’s unsubsidized range.

Could carbon pricing do it instead? Grey hydrogen emits 9–10 kg of CO₂ per kilogram produced. To close a $3 gap on carbon alone you need a price around $300 per tonne. No major carbon market is remotely close. The honest conclusion is uncomfortable and worth saying plainly: this industry does not currently run on economics. It runs on policy.

It Is a Bet on Your Grid

That $5 figure is the harsh case — PEM strapped to bouncy solar. Change the grid and you change the answer.

Solar is the trap that catches people. It is the cheapest electricity on paper, but a solar plant delivers only about a quarter of its nameplate over a year. Bolt an electrolyzer straight onto panels and your expensive machine idles three quarters of the time — and idle capital is exactly what wrecks $/kg. You have two ways out and both cost money: oversize the array and add a large battery to buy running hours, or pull grid power around the clock and accept both a higher price and a dirtier carbon footprint. The real constraint is not the cost of cheap solar. It is the availability of cheap solar, plus the battery bill.

Now put the same electrolyzer on firm hydro — Quebec, Manitoba, British Columbia, Washington State, Norway. Power is cheap and it does not stop. The machine runs near flat out, idle cost collapses, and because output is steady you do not need the fast-ramping premium machine at all: a cheaper alkaline system will do. Put those together and hydrogen lands near $3/kg rather than $5. Still above grey’s $2 — but now inside the range where today’s subsidy, or a modest carbon price, closes the gap.

Which is why green hydrogen has never been one price, and why the projects actually operating cluster where power is both cheap and firm.

The Ladder

The second answer is about use. The most useful framework here is Michael Liebreich’s hydrogen ladder, and the sorting question is simple: does this job actually require the molecule, or is there something cheaper and simpler already doing it?

At the top sit the uses with no substitute — ammonia, refining, methanol. The chemistry needs that atom; a battery cannot make fertilizer. In the middle, green steel: genuinely promising, but competing routes want the same prize. At the bottom, passenger cars, where batteries have already won, and home heating, where a heat pump delivers about six times the heat from the same kilowatt-hour. Liebreich’s own line is hard to improve on: heat pumps do multiplication, hydrogen does division.

The pattern is almost embarrassingly simple. Hydrogen wins where it is the ingredient — where you genuinely need the molecule to make something else. It loses where it is just a courier for electrons. And look at what sits at the top of that ladder: the 100 million tonnes we already make. The market was on the ladder the whole time; we kept staring at the bottom rungs.

Follow the Money

Capital has read the ladder. Announced clean-hydrogen capacity for 2030 now totals 37 Mt/yr, down from 49 Mt a year earlier. Of everything announced, only about 9% has reached final investment decision — the point where money actually moves.

The bottleneck is almost quaintly old-fashioned. Nobody signs the offtake. New agreements fell from 2.4 to 1.7 Mt/yr, close to a third. Meanwhile roughly 80% of the projects that are moving forward have a committed buyer behind them. Everyone wants green hydrogen to exist; almost nobody wants to sign a ten-year contract at $5/kg.

Hence 2025’s exits — around sixty large projects cancelled, a 26 GW Australian megaproject dropped, three US projects cut by one of the world’s largest hydrogen producers. That is not market failure. That is capital repricing hype at cost.

Governments show their real thinking in how they pay. The US writes the biggest cheque — up to $3/kg on a four-tier carbon-intensity ladder, with the top tier demanding power that is new, nearby and hourly-matched — but it shuts to projects not under construction by the end of 2027. Canada runs the same logic through a refundable investment tax credit of 40/25/15% by carbon intensity, plus another 15% for ammonia-conversion equipment. Both pay for cleanliness, not for a colour name. Europe does the most honest thing of all: it auctions the subsidy and lets producers bid the minimum they need, which surfaced clearing prices under €0.50/kg in the first round and around €0.60 in the second — and quietly revealed that the best projects can nearly stand up while average ones cannot.

China took the most predictable path and simply won the machines: roughly 60% of global electrolyzer manufacturing capacity, about two-thirds of installations, alkaline systems at $600–1,200/kW against roughly $2,000 in the West. The manufacturing war is over. But winning the factory is not the same as winning the molecule — the world’s largest green hydrogen plant has run at roughly 20% of nameplate, because when renewable output dips you throttle back for safety. That is the utilization trap, and it is the same trap as the solar one. Notably, China’s next wave — including a multi-billion-dollar wind-to-hydrogen-to-ammonia complex — aims squarely at the top of the ladder.

The Verdict

Maturity 6/10 — the machines are commercial and for sale, but gigawatt-scale reliable operation is unproven. Economics 2/10 — two to three times the incumbent, with two exits only: ~$300/tonne carbon, or a subsidy with an expiry date. Scale-up 3/10 — enormous ceiling, but 9% FID is 9%. Moat 4/10 — serious players and serious capital, yet equipment is commoditizing and long-term contracts go unsigned. Timing 3/10 — the shakeout is not over and the biggest subsidy is counting down.

So, two stamps. As the fuel of everything — driving, heating, stuffing the grid into a tank — overhyped: physics takes a third off the top before you start, every bottom-rung use already has a cheaper winner, and capital has noticed. As the cleanup of the ~100 Mt of dirty hydrogen we make every year — ammonia, refining and methanol as the base, green steel next in line — a real bet: the demand already exists, the big subsidies pay by carbon intensity and point straight down this road, and the projects that survive cluster where power is cheap and steady.

Green hydrogen is not the fuel of the future. It is the cleanup of the present — and honestly, the cleanup is the better business.

Green hydrogen is not the fuel of the future. It is the cleanup of the present.