In a take-or-pay hydrogen offtake contract, the number that decides feasibility is not the buyer's headline volume but the minimum it must pay for whether or not it takes delivery. In the worked 100 MW example below, a project only covers its capital cost in a downside year if buyers are committed to pay for about 70% of the plant's output at a $7/kg contract price.
Put simply: contracted volume tells you how much hydrogen you might sell; the take-or-pay minimum tells you how much revenue you can count on. A 100% contracted plant with a 60% minimum can have a weaker revenue floor than a 75% contracted plant with a 90% minimum. Developers, buyers and lenders should all run that arithmetic before agreeing terms.
This is a spoke article in our offtake cluster. For the full list of contract terms to check before final investment decision (FID), start with the hydrogen offtake agreement checklist.
The real question: how much revenue survives a bad year?
A buyer committing to "5,000 tonnes a year" can mean very different things. It may be an option to buy up to 5,000 tonnes, a firm obligation to take every tonne, or something in between: an obligation to take, or pay for, a minimum share of the nominated volume.
Firm commitments are still scarce. The IEA's Global Hydrogen Review 2026 reports that "New offtake agreements for low-emissions hydrogen reached 1.7 Mtpa in 2025, as in 2024," and that "One-fifth of all new agreements were firm offtakes" (IEA, Global Hydrogen Review 2026 — Demand). Most signed volume, in other words, is not yet a commitment to pay.
That is why take-or-pay terms matter. Law firm Norton Rose Fulbright (practitioner commentary, not a primary source) describes it plainly: "With a take-or-pay obligation, the buyer is required either to take the hydrogen and pay for it or just pay for it," which "ensures that a project can expect a minimum level of revenue" (Norton Rose Fulbright, 2023). The same article notes that "Financiers are likely to amortize the debt based on the length of the agreement."
Buyers push back for understandable reasons. A 2025 note from UK law firm MFMac (also practitioner commentary) reports suppliers "insisting on including 'take or pay' obligations on offtakers," while "Offtakers are likely to resist such strict obligations from the outset," and suggests that "A balance might be struck with fluctuating take or pay arrangements or production thresholds" (MFMac, 2025).
Take-or-pay hydrogen offtake: the three numbers that set the revenue floor
Before modelling anything, pin down three inputs from the term sheet:
- Contracted share: the buyer's annual contract quantity as a share of the plant's expected annual output.
- Take-or-pay minimum: the share of that contract quantity the buyer must pay for even if it does not take it.
- Contract price and its boundary: $/kg at the plant gate or delivered, fixed or indexed.
Multiply the first two and you get the share of plant output that is paid for in a downside year. Everything above that line depends on the buyer actually wanting the hydrogen, or on finding another buyer. Norton Rose Fulbright's article states that "There is no merchant market for hydrogen" in the way there is for power or gas, so a conservative screening case should assume uncontracted volume earns nothing.
Worked example: a 100 MW plant under six offtake structures
All inputs below are illustrative assumptions, not quotes, project data or H2Hub outputs.
Stated assumptions:
- Electrolyzer: 100 MW, installed cost $2,200/kW ($220 million). This sits inside the IEA's 2025 global-average installed range of $1,920–2,500/kWe, which includes the electrolyser system, electrical equipment, gas treatment and EPC (IEA GHR 2026 Assumptions Annex).
- Fixed operating cost: 3% of capital cost a year, the IEA annex assumption ($6.6 million a year).
- Electricity use: 55 kWh/kg, DOE's 2022 system status value for PEM electrolysis (DOE).
- Power price: $45/MWh, paid only on electricity used ($2.48 per kg produced).
- Full-load hours: 5,000 a year, giving about 9,091 tonnes of hydrogen a year.
- Finance: 8% discount rate over 20 years, capital recovery factor 0.10185, or $22.41 million a year.
- Contract price: $7.00/kg at the plant gate.
- Downside year: buyers take and pay for only their take-or-pay minimum, and uncontracted output finds no buyer.
On these inputs the plant's levelized cost of hydrogen is $5.67/kg ($2.46 capital, $0.73 fixed operating cost, $2.48 power). If the LCOH build-up is new to you, see what LCOH is and how it is calculated.
The test we apply is capital-recovery cover: downside-year revenue, minus the power to produce the minimum volume, minus fixed operating cost, divided by the $22.41 million annual capital recovery. Above 1.0, the floor revenue repays the capital at 8%. Below 1.0, it does not.
Results
| Contracted share | Take-or-pay minimum | Paid volume (t/yr) | Floor revenue ($m/yr) | Cash after power and fixed O&M ($m/yr) | Capital-recovery cover |
|---|---|---|---|---|---|
| 100% | 60% | 5,455 | 38.2 | 18.1 | 0.81 |
| 100% | 80% | 7,273 | 50.9 | 26.3 | 1.17 |
| 100% | 90% | 8,182 | 57.3 | 30.4 | 1.36 |
| 75% | 80% | 5,455 | 38.2 | 18.1 | 0.81 |
| 75% | 90% | 6,136 | 43.0 | 21.2 | 0.94 |
| 50% | 90% | 4,091 | 28.6 | 11.9 | 0.53 |
With no take-or-pay minimum at all, the downside year earns nothing and the plant still pays $6.6 million of fixed operating cost, a cover of −0.29.
What the numbers say
- Volume and minimum trade one for one. 100% contracted at a 60% minimum and 75% contracted at an 80% minimum both leave 5,455 tonnes paid for, and the same 0.81 cover. A buyer that cannot commit to more volume can still make a project work by accepting a higher minimum, and vice versa.
- The break-even line is about 70% of output. At $7.00/kg, the plant needs paid volume of 70.5% of expected output to cover capital recovery and fixed operating cost in a downside year. That is a useful single number to take into a negotiation.
- Price can substitute for volume, at a cost. At 75% contracted and an 80% minimum, the contract price would need to be $7.79/kg for the floor to reach 1.0 cover. With 100% contracted at an 80% minimum, $6.46/kg would be enough.
- Half-contracted plants struggle even in a good year. At 50% contracted, even if the buyer takes every contracted tonne, cover is only 0.62. The uncontracted half has to find a buyer for the project to work.
Two details favour the seller slightly. If a buyer pays for hydrogen it does not take, the plant can avoid producing it and save about $2.48/kg of power, so our floor (which assumes the buyer takes the minimum) is conservative. And a real project would also earn something on volume above the minimum in most years.
Capacity matters too. Running fewer hours lowers output and raises the cost per kg; see how electrolyzer utilization changes the cost of hydrogen. To find which other input could break the same project, the LCOH sensitivity analysis walks through seven of them.
What the example leaves out
This is a single-year screen, not a lender's model. It ignores:
- Debt structure. A lender's debt service coverage ratio uses actual debt payments and taxes, not a capital recovery factor on the whole investment.
- Supply-side obligations. Take-or-pay usually comes with a seller's delivery obligation. A plant on variable renewable power may need storage or firming to meet a daily nomination, which adds cost.
- Shortfall and force majeure. Exceptions for plant outages or buyer shutdowns can reduce the effective minimum below its headline value.
- Credit. A take-or-pay minimum is only as good as the buyer's ability to pay it. Norton Rose Fulbright notes that "Financiers will require a creditworthy buyer or a buyer with a creditworthy parent."
- Price indexation, carry-forward and make-up rights, which shift money between years.
Five questions to ask before you sign
- What share of expected output is paid for in a bad year? Multiply contracted share by the minimum and compare it with your own break-even line.
- Is the minimum measured annually, quarterly or monthly? Shorter periods protect the seller more but are harder for the buyer to accept.
- Can the buyer carry unused paid volume forward? Make-up rights reduce the buyer's cost but delay the seller's revenue.
- What can the seller do with volume the buyer declines? A right to resell it, and a realistic buyer to sell it to, changes the downside case.
- Does the contract term cover the debt term? If financing is sized on the contract length, a short tenor shrinks the debt a project can raise.
Try it in H2Hub
Start a free H2Hub project with the same inputs as the example: a 100 MW electrolyzer, $2,200/kW installed, 55 kWh/kg, power at $45/MWh and about 5,000 full-load hours. Compare your project's cost per kg with the contract price you are negotiating, then change one number, such as hours or power price, to see how much headroom the take-or-pay floor needs. You can start without an account or a card; your first full project and its PDF report are free, and downloading that PDF needs a free sign-in. Re-running with new inputs needs a subscription.
Subscribers can go further in the Offtake Lab and the Project Finance Lab, which are subscriber tools.
Pre-feasibility caveat
This is general screening guidance, not legal, financial or investment advice. Take-or-pay terms, their enforceability and their accounting treatment vary by jurisdiction and contract. Have counsel and lenders review actual documents before relying on any revenue floor.
Frequently asked questions
What is a take-or-pay clause in a hydrogen offtake agreement?
It obliges the buyer to pay for a minimum quantity of hydrogen in each period whether or not it takes delivery. It gives the producer a predictable revenue floor, which lenders look for before financing a project.
What take-or-pay percentage is typical for hydrogen?
There is no published market standard, and the practitioner sources we checked give no figures. The worked example shows how to find the minimum your own project needs: in that case, buyers had to pay for about 70% of expected output.
Why do buyers resist take-or-pay terms?
Buyers carry the risk that their own plant needs less hydrogen than planned, or that cheaper supply appears later. Practitioners report offtakers resisting strict obligations and suggest flexible thresholds as a compromise.
Is a take-or-pay contract enough to finance a hydrogen project?
Not on its own. Lenders also look at the buyer's credit, the contract term, exceptions such as force majeure, and whether the plant can reliably deliver the nominated volume.
Primary sources and further reading
- International Energy Agency — Global Hydrogen Review 2026: Demand. 2025 offtake volumes and the firm share. Checked 12 November 2026.
- International Energy Agency — Global Hydrogen Review 2026: Assumptions Annex. Installed electrolysis capital cost and fixed operating cost.
- U.S. Department of Energy — Technical Targets for Proton Exchange Membrane Electrolysis, modified 29 May 2026. System electricity use.
- Norton Rose Fulbright (law firm commentary) — Key issues for hydrogen developers, James M. Berger, 23 June 2023.
- MFMac (law firm commentary) — Hydrogen offtake arrangements: are they taking off?, Jennifer Kerr, 24 June 2025.
Editorial verification: 12 November 2026. Worked examples are calculations from stated assumptions, not forecasts, market prices or H2Hub-generated results.
