Beyond China: How to Build a Resilient SAF Feedstock Supply Chain
Almost all of the Sustainable Airline Fuel (SAF) sold today is made from used cooking oil. That oil is becoming harder to buy, and China is the main reason. SAF production paths differ in how exposed they are to feedstock costs, and in how much of the supply chain the producer controls. The part left uncontrolled now runs through China.
On 6 August American Airlines flew paying passengers from Corpus Christi to Dallas Fort Worth on jet fuel made from waste carbon dioxide and renewable electricity. It was the first fuel of that kind delivered into a US commercial airport. The day before, Caixin reported sixteen new Chinese sustainable aviation fuel projects in the first half of 2026. The rest of the world announced four.
Both stories are about SAF feedstock, which is the raw material a fuel is made from. The majority of SAF cost is determined by its raw material costs. The processing step is comparatively cheap, and the market is far too small to set a price of its own.
Nearly all SAF comes from one raw material
The dominant production route is called HEFA, short for hydroprocessed esters and fatty acids. A refinery takes waste fats and oils, adds hydrogen, and turns them into a fuel that behaves like ordinary kerosene. Used cooking oil is the cheapest suitable input, and it supplies around 80% of the SAF feedstock used in commercial production today.
Everything else is small. Global SAF production reached 1.9 million tonnes in 2025, or 0.6% of the jet fuel the world burns. HEFA accounts for most of that volume, so the price of used cooking oil largely determines the price of SAF (70-85% of production cost).
What the 10% price rise actually measures
BNEFs latest survey put the average SAF price at $1,817 per tonne. That is about 10% higher than a year earlier, and five times the price of ordinary jet fuel. The figure has been widely quoted as evidence that the raw material has run short. BNEF gives two other reasons. Ordinary jet fuel became more expensive after the Strait of Hormuz closed, and SAF sells at a premium over it, so the premium rose too. Demand also grew where blending rules oblige suppliers to sell a set percentage of SAF.
The year-on-year move therefore reflects the oil market more than anything. Northwest Europe is the largest marketplace for SAF. The fuel averaged $2,830 per tonne there in the second quarter of 2026. That is up 31% since the Iran war began, against ordinary jet fuel at $1,404. BNEF expects the European SAF average to stay near $2,746 into early 2027.
The SAF feedstock shortage is real, and it works on a longer horizon than the oil market does. It sets the floor under the price, and the oil market moves the price around above that floor. The evidence for it appeared at the same time from China.
China is taking the SAF feedstock Europe depends on
In late 2024 Beijing removed a 13% export tax rebate on used cooking oil. Exports became less profitable, and more oil stayed at home. Chinese refiners then started building. Sixteen new SAF projects were announced in the first half of 2026. State-owned companies took close to half of the new capacity. By July 2026 Sinopec had completed a takeover of China National Aviation Fuel Group, so one company now controls the chain from refinery to airport fuel tank.
Where that fuel goes makes this a European problem. About 99% of Chinese SAF is exported, and European blending rules are the reason there is a buyer for it. China has no blending rule of its own, so its refiners have had no domestic use for the fuel. Sixteen new plants give them one. Caixin’s reporting points to a worldwide shortage of used cooking oil by 2028, well before the loans on a European plant built today would be repaid.
What the European rules do and do not restrict
European project documents often claim that the law will eventually force refiners off used cooking oil. That rests on a misreading of two separate laws. The EU keeps a list of approved raw materials in Annex IX of its renewable energy directive. Part A covers advanced materials such as straw, manure and non-food plant matter. Part B covers used cooking oil and animal fats. RED III, the current version of the directive, limits Part B materials to 1.7% of the energy used in road and rail transport. Aviation falls outside that limit.
The aviation rules work differently. ReFuelEU Aviation obliges fuel suppliers to blend a rising share of SAF at EU airports, starting at 2% in 2025 and reaching 6% in 2030. It limits aviation biofuels made from food crops to 3% of a supplier’s volume. Part A and Part B materials face no such ceiling. The pressure on used cooking oil in aviation is commercial. It comes from price and availability, on a timetable no legislator set.
Two routes that escape the raw material problem
The American Airlines flight used e-SAF, made from captured carbon dioxide and hydrogen produced with renewable electricity. It uses no biological raw material, so it sidesteps the SAF feedstock market completely. It went into the airport’s shared tanks with nothing modified on the aircraft or the ground. The flight showed that it works. It left open the questions of volume and cost. Infinium’s next plant should make over 5 million gallons a year at full output. The world burns roughly 100 billion gallons of jet fuel a year. On cost it is the most expensive of the three routes, because making hydrogen from renewable electricity and capturing carbon costs more than buying waste oil or growing biomass.
The second route puts the SAF feedstock under the producer’s own control. That is the model behind the SAF project MVAventures is developing in Brazil. A developer can plant and harvest the biomass its refinery will run on. Grassy energy crops with a low starch content fall inside Part A of Annex IX, even when they are grown as a main crop. Miscanthus, switchgrass, giant cane and sorghum all qualify. They face no ceiling under the aviation rules, and they count toward the same 6% blending obligation as used cooking oil. A plant with its own planted acreage sets its input cost through its own agronomy. A HEFA refinery buying on the open market has no comparable certainty.
The biomass route
That certainty has a cost, in the conversion step. Turning plant matter into fuel is dearer than processing waste oil, because the cellulose has to be broken down first. Production costs have been modelled across all three routes. The biomass route carries the higher conversion cost but the smaller exposure to feedstock price.
The supply side moves faster than it once did. Sorghum is an annual grass, and field trials in North Carolina put the average season from planting to harvest at 140 days, with the crop taken in a single cut. The land and the refinery can therefore be financed on the same timetable.
Europe has started building. Axens announced its ERA project in July, France’s first industrial-scale SAF unit running on advanced bioethanol from European biomass. Design capacity is 50,000 tonnes a year, with start-up expected in 2030. No site has been chosen yet, and the financing partners are still being assembled.
However, the market has caught up with only one of these routes. Every index Argus launched on 3 August for aviation emissions costs is built on HEFA, because that is where enough deals happen to see a price.
What this means for capital
If a European HEFA project reaches your desk, the raw material contract is the document to read first. Price forecasts are of limited use, because the price will keep moving with the oil market for reasons unconnected to the project. Ask how many tonnes of SAF feedstock are contracted, for how many years, and from which country. Then ask what happens to that contract when Chinese refiners want the same material. A plant buying used cooking oil on the open market is exposed to a decision taken in Beijing, and it earns nothing for carrying that exposure.
A project growing its own biomass needs different questions. Ask what acreage is owned or under long contract, and what the yield assumptions are. Check that the crop qualifies under Part A. The risk in this model sits in agronomy and execution, which a developer can manage directly.
For an e-SAF project the commercial case rests on one rule. ReFuelEU obliges suppliers to include a small share of synthetic fuel from 2030, set at 1.2% averaged over 2030 and 2031. That obligation is the whole market, so the sale contract has to be priced against it.
The plants still running at the end of this decade will be the ones that secured their SAF feedstock early, by contract or by planting it.
If you work in aviation, shipping, project finance, or the energy transition and want to understand what investment-grade green fuel project development looks like in practice, contact us .
