Insight, Industrial Policy

The Mandate Is the Mine

A R47 billion synthetic aviation fuel project in South Africa's Northern Cape is being built against demand that does not exist yet and cannot exist without a European quota. The resource underneath this project is not sunlight. It is a paragraph in EU legislation.

Line-art sequence: a solar array and electrolyser in an arid landscape on the left, a tank of synthetic fuel and a cargo ship in the middle, and on the right, in place of a market or a refinery, an open legal statute book with a percentage figure on the page, drawn at the same scale as the industrial objects.
Every project has a resource at one end. This one has a quota.

What a Northern Cape fuel plant is actually built on

Engineering details published at the end of August show Phelan Green Hydrogen's Northern Cape project moving toward construction, combining renewable electricity, captured carbon dioxide and green hydrogen to manufacture electro-sustainable aviation fuel. The wider project is estimated at around 47 billion rand (as reported), construction is expected to begin in the fourth quarter of 2026, and technology is being supplied by Johnson Matthey, Honeywell UOP and Sungrow Hydrogen.

It sits alongside a broader shift in South African positioning. The SA-H2 Fund reached a first close at roughly 3 billion rand, around 182 million dollars (as reported), designed to invest not only in hydrogen production but in green ammonia, methanol, fertiliser, steel and other hard-to-abate industries, and the September Africa Green Hydrogen Summit is being framed around implementation, industrial demand and regional value chains rather than project announcements.

That shift is right, and we have argued for it: hydrogen is worth more as an industrial feedstock than as an export commodity, because the value sits in what you make with it.

This piece is about something else. It is about where the demand for electro-sustainable aviation fuel comes from, because the answer determines what kind of asset this is.


There is no market for this fuel

Synthetic aviation fuel is not a product that airlines buy because they prefer it. The International Council on Clean Transportation estimated the production cost of e-kerosene in the EU at roughly ten times that of fossil kerosene in 2020, narrowing to about two and a half times by 2050 as the technology matures.

Nobody buys a fuel at several times the price of the identical-performing alternative unless they are required to.

Stepped line chart, 'The synthetic fuel sub-mandate, share of fuel supplied at EU airports'. Cobalt steps rising across a horizontal axis labelled 2030, 2032, 2035, 2040, 2045, 2050 with values printed beside each step: 0.7 per cent, 1.2 per cent, 5 per cent, 10 per cent, 15 per cent, 35 per cent. Vertical axis 0 to 40 per cent. A coral vertical marker at 2030 annotated 'the year this plant needs a buyer'. Footnote: ReFuelEU Aviation, Regulation (EU) 2023/2405. Schedule as published by EASA.
The demand curve for this plant is a schedule of numbers in a regulation, not a price signal from an airline.

They are required to. Under ReFuelEU Aviation, fuel supplied at EU airports must meet a rising sustainable aviation fuel share, and within that a specific sub-mandate for synthetic fuels: 0.7 per cent from 2030, 1.2 per cent from 2032, 5 per cent from 2035, 10 per cent by 2040 and 35 per cent by 2050. The European Commission estimates that meeting the 2035 targets will require 6.8 million tonnes of aviation and maritime e-fuels.

So the demand curve for this plant's product is drawn by a regulation. Remove the sub-mandate and the product has no buyer at any volume that would justify 47 billion rand.

That is not a criticism. It is a description of the asset. Policy-created markets are legitimate and often necessary, because a technology with a large public benefit and no private premium will not otherwise be built. But a project financed against a regulatory quota is holding a different risk from a project financed against consumer demand, and it should be underwritten as such.

The useful way to put it: in a conventional resource project, the thing underneath the asset is geological. Here, the thing underneath the asset is a paragraph in another jurisdiction's statute book, and paragraphs can be amended.

Line-art diagram split by a hairline vertical rule. Left half: a conventional mine headframe drawn above layered rock strata in forest and butter, labelled beneath 'the resource is geological'. Right half: an electrolyser and a hydrogen storage tank drawn above the open pages of a legal statute book with a single highlighted paragraph in coral, labelled beneath 'the resource is legislative'. Both structures drawn at exactly the same size so the comparison is visual.
Same scale, different substrate. One can be surveyed. The other can be amended.

How solid is the paragraph

This is an empirical question with an uncomfortable partial answer, and it is answerable now rather than in 2032.

On the positive side, the architecture is unusually well built. ReFuelEU is a regulation rather than a directive, so it applies directly rather than through national transposition. The trajectory runs to 2050. Member States were required to set non-compliance penalties, and where they have, the numbers are serious: Germany has legislated penalties of 4,700 euros per tonne for missing sustainable aviation fuel and 17,000 euros per tonne for missing e-kerosene volumes, and has raised its own 2030 synthetic fuel quota above the EU floor. A financial commitment has been assembled around it, with the Commission expecting to mobilise 2.9 billion euros through 2027, including 153 million for synthetic aviation fuels through the Innovation Fund and 300 million via the European Hydrogen Bank.

On the negative side, the enforcement layer was late and incomplete. Member States were required to communicate their penalty regimes by the end of 2024. In June 2026 the Commission launched infringement proceedings against thirteen Member States for failing to do so.

Sit with that. Thirteen Member States, eighteen months after the deadline, had not told the Commission what the penalty for non-compliance would be. A quota without a penalty is a target. The mechanism that converts the regulation into a purchase order was, across half the Union, not in place.

This is the part a project sponsor should be pricing, and it is not the headline risk. The headline risk is that the mandate is repealed, which is unlikely. The real risk is softer and far more common: a review clause exercised, a compliance deadline extended, a flexibility mechanism introduced, penalties set low enough to make paying them cheaper than buying the fuel, or a shortfall permitted to be made up in a later period. Each of those leaves the regulation formally intact and destroys the price.

There is precedent for exactly this in European renewable support. Several Member States revised feed-in tariff arrangements after investment had been committed, in ways investors regarded as retroactive, and the resulting arbitration cases ran for a decade. The lesson those disputes taught was not that European regulation is unreliable. It was that the durability of a support scheme is a function of what it costs the government to keep it, and that cost rises when the technology stays expensive.

Line-art diagram: a sturdy upright pillar labelled at its base 'the mandate, formally intact'. Around the pillar, five small wedges are being inserted at different heights in orange, blue, yellow, sky and forest, each labelled: 'review clause exercised', 'deadline extended', 'flexibility mechanism added', 'penalty set below the fuel price', 'shortfall carried to a later period'. The pillar remains upright and undamaged. To one side, a small figure with a clipboard looks at the pillar and not at the wedges.
The mandate does not have to fall for the price to. Each wedge is compatible with the regulation staying on the books.

What follows for the South African project

Four things, and they are structural rather than a judgement on the project.

The offtake contract is the asset, not the plant.

A binding, long-dated purchase agreement with a fuel supplier obligated under ReFuelEU converts regulatory risk into counterparty risk, which is a risk a lender knows how to price. Without it, the project is holding the regulatory risk directly. The Commission's own coalition has noted that investment decisions and construction must begin by 2027 to avoid missing 2030 targets, which is leverage for a developer that is available now and will not be available later.

The carbon source determines the compliance life of the product.

Electro-sustainable aviation fuel requires carbon dioxide, and where it comes from matters under EU sustainability criteria. Carbon captured from an industrial or fossil source has a different and more time-limited status than carbon captured from air or from biogenic sources. In a country with South Africa's generation mix this is not a detail, and any assessment of the project should establish it early.

Value capture depends on what stays in the country.

The technology is supplied by three international licensors. The fuel is destined for a European compliance market. The question this leaves is what South Africa retains: engineering, operations, maintenance, catalyst handling, chemical inputs, and the capability to build the second plant with less external help. That is the local integration question that determines whether this is industrial development or a well-located conversion facility, and it is decided in procurement and training decisions taken now.

Domestic demand is the hedge.

The strategic case for South African hydrogen is stronger when the molecules anchor domestic fertiliser, steel and chemicals than when they leave as fuel. The SA-H2 Fund's mandate across ammonia, methanol, fertiliser and steel is the right shape for exactly this reason. Domestic industrial demand is not a consolation prize if the export market wobbles. It is the thing that makes the export market survivable.


The general point, which is not about hydrogen

A growing share of African green industrial investment is being underwritten against demand created by European regulation. Carbon border adjustment, deforestation-free supply chain rules, battery regulation, sustainable aviation fuel quotas and corporate reporting requirements all create markets for products that would not otherwise have buyers.

This is a real opportunity and it is a real dependency, and the second half is rarely said out loud. A country building industrial capacity against another jurisdiction's compliance schedule has acquired an exposure to a political process in which it has no vote. We described the same asymmetry from a different direction in European capital and African growth, and it applies with more force here, because the commitment is a physical asset with a thirty-year life.

The mitigations are not exotic. Long-dated offtake contracts with obligated parties. Diversification across compliance regimes rather than dependence on one. Domestic and regional demand developed in parallel rather than after. And an honest reckoning, at the point of financial close, of what the plant is worth if the quota moves. The same pattern of a long-lived asset built for a single counterparty is the argument of The Smelter Contract.

None of that requires pessimism about European climate regulation. It requires treating a legislative trajectory as what it is: the most valuable and the least controllable input in the whole project.

The Lab works on this in local manufacturing and energy access, and on what a large industrial arrival means for the region around it through entering a new context.

If you are financing an industrial project whose demand comes from somebody else's statute book, tell us what you need to know.


Sources


This is an independent insight piece by Transitions Lab. For the Lab's applied work, see Local Manufacturing & Supply Chains. See also The Value Is in the Second Step on why processing rather than export is where the value sits, Capability Is the Slow Part on what determines whether a plant leaves capability behind, The Smelter Contract on long-lived assets built for a single counterparty, and Strategic Is Not the Same as Financeable on why designation and subsidy do not move a final investment decision when the binding constraint is variance rather than expected return. To discuss a study, see Contact.

Read more Articles & insights See all articles → See it in the field Case studies See all case studies →