The African Development Bank is putting $20 million of project preparation support behind four green hydrogen projects representing $23 billion of prospective investment. The ratio is the story, and so is one word in the instrument: reimbursable.
The African Development Bank has selected four green hydrogen projects in Egypt, Morocco, Namibia and South Africa for $20 million of reimbursable project development support, subject to board approval. The four represent roughly $23 billion of prospective investment, around 20 gigawatts of renewable generation and 7 gigawatts of electrolysers, with applications including aviation and maritime fuels and low-carbon iron.
Alongside it, a Team Europe package commits more than €57.8 million to South African green hydrogen, critical minerals, processed metals and battery value chains, including a €20.5 million blended finance facility explicitly aimed at making hydrogen projects bankable, together with support for regulation, research, skills and domestic processing.
The ratio in the first package is roughly one to a thousand, and it points at something the sector has been slow to say plainly. Africa does not have a shortage of proposed green hydrogen projects. It has a shortage of projects capable of surviving the distance between an announcement and a final investment decision.
The work between a memorandum of understanding and a financeable project is specific, expensive and almost entirely invisible.
Resource measurement to bankable standard, which for wind means a met mast and two years of data rather than a satellite reanalysis. Front-end engineering design detailed enough for a lender's technical adviser to price. Environmental and social impact assessment to the standard a development finance institution will accept rather than the standard the host regulator requires. Grid and water studies. Land assembly and the legal work to make title clean. Permitting across several agencies. A term sheet with an offtaker. Legal, tax and structuring work across jurisdictions.
That package typically costs between two and five per cent of a project's capital cost, which for a billion-dollar facility is a figure in the tens of millions, spent entirely before anything is certain and with a meaningful probability of being written off.
Which produces a specific financing gap. Equity investors want to enter after the main uncertainties are resolved. Lenders cannot enter before. Developers with balance sheets can fund it themselves and are mostly the large international players. A domestic or first-time developer with a genuinely good site cannot, and the project does not fail on its merits so much as on the absence of anybody willing to pay for finding out whether it has any.
That is why preparation capital is the highest-leverage money in the chain, and it is also why there is so little of it. A funder writing a cheque for a feasibility study has no asset, no visible output and a strong chance of nothing at all.
The AfDB instrument is described as reimbursable, and that single term changes its character considerably.
A reimbursable preparation facility is repaid, typically out of the project's financing, if the project reaches a final investment decision. If the project does not proceed, the money is not recovered. The funder is therefore taking a loss distribution shaped like equity at the riskiest stage of the project's life, while the upside is capped at repayment of principal and perhaps a fee.
That structure has two consequences worth naming.
It is better aligned than a grant. A developer who must repay has an incentive to spend the preparation budget on things that move the project rather than on consultants' reports that satisfy a reporting requirement. The repayment obligation is a discipline a grant does not impose.
And it introduces a selection problem. A facility that is repaid only on success will, if it is managed sensibly, be steered towards projects most likely to reach a decision. Those are the projects with strong sponsors, mature sites, existing offtake conversations and experienced teams, which is to say the projects least dependent on the facility existing.
The Team Europe commitment is structurally different and worth reading alongside, because it funds the things around the project rather than the project itself: regulation, research, skills, investment facilitation and domestic processing capability.
That matters because preparation money assumes a system capable of absorbing it. A feasibility study requires a regulator that can issue a permit, a grid operator that can produce a connection study, a cadastre that can confirm title and a pool of engineers who can review the work. Where those are missing, preparation capital funds a study that cannot be acted on, which is the capability constraint arriving one layer down.
The South African package is explicit that hydrogen should feed fertiliser, green iron and steel, fuels and equipment manufacturing rather than leave as a commodity, which is the right ambition and the one that makes the preparation question harder rather than easier. An export ammonia project needs an offtaker. An industrial hydrogen project needs an offtaker, a co-located industrial user, a port, a tariff and a skills pipeline, and each of those is a separate preparation workstream.
Conversion rate, and the reason for each failure. Of projects receiving preparation support, how many reach a final investment decision, and for those that do not, which specific workstream stopped them. Resource, offtake, permitting, grid, land or sponsor capacity. That register would be the most useful document in African project finance and no facility publishes one.
Cost of preparation as a share of capital cost, by project and by country. If it varies widely, the variation locates where the institutional friction sits.
Who the sponsors are. Whether preparation facilities are reaching domestic and first-time developers or subsidising the preparation budgets of international firms that could fund it themselves. This is the additionality test in its practical form.
And time. Months from preparation funding to financial close, tracked against a comparison group that did not receive support. Faster is the claim, and it is checkable.
The ratio of twenty million to twenty-three billion will be used in communications material, and it deserves to be. It is also a prediction rather than a result, and the thing worth watching is not whether the four projects are good but how many of them are still alive in three years, and whether the facility was brave enough to lose money on the ones that were not.
The Lab works on this in financial inclusion and green industrialisation and local manufacturing, and on the distance between a commitment and an operating asset through measuring change.
If you are running a preparation facility and want conversion, additionality and failure reasons measured rather than assumed, tell us what you need to know.
If this reads close to a decision you are weighing, we would like to hear about it. A short note is enough to start.
The Lab is an independent research team, not an advocacy organisation. We do not campaign, endorse products, or accept commissions where a finding is expected in advance. Where a partner's own theory of change is what needs testing, that is a legitimate engagement, and the Lab tests it honestly.