Insight, Finance & Minerals
Twenty-three of the EU's sixty strategic critical mineral projects have told Brussels they may not reach a final investment decision. The instinct is to conclude they need more money. The economics say something more uncomfortable, which is that they may need less uncertainty.
Twenty-three of the sixty projects the European Union has designated as strategic critical mineral projects have warned the Commission that liquidity and financing constraints could prevent them reaching a final investment decision (as reported). A framework intended to mobilise around 1.7 billion euros exists, developers argue capital is arriving too slowly, and some projects are already suspended.
The immediate reading is that the sums are too small and the disbursement too slow, and both may well be true. But it is worth being precise about what a final investment decision actually is, because the standard remedies are aimed at a variable that may not be the binding one.
The textbook rule is that a firm should invest when the net present value of a project is positive. Avinash Dixit and Robert Pindyck's Investment under Uncertainty set out why that rule systematically fails to describe what firms actually do, and the explanation has three components.
Most investment is irreversible. The money spent on a processing plant cannot be recovered if the decision turns out badly. The environment is uncertain, and it stays uncertain: commodity prices, offtake demand, policy and permitting all move. And crucially, the firm can wait.
Those three together mean an investment opportunity behaves like a financial option. Holding it has value, because waiting produces information, and exercising it destroys that option. So the correct decision rule is not to invest when net present value turns positive. It is to invest when net present value exceeds the value of continuing to wait, which is a higher bar and sometimes a much higher one.
The approach recognises the option value of waiting for better, though never complete, information, and it explains a behaviour that puzzles policymakers constantly: firms applying hurdle rates far above their cost of capital, and investing only in projects that are deep in the money.
Read the twenty-three warnings through this and they say something different from what they appear to say.
A project that will not reach a final investment decision is not necessarily a project that cannot be financed. It may be a project whose owners have concluded that waiting is worth more than building.
If the binding constraint is the option value of waiting, then interventions divide sharply into two kinds, and only one of them works.
Instruments that improve expected returns. Grants, concessional debt, equity co-investment, capital allowances. These raise net present value. They also, in a volatile environment, raise the value of the option to wait, because a larger prize is more worth being patient for. Their effect on the timing of a decision is therefore much weaker than their effect on the arithmetic, and a developer can accept a grant and still not build.
Instruments that reduce uncertainty. Offtake guarantees, price floors, contracts for difference, binding permitting timetables, and firm public procurement commitments. These narrow the distribution of outcomes. Narrowing the distribution reduces the value of waiting directly, because there is less information to be gained by delaying.
That is the practical implication and it is counterintuitive: for a project stuck before a final investment decision, a smaller instrument that removes uncertainty will often move the decision when a larger one that improves returns will not.
It also explains why critical minerals are particularly badly served by the standard toolkit. Mineral prices are exceptionally volatile, and lithium's recent history is the obvious example. Volatility is the input that makes waiting valuable. A subsidy calibrated against an average price does almost nothing to the variance, and it is the variance that is holding the decision.
The technology readiness scale describes a device and says nothing about whether anybody will build it. That gap is exactly what these twenty-three warnings expose, and it recurs across the series: a technically proven system that no institution can absorb, no market will pay for, or no committee will approve.
A fuller readiness picture for capital-intensive transition infrastructure would need at least four dimensions, and only the first is currently assessed.
Technical. Does it work at scale. Well covered.
Regulatory. Is there a permit, on a timetable somebody can rely on. Partially covered, and increasingly the subject of political argument in Europe.
Market. Is there a buyer, at a price, under a contract. Barely covered, and this is where most of these projects sit.
Financial. Is the distribution of outcomes narrow enough that a committee will exercise the option rather than hold it. Not assessed anywhere.
The fourth is not the same as the third. A project can have a buyer and still fail to reach a decision if the price in that contract floats with a volatile index, because the contract transfers volume risk and leaves price risk where it was.
This is the argument for treating readiness as a property of the setting rather than the technology, extended into the financial layer. It is measurable. It requires asking investment committees what they would need to see, which is a straightforward research exercise that nobody appears to be running.
There is an unusually clean study available here and the data exists.
Sixty designated projects, of which twenty-three have warned and some have already stalled. That is a natural comparison group. What distinguishes the projects that reached a decision from those that did not is answerable by looking, and the candidate explanations are testable against each other: capital availability, offtake structure, permitting timeline, commodity exposure, sponsor balance sheet, and whether any instrument reduced variance rather than raising the mean.
The answer would be worth a great deal, because the current policy debate is being conducted on assertion. Developers say capital is too slow. Officials say projects are not bankable. Both may be describing the same thing from different sides, and neither has published the comparison.
Extending it beyond Europe would sharpen it further. Lithium, copper and battery material projects in Africa and Latin America face the same volatility with much weaker public support and, in some cases, reach financial close anyway. Understanding why would say something about whether the constraint is genuinely capital or is specific to how European instruments are designed. That is the sort of question where evidence from outside Europe is analytically necessary rather than decorative, and it is the same argument we made about adoption research.
Time from designation to final investment decision, for all sixty, with the stalled ones treated as censored observations rather than excluded. This is basic and it is not published.
What each project actually asked for, and what it received. Whether the instrument offered addressed returns or uncertainty, and whether the projects that received variance-reducing support moved faster.
The hurdle rate. Investment committees know theirs. Asking a sample of developers what return they require, and how that compares to their cost of capital, would locate the option premium directly.
What happened to the projects that were not designated. The list creates a treated group and an untreated one. If undesignated projects reach decisions at similar rates, the designation is doing less than assumed, and that is worth knowing before the next list.
Europe has been clear that these projects matter. The harder step is accepting that a committee deciding whether to build a plant is not asking whether the project is important. It is asking whether it knows enough yet, and money does not answer that question.
The Lab works on this in local manufacturing and financial inclusion, and on the distance between a policy commitment and an operating asset through measuring change.
If you are designing support for projects that are stuck before a final investment decision, tell us what you need to know.
This is an independent insight piece by Transitions Lab. For the Lab's applied work, see Financial Inclusion & Investment. See also The Hurdle Is Not the Risk on why a de-risking instrument may address the wrong band, The Mandate Is the Mine on demand that exists only because a regulation created it, and Capability Is the Slow Part on the difference between a commitment and an operating industry. To discuss a study, see Contact.