Price floors as credit support for critical minerals

A project lender advances against cash flow it can rely on, not against the price a commodity might fetch. On a material whose price swings widely, the revenue a merchant producer will actually earn is a distribution, not a number, and a lender sizes debt against the bottom of that distribution — which, uncapped, can be close to zero.

A price floor changes the shape of what is being underwritten. It guarantees a minimum price for the output: if the market price falls below the floor, a counterparty pays the shortfall; if it rises above, the producer keeps some or all of the difference. The producer still carries the risk of producing the volume, but the price beneath that volume no longer falls away.

What that does to the structure is concrete. The lender sizes the debt, and sets the coverage it requires, against the floor rather than the spot price, because the floor is the revenue that survives a downside. A credible floor therefore raises the debt a project can carry and lowers the risk premium on it, because the tail of the price distribution — the part that made the project unbankable — has been transferred to whoever wrote the floor. The floor is not a subsidy to the economics; it is a change to what the economics can be borrowed against.

Price volatility is ordinary, and mature commodity markets have instruments for it — futures, swaps, hedges — that a lender will accept in place of a floor. Critical minerals are different in a way that disables most of those instruments, and the difference is not geology.

The processing and refining of these materials is concentrated in a small number of jurisdictions, and in the case of rare earths overwhelmingly in one. (International Energy Agency, as of August 11, 2026) A supplier with that position can lower prices below the cost of any newly built Western competitor for as long as it takes to make the competitor uneconomic — not because of a market cycle, but as a matter of policy. Against that, a hedge struck at a market price offers little: the market price itself is the variable being driven, and there may be no liquid forward market to hedge into at all.

So the risk a critical-minerals project actually needs covered is not ordinary price volatility. It is the risk that a state actor sets the price below the project's cost deliberately — which is a geopolitical risk wearing a commodity's clothes. That reframing is the whole reason the instrument on this page matters: a floor transfers exactly that risk, and it is worth what the party bearing it is good for. The supply-concentration mechanics behind the exposure are their own subject; this page is about the instrument that makes a project financeable in spite of them.

A floor is one of a small family of contractual supports, and they are routinely conflated. They do different things, and a lender reads them differently because they allocate different risks.

A floor is only ever as good as the party standing behind it, because the floor pays precisely when prices are low — which is precisely when a weak counterparty cannot. This is the analysis a lender actually runs on a floored project, and it collapses to one question: is the floor writer good for the payment in the state of the world where it is owed?

That question is why the significant critical-minerals floors are being written by governments rather than by commercial buyers. A government floor substitutes sovereign credit for commodity-price risk, and it is written by the party that captures the strategic benefit of the domestic supply existing at all — which is what makes it durable rather than opportunistic.

The pattern is now visible in live transactions. A US defence agency has written a ten-year floor for a domestic rare-earth producer, settled as a periodic cash payment of the shortfall to the floor, alongside an equity stake — and roughly a billion dollars of private bank financing was committed to the same expansion once the floor was in place, which is the crowd-in effect the floor exists to produce. (MP Materials, as of August 11, 2026) A government-backed Japanese offtaker has set a floor at the same level for an Australian producer on a contract running more than a decade. (Lynas Rare Earths, as of August 11, 2026) And the toolkit is widening beyond floors into government purchase and storage: a national strategic reserve is being structured so that committed access to it can serve as the demand anchor a project uses to reach financial close — a stockpile functioning as a bankable offtake. (Export-Import Bank of the United States, as of August 11, 2026)

The common thread is that the credit support and the strategic motive sit with the same party. That is what a commercial buyer, however large, cannot replicate: it can guarantee a price, but not for the reason that keeps the guarantee good through the downside it is written for.

A floor in a term sheet is not yet credit support. What makes it bankable is a short list of properties, and the failure is almost always that one of them was assumed rather than confirmed.

• The writer's credit, in the downside. The floor is a promise to pay when prices are low. A floor from a party that is itself impaired when that happens adds nothing. A sovereign or government-agency floor is strong for exactly this reason; a floor from a thinly capitalised buyer needs its own credit support behind it — a guarantee, a standby letter of credit, a reserve.
• Enforceable and measurable. The benchmark the floor references has to be one both sides can observe and neither can manipulate, and the payment mechanic has to be defined precisely enough to compel. A floor against an index that is thin or set inside the concentrated market it is meant to protect against is weaker than it reads.
• Assignable to the lenders. A floor that cannot be assigned as security, or that falls away on the events that would trigger it, is a benefit to the producer and not to the financing.
• Aligned with the debt. The floor's tenor, its payment timing and its covenants have to line up with the debt service they are meant to cover. A floor shorter than the loan, or paying on a cycle the debt service does not match, leaves a gap and can create refinancing pressure the structure did not intend.
• Sized against the real cost curve. A floor below the project's all-in cost of production guarantees a loss rather than a margin. The floor has to clear the cost of producing the volume, not merely be positive.

The durable point is that a floor moves risk; it does not delete it. A lender that has satisfied itself on the five properties above is underwriting the floor writer's credit in place of the commodity's price — which is a better risk, but a different one, and it should be underwritten as such rather than treated as though the price risk had simply gone away.

How a floored project disappoints, and each is avoidable at structuring:

• The floor writer could not pay in the downside. A minimum price guaranteed by a party that is itself distressed when prices fall is a clause, not a cash flow. The whole value of a government floor is that it does not have this failure mode.
• The floor was sized below the cost curve. A guaranteed price that does not clear the cost of production locks in a loss. A floor is only credit support above all-in cost, and a floor set for political optics rather than project economics is not the same instrument.
• The benchmark was weak. A floor referencing a thin or manipulable price — or one observed inside the very market the floor is meant to protect against — is measurable in theory and contestable in practice.
• The floor and the debt were misaligned. A floor whose term, timing or covenants do not match the debt service leaves a gap between the support and the obligation it was meant to cover, and the gap is discovered at a refinancing.
• An announced policy was treated as an executed floor. Proposed price-floor frameworks and reference-price schemes are announced ahead of the mechanisms that would make them operative. A government's stated intention to support a price is not a signed agreement a lender can size debt against, and the two must not be conflated in a model.
• The floor was mistaken for a solution to volume. A price floor guarantees price, not offtake. A project still has to sell — or be paid for not selling under a take-or-pay — and a floor over uncontracted volume protects the price of material nobody is obliged to buy.

Continuum structures and arranges financings around price-floor, offtake and prepayment support of this kind, and coordinates the parties to them. It does not trade, source, take title to or hold materials, is not a principal, a bank, a broker-dealer or a direct lender, and does not hold client funds. The underlying materials, their processing and their end markets are covered by Pantheon, not here.