The Royalty Playbook, Applied to the Rarest Dirt

In 1986, two Canadian geologists named Seymour Schulich and Pierre Lassonde did something the mining industry thought was eccentric: they stopped trying to find gold and started buying the right to be paid when other people found it. Their company, Franco-Nevada, paid miners cash up front in exchange for a royalty — a small percentage of whatever the mine produced, for as long as it produced. Franco-Nevada never bought a truck, never hired a geologist to run a drill, never worried about a cost overrun. Forty years later it is the largest royalty company in the world, with a market capitalisation of roughly $35 billion as of September 2026. The toll road beat the mine.

The model has since been refined twice. Wheaton Precious Metals, the pure streaming company, buys a fixed share of a mine's future output at a steep discount to the spot price — commonly around 20% to 25% of spot for its streams — and resells at the market. Royal Gold, founded in Denver in 1981, runs a hybrid of both. Wheaton's third-quarter 2025 revenue was $476 million, up 55% year over year. Royal Gold's revenue in the first half of 2026 was $919.7 million, and in October 2025 it acquired Sandstorm Gold and Horizon Copper for $4.15 billion. The royalty and streaming business is, by now, a proven way to own the economics of mining without owning the mine.

01 — How the toll is collected

The mechanics are simple enough to fit in a paragraph. A net smelter return royalty — the standard form — typically pays its holder between 1% and 5% of the value of the metal a mine produces, after smelting and refining charges. A stream is different: the streaming company pays a large sum up front and then buys a defined share of future production at a fixed, low price for the life of the mine. In both cases the holder's costs are close to zero once the deal is signed. The mine spends the money; the royalty holder collects the percentage. It is the closest thing in mining to rent.

Now apply that arithmetic to rare earths. The International Energy Agency expects total rare-earth demand to grow from 91 kilotonnes in 2024 to 123 kilotonnes by 2030, with clean-energy demand roughly doubling from 19 to 38 kilotonnes over the same period. The United States mined about 51,000 tonnes of rare-earth oxides in 2025, 13% of world production, and Australia about 29,000 tonnes. The two Western producers of scale are MP Materials, which produced a record 2,599 tonnes of neodymium-praseodymium in 2025 — more than double its 1,294 tonnes in 2024 — and Lynas Rare Earths, which produced 7,260 tonnes of NdPr in its 2026 financial year on revenue of A$977.9 million, up 76%. A 2% royalty on a single 2,000-tonne NdPr operation at the $110-per-kilogram floor is roughly $4.4 million a year, every year, with no operating costs attached.

02 — Why rare earths have no royalty giant yet

If the model is so good, why is there no Franco-Nevada of rare earths? The answer is in the structure of the industry. Gold has thousands of mines, a transparent spot price, and a century of deal flow. Rare earths have a handful of producers, a price that has historically been set in China, and a supply chain so concentrated that the top three refining countries accounted for 97% of refined supply in 2024. A royalty company needs a pipeline of deals to underwrite, and in rare earths the pipeline has been thin: one or two mines of scale outside China, a scattering of development projects, and a price series that, until recently, could fall by half in a quarter.

That thinness is also the opening. The deals that do exist are getting bigger and better structured. The Pentagon's $110-per-kilogram NdPr floor with MP Materials runs for ten years and is matched by Lynas's twelve-year offtake with Japan Australia Rare Earths at the same floor. The floor removes the single biggest objection a royalty buyer has always had to rare earths — that the price cannot be modelled. And the capital need is real: new rare-earth mines and separation plants are measured in hundreds of millions of dollars, and the governments writing the cheques would rather see private capital share the load. A royalty firm that can write a nine-figure cheque in exchange for a percentage of a mine's life is offering exactly what a developer needs and what a government wants to see.

The market is also getting easier to price. The most recent SMM weekly wrap-up, published November 2, 2026, quoted praseodymium-neodymium oxide at $115 to $121 per kilogram on a free-on-board basis, down $3 on the week, with neodymium oxide at $138 to $140 and terbium oxide at $1,150 to $1,238. These are weekly, published, third-party prices for the material that a royalty would be written on. A royalty business does not need prices to be high. It needs them to be visible. They are becoming visible.

03 — The honest caveat

None of this makes a rare-earth royalty a safe investment. The counterargument is blunt: a royalty is only as good as the mine it is written on, and most rare-earth projects outside China are not mines yet. They are drill results, feasibility studies, and permitting applications. A royalty on an unbuilt mine is not income; it is an option, and options expire. The two producers of scale — MP Materials and Lynas — are the only royalty-grade counterparties in the West, and both already have their offtakes spoken for. A new royalty firm would be competing for a very small number of very large deals, against governments that are writing the biggest cheques of all.

There is also the concentration problem. Gold royalties diversify across dozens of mines in a dozen countries. A rare-earth royalty portfolio would, for the foreseeable future, be two names and a handful of development projects. A single permitting delay, a single plant fire, a single change of government policy would hit the whole book. And the price floor that makes the model underwritable is a contract with a government — the same kind of contract that can be renegotiated when the politics change.

But concentration cuts both ways. The reason the royalty giants are worth tens of billions is that they were early to a model nobody else was running. In gold, being early meant the 1980s. In rare earths, early means now: the floor is eighteen months old, the offtakes are being signed, the demand curve — 20.7 million electric cars sold in 2025, an IEA forecast of roughly 23 million in 2026 — keeps steepening, and there is still no dominant royalty brand in the space. The first firm to build that brand will need a name that says what it does before anyone reads the pitch deck. RareEarthRoyalties.com is that name.