Materials Edges · Explainer · Business models
Royalties, Streams and Generators
Two mining companies can own the same ore body and make money in completely different ways — or fail to. The structure decides who pays for the drilling, who carries the risk, and who gets diluted. This is a plain-English guide to the models and deal terms you meet across the sector, and what each one is really trading away.
Reference page · Updated 1 September 2026 · Research and education, not investment advice
Almost every structure in mining is an answer to one question: a mine costs enormous sums long before it earns anything — so who pays, and what do they get for it? Exploration is cheap and usually fails. Development is expensive and usually late. Production, if you reach it, throws off cash for decades. The models below all shuffle that timing problem between different parties.
Read them as trades, not as labels. Each one buys something and gives up something, and the thing given up is often the part that matters most in the outcome you actually get.
Owning the metal without owning the mine
Revenue model
Royalty
A right to a slice of a mine's output or revenue, for the life of the mine, without any obligation to pay operating or capital costs. Usually created when someone sells a property but keeps a royalty, or when a royalty company buys one for cash.
- Who pays to build
- The operator. The royalty holder contributes nothing after the initial purchase.
- How you get paid
- A percentage off the top, for as long as the mine produces.
- Main appeal
- Cost inflation does not touch you. If capex doubles, your cheque is unchanged.
The three letters matter enormously:
- NSR
- Net Smelter Return — a percentage of revenue after smelting, refining and transport are deducted. The sector standard, typically 0.5–3%.
- GRR
- Gross Revenue Royalty — a percentage of gross revenue with no deductions. Better for the holder than an NSR of the same headline rate.
- NPI
- Net Profit Interest — a share of profit after all costs. Sounds generous; behaves very differently, because a mine can produce for years and report little profit.
What to watch: whether the operator holds a buy-back right (many royalties can be repurchased, capping your upside exactly when the asset works), and whether an NPI is disguised as a headline percentage several times larger than the NSR it is competing with.
Revenue model
Streaming
A large payment up front in exchange for the right to buy a fixed share of future production at a price set far below market — often around a fifth of spot, sometimes a nominal few dollars an ounce. Streams are frequently written over a by-product, so a copper mine sells its gold or silver stream to fund the build.
- Who pays to build
- The streamer funds a chunk of construction up front, in cash.
- How you get paid
- By buying metal cheaply for decades and selling it at market.
- Versus a royalty
- A royalty is a slice of revenue and nothing more. A stream keeps paying — the discounted purchase price — so it is closer to a very long-dated supply contract.
What to watch: from the miner's side, a stream is expensive equity dressed as financing. It does not dilute the share count, which is why boards like it, but it permanently removes a share of the best revenue the mine will ever produce. A company that streams away its by-product credits has sold the part of the orebody that made the economics work.
Getting exploration paid for by someone else
Funding model
Prospect generator (project generator)
Instead of drilling its own ground with its own cash, the company acquires prospective land, adds early value cheaply — mapping, geophysics, first holes — then brings in a partner to fund the expensive work in exchange for the majority. The generator keeps a minority interest, often a royalty, and sometimes charges a fee to run the programme.
- Who pays to drill
- The partner, typically a major or mid-tier miner.
- What you keep
- A carried or diluting minority, frequently a royalty, occasionally management fees.
- Main appeal
- Many shots on goal without funding each one, and far less dilution than a conventional junior.
What to watch: the honest test is whether fee and royalty income covers head-office overhead. If it does, the company can wait out a bad market without issuing shares — the whole point of the model. If it does not, it is a conventional junior with a better story, and it will still be raising money at low prices when the cycle turns.
Deal term
Earn-in (option agreement)
The mechanism that makes generators work. A partner earns a defined percentage by spending an agreed amount on the ground over an agreed period, usually in stages, often with cash payments and share issues alongside. Miss a milestone and the earn-in lapses; the ground reverts.
- Typical shape
- Spend a set sum over several years to earn a first tranche, then more spending or a study to earn a second.
- Why staged
- It lets the partner walk away cheaply after bad results — which is exactly why early stages are usually funded and later ones often are not.
What to watch: announced earn-ins are stated at their full headline value, but that figure is the sum of every stage including ones the partner may never reach. The money actually committed is stage one. Read which tranche is firm and which is optional.
Deal term
Carried interest
Your share of costs is paid by someone else up to an agreed point — often a production decision or first pour. You keep your percentage without writing cheques.
- Free-carried
- The costs are never recovered from you. The strongest version, and rarer than the phrase suggests.
- Carried to repayment
- Costs are recovered later out of your share of cash flow, so your first years of production pay off the carry before you see anything.
What to watch: "fully carried" and "free-carried" are not the same claim, and the difference can be years of cash flow. Check what happens after the carry ends too — that is usually the moment a small holder must start funding its share or begin diluting.
Deal term
Joint venture, dilution and back-in rights
Once an earn-in completes, the parties usually form a joint venture. In a participating JV each side funds its pro-rata share of every budget. Fail to fund and you dilute, typically on a straight-line formula.
- Dilution floor
- Many agreements convert a partner's interest to a royalty once it falls below a threshold — a small stake turning into, say, a 2% NSR.
- Back-in right
- A former owner or a state body may reclaim a percentage after a milestone by repaying a multiple of costs. Common in some jurisdictions and easy to overlook.
What to watch: a minority partner in a participating JV with no treasury is on a slow path to a royalty whether it intends to be or not. That is not necessarily bad — a royalty with no funding obligation may be the better outcome — but it should be a decision, not a surprise.
Getting the mine built
Construction finance
Offtake and prepayment
An offtake commits future production to a buyer — a trader, smelter or industrial consumer. A prepayment attaches cash to it: the buyer funds construction now and is repaid in metal or concentrate later.
- Who pays to build
- The buyer, partially, ahead of production.
- Cost to the miner
- Pricing terms, treatment and refining charges, and a customer that must be delivered to on schedule.
What to watch: committing all of your output to a single counterparty removes commercial flexibility precisely when you would most want it, and prepayments are debt in commercial clothing — repayable in product, secured, and unforgiving if the ramp-up slips.
Construction finance
Strategic investor, equity and debt
The conventional routes. A strategic investor — often a producer or a sovereign vehicle — takes a stake on the register, bringing validation and sometimes technical help. Equity raises dilute everyone. Debt does not dilute but demands repayment on a schedule an early mine may not meet.
What to watch: a strategic holder is genuine third-party validation and worth weighing, but check the price and the rights attached. A stake taken at a deep discount, with board seats and anti-dilution protection, is a very different signal from one bought at market.
Corporate
Spin-out
Non-core ground is placed into a separate listed vehicle and distributed to shareholders, so assets the market was not valuing get their own story, treasury and management.
What to watch: spin-outs surface value when the parent is genuinely too complex to price, and destroy it when they exist mainly to generate promotion and fees. The tell is whether the new entity is funded well enough to actually do work.
The models side by side
| Model | Who funds the work | Who carries the risk | How the holder is paid | Main weakness |
| Royalty | The operator | The operator | Percentage of revenue, life of mine | No control; often buy-backable |
| Stream | Streamer funds part of the build | Shared — streamer risks the mine failing | Buying metal far below market | Permanently removes the best revenue |
| Prospect generator | Partners | Partners, mostly | Carried stake, royalties, fees | Slow; small share of any success |
| Earn-in | The incoming partner | The partner, stage by stage | Retained interest once earned | Later stages may never be funded |
| Carried interest | The paying partner | The payer | A stake without cheques, to a milestone | Repayment terms; funding cliff after |
| Participating JV | Both, pro-rata | Both | Direct share of the asset | Dilution if you cannot fund |
| Offtake / prepayment | The buyer, in advance | The miner | Cash before production | Debt-like; single counterparty |
| Conventional junior | Shareholders | Shareholders | Full ownership of the outcome | Serial dilution through the cycle |
The other vocabulary: how proven is it?
Separate from how a company is financed is how far its ground has been proved up. These terms are defined under reporting codes such as JORC and NI 43-101, and they are not interchangeable marketing words.
InferredResource implied by limited drilling. Geologically reasonable, economically unproven.
IndicatedEnough drilling to support planning and, with a study, reserves.
MeasuredThe highest resource confidence.
ReserveThe part shown to be economically mineable. Requires a study — not just geology.
PEA / scopingFirst economic sketch. May use inferred material; cannot support reserves.
Pre-feasibilityThe first study that can convert resources into reserves.
FeasibilityBankable detail — the basis lenders will fund against.
ConstructionFunded and being built. Capital risk is now execution risk.
Resource is not reserve. A large inferred resource with no study behind it is a geological statement, not an economic one. Plenty of companies restart or expand mines with no reserves and no feasibility study — that is a legitimate decision, but it means grade, recovery and throughput are being assumed rather than demonstrated. When you read a headline resource number, the useful question is which category it sits in and what study, if any, stands behind it.
How we use this
Our de-risking score measures one thing deliberately: how advanced and proven a company's flagship asset is — its stage, jurisdiction, resource and economics. It is built for single-project miners, and on those it works well.
It systematically understates the models on this page. A prospect generator's real risk reduction is diversification, other people's money and recurring fee or royalty income — none of which a single-asset meter can see. A royalty company may hold interests over a dozen mines it will never operate and score poorly on every line. That is a limitation of the measure, not a verdict on the business.
So we read the score alongside the structure. Where the income comes from, who is funding the next hole, and what happens when the carry ends usually tell you more about a company's ability to survive a bad two years than any single number can.