A Company With a Country's Trade Statistics: Why Brazil's Vanadium Market Is One Mine in Bahia

There is a sentence you can find in almost any summary of the vanadium market: Brazil supplies 47 per cent of the vanadium pentoxide the United States imports. It is true. It is also, on its own, misleading in a specific and useful way, because the same country supplies 6.4 per cent of the world’s mined vanadium. Set the two numbers beside each other and they refuse to reconcile — one describes a trade route, the other describes an orebody.

The resolution is that Brazil is not a national industry. It is one mine in Bahia State, 99.94 per cent owned by a single NASDAQ- and TSX-listed company, covering 48,954 hectares, employing 494 people, with a 31-year life under a plan dated October 2024. Everything a buyer might infer from the phrase “Brazilian supply” — multiple counterparties, substitutable origins, a market that behaves like a market — has to be re-derived from that single fact.

Two shares, one mine

The arithmetic is worth doing in the open, because it is the whole argument. USGS puts Brazilian mine production at 6,400 tonnes of contained vanadium against a world total near 100,000. Largo’s own disclosure for the Maracás Menchen mine reports 9,150 tonnes of V2O5 equivalent produced in 2025. At 56 per cent vanadium content, that is 5,124 tonnes of contained metal — 80.06 per cent of the national figure.

Square that share and you get a Herfindahl–Hirschman index of 6,410, roughly two and a half times the threshold at which a market is conventionally called highly concentrated. Pheonix records one tier-one player in this market, which reads oddly in a country report and is nonetheless the honest number: a market with a single double-digit participant has one, not zero.

The United States sits on the other side of that position with no primary production of its own, 58 per cent import reliance and about 94 per cent of reported consumption going into metallurgical applications. Combine the two published ratios — Brazil’s 47 per cent of imports, America’s 58 per cent import reliance — and Brazil covers roughly 27 per cent of US vanadium consumption. That figure is a bridge between two ratios struck on different bases, and it should be labelled as one rather than presented as a measurement. The full report says so on the page.

What the pool actually is

Pheonix values Brazilian V2O5 at 11,400 tonnes and USD 130.69 million in 2026, reaching 14,300 tonnes and USD 220.68 million by 2033. Volume compounds at 3.29 per cent. Value compounds at 7.77 per cent. The realised price moves from USD 11,464 to USD 15,432 a tonne, or USD 5.20 to USD 7.00 a pound.

Value therefore compounds 2.36 times faster than volume, and price does 56 per cent of that work. That split is the first thing to understand about this market, because it means the forecast is not a growth story that happens to include a price line. It is a price recovery with a volume cap bolted to it, and the two lines are answering different questions. Volume is answering “how much ore is there and what does it cost to move through the plant”. Value is answering “what is a tonne of pentoxide worth in 2033” — and the honest answer, from the operator’s own cost disclosures, is less than it costs to produce.

The trough, in the operator’s own arithmetic

In 2025 the mine produced 9,150 tonnes of V2O5 equivalent — 3.68 per cent below the floor of its own guidance of 9,500 to 11,500 — and sold 8,686 tonnes at an 80.1 per cent recovery rate. Cash operating cost guidance was USD 4.50 to USD 5.50 a pound. The Chinese benchmark in September 2025 was USD 4.86.

Read those two lines together. At the top of its own guidance range, the producer carrying 80 per cent of the national market was selling below cash cost. At the bottom of the range it kept roughly USD 793 a tonne. There is no version of that spread that funds a mine.

Add depreciation, impairment and financing and the picture sharpens. The company reported a net loss before tax of USD 51.7 million, which is about USD 5,949 on each of the 8,686 tonnes sold. Against revenue of USD 10,714 a tonne, that implies an all-in unit cost near USD 7.56 a pound, or USD 16,663 a tonne. The product page tells you how the company responds to that: a guaranteed 99.0 per cent vanadium content against an industry standard of 99.0 per cent, with a typical content of 99.5, positioned at catalysts and at electrolyte for flow batteries. When you cannot move the price of a commodity, you move the specification.

A forecast that never reaches breakeven

The base case takes the realised price to USD 7.00 a pound by 2033. That is 7.98 per cent below the implied all-in cost of USD 7.56. At 2033 volumes the same tonnes are worth USD 220.68 million at market and USD 238.29 million at the producer’s cost — a gap of USD 17.61 million.

It is worth being precise about what that means, because it is easy to file under pessimism. It is not a prediction that the mine fails. It is a statement that the market clears, across the whole forecast window, at a price that does not repay the full cost of the asset producing 80 per cent of it. A producer in that position does not expand on schedule, because expansion capital comes out of margin and there is no margin. The volume line is capped as a consequence of the price line, not independently of it.

That is also why the scenarios are described the way they are. The conservative case at USD 171.96 million — 22.08 per cent below base — is not a demand failure. It is what happens if recovery stalls at 80.1 per cent and the price recovers only to the top of the producer’s cash cost range. The optimistic case at USD 274.78 million, 24.52 per cent above, requires two things at once: throughput capital the base case says will not be financed, and a battery-grade premium that is contracted rather than spot. The band looks symmetric in value. The outcomes are not. The downside is a continuation; the upside is an achievement.

Where 2,900 tonnes comes from

Adding 2,900 tonnes over seven years on a single orebody is not a demand question, and the report decomposes it accordingly. Moving recovery from 80.1 to 90 per cent supplies 1,132 tonnes with no new ore — 39 per cent of the growth. The remaining 1,768 tonnes, 61 per cent, requires throughput capital.

So watch where the capital actually went. Largo’s ilmenite concentration plant, completed in 2023, takes feedstock from the same vanadium operations. Flotation circuits installed from September 2025 were designed to lift concentrate capacity from 42,000 to 115,000 tonnes a year — a 2.74-fold increase — against a vanadium volume forecast that adds 25.4 per cent over seven years. In the worst price environment the metal has seen in a decade, the funded expansion went to the titanium co-product, on the same orebody, through the same concentrator, with the same people.

That is not diversification. It is fixed-cost absorption, and it is the only move available to a single-asset primary producer in a trough. Anyone forecasting 2,900 tonnes of vanadium growth while reading a press release about ilmenite expansion is reading two halves of the same decision.

The price of a specification

The pool splits by product form into a narrow commodity band and one genuinely differentiated line. Metallurgical products — flake, ferrovanadium, vanadium nitride — are 80.9 per cent of the value and all sit inside a 1.19-times band. Flake, the traded intermediate, carries 55.6 per cent of value at a 1.00-times multiple and USD 10,989 a tonne. Battery-grade pentoxide and electrolyte realise USD 21,053, a 1.92-times multiple, on 6.0 per cent of the tonnage and 11.0 per cent of the value. On the end-use cut, battery-grade material and aerospace master alloys together take 13.0 per cent of the tonnage and 20.35 per cent of the value.

Two conclusions follow and they point in opposite directions. The first is that alloying is a conversion business with a thin, published premium: no amount of operating excellence moves a tonne of flake far from the benchmark. The second is that the only line capable of repricing the pool is too small to do it today — 6 per cent of the tonnage, 11 per cent of the value — and is the only line with any pricing freedom at all.

The trap for a buyer is to negotiate on the average. A tonne of metallurgical flake at USD 10,989 and a tonne of electrolyte-grade pentoxide at USD 21,053 are both “a tonne of V2O5” in a contract schedule and are not the same purchase. The whole negotiation is which specification the price refers to, and the average is the number that hides it.

Map the asset, not the country

Which brings the argument back to where it started. The United States takes 41.65 per cent of Brazilian tonnage and 44.14 per cent of the value, and the destination set is not assumed: it is cross-checked against the five jurisdictions in which the producer publishes pentoxide safety data sheets — North America, Europe, China, Brazil and South Korea. A company does not maintain regulatory documentation in a market it does not sell into. That is as close as this market comes to a published customer list.

Behind the supply line sits a balance sheet that sets the timetable rather than the strategy: USD 9.7 million of cash against USD 107.1 million of debt, a USD 84.2 million deferral running to March 2026, and a rollover to September conditional on raising USD 22 million. The cost-to-price gap of USD 17.61 million is 80.0 per cent of that condition. Meanwhile the regulatory setting is benign and price-linked — titles administered by the Agência Nacional de Mineração, CFEM levied on gross revenue rather than profit, a critical and strategic minerals programme that signals intent rather than writing instruments. Regulation is not the constraint on this market between 2026 and 2033. The price is.

For a buyer, the practical consequence is uncomfortable and simple. Country-level due diligence will tell you Brazil is a stable, low-cost, export-oriented vanadium jurisdiction, and every one of those adjectives is defensible. It will not tell you that a single production interruption at one mine in Bahia would remove four-fifths of the national pool and roughly a third of US pentoxide imports in the same month, because that fact only appears when a supply chain is mapped to assets rather than to countries.

Brazil’s vanadium market is not a country. It is a company with a country’s trade statistics — and a country report that does not name the mine is a trade statistic with a forecast attached.

Comparable concentration and co-product problems are treated in the global manganese mining market, cobalt mining, graphite mining, lithium mining and nickel mining reports. The demand end of the metallurgical chain is in the tool steel and speciality alloy steel market, and the buyer at the other end of the electrolyte line in stationary energy storage.