Every few years lithium reminds everyone that it is a commodity, not a religion. The price runs to the moon on an EV-demand panic, every junior with a claim and a press release gets a billion-dollar valuation, the supply response lands a year late and all at once, the whole thing comes down the elevator while the bulls take the stairs — and then, just when the obituaries are written, it rips back. We are now in the rip-back. After bottoming near a nine-year low in 2025, battery-grade lithium carbonate roughly doubled through the first quarter of 2026 and sits up on the order of 30% year on year. The knife stopped falling. That changes the question, but not the discipline.

First, the units, because the units win every argument

You cannot have a lithium opinion without these:

  • Brine vs hard-rock (spodumene). Brine is cheap and slow; hard-rock is faster and pricier. Different cost curves, different ramp behaviour, different winners at different prices.
  • Carbonate vs hydroxide. Different products for different cathodes. The price you quote has to match the product, or you're comparing nothing to nothing.
  • LCE. Lithium carbonate equivalent — the common unit. If a resource statement isn't normalised to LCE, normalise it before you believe the comparison.

Half the bad lithium takes on the internet are just unit errors with confidence — and a doubling price makes the bad takes louder, not rarer.

A rally is doing a lot of work now

On the way down, "cheap" was doing the heavy lifting — a junior down 80% was down 80%, not cheap. On the way back up, the word to watch is "leverage." A rebound lifts everyone's quote; it does not lift everyone's economics. The question is the same one that mattered in the trough, just inverted: where do you sit on the cost curve, and how much of this price actually drops to your margin?

On the way down the cost curve is the judge and the balance sheet is the jury. On the way up it's the same court — it just hands out rewards instead of sentences.

The lowest-cost producers are the ones that printed through the trough and now get the upside first. Albemarle ($ALB)'s slice of Greenbushes — the best hard-rock mine on the planet — and SQM ($SQM)'s Atacama brine sit at the bottom of the cost curve, so the move from survival to fat margin is fastest for them. Pilbara Minerals ($PLS), the largest independent spodumene play, is the cleaner pure-play tell: when the rally is real, Pilgangoora's economics swing hardest, and when it's a head-fake, so do they. The high-cost hopefuls that diluted to survive 2025 get a bounce too — but a bounce is not a business.

The take

We are constructive on lithium demand over the long run and unsentimental about lithium equities at any point in the cycle. A doubling does not change that — it tests it. The metal has a future; a recovering price just means more companies will claim to be the ones delivering it. Our job is the same as it was at the bottom: tell you who actually sits low on the cost curve and who is renting a chart, in the right units.