Tearing apart Indonesia's nickel quota machine

Nickel.

The financial press is covering the Indonesia supply cut story the same way they cover every commodity story: a headline number, a price chart, a quote from a strategist in London. "Jakarta slashes mining quotas. Nickel rallies 30%." That's the dashboard. The engine is different — and considerably more interesting.

Here's what's actually happening. Indonesia controls roughly 60% of global nickel mine supply. Not 20%. Not 30%. Sixty. They achieved this in about a decade by banning raw ore exports, which forced Chinese capital to build smelters inside Indonesia itself — 35 RKEF furnaces, multiple HPAL plants, the works. The country went from two nickel smelters in 2013 to a processing complex that now needs 340 to 350 million wet metric tonnes of ore per year just to run at normal utilization.

Then in late 2025, Jakarta did something unusual. Instead of approving another record quota, they cut it — hard. The 2026 RKAB came in at 260 to 270 million wet metric tonnes, down from 379 million in 2025. The market called it a supply cut and bought the headline. That's not quite what happened.

What happened is that Indonesia's quota system was revealed to be something between a planning tool and a fiction. In 2025, the approved RKAB was 379 million tonnes. Actual production was around 265 million. That's a utilization rate of 55%. The "surplus" everyone was pricing wasn't a surplus — it was zombie allocations. Companies hoarding permits they had no intention of using. The market spent a year pricing a glut that existed mostly on paper. Now Jakarta is aligning the quota to something closer to reality. It's not a cut. It's an audit.

Story off. Numbers on.

Indonesia's smelter complex runs at around 2.7 million tonnes of RKEF and HPAL processing capacity annually. At 90% utilization — where it ran last year — it needs roughly 340 to 350 million wet metric tonnes of ore to feed. The 2026 RKAB caps domestic mining at 260 to 270 million. That's not a rounding error. That's an 80 to 90 million wet metric tonne gap. FINI, the Indonesian smelter industry group, has already said utilization will drop to 70–75% this year. That's 270,000 to 400,000 fewer tonnes of processed nickel entering the supply chain.

The importation plug-in isn't clean. Indonesia needs to import 50 million tonnes from the Philippines in 2026 — more than triple the 15 million it imported in 2025. The Philippines' output fell 24% in 2025. It's being asked to triple exports while its own production is declining. That math doesn't close.

Meanwhile, LME nickel sits at around $19,100 per tonne as of the first week of May. That's up from lows of $13,865 in 2025. The rally already happened. But the physical mechanism — the actual reduction in processed nickel hitting the market — hasn't fully hit the delivery chain yet. Smelters have been drawing down stockpiled ore buffers, typically one to two months of inventory. Those buffers are finite. Eramet announced a temporary production halt at Weda Bay Nickel starting May 2026. Vale Indonesia cut its quota application by 30%. The machine is slowing down. The LME price is still catching up to why.

This is not a supply cut story. This is a structural feed gap inside the world's largest integrated nickel system, materializing in real time while the financial market debates whether Jakarta will roll back the policy by July. The July RKAB revision window is real — miners can apply for quota increases after Q2 reporting. Whether they get them is a different question. The government's royalty regime was restructured so that every $1,000 per tonne increase in nickel price generates roughly $250 more per tonne in tax revenue. Jakarta now has a direct fiscal incentive to keep ore tight. They're not going to loosen the tap for free.

Nobody is talking about the sulphur.

The HPAL process — high-pressure acid leach — is how Indonesia extracts battery-grade nickel from laterite ore. It needs large volumes of sulphuric acid. Sulphuric acid is made from sulphur. And here's the thing SMM published in their year-end review that almost nobody in the western financial press picked up: sulphur cost has now overtaken ore cost as the primary input cost for MHP production. Not tied with it. Overtaken it.

The market models that priced nickel at $14,000 last year were built on an assumed cost floor. Raising royalties helped push that floor up. But the sulphur issue creates a second floor, embedded in the processing chemistry, that doesn't respond to government policy at all. When your biggest variable cost is the price of sulphur — a global market driven by fertilizer demand and refinery configurations that have nothing to do with nickel — you've lost control of your margin at the bottom of the cycle. If nickel goes back to $13,900, a large portion of Indonesian HPAL capacity is uneconomical. Not marginally. Operationally.

The second weak link: ore grade decline. This one's geological. Indonesia's saprolite — the laterite layer used for NPI and ferronickel — has been "triple high-graded" for years, meaning mines have been selectively pulling the best ore, the highest nickel content, the best nickel-to-iron ratio. Grade is declining. To maintain output, you need more tonnes of lower-quality ore. More ore means more quota. More quota is exactly what Jakarta is restricting. I don't think the market has done this math all the way through yet.

There's one more thing. The zombie allocation problem cuts both ways. The RKAB revisions in 2025 created perverse incentives — companies applied for far more quota than they could use. ICBC put out a note in December saying those grossly inflated approved quotas bore "little resemblance to physical market realities." So when Jakarta cut the 2026 quota from 379 million to 265 million, they weren't really cutting supply by that much — they were cutting the fiction. Actual 2025 production was around 265 million tonnes already. The real cut comes from what happens when quota is tight and miners can't hoard buffer. That's the transmission mechanism nobody is modelling cleanly.

The sequence matters more than the direction here, because there are two forces working on different timescales and they're going to pull against each other before they align.

The July revision window is the hinge. If Jakarta approves meaningful quota increases — and historically they often have, because downstream industry pressure is intense — prices pull back. The market sells the news. If Jakarta holds firm, or grants only token revisions, smelter utilization stays suppressed, processed nickel output stays below trend, and the tightness we've been pricing as a policy signal becomes a physical reality in the back half of the year.

The key variable is the fiscal incentive structure. Jakarta restructured its royalty regime to ad valorem in April 2025 — roughly $250 more per tonne in tax revenue for every $1,000 increase in nickel price. That's not a small number. The government has approximately $250 million per year in additional revenue at stake for every thousand dollars they can hold the price above its old floor. They are, as one analyst at Mining.com put it, behaving like the "OPEC of one." The analogy isn't perfect — they can't control processing demand from their own smelters — but the incentive alignment is real and it's new. It wasn't there before 2025.

The broad nickel ETFs won't capture this cleanly. They're too diluted. The more surgical play is in western sulphide projects — Canada Nickel's Crawford, for instance, is a first-quartile cost structure with carbon sequestration that makes it strategically interesting to governments trying to build supply chains outside the Indonesia-China nexus. These aren't leveraged to spot the same way Indonesian producers are. But they're not exposed to the ore grade problem, the sulphur cost problem, or Jakarta's quota calendar either.

One honest caveat: the zombie allocation dynamic cuts in both directions. If actual 2025 production was already 265 million tonnes while the approved quota was 379 million, then Jakarta's "cut" to 265 million isn't really a cut at all — it's matching the quota to reality. The feed gap exists regardless, because the smelters that were built on the assumption of future quota growth now find the ore isn't there. But if the market rerated on the fiction of the cut rather than the fact of the gap, there's a risk the repricing has run further than the physical fundamentals justify — at least until Q3 when the buffer stocks run out. Watch the smelter utilization numbers out of FINI in June. That's where the thesis gets confirmed or it doesn't.