Ecopro BM is putting ₩765 billion into a nickel smelter. The plant, called BNSI, is under construction in Sulawesi, Indonesia, and will produce 90,000 tonnes of nickel a year. Total project cost is around $1.84 billion, and the Ecopro group intends to hold 39% of it, the largest single stake.
Ecopro BM Co., Ltd. (KOSDAQ:247540) does not smelt anything today. It buys metal, converts it into high-nickel cathode powder, and sells that to Samsung SDI and SK On. Going upstream into ore is a different business with different economics and a different cycle.
The company has now put a number on what it expects from that business. Its medium-term plan calls for roughly ₩2.5 trillion of nickel sales around 2030, with ₩530 billion of operating profit. That is a 21.2% operating margin.
Hold that figure next to the company's own history for a second.
Ecopro BM's best operating margin in the eight years of accounts on file is 8.5%, set in 2018 on revenue of ₩589.2 billion. In the 2022 boom it managed 7.1%. In 2025 it was 5.7%. In the June quarter of 2026 it was 3.1%.
The cathode business has never earned 21%. It has never earned half of 21%.
The absolute number is stranger still. ₩530 billion of operating profit from nickel would beat the ₩380.7 billion the entire company earned in 2022, its record year, on ₩5,357.6 billion of revenue. And ₩2.5 trillion of nickel sales would roughly match the ₩2,531.6 billion the whole company sold in 2025.
Management is describing a second business, arriving in about four years, that is larger and far more profitable than the one it has spent a decade building. That claim may be right. It is not a modest claim, and it deserves to be treated as the centre of the investment case rather than a footnote to a rights offering.
The strategic logic here is better than the margin target makes it sound.
Cathode is a pass-through business. Ecopro BM buys metal at market and sells powder at metal cost plus a conversion fee. That structure caps the upside and leaves the maker exposed to timing losses when prices fall between purchase and sale. Owning the nickel converts part of that pass-through into an owned margin.
The second reason is where the metal comes from. American battery tax credits turn on sourcing rules that exclude material tied to certain foreign entities, and most Indonesian nickel capacity is Chinese-financed. A Korean-controlled smelter in Indonesia is precisely the structure that answers that question. Taking 39% and the largest shareholding, rather than an offtake contract, looks deliberate on exactly this point.
There is a third reason, less flattering. Ecopro BM's conversion margin has been squeezed for three years. Moving up the chain is what a converter does when the conversion fee stops paying.
Nickel is not a good market right now, and nobody serious argues otherwise.
Combined LME and Shanghai exchange inventories reached a record 468,600 tonnes in mid-2026, about six weeks of global consumption. Forecasts put the 2026 average price near $15,250 a tonne. Prices spiked in May on Indonesian mining restrictions, then fell 14% in June to around $16,395.
The cost side is the problem. Estimates put total nickel pig iron production cost near $15,740 a tonne after a recent $500 increase. When the marginal producer's cost is level with the price, the least efficient smelters shut. Analysts have been flagging curtailment risk at rotary kiln electric furnace plants for exactly this reason.
Indonesian producers are the low-cost end of that curve, and integrated operations there have held their margins better than most. So BNSI is being built in the right country. But 21.2% is a good-times margin in a commodity that is currently oversupplied by as much as 261,000 tonnes on some estimates, and the plan reaches its target in 2030, four years out. Commodity forecasts at that horizon are guesses with decimal points.
The money comes from shareholders. Ecopro BM is raising ₩1.2 trillion, about $845 million, in a rights offering. The first issue price has been set at ₩89,500, which is 26% below the estimate the company published on June 30.
That price is worth pausing on. The stock's 52-week low is ₩91,142. The rights are being offered below anything the shares have traded at in a year.
An earlier read of this deal noted that around ₩915 billion of the proceeds was earmarked for securities issued by other companies rather than for cathode plant. The nickel smelter is a large part of what those securities are. The reported figures do not reconcile cleanly, ₩765 billion against ₩915 billion, and the difference is probably scope: group versus subsidiary, or equity versus total commitment. The company has not laid the two side by side in one place.
The balance sheet explains the urgency. Total liabilities were ₩3,334.9 billion at June 30 against ₩2,096.5 billion of equity. Cash was ₩467.1 billion. Operating cash flow was negative in both quarters of 2026. This is not a company funding a $1.84 billion project out of surplus.
The strongest counter is that vertical integration in batteries has worked before. CATL and the big Chinese cathode makers moved upstream into Indonesian nickel years ago and it is a large part of why their costs are what they are. Ecopro BM is late to a trade that has already paid, not early to an untested one.
Second, the 21.2% margin may not be a forecast of nickel prices at all. If BNSI's output is contracted internally at a transfer price, the margin becomes an accounting choice about where profit sits between smelter and cathode plant. Group profit would not change. That would make the number less impressive and also less risky than it reads.
Third, sourcing rules could make Indonesian nickel from a Korean-controlled smelter genuinely scarce and worth a premium that has nothing to do with LME prices. If that happens, 21% is conservative.
Against all of that, the same management team guided this company into ₩2.36 trillion of cathode capex through a demand downturn and now runs plant at a 3.9% operating return. Forecast accuracy is part of the track record.
The disclosure that matters most is the final structure of the Ecopro stake in BNSI, because that determines whether the 21.2% is a real third-party margin or an internal transfer. The rights offering documents should carry it, and the final issue price arrives with them.
After that, the checkable items are dull and useful. BNSI's construction milestones and first-production date will tell you whether 2030 is realistic. Ecopro BM's own operating margin, 3.1% in the June quarter, will tell you whether the cathode business is stabilising while the nickel plan is built. And the LME nickel price through 2027 will show whether the surplus of 2026 was the cycle low or the new normal. All three are public, and none require taking management's word for anything.
kstock reads DART every morning and writes up what moved — the contract, the buyback, the number that does not add up. The daily post and a Saturday roundup, by email.