Two things about Korea Zinc (010130) have been true at the same time for about a year, and they don't fit together.
The first is that the business got dramatically better. Trailing twelve-month net income through June is ₩983.6bn. In FY2024 the company earned ₩304.7bn. That's 3.2 times, and it isn't an accounting artefact — trailing operating income is ₩2,035.1bn against ₩723.5bn in FY2024.
The second is that the stock went the other way. From a 52-week high of ₩2,115,000 the shares are down 39.6% to ₩1,277,000, which includes a 7.13% intraday fall on 31 August after the Supreme Court decision, according to Joongang Economy News.
Earnings tripled, the price fell by nearly 40%. That combination usually means one of two things: the market thinks the earnings are temporary, or the price was never about earnings in the first place. I think it's mostly the second, and the way to see it is the return on equity.
Here is net income divided by year-end equity for every fiscal year on file.
FY2015: 9.85%. FY2016: 10.50%. FY2017: 10.38%. FY2018: 8.39%. FY2019: 9.34%. FY2020: 8.11%. FY2021: 10.49%. FY2022: 6.84%. FY2023: 4.68%. FY2024: 4.11%. FY2025: 7.00%.
Trailing twelve months: 8.48%.
Eleven years plus the current period, and the number has never once reached 11%. It clusters between 8% and 10.5% in good years and falls to 4% in bad ones. That is exactly what you'd expect from a custom smelter — a business that buys concentrate, charges a fee to refine it, and earns by-product credits, with heavy fixed plant and no ore of its own. It is not a business with pricing power over its own returns.
Now put the multiple next to it. Book value at 30 June was ₩11,593.9bn, or ₩555,458 a share. At ₩1,277,000 the stock is at 2.30 times book.
Price-to-book equals return on equity times the price-to-earnings ratio. Run it: 8.48% times 27.1 gives 2.30. The arithmetic is consistent, which means the 2.30 is not an independent judgement about asset value. It is the mechanical consequence of paying 27 times earnings for a company earning 8.5% on its equity.
Suppose you wanted a 10% earnings yield on what you pay — a modest requirement for a cyclical industrial. At 2.30 times book, that needs a return on equity of 23.0%.
This company has never earned half of that.
Be generous instead and use the run rate rather than the trailing year. The last two quarters produced ₩354.0bn and ₩351.5bn of net income. Annualise them and you get ₩1,411.2bn, a 12.2% return on equity, and about 18.9 times earnings at the current price. Twelve percent would be the best return in the company's recorded history, achieved on an equity base that is now ₩11,593.9bn rather than the ₩5-7tn it earned those historical returns on.
Nineteen times earnings for a smelter running at its best-ever return, in a metals upswing, is still a growth multiple. It might be the right one if zinc, lead and silver stay where they are and the treatment-charge recovery holds. It is not a cheap one.
Now go back to the 52-week high. At ₩2,115,000 the market value was ₩44,146.4bn. Set that against the ₩304.7bn the company earned in FY2024 and you get 144.9 times. Against the ₩983.6bn it has earned in the last twelve months, 44.9 times, and 3.81 times book.
Nobody pays 3.8 times book for a business that has never earned 11% on equity because they've modelled the treatment charge. They pay it because shares in a company whose control is contested are worth more to the people contesting it than they are to anyone else. That premium is not fundamental, it is not permanent, and it drains when the contest resolves.
At the 52-week low of ₩804,000 the same shares were at 1.45 times book and 17.1 times trailing earnings. That's roughly where a smelter earning 8.5% on equity would normally sit. So the year's range — 1.45 to 3.81 times book, a 2.6-fold span on a book value that barely moved — is a fairly clean picture of the control premium arriving and leaving.
Today's 2.30 sits between the two. Some of the premium has gone. Not all of it.
The strongest case against me is that the earnings recovery is structural rather than cyclical. Operating income of ₩1,333.2bn in six months, against ₩1,231.9bn in the whole of FY2025, is not a rounding move, and the improvement came from more than metal prices passing through — I've written separately about the margin expansion underneath it.
If the treatment-charge environment has genuinely reset higher, and if the battery-materials and copper-foil businesses eventually contribute, then the historical 8-10% return ceiling was a feature of the old business mix rather than a law. A company earning 15% on equity would deserve something close to 2.5 times book, and the multiple would be right for a reason nobody has priced yet.
The second point is that book value is understated. Equity rose ₩3,593.3bn in FY2025 on new issuance, and inventories carried at cost in a rising metals market are worth more than they're carried at. A smelter's book is not a fair-value figure.
Both of those are real. Neither has shown up in the return on equity yet, and the return on equity is the only number that reconciles the price to the business.
The obvious one is that the control fight doesn't resolve on 9 September. If item three — the audit-committee seat contested between the company's nominee and the one proposed by Young Poong and its allies — goes narrowly either way, the premium doesn't drain, it persists into the next round. In that case my whole framework mistimes rather than misjudges.
The second is metal prices. Trailing revenue is ₩21,374.3bn, up enormously, and a large part of that is the price of zinc and silver rather than volume. If those hold, earnings hold and the 19 times forward multiple compresses on its own. If they fall, the return on equity goes back toward 4% and 2.30 times book becomes indefensible fast.
The vote on 9 September, and specifically the margin. A decisive result removes the reason anyone pays above the earnings multiple, and the price then has to be justified by the treatment charge alone.
After that, the third-quarter return on equity. Two consecutive quarters near ₩350bn of net income would put the annualised figure at 12.2%, the best this company has ever managed. A third would start to make the case that the ceiling moved. Anything materially below ₩300bn and the run-rate argument collapses back to the trailing 8.48%, which does not support 2.30 times book on any reading of this company's eleven-year record.
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.