Jusung Engineering Co., Ltd. (KOSDAQ:036930) closed at ₩173,700 on September 2. Its 52-week low is ₩26,450. That is a 6.6-fold move inside twelve months, and it has taken the market capitalization to ₩8.07tn — about $5.9bn at roughly 1,370 won per dollar, which is a rough rate.
Over the same stretch, the business shrank. First-half revenue was ₩114.8bn against ₩199.6bn a year earlier, down 42%. Operating profit went from ₩40.5bn to a loss of ₩5.6bn. The March quarter lost ₩7.0bn at the operating line; the June quarter scraped back to ₩1.4bn.
So the stock and the income statement are describing different companies. Working out which one is describing the future is the entire question, and it turns on a piece of Korean market plumbing that most US readers won't have encountered.
Take the accounts at face value first. Revenue peaked at ₩437.9bn in 2022, fell to ₩284.7bn in 2023, recovered to ₩409.4bn in 2024, then dropped to ₩310.7bn in 2025. Operating profit ran ₩123.9bn, ₩28.9bn, ₩97.2bn, ₩31.3bn across those four years. That is not a growth company. It is a supplier whose results are a direct function of when two customers press the button on a fab.
The trailing four quarters make it starker. Revenue of ₩225.8bn is the weakest run rate since the 2020 trough. Trailing operating income is negative ₩14.8bn. Trailing net income of ₩7.2bn survives only because finance income and a tax credit sit below the operating line.
Cash has been leaving too. Cash and equivalents were ₩250.7bn at the end of 2024 and ₩123.0bn in June. Operating cash flow was negative in three of the last four quarters and negative ₩30.7bn for all of 2025.
If you priced this company on what it has reported, you would not arrive anywhere near ₩8tn.
Look at where the loss comes from. Gross margin in the June quarter was 50.5%, and 57.1% for full-year 2025. Jusung is not losing money because its tools are cheap to build. It is losing money because selling, general and administrative expense — which carries most of its development spending — ran ₩64.5bn in the first half against ₩114.8bn of revenue. Fifty-six cents of every revenue won.
That cost base barely moves with volume. In 2024, SG&A was ₩151.8bn on ₩409.4bn of revenue. In 2025, ₩146.1bn on ₩310.7bn. Revenue fell nearly a quarter; costs fell 4%.
Which means the operating leverage is violent in both directions, and the company has already demonstrated it. In 2020 Jusung did ₩118.5bn of revenue and lost ₩25.0bn at the operating line. In 2021 it did ₩377.3bn and earned ₩102.6bn — a 27% operating margin, from a loss, in one year. Nothing about the cost structure changed. The orders arrived.
That single pair of years is the whole bull case, and it is a legitimate one. A company with 55% gross margins and a fixed cost base doesn't need a good market. It needs one good year.
Here's the structural point, and it's the reason the filed quarterly numbers are a poor guide to this stock.
Semiconductor equipment revenue is generally recognized when the customer accepts the tool, not when the order is signed. For fab equipment that gap runs several quarters. So a 2026 order shows up as 2027 revenue, and the income statement you are reading always describes a decision made a year ago.
Korean disclosure rules partly fix this in a way US rules do not. A listed Korean company must file a specific disclosure — a single sales or supply contract notice — whenever one contract exceeds a set share of the prior year's revenue. For a KOSDAQ company the threshold is 10%. On Jusung's 2025 revenue of ₩310.7bn, that means any order above roughly ₩31bn becomes a public filing on the day it is signed, with the counterparty sometimes named and the delivery period stated.
A US supplier of comparable size discloses a material contract if it's material, on a timeline of its own choosing, often in a quarterly filing months later. Korea makes the order itself the disclosure event. For a company like this one, the running tape of those filings is a better instrument than the earnings release.
What the trade press has picked up in 2026 fits the pattern: Jusung said in May it had supplied atomic layer growth equipment to a global memory maker, describing it as a world first, and its shares moved again in August when SK hynix laid out fab expansion at Yongin and Cheongju. Reported order backlog figures around ₩233.6bn have circulated, though the standardized quarterly financial data doesn't carry a backlog line, so that isn't a number you can pull from the filings yourself.
Against equity of ₩591.9bn, ₩8.07tn is 13.6 times book. Against trailing revenue of ₩225.8bn, roughly 36 times sales. Trailing earnings of ₩7.2bn produce a P/E above 1,000, which is a way of saying the ratio has stopped functioning.
Even on optimistic sell-side estimates circulating for 2026 — around ₩450bn of revenue and ₩121bn of operating profit — the shares would sit near 18 times revenue and somewhere in the sixties on operating profit. Those are not estimates I can verify against anything filed, and the first half at ₩114.8bn makes ₩450bn for the year a demanding second half.
Put differently: the market is not paying for 2026. It is paying for the 2027 that a 2026 order book would produce, on the template of 2021.
Customer concentration is the obvious one. Orders come mainly from Korean memory makers, with some Greater China memory exposure. Two customers deciding to push a process transition out two quarters is enough to make the current run rate persist into next year, and neither of them announces that decision in advance.
The China exposure carries its own problem. Equipment sales into Chinese memory have been subject to shifting export controls, and a small Korean supplier has no ability to influence how those rules land.
The 52-week range itself is a risk statistic. A stock that traded at ₩26,450 within the last year and now trades at ₩173,700 has a shareholder base that arrived at very different prices, and KOSDAQ small caps can retrace a move like this quickly. The high is ₩250,500, so the shares are already 31% off it.
And the cash burn. Non-current liabilities have sat near ₩229bn for three years, which is a real fixed obligation against ₩123bn of cash and negative operating cash flow. Jusung is not fragile — equity is ₩591.9bn — but another four quarters like the last four would put financing back on the table.
Watch DART, not the earnings calendar. Individual supply-contract filings above roughly ₩31bn are the earliest hard evidence that the 2027 revenue exists, and they arrive on the day the contract is signed rather than in November.
In the results themselves, the line to check is the September quarter's revenue against ₩58.8bn a year earlier. Two consecutive quarters of sequential growth would mean acceptances have restarted; a third quarter near ₩55bn would mean the order cycle everyone is pricing hasn't reached the shipping dock yet.
Then SG&A. If it stays near ₩30bn a quarter while revenue climbs, the 2021 operating margin repeats almost mechanically. If the company spends the upturn on development the way it spent 2024, the leverage that justifies this multiple never shows up in the profit line at all.
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.