At the end of December 2025, HD Korea Shipbuilding & Offshore Engineering Co., Ltd. (KRX:009540) held ₩3,742.0bn of cash and equivalents. Six months later it held ₩10,996.0bn. The balance nearly tripled in half a year, and the company earned ₩2,735.6bn of net income across that period.
Profit did not put ₩7.3tn in the bank. Customers did.
Operating cash flow for the first half, on the cumulative basis Korean interim statements use, was ₩4,379.5bn — 1.6 times net income. For a manufacturer that is unusual. For a shipbuilder taking record orders it is exactly what should happen.
Ships are paid for in instalments tied to milestones: contract signature, steel cutting, keel laying, launch, delivery. A yard signing a lot of new contracts collects the early instalments long before it books the revenue, because revenue is recognized as construction progresses while cash arrives on the schedule the contract sets. Order intake in the first half of 2026 was $16.38bn, which the company put at 96.2% of its full-year target achieved in six months. That order intake is the cash.
The same effect ran in reverse a few years ago. FY2019 operating cash flow was negative ₩1,464.1bn and FY2020 negative ₩90.0bn, when the order book was thin and the yards were consuming advances collected earlier. Cash flow at a shipbuilder leads the income statement by about two years in both directions, and reading it as a quality-of-earnings signal in either year would be a mistake.
The liability side makes the point more clearly than the asset side. Total liabilities were ₩27,800.9bn at the end of June. Of that, ₩24,882.9bn is current and only ₩2,918.1bn is non-current.
A company with ₩18,923.9bn of equity and less than ₩3tn of long-term liabilities is not a leveraged company in any ordinary sense. What it has is an enormous short-term liability stack, and in shipbuilding that stack is mostly contract liabilities — money received against work not yet performed — plus trade payables to a large subcontractor and steel-supplier base.
That distinction matters for how you read the cash. Money held against contract liabilities is not free cash. It is a customer's money that the yard will consume as it builds. If orders stop, the cash drains without any deterioration in profitability, because the advances keep being spent while new ones stop arriving. That is precisely what happened in 2019 and 2020.
The consolidated summary does not split current liabilities into contract liabilities and payables, which is the single most useful disclosure for anyone trying to size how much of the ₩11tn is genuinely the company's. It is in the notes to the half-year report, and it is worth the trip.
Two other flows in the half deserve attention.
Investing activities produced a net inflow of ₩1,422.2bn, while purchases of property, plant and equipment were only ₩380.4bn. So roughly ₩1.8tn came back out of deposits and financial assets. Financing activities produced a net inflow of ₩1,345.8bn, and that is after paying ₩783.0bn of dividends, which implies something on the order of ₩2tn of net new borrowing.
A company sitting on ₩11tn of cash that also draws down financial assets and raises debt is doing one of a few things: refinancing at the subsidiary level, pre-funding something it has not announced, or managing currency exposure by matching dollar liabilities against dollar receivables. Shipbuilders do the last one routinely, since contracts are priced in dollars and costs are largely in won. None of those is alarming. All of them mean the headline cash figure overstates net liquidity, and the company should be asked which it is.
This is the part of HD Korea Shipbuilding's statements that most deserves a warning label.
In FY2024 the company reported ₩1,434.1bn of operating income, ₩1,877.5bn of finance income and ₩5,179.0bn of finance costs. Add those three and you get a pretax loss of about ₩1,867bn. The reported pretax figure was positive ₩1,822.7bn. Roughly ₩3.7tn sits between them, in items the summary income statement does not name.
The quarterly series is just as violent. Finance income was ₩2,599.8bn in the second quarter of 2025 and negative ₩140.3bn in the fourth. Finance costs were ₩3,144.4bn in the fourth quarter of 2024 and ₩667.7bn a year later. These are currency and derivative movements on a book of dollar-denominated contracts hedged years forward, and they can be larger than the operating result in any given quarter.
The practical consequence: net income at this company is close to useless as a quarterly signal. Operating income, which was ₩1,645.1bn in the second quarter on an 18.4% margin, is the number that reflects the business. Anyone valuing the shares on a trailing price-to-earnings multiple is valuing a currency position as much as a shipyard.
Dividends paid were ₩3.1bn in FY2023 and ₩1.0bn in FY2024 — nothing, effectively, after the loss years. FY2025 brought ₩715.3bn. The first half of 2026 alone shows ₩783.0bn.
That is a genuine capital-return restart, and it is being funded from a position of strength: retained earnings reached ₩18,158.8bn at the end of June against ₩14,300.4bn at the end of FY2022. A holding company whose subsidiaries are all generating cash simultaneously, for the first time since the last cycle, has a straightforward decision to make about what to do with it.
The caution is the same as with the cash. Paying dividends out of a balance sheet inflated by customer advances is fine while orders keep coming and dangerous if they stop. The current backlog — more than 500 vessels, roughly three and a half years of work extending into 2029 and 2030 — makes that a distant concern rather than a near one. But it is the concern.
The strongest counter is that I am describing a normal shipbuilding upcycle balance sheet as though it were a puzzle. Advance payments funding working capital is how the industry has always worked, and a yard with three and a half years of backlog has more visibility than almost any industrial company. Nobody worries about a defence contractor holding progress payments.
The FX volatility argument also cuts the other way. A Korean shipbuilder that hedges dollar receivables years forward is doing the responsible thing, and the accounting noise is the price of removing the actual economic risk. The alternative — leaving a multi-year dollar book unhedged — would produce real losses rather than accounting ones.
And the low long-term debt is genuinely reassuring. With ₩2,918.1bn of non-current liabilities against ₩18,923.9bn of equity, this company could absorb a bad cycle without a financing problem, which is more than it could say in 2021.
The contract liabilities line in the third-quarter balance sheet notes. If it has grown roughly in step with the cash, then the ₩11tn is customer money and the right way to think about it is as deferred revenue with a matching obligation. If contract liabilities are flat and cash still climbed, then the company is genuinely accumulating its own money and the capital allocation question becomes live.
Second, whether order intake in the second half slows now that 96% of the annual target was booked in the first. A yard that stops signing while continuing to build starts consuming advances rather than collecting them, and the cash line turns first — well before anything shows up in reported profit.
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.