S-Oil Corporation (KRX:010950) reported ₩1,231.1bn of operating income in the first quarter of 2026 and ₩965.0bn in the second. Against a first half of 2025 that produced an operating loss of ₩365.5bn, that is one of the sharper reversals a large Korean industrial has posted in recent years.
The cash flow statement for the same six months shows ₩567.0bn going out the door rather than coming in.
Both facts come from the half-year report filed with DART on 14 August, and reconciling them is the most useful thing an investor can do with S-Oil's numbers right now.
Look at the balance sheet either side of the period. Inventories were ₩4,164.4bn at the end of December 2025. By the end of June 2026 they were ₩6,611.4bn. That is ₩2,447.0bn of additional stock, a 59% increase in six months. Trade receivables went from ₩2,005.1bn to ₩2,805.6bn, another ₩800.6bn.
Add those and working capital consumed roughly ₩3.2tn while the income statement was producing ₩2.2tn. That more than accounts for the negative operating cash flow, and it means the profit exists on paper long before it exists in the bank.
Refiners work this way, and a US investor who has followed Valero or Phillips 66 through a crude rally will recognize the pattern. A refinery buys crude, holds it for weeks in tanks and pipes, processes it, and sells the products later. When crude prices rise across that holding period, the cost of goods sold reflects barrels bought at yesterday's cheaper price while revenue reflects today's. The difference shows up as margin. It is real in the sense that it is a genuine gain on inventory held. It is not a repeatable operating margin, and it reverses when prices fall.
The reverse case is on file too. S-Oil's gross profit in the second quarter of 2025 was negative ₩150.8bn — the company sold products for less than the accounted cost of the crude that made them. FY2024 gross profit was ₩1,225.9bn on ₩36,637.0bn of revenue, a 3.3% gross margin. Those were the quarters when the inventory effect ran the other way.
Read the interim cash flow figures carefully, because Korean filers present them cumulatively rather than as discrete quarters. Cumulative operating cash flow was negative ₩523.9bn at the end of March and negative ₩567.0bn at the end of June. The second quarter on its own, therefore, was roughly breakeven — an outflow of about ₩43bn.
That matters. The cash drain was concentrated in the first quarter, and the second quarter had close to stopped it even while inventories kept climbing. Anyone reading a headline that says S-Oil burned cash all half should know the trend inside the half was improving.
The income statement across the same two quarters moved the other way. First quarter: revenue ₩8,942.7bn, gross profit ₩1,449.8bn, a 16.2% gross margin. Second quarter: revenue ₩11,343.5bn, gross profit ₩1,179.0bn, a 10.4% margin. Revenue jumped 26.8% while gross profit fell 18.7%.
That is what a fading inventory tailwind looks like. Volume and price both rose in the second quarter, but the spread between what the crude cost and what the products fetched narrowed. Operating income followed: ₩1,231.1bn down to ₩965.0bn. So profit quality and cash quality moved in opposite directions, which is exactly what you would expect if the first quarter's margin was substantially a holding gain.
Management's own framing supports part of this. The second quarter was described on the earnings call as driven by strong refining margins and record lube base oil spreads, with the lubricants business alone contributing ₩477.4bn of operating profit. That figure is worth sitting with. Lubricants, the smallest of S-Oil's three segments, produced roughly half the company's operating income in the quarter. The refining business produced the other half.
Total liabilities reached ₩20,156.5bn at the end of June against ₩10,085.6bn of equity. That is a ratio of 2.00. At the end of FY2022 it was ₩11,129.7bn against ₩8,483.5bn, or 1.31. Current liabilities of ₩14,109.7bn now exceed current assets of ₩11,497.1bn by ₩2,612.6bn.
Some of that is the Shaheen petrochemical project, which has been absorbing capital for three years. Some of it is the inventory build itself, which has to be funded. Financing activities brought in a cumulative ₩1,183.4bn across the half, against ₩1,235.6bn of net income.
Borrowing while profitable is not a crisis. Borrowing while profitable, with negative operating cash flow and a working capital position that has gone negative, is a different setup from the one the headline earnings suggest. And the Bank of Korea raised its base rate to 3.00% on 27 August, its second consecutive increase, which makes the cost of carrying that balance sheet higher than it was when the borrowing was arranged.
Cash and equivalents fell from ₩1,834.7bn at year-end to ₩1,550.2bn.
The interim dividend declared on 10 August gives a clean read on the share price without needing a quote. The board set ₩800 per share and stated a 0.6% yield against the week's average price, which implies a common share around ₩133,000. On roughly 112.6m common shares, derived from the ₩90.1bn common dividend total in the same filing, that puts common equity near ₩15tn (~$11bn at a rough ₩1,400 to the dollar).
Against equity of ₩10,085.6bn that is about 1.5 times book. Against trailing four-quarter net income near ₩1,523.9bn it is roughly 10 times earnings. Neither number is demanding for a refiner in an up-cycle. Neither is cheap for one whose profit is partly an inventory effect and whose largest capital project has not started producing.
The strongest counter is that a mid-year inventory build ahead of a major startup is exactly what you would expect and says nothing bad about margins. S-Oil's Shaheen complex is scheduled to begin commissioning in the fourth quarter of 2026 with commercial operation targeted for early 2027. A plant of that size needs feedstock in the tanks before it runs, and buying that feedstock is a one-time cash outflow that converts to product later. Read that way, the ₩2.4tn inventory increase is pre-funding a startup rather than accumulating unsold refined product.
I find that partly convincing and cannot separate the two from the consolidated statements. The half-year report gives a single inventory number, not a split between crude, work in process, finished product and Shaheen feedstock. If the segment detail eventually shows most of the increase was feedstock, my concern shrinks considerably.
The lube performance is also genuinely good rather than accounting-driven. Base oil spreads are a physical margin between crude and a specialty product, and a record quarter there reflects a tight market S-Oil is well positioned in. That part of the ₩965.0bn is high quality.
The third-quarter cash flow statement, specifically the change in inventories line. If inventories flatten or fall while operating income stays above ₩700bn, the profit was real margin and the first half was a working capital timing effect. If the cumulative operating cash flow line is still negative nine months into the year with inventories higher again, then S-Oil is funding an increasingly large stock position with borrowed money into a rising rate environment, and the earnings quality question becomes a balance sheet question.
The second thing is the gross margin trend. It went 16.2% then 10.4%. A third-quarter figure near 10% or above suggests a durable spread. Materially below that, with crude having stopped rising, would confirm how much of the first quarter was inventory gain.
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.