S-Oil: profit up 23-fold, but the whole gain rests on refining margin
Contents
S-Oil reported third-quarter 2026 results. Revenue rose 42.2% year on year to KRW 11,968.7bn, EBITDA by 287.6% to KRW 1,667.9bn and net profit 23.7-fold to KRW 1,435.6bn. The EBITDA margin climbed from 5.1% to 13.9%, and that is the main source of profit. Our conclusion: rather attractive – EV/EBITDA of 4.0 against its own three-year average of 11.4, a 65% discount to history. Our view: the share is rather attractive at the current price; no portal model for this share; EV/EBITDA 4.0 against a 3-year average of 11.4, a 65% discount to its history.
Our view: is there upside
Our view in numbers
| Metric | Our read |
|---|---|
| Verdict | Rather attractive |
| Upside on the portal model | no portal model for this share |
| EV/EBITDA now and 3-year average | 4.0 vs 11.4 – a 65% discount to its history |
| Dividend yield, 12m | 1.1% |
| Net debt / EBITDA | 1.17 – low leverage |
| Strongest case for | EV/EBITDA of 4.0 against its own three-year average of 11.4 is a 65% discount that leaves room if the margin holds even partially. |
| Strongest case against | The 13.9% EBITDA margin against 5.1% a year earlier is a refining spread, and its return to normal would cut EBITDA fourfold. |
| What would change the view | Holding the EBITDA margin above 10% for two more quarters would raise the verdict a step; a fall below 7% would lower it. |
Key takeaways
— Revenue added 42.2% year on year, and growth accelerated from 40.9% a quarter earlier
— EBITDA rose almost fourfold, and the margin climbed from 5.1% to 13.9%
— Net profit of KRW 1,435.6bn is 23.7 times the year-ago figure
— Leverage of 1.17x EBITDA LTM is a level, not a trend
— Dividend yield of 1.1% against profit three times the year-ago level
— EV/EBITDA of 4.0 against its own average of 11.4 – a 65% discount to history
Key figures, KRW bn
| Metric | Q3 2025 | Q3 2026 | Change |
|---|---|---|---|
| Revenue | 8 415 | 11 969 | +42.2% |
| EBITDA | 430 | 1 668 | +287.6% |
| Operating profit | 229 | 1 668 | +627.6% |
| Net profit | 63.2 | 1 436 | +2171.3% |
| Operating cash flow | 891 | — | — |
| Capex | 919 | — | — |
| EBITDA margin | 5.1% | 13.9% | +8.8 pp |
| Net margin | 0.8% | 12.0% | +11.2 pp |
Revenue added 42.2% year on year, and growth accelerated from 40.9% a quarter earlier
In the third quarter of 2026 revenue came to KRW 11,968.7bn, 42.2% more than a year earlier. A quarter earlier growth was 40.9%, an acceleration of 1.3 percentage points. In absolute terms this was the best quarter in the available history: KRW 11,968.7bn against KRW 11,343.5bn in the second quarter.
The acceleration runs off a low base: a year ago revenue was falling 4.8%, and in the second quarter of 2025 it was falling 15.9%. Comparison with a weak period flatters the rate, and that matters for judging the durability of growth.
The driver of revenue is prices, not volumes: in refining, revenue follows the cost of feedstock and products almost entirely. The facts give no utilisation data, so we do not claim volumes as the cause.
Revenue and EBITDA by quarter
| Period | Measure | bn KRW |
|---|---|---|
| Q4 2024 | Revenue | 8917.081 |
| Q1 2025 | Revenue | 8990.533 |
| Q2 2025 | Revenue | 8048.504 |
| Q3 2025 | Revenue | 8415.367 |
| Q4 2025 | Revenue | 8792.553 |
| Q1 2026 | Revenue | 8942.689 |
| Q2 2026 | Revenue | 11343.488 |
| Q3 2026 | Revenue | 11968.692 |
| Q4 2024 | EBITDA | 414.9 |
| Q1 2025 | EBITDA | 173.435 |
| Q2 2025 | EBITDA | −141.789 |
| Q3 2025 | EBITDA | 430.326 |
| Q4 2025 | EBITDA | 580.657 |
| Q1 2026 | EBITDA | 1442.074 |
| Q2 2026 | EBITDA | 1177.693 |
| Q3 2026 | EBITDA | 1667.875 |
EBITDA margin
| Period | Measure | % |
|---|---|---|
| Q4 2024 | EBITDA margin | 4.7 |
| Q1 2025 | EBITDA margin | 1.9 |
| Q2 2025 | EBITDA margin | −1.8 |
| Q3 2025 | EBITDA margin | 5.1 |
| Q4 2025 | EBITDA margin | 6.6 |
| Q1 2026 | EBITDA margin | 16.1 |
| Q2 2026 | EBITDA margin | 10.4 |
| Q3 2026 | EBITDA margin | 13.9 |
EBITDA rose almost fourfold, and the margin climbed from 5.1% to 13.9%
EBITDA for the third quarter of 2026 was KRW 1,667.9bn against KRW 430.3bn a year earlier, up 287.6%. The EBITDA margin rose from 5.1% to 13.9%, or 8.8 percentage points. Operating profit equals EBITDA at KRW 1,667.9bn; depreciation is not broken out for the reported period.
The entire profit gain comes from margin, not scale: revenue rose 42.2%, EBITDA almost fourfold. That is the typical picture for a refiner when the spread between feedstock and products widens: each point of margin falls on the whole revenue line.
The durability of that margin is the main question. The facts give no refining-spread data, so we do not name the cause of its widening. But whether the quarter repeats depends on it.
Net profit by quarter
| Period | Measure | bn KRW |
|---|---|---|
| Q4 2024 | Net profit | −131.72 |
| Q1 2025 | Net profit | −44.558 |
| Q2 2025 | Net profit | −66.782 |
| Q3 2025 | Net profit | 63.206 |
| Q4 2025 | Net profit | 225.086 |
| Q1 2026 | Net profit | 720.991 |
| Q2 2026 | Net profit | 514.584 |
| Q3 2026 | Net profit | 1435.596 |
Net profit of KRW 1,435.6bn is 23.7 times the year-ago figure
Net profit for the third quarter of 2026 was KRW 1,435.6bn against KRW 63.2bn a year earlier, up 2171.3%. The net margin rose from 0.8% to 12.0%. Profit almost caught up with EBITDA: the gap between them is KRW 232.3bn, which is interest and tax.
The year-ago base was almost nil: KRW 63.2bn of net profit on revenue of KRW 8,415.4bn. So the 23.7-fold multiple reflects the weakness of last year as much as the strength of this quarter.
The facts name no one-off items, so we do not attribute profit to asset sales or revaluation. It was all earned on operating margin, which makes its quality higher than profit assembled from one-off gains.
Net debt
at each reporting date, current net debt / EBITDA (LTM) = 1.17
| Period | Measure | bn KRW |
|---|---|---|
| Q4 2024 | Net debt | 5860.18 |
| Q1 2025 | Net debt | 5663.329 |
| Q2 2025 | Net debt | 6258.672 |
| Q3 2025 | Net debt | 6380.218 |
| Q4 2025 | Net debt | 5700.681 |
| Q1 2026 | Net debt | 6802.659 |
| Q2 2026 | Net debt | 7531.631 |
Leverage of 1.17x EBITDA LTM is a level, not a trend
Net debt at the latest annual balance-sheet date is KRW 5,700.7bn, or 1.17x EBITDA over the last twelve months. That is moderate for a refiner: EBITDA LTM is KRW 4,868.3bn, and debt is covered by it in a little over a year.
Comparison with a prior level is not possible: the facts give no earlier value for the ratio. Absolute net debt did move – from KRW 5,860.2bn at end-2024 to KRW 5,700.7bn at end-2025, a decrease of KRW 159.5bn. The debt build in the second and third quarters of 2026 is not shown as a separate line in the facts.
Interest burden shows indirectly: the gap between EBITDA and net profit for the quarter is KRW 232.3bn. On that base debt service does not look like a threat, but there is little cushion if the margin falls.
Valuation vs its own history
now vs the 3-year average
| Period | Measure | × |
|---|---|---|
| EV/EBITDA | Now | 3.9623933408773784 |
| EV/EBITDA | 3-year average | 11.444848578209058 |
Dividend yield of 1.1% against profit three times the year-ago level
Dividend yield over the last twelve months is 1.1% on the current price. On a market capitalisation of KRW 11,758.5bn that is about KRW 131.6bn of payouts. For a company with net profit LTM of KRW 2,896.3bn the payout ratio works out modest – around 4.5%.
Our estimate: if the payout stays at this level, the 2026 dividend could yield about 1.5% on profit that has grown for three quarters running. The estimate rests on two assumptions – the payout ratio holding and no large one-off charges cutting the profit base.
A yield of 1.1% is below the policy rate and gives the holder no cushion. The dividend here is a pleasant addition to profit growth, not the main argument. A fall in the refining margin could cut the payout: the quarter's profit depends almost entirely on it.
EV/EBITDA of 4.0 against its own average of 11.4 – a 65% discount to history
EV/EBITDA over the last twelve months is 4.0; the three-year average of that multiple is 11.4. The share trades 65% below its own history. P/E LTM is 4.1, which corresponds to net profit LTM of KRW 2,896.3bn on a market capitalisation of KRW 11,758.5bn.
The discount is explainable: the market does not believe the refining margin will hold at 13.9%. If EBITDA returns to the year-ago level, the multiple at the current price would rise roughly fourfold to 16, and the discount would turn into a premium. That is the main valuation risk.
ROE of 53.4% on such profit and low debt looks exceptionally high, but it is calculated at the peak of the margin. A sustainable ROE will be lower, and it cannot be taken as the norm.
Valuation on the latest reported figures
| Metric | Value |
|---|---|
| Market cap | 11 758 bn KRW |
| P/E (LTM) | 4.1 |
| EV/EBITDA (LTM) | 4.0 |
| P/B | 1.32 |
| Net debt / EBITDA (LTM) | 1.17 |
| ROE | 53.4% |
| Dividend yield (12m) | 1.1% |
| EV/EBITDA, 3-year average | 11.4 |
Bottom line
The quarter was strong: revenue +42.2%, EBITDA almost fourfold, net profit 23.7-fold, and all of it earned on operating margin rather than one-off gains. The weak spot is the source: a margin of 13.9% against 5.1% a year earlier is a refining spread, not a durable advantage. Debt at 1.17x EBITDA LTM is moderate, the 1.1% dividend modest. The holder's question is simple: will the margin hold at least half its current level. Our conclusion: rather attractive – EV/EBITDA of 4.0 against its own three-year average of 11.4, a 65% discount to history. Our view: the share is rather attractive at the current price; no portal model for this share; EV/EBITDA 4.0 against a 3-year average of 11.4, a 65% discount to its history.