KEPCO Q2: Analysts Saw the Cut Coming — Just Not Half of It

Not financial advice. TheGatBull explains Korean market structure for global investors; it does not recommend trades.

Korea Electric Power Corporation (KRX: 015760 / NYSE: KEP) filed its second-quarter provisional results after Wednesday’s close. Consolidated operating profit came in at ₩1.13 trillion (about $795 million) — down 47.2% from a year earlier and 42.5% below the analyst consensus of ₩1.96 trillion (about $1.38 billion), a brokerage average as compiled and reported by Hankyung. All won figures here are converted at ₩1,419.4 per dollar, the Seoul 15:30 close on August 13.

The decline itself was not a surprise. Brokerages had been pencilling in full-year profit declines of 15–20% since June. What nobody pencilled in was the size of this quarter. KEPCO delivered 57% of what the market had expected for it.

The next day, as the KOSPI surged 3.56%, KEPCO fell 6.63% to ₩33,100 (about $23.30). Eight of the ten brokerages that published notes cut their price targets.

This piece takes that ₩836 billion gap between consensus and reality apart, using the filing’s own numbers.

What this means for EWY and KEP holders. KEP trades directly on the NYSE as an ADR, so a US account can hold it outright. If you own EWY, your exposure is minimal — 0.37% of the fund (iShares holdings, Aug 6). This earnings report will not move EWY. But it is the cleanest illustration available of how a utility gets priced in a country where the electricity tariff — the retail rate customers pay, not a trade duty — is a political decision.

Under the Gat. Surprises in US utility earnings come from demand. KEPCO’s surprises always come from cost — because the switch on the revenue side is held by the government, not the company.

1. What the filing said

August 12 was the peak day of Korea’s earnings calendar — 14 filings in a single session, per KIND. We mapped the August schedule here.

The table below is taken from KEPCO’s provisional results filing with Korea’s Financial Supervisory Service (DART), dated August 12. These are provisional figures prepared before the external auditor’s review and may change. The operating detail that follows — fuel costs, generation mix, sales volume — comes from the company’s briefing the same day, as reported by Korean media.

Consolidated	2Q26	vs 1Q26	vs 2Q25
Revenue	₩21.92tn (~$15.4bn)	−10.2%	−0.1%
Operating profit	₩1.13tn (~$795m)	−70.2%	−47.2%
Net profit	₩277.5bn	−89.0%	−76.4%

For the first half: revenue ₩46.32tn (+0.3%), operating profit ₩4.91tn (~$3.46bn), down 16.6%, net profit ₩2.80tn (−21.0%).

Two things in that table deserve more attention than the headline.

First, revenue barely moved while operating profit halved. That asymmetry is the entire structure of this company in one line.

Second, the sequential drop is worse than the year-on-year one. Operating profit fell 70.2% from the first quarter — a figure that went largely unreported. The second quarter, April through June, sits between Korea’s heating and cooling seasons; even so, a 70% sequential fall is steep.

Per the company’s briefing, electricity sales volume actually fell 0.6% in the first half, and the average selling price was roughly flat. In the country building some of the world’s most power-hungry AI infrastructure, volume went down — not because demand is absent, but because the heat-wave season had not yet reached the books.

One more line in the same filing matters. KEPCO reports twice: consolidated (the parent plus its generation subsidiaries) and separate-entity (the parent alone — the grid and retail business that actually sells you electricity at the government-set rate). The separate-entity numbers are worse: first-half operating profit of ₩2.12tn, down 25.7%, against −16.6% consolidated. The part of the company most exposed to the frozen rate is the part hurting most.

2. The gap has a name: coal

Fill the space between the ₩1.96tn consensus and the ₩1.13tn result with the cost lines from the company’s briefing, and most of it is fuel.

• Subsidiary fuel costs +₩818bn (+8.8%) — first-half fuel bill ₩10.14tn (~$7.1bn)
• The price of imported thermal coal rose to $128.2 a tonne — up 24.3% year on year
• Power purchase costs −₩151bn (−0.9%); other operating costs +₩453bn (+3.3%)

And here is the irony of the quarter. Fuel costs rose for two reasons stacked on top of each other: coal got more expensive, and KEPCO burned more of it. The company attributes the increase to preemptive market-stabilization measures, including raising the output caps on coal generation (translated from the Korean).

Put plainly: during an LNG price spike driven by Middle East conflict, KEPCO leaned harder on relatively cheap coal to protect the electricity bill — and then that coal turned out to be 24% more expensive too. Buying less power from private LNG generators did trim power-purchase costs slightly. The defense worked; the price of the defensive weapon went up.

Under the Gat. There are very few lines on this income statement the company controls. The price of coal is set by international markets; the price of electricity is set by the Ministry of Economy and Finance. A KEPCO shareholder has essentially bought the spread between those two.

3. The mirror flipped

Thursday’s price action was textbook — and the exact inverse of a week earlier.

Session	KOSPI	Chipmakers (electrical & electronics)	Utilities (electricity & gas)
Aug 6 (selloff)	−4.58%	−7.49%	+0.73% — the shelter
Aug 13 (rally)	+3.56%	+3.16%	−5.25% — the laggard

KRX sector indices, regular-session closes. KEPCO itself traded down as much as 7.48% intraday before closing down 6.63%.

The target-price cuts piled up, all in won: KB ₩54,000 → ₩48,000; iM ₩53,000 → ₩48,000; Samsung Securities ₩63,000 → ₩52,000; NH ₩63,000 → ₩55,000. That is an average target of about ₩50,750 — still roughly 53% above Thursday’s ₩33,100 close. None changed their rating direction, and most kept “buy.” The Street cut how much it expects, not whether it expects.

What lifted the KOSPI 3.56% that day were the softer US July CPI print and a recovery in semiconductor sentiment: Samsung Electronics +4.89%, SK Hynix +5.92% (Seoul closes, KRX regular session). So KEPCO’s −6.63% cannot be blamed on the market. It was priced entirely off the filing.

4. What this actually says — the arithmetic of an administered price

This is why the piece exists. KEPCO’s costs are market prices. Its revenue is an administered price.

Coal and LNG are repriced by global markets in real time. The electricity tariff requires government approval, and this year it has been frozen under an inflation-control policy. iM Securities put it plainly in its note: given the government’s price-stability stance, a tariff increase this year is hard to expect. (An analyst view, not company guidance.)

Regulated US utilities face rate regulation too — but through contracts with state commissions that spell out fuel-cost pass-through mechanisms and allowed ROE bands. In Korea, central government inflation policy sits in that seat. Not a perfect parallel — we took that difference apart in our KEPCO vs Southern Company comparison. It is also why the path from AI power demand to utility re-rating, which worked in the US, has not repeated in Korea — see our piece on Korea’s AI power valuation gap.

On top of the tariff sits the balance sheet, in the company’s own figures: debt of ₩210.7tn (~$148bn), borrowings of ₩133tn (~$94bn), and daily interest expense of ₩11.5bn (~$8.1m) — the ledger left behind by years of selling power below cost during the Russia-Ukraine energy shock. KB Securities added the coming grid and generation capex tied to government megaprojects and concluded dividend capacity looks constrained — an analyst view, not company guidance. That lands directly on a dividend that only resumed last year after a four-year gap.

5. Next quarter: two invoices arrive together

The third quarter is KEPCO’s seasonal peak. This year’s record heat wave sends cooling demand into that quarter’s revenue.

But the cost side arrives in the same quarter. KB Securities expects the profit decline to continue through the second half on LNG price increases reflected with a lag from the third quarter (translated from the Korean).

Three things decide next quarter, and almost nothing else does:

1. Quarterly fuel unit costs (LNG and thermal coal)
2. Nuclear utilization rate — nuclear is KEPCO’s cheapest generation, so the share of the year its reactors actually run moves the fuel bill directly
3. Whether the tariff moves at all this year

Under the Gat. In the third quarter, the heat wave arrives as revenue and the delayed LNG settlement arrives as cost, in the same ledger. The question for next quarter is not “will they be profitable” — it is “which invoice is bigger.”

To put it plainly: KEPCO’s second quarter was an expected decline of an unexpected size. The gap has a name — coal — and the structure has a name too: an administered price. Here is the fact about the Korean market you cannot see without looking at this company. Even the strongest AI-power-demand narrative in the world does not automatically translate into utility profit in a country where the tariff is set by the government. Only an investor watching both sides of that spread — the international fuel market and the government’s tariff switch — can read this company’s next quarter.

FAQ

Why did KEPCO miss consensus by so much in Q2 2026?

Consolidated operating profit was ₩1.13tn against a consensus of ₩1.96tn, a 42.5% miss. Per the company’s briefing, the main driver was fuel: the price of imported thermal coal rose 24.3% year on year while KEPCO also burned more coal under market-stabilization measures, pushing fuel costs up 8.8%. With the electricity tariff frozen, none of that could be passed through.

Why was revenue nearly flat while profit halved?

The average selling price was roughly unchanged (tariff freeze) and sales volume fell 0.6%. The second quarter covers April to June, before Korea’s summer cooling demand — that arrives in the third-quarter books.

Can US investors buy KEPCO?

Yes. KEPCO trades directly on the NYSE as an ADR under the ticker KEP. It is also held in EWY, but at just 0.37% of the fund (iShares holdings, Aug 6).

What about the dividend?

KEPCO resumed dividends last year after a four-year suspension (₩1,540 per share for FY2025 — about $1.08, a 4.7% yield against Thursday’s close). KB Securities judged dividend capacity constrained given grid investment burdens — an analyst estimate, not company guidance.

Is KEPCO profitable in 2026?

Yes, but shrinking. First-half consolidated operating profit was ₩4.91tn, down 16.6% year on year; net profit ₩2.80tn, down 21.0%. On a separate-entity basis, first-half operating profit fell 25.7%.

What is the next checkpoint?

August 14 is the statutory deadline for Korean half-year reports, which carry the confirmed financial statements. After that, third-quarter results — the quarter where heat-wave revenue and lagged LNG costs arrive together.

Sources

• KEPCO provisional consolidated results filing, DART (Aug 12, 2026) — primary source for the consolidated and separate-entity income statement figures
• Hankyung (Aug 13, 2026) — consensus estimate and brokerage target-price changes
• Newspim, BusinessPost (Aug 12, 2026) — company briefing detail on fuel costs, generation mix and sales volume
• Etoday (Aug 13, 2026) — KRX index, sector and flow data
• Yahoo Finance chart API — KEPCO (015760.KS), KOSPI (^KS11), KOSDAQ (^KQ11), Samsung Electronics (005930.KS), SK Hynix (000660.KS); KRX regular-session closes, retrieved Aug 13, 2026
• iShares MSCI South Korea ETF (EWY) — holdings as of Aug 6, 2026

This article is for information only and is not investment advice. All investment decisions and their consequences rest with the reader. The author holds no position in any security mentioned.

---