LS ELECTRIC Stock (010120): The Korean Grid-Equipment Maker Whose Biggest Customer Is Its Own US Subsidiary

Every English-language story about Korea’s grid-equipment boom is really a story about transformers being sold out. LS ELECTRIC sits at the centre of that story — a $23.5bn company whose order book grew 38.5% in six months. Then you open its own half-year filing and find its extra-high-voltage transformer plants ran at 66.5% and 69.9% of capacity, its gross margin went down, and 41.7% of its power-infrastructure sales went to a customer it owns outright.

🏢 What Is LS ELECTRIC, and What Does It Actually Sell?

LS ELECTRIC is the Korean maker of the equipment that sits between a power plant and a wall socket — circuit breakers, switchgear, meters, protective relays, extra-high-voltage transformers and HVDC converter gear — plus a smaller factory-automation business selling PLCs and inverters. It is not a utility, and it does not generate electricity. It sells the hardware that moves and protects it.

The company was established on 24 July 1974 and listed on the Korea Exchange in July 1994. It is headquartered at LS Tower in Anyang, just south of Seoul, with domestic plants in Cheongju (Chungbuk), Cheonan (Chungnam) and Busan. Registered capital is ₩150.0bn ($111.1M). As of 31 December 2025 the parent company employed 3,482 people with average tenure of 16.0 years. Co-CEOs are Koo Ja-kyun (chairman) and Chae Dae-seok. (Source: DART annual report FY2025, filed 18 Mar 2026, rcpNo 20260318001243.)

One naming point matters before you search for anything: the company was called LS산전 — LS Industrial Systems — until March 2020, when it renamed itself 엘에스일렉트릭 / LS ELECTRIC. Most English-language material older than 2020 is filed under the old name. There is an English Wikipedia entry for LS Electric if you want a neutral entity anchor, and the company runs an English corporate site.

The filing names its own customers plainly: for the power business, “domestic and overseas power authorities, exclusive dealers, data centres and energy-intensive industrial companies”; for automation, “automotive, electronics, semiconductor and display production-equipment makers, IT companies and smart-factory businesses.” The consolidated group contained 42 companies at 30 June 2026 — and it also contained 42 at end-2025 and 42 at end-2024, up from 28 at end-2023. The count has been flat for three reporting periods, which matters a great deal in a moment, because something else about the group’s internal trading did not stay flat. (Source: DART annual report FY2025, rcpNo 20260318001243, consolidated-subsidiary overview and the 52nd/51st/50th fiscal-year comparison table.)

📊 Where Does LS ELECTRIC’s Revenue Actually Come From?

About four-fifths of it is the power business, and you cannot read that share off the percentages the filing prints — because the filing states the power segment at 110.8% of total revenue. That is not an error. Korean segment disclosures report each segment before eliminating sales between group companies, then reconcile to consolidated revenue with a single negative line. Add the segments up and you get more than 100%.

Here is the honest version, with both denominators shown so you can see the gap.

Two things jump out. First, automation is 7.1% of segment revenue and 2.7% of segment pre-tax profit. Anyone describing LS ELECTRIC as a diversified power-and-automation company is describing a rounding error: on FY2025 numbers the power segment produced 94.2% of segment pre-tax profit. Second, the elimination line is growing much faster than revenue — from 18.5% of consolidated revenue in FY2024 to 28.5% in FY2025 to 35.5% in H1 2026. Sales inside the group are nearly doubling as a share of the business. We will come back to why.

Geographically, the filing gives a split that it explicitly labels as before intra-group elimination — its own note reads “the above financial information is stated before the elimination of internal transactions.” On FY2025 figures totalling $4,727M, Korea was $2,017M, North America $1,391M, other Asia $735M, Europe $286M, China $217M and other $80M. The standout is North America, which grew 44.5% year-on-year (from $963M), against 10.0% for Korea.

Treat that 44.5% as a gross figure, not a demand figure. The pre-elimination North America line is exactly the line that carries the parent’s sales to its own US subsidiary — the subject of the next-but-one section — so it cannot be read as end-market growth. The parent-only (separate) accounts give a second reading of the same market on a narrower basis, and it points the same way: North America revenue of $654M in FY2025 against $436M in FY2024, up 49.95%. Both bases are gross of the intercompany billing. Neither is evidence about the end customer; both are evidence that a great deal more product is moving into North America. (Source: DART annual report FY2025, rcpNo 20260318001243 — consolidated and separate segment notes, “revenue by region”. Won amounts of ₩6,383.0bn / ₩2,724.1bn / ₩1,879.1bn / ₩993.1bn / ₩385.6bn / ₩293.4bn / ₩107.8bn and ₩1,300.1bn consolidated; ₩882,949mn and ₩588,829mn separate.)

The filing also states there was no single customer above 10% of consolidated revenue in FY2025 or FY2024 — a claim we are going to test directly.

🗓️ How LS ELECTRIC Got Here: A Short History

Three things rebuilt this company: a 2022 reorganisation that put its Americas businesses under a single US holding entity, a 2024 acquisition that bought it a transformer factory, and roughly ₩100.8bn of capacity added at Busan in 2025 for extra-high-voltage transformers. The order wins followed the factories, not the other way round.

🌍 Is LS ELECTRIC Actually a Global Leader in Grid Equipment?

It is the dominant player in its home market — a company-estimated 63.8% of Korean low- and high-voltage power devices in FY2025 — but it publishes no market share at all for extra-high-voltage transformers or HVDC equipment, which are the products the “Korea is winning the transformer race” narrative is actually about. Those are two different claims, and they are being traded as one.

The company does disclose market share, and the tables are easy to miss because they sit one subsection below the business description. There are five of them, all labelled as the company’s own estimates, all limited to Korea, and all limited to mass-produced items — the filing says so in the header: “the domestic market shares of principal products as estimated by the company at the end of the current period, recorded only for mass-produced products the company manufactures.”

Two things in that table are worth drawing out. The switchboard share went from 26.0% to 32.6% in two years, which is real domestic share gain in a competitive product. And the stainless-tube share went the other way, 22% to 16%. Neither of those is a transformer number.

What the filing gives no share figure for is the export franchise: extra-high-voltage transformers, HVDC converter equipment and gas-insulated switchgear. Those are made to order, not mass-produced, so they fall outside the stated scope of every table above. The company also omits prices for the same reason, writing that order-based power products are priced to each customer’s specification — so there is no price series either. If you have read a global market-share number for LS ELECTRIC in transformers, it did not come from this filing; we looked for it, and the closest thing on offer is a domestic share of a different product.

Two other disclosures are worth having, because both are company-stated and checkable:

• Data-centre revenue, company-disclosed: “approximately ₩546.8bn arose in 2025 related to data centres” — $405M, or 11.0% of consolidated FY2025 revenue. That is the honest size of the AI-data-centre exposure everyone is buying.
• Export recognition: Korea’s Trade Day awards — a $700M export tower in 2023, a $900M export tower in 2024.

So: a dominant domestic incumbent, a genuine and fast-growing US position, and no disclosed share in the product the global narrative rests on. The 11.0% data-centre figure is the number to anchor on, and it came from a paragraph in a Korean-language filing, not a press release. For how the same kind of gap shows up across the Korean power complex, see our decode of KEPCO KPS.

📈 Is the Business Getting Better or Worse?

Better on every headline line — revenue up 32.7% and net profit up 74.5% in H1 2026 — but the improvement is thinner than it looks, because gross margin fell and the gap was closed by overhead leverage and a lower tax rate.

Now the decomposition, which is where the story is:

Read down that table and the cascade is clear: operating profit grew 55.7%, pre-tax profit 65.6%, net profit 74.5%. Each step amplifies the one before — not because the business got better at each level, but because non-operating items and a 3.48-point lower tax rate stacked on top. The operating engine grew 55.7%; the 74.5% headline is that plus tax.

And the gross-margin line is the one that should stop a bull. If extra-high-voltage transformers were genuinely scarce and LS ELECTRIC were pricing accordingly, gross margin would rise. It fell, slightly, on 32.7% more revenue. Every basis point of the operating-margin improvement came from the denominator.

🔍 Who Is LS ELECTRIC’s Biggest Customer? Its Own American Subsidiary

LS ELECTRIC AMERICA INC. — a company LS ELECTRIC owns — accounted for 41.7% of power-infrastructure segment sales in 2026 year-to-date, up from 31.2% in FY2025, and it is the named counterparty on every single one of the seven US supply contracts LS ELECTRIC has disclosed since March 2025. This is the most consequential thing in the filings and it does not appear in English anywhere.

Start with the customer table the company publishes for its power-infrastructure sub-segment:

Note what has happened to KEPCO. Korea’s national utility — the customer every foreign reader assumes is the anchor of a Korean grid-equipment business — has gone from 5.1% to 2.7% of power-infrastructure sales. If you want to understand how KEPCO’s own supply chain works, our decode of KEPCO KPS, the utility’s maintenance contractor, sits on the other side of the same meter.

Now the contracts. Korean issuers must file a “single sales/supply contract” disclosure when a deal passes a size threshold, and those filings name the counterparty. Here is every US one since March 2025, in the dollars the filings themselves state, excluding VAT as the filings themselves specify:

Every one of those filings describes the same structure in almost identical words: LS ELECTRIC AMERICA Inc. won the end-customer project, and LS ELECTRIC the parent is supplying it. The end customers appear only in the narrative — Bloom Energy by name, otherwise “a North American super-large investor-owned utility”, “a global hyperscaler”, “Big Tech”.

Three consequences a foreign investor should hold onto:

• These are transfer-priced intercompany purchase orders, and they are eliminated on consolidation. The $1.16bn does not add to consolidated revenue at that value. It is the amount the parent bills its own subsidiary — which is exactly why the elimination line grew from 18.5% to 35.5% of consolidated revenue.
• The consolidated “no customer above 10%” statement and the parent-only accounts contradict each other, correctly. The separate (parent-only) statements disclose one customer at 22.62% of standalone revenue in FY2025 — $580M, up from 12.13% and $279M a year earlier — so revenue from that single counterparty more than doubled. The filing anonymises it as “Company A”. The related-party note in the same document shows the only counterparty of that magnitude is LS ELECTRIC Americas Holdings at $605M of goods and services. That identification is our inference from the size, not a statement in the filing — but the next-largest related party is $68M, an order of magnitude smaller, and LS ELECTRIC’s holding in Americas Holdings is 100%. Total related-party sales were $886M, or 34.6% of standalone revenue. (Won amounts as filed: ₩783.3bn, ₩377.0bn, ₩817.3bn and ₩1,196.9bn.)
• You cannot verify end demand from the parent’s numbers. A purchase order from your own distributor tells you the distributor expects to sell; it does not tell you it has. The place that shows up, if it shows up, is inventory and receivables — and both grew faster than revenue in H1 2026 (receivables +43.0%, inventory +32.9%, against annualised revenue growth of 19.0%).

🏭 Are the Transformer Plants Sold Out? Its Own Filing Says 66–74%

No. In the six months to June 2026 LS ELECTRIC’s extra-high-voltage transformer plant at Busan ran at 69.9% of capacity, its Cheongju No.2 plant — EHV gas-insulated switchgear and transformers — at 66.5%, and subsidiary LS Power Solution’s Ulsan EHV plant at 73.7%. The company publishes these numbers itself, in a table almost nobody reads.

Two readings are possible and the filing supports the kinder one. Capacity is being added faster than it is being filled: Busan’s stated capacity in H1 2026, annualised, is roughly 82% above its FY2025 full-year figure, and Ulsan’s about 56% above — consistent with the ₩100.8bn Busan No.2 building that only completed in October 2025. [Those annualisation figures are our estimate, on three assumptions: that we can double a six-month capacity figure, that capacity is spread evenly through the year, and — for Busan and the LS Power Solution plants, whose capacity the filing states in won rather than hours — that won capacity is a fair proxy for physical capacity, which it is not if product mix or price moved. The utilisation percentages themselves are the company’s, not ours.] A plant that came online nine months ago running at 70% is a plant filling up, not a plant without demand.

But that is precisely why “sold out” is the wrong frame. The constraint on LS ELECTRIC’s transformer revenue in H1 2026 was not capacity. It had roughly 30% of its EHV transformer capacity idle at the same time as its order book grew 38.5%. That combination is the opposite of a pricing-power story, and it is consistent with the gross margin falling rather than rising.

One more line in the same table deserves a mention, because it cuts against the diversification narrative: annualised, the Cheonan automation plant’s capacity shrank about 18% versus FY2025. The company is not building out automation. It is building out transformers.

📋 What Is Actually in the Order Book — and On What Basis?

$6.66bn (₩8,989.4bn), summed across seven separate legal entities as of 30 June 2026 — and LS ELECTRIC never publishes a single consolidated backlog figure, so every number you see quoted is somebody’s choice of which entities to add up.

This is the part of a Korean filing where foreign analysis most often goes wrong, because the backlog disclosure has five independent axes stacked on top of each other: gross orders versus delivered versus remaining; consolidated versus parent-only; current-period versus cumulative; VAT-inclusive versus not; and which segments are in scope. Miss any one and the number changes. We have written up the same problem in Hanwha Aerospace’s order book, where the axes were also unlabelled.

Four things to be careful about, all of which we hit while building that table:

• “Delivered” is not a backlog decline. The 기납품액 column is work completed and handed over — normal progress, and the mechanism by which backlog becomes revenue. Reading it as a cancellation is a mistake we have made before and it produces a number that is wildly wrong.
• The VAT basis of the backlog is not stated. That matters because at least two large domestic HVDC contracts inside it say in their own filings that the value is VAT-inclusive: the ₩561.0bn ($415M) East Coast–East Seoul converter deal, and the ₩212.6bn ($157M) East Coast–Sin-Gapyeong converter station, whose filing states plainly that “the contract value is an amount inclusive of value-added tax.” Strip 10% out of the first and it is ₩510.0bn ($378M) — a ₩51bn ($38M) difference on one line item, and that ex-VAT figure is our own division, not a number the filing prints. The UK Widow Hill battery contract is worse: GBP 74,232,000 “including 20% VAT”, or GBP 61,860,000 without. We cannot tell you how much of the $6.66bn is VAT, and neither can the filing.
• The parent’s order book covers only about two-thirds of the parent’s revenue. Parent deliveries of $1,098M in H1 2026 against parent revenue of $1,651M implies roughly 66.5% came through the order book (₩1,483.3bn against ₩2,230.0bn as filed); the rest is catalogue product sold through distributors, which never appears as backlog at all. Backlog is not a forecast of total revenue.
• Nothing in the order book is EHV-transformer-specific. EHV transformers sit inside the parent’s “Power T&D” line and inside LS Power Solution’s $420M. If you have seen a figure for LS ELECTRIC’s extra-high-voltage transformer backlog specifically, it did not come from this filing — we looked, and it is not disclosed.

The company’s own delivery dates have moved — repeatedly

Here is a dataset that exists only in Korean. Korean issuers must file an amendment when a disclosed contract’s terms change, and the amendment must state the reason and print the old value beside the new one. Of the 23 supply-contract filings LS ELECTRIC made between January 2024 and September 2026, ten were amendments. We read all ten: seven changed the contract period, two changed the contract value, and one changed payment terms. The result is a public record of the company’s own schedule moving — and one contract still in progress, which predates that window, has moved further than anything inside it.

The honest counterpoint, stated with its limits: none of the 2025–26 US data-centre or transformer POs has yet been amended for a date slip. Their end dates stand as originally filed. One of them was amended — the Bloom Energy fuel-cell order, on 14 May 2026 — but only to record that an advance payment now exists; the end date did not move (rcpNo 20260514800292). The slips are concentrated in EPC and infrastructure work, where LS ELECTRIC depends on someone else’s permitting and civil works.

But look at what the Sin-Gapyeong line does to the comfortable version of that argument. HVDC converter equipment is not adjacent to the bull case — it is the bull case, the product group the GE Vernova localisation MOU and the Busan valve line are built around. That contract’s completion date has moved by five and a half years and it is still open. So “EPC slips, equipment doesn’t” is not a distinction this company’s own filings support.

And the US orders are young. The oldest is March 2025; the longest-dated runs to October 2029. On four of the seven, less than a year has elapsed. “Not yet amended” is a statement about elapsed time, not about risk profile — and the amendment record is the only place you will ever see the difference.

💰 What Do You Actually Pay for LS ELECTRIC Today?

₩212,000 a share, a $23.55bn market capitalisation, and 109.9× FY2025 earnings — which makes it both the least profitable and by far the most expensive of Korea’s three large listed power-equipment makers.

Now the comparison that makes the case.

The single line to take away: the market pays almost exactly the same multiple of sales for LS ELECTRIC as for HD Hyundai Electric — 6.40× versus 6.31× — but HD Hyundai Electric converts 24.40% of those sales into operating profit and LS ELECTRIC converts 8.59%, a 2.84× gap in margin. On FY2025 earnings that is a 3.1× valuation gap in LS ELECTRIC’s disfavour. If you want the other side of that trade read from the same filings, we decoded HD Hyundai Electric’s backlog and its 0% US dumping margin and Hyosung Heavy’s two-sized order book separately.

The bull rebuttal is legitimate and you should weigh it: trailing FY2025 is stale. LS ELECTRIC’s H1 2026 net profit grew 74.5%, and doubling the company’s disclosed H1 2026 EPS of ₩1,600 gives roughly 66.3× [our estimate, six months doubled]. That is a large improvement and still nearly double HD Hyundai Electric’s FY2025 multiple. For the gap to be justified, LS ELECTRIC’s margins have to converge on its peers’ — which brings you straight back to a gross margin that fell and EHV plants at 66–74%.

🏷️ Which LS Ticker Are You Actually Buying?

Only 010120 is LS ELECTRIC, the operating company that makes grid equipment; 006260 is LS Corp, the holding company that owns 48.46% of it and also owns the cable and metals businesses. They are different securities with different economics, and they are easy to confuse because both are “LS”.

One search note, because it will waste your time otherwise: the English word “transformer” collides with a film and toy franchise, so generic queries about Korean transformer makers return cartoons. Search for power transformer, substation, switchgear or grid equipment instead.

The split that will break your spreadsheet

On 5 February 2026 the board approved a 1-for-5 face-value split, cutting par value from ₩5,000 to ₩1,000. Shareholders approved it on 26 March, it took effect on 10 April 2026, trading was suspended 8–10 April, and the new shares listed on 13 April 2026. Shares outstanding went from 30,000,000 to 150,000,000. The stated purpose was to expand the float — relevant if you care about index eligibility and liquidity screens, which we cover in our piece on how Korea’s index constituents get picked.

Here is the trap, and the company’s own filings demonstrate it. LS ELECTRIC’s FY2025 earnings per share is printed as ₩9,647 in its March 2026 annual report and as ₩1,929 in its August 2026 half-year report. Both are correct — the second restates the first onto the post-split share count (₩9,647 ÷ 5 = ₩1,929.4). But if your data source mixes filings, or splices an unadjusted price series across 13 April, you will produce numbers that are wrong by a factor of five. The raw close went from ₩718,000 in late March to ₩212,000 in September, which looks like a 70% collapse and was in fact a gain. We know because we walked into a version of this ourselves while researching this article; the reliable method is to compute returns from market capitalisation, or to divide every pre-13-April price by exactly 5.

🌏 Is Foreign Money Buying or Selling LS ELECTRIC?

Buying, in all three windows we can measure — but the 20-session figure is larger than the 60-session figure, which means the 40 sessions before late August were net selling, and four-fifths of the recent buying happened on three days.

The discipline here matters more than the numbers. All three foreign windows are positive, so “foreigners are buying LS ELECTRIC” survives the multi-window test — which is not something we can usually say. But look at the arithmetic: 20 sessions of +921,435 shares inside 60 sessions of +484,633 means the earlier 40 sessions were net selling of roughly 436,802 shares. And the 20-session total is not a trend either: three sessions — 27 August (+283,435), 18 September (+237,875) and 9 September (+219,206) — account for 80.4% of it. Strip those three out and the other 17 sessions net to +180,919 shares — about 10,600 a session against average daily volume of 470,010 over the same window, or roughly 2% of a day’s trading. That is three repricing events, not accumulation.

Institutions went the other way over both longer windows. We would not build a thesis on that: Korean pension money sells most large caps most weeks, so an institutional net-sell figure needs a peer group before it means anything. For why these flows move Korean prices at all, see how foreign and institutional flows move Korean stocks, and our weekly Foreign Flow Watch.

🏛️ Who Controls LS ELECTRIC, and What Do Minority Holders Get?

LS Corp, the listed holding company, owns 48.46% — and minority holders get a 0.28% dividend yield and a treasury-share programme that has never cancelled a single share.

As of 31 December 2025 the register showed LS Corp with 14,539,000 shares, 48.46%, and the National Pension Service with 2,814,746 shares, 9.38%. The largest-shareholder group including affiliated persons came to 49.33%. There were 108,607 small shareholders holding 40.10% between them. The largest shareholder has not changed during the disclosure period. The board is nine directors — four executive and five outside, so outside directors are a majority. Chairman Koo Ja-kyun’s FY2025 compensation was $5.3M (₩7,151mn). (Source: DART annual report FY2025, rcpNo 20260318001243.)

On shareholder returns, three facts and one quotation.

• The dividend is small and the company says it intends to keep it that way. FY2025 declared ₩3,000 per pre-split share — ₩600 post-split, a 0.28% yield at ₩212,000 — totalling $66.1M (₩89,215mn) at a 31.1% consolidated payout ratio. The dividend-policy section states that rather than expanding cash dividends in the short term and depleting investment resources, the company will “manage the payout ratio at a stable level” while pursuing strategic investment. That is a direct statement of intent, and it is the opposite of what Korea’s Value-Up programme asks for.
• No share has ever been cancelled — and that is a cumulative, since-inception fact, not a one-year one. The FY2025 annual report’s “total number of shares” table (rcpNo 20260318001243) has a section for shares retired since incorporation, with lines for capital reduction, retirement out of profits, redemption of redeemable shares and other. All four read zero, and shares issued equals shares outstanding at 30,000,000 on a pre-split basis. Meanwhile the treasury holding fell from 298,739 to 261,509 shares in FY2025, and the cancellation column of the acquisition-and-disposal table is also blank: all 37,230 shares went to staff — 3,290 on RSU vesting, 2,140 as executive long-term incentive, 31,800 for “employee ownership motivation”. The pattern repeated in 2026 with two fresh disposal decisions: 22,581 shares at ₩844,000 (pre-split) on 26 March (rcpNo 20260326001137, ₩19,058,364,000) and 32,520 shares at ₩187,700 (post-split) to 3,252 employees on 15 July (rcpNo 20260715000510, ₩6,104,004,000). Treasury stock here is a compensation currency, not a return of capital. That distinction is central to the Korea Discount debate: a buyback that parks shares for later reissue to insiders does not shrink the share count.
• One genuine governance improvement. FY2025 is the first year LS ELECTRIC set the dividend record date after the dividend was decided and disclosed — board resolution 5 February 2026, record date 27 February, AGM approval 26 March. Its own table marks dividend predictability as provided for FY2025 and not provided for FY2024 or FY2023. Korean investors historically had to buy before knowing the dividend. That is fixed here, and it is a real, dated, verifiable improvement.

One structural note for anyone weighing 010120 against 006260: the holding company trades at 0.32× sales and 33.0× its own disclosed earnings per share against the operating company’s 6.40× and 109.9×. A holding-company discount of that width is normal in Korea and is not by itself an arbitrage — LS Corp’s earnings are a blend of cable, metals and electric, most of the group’s profit accrues to minority holders of the subsidiaries rather than to LS Corp’s own shareholders, and LS Corp’s minorities do not control the dividend policy of any of them.

📚 Lingo Check

🎯 Why It Matters for K-Export Stars

Every figure in this article came from free, public documents that happen to be written in Korean — which is the whole of our edge, and the whole of the reason this analysis does not already exist in English. Nothing here is a scoop or a leak. Every number came from documents any investor on earth can download for free — LS ELECTRIC’s half-year report, its annual report, and 23 contract filings. They are simply in Korean, and they are long.

The wires reported the orders: $312M here, $216.5M there, a growing backlog, a Korean company riding the AI-data-centre build-out. All of that is true. What the wires did not do was open the counterparty field on those filings and notice it says subsidiary seven times out of seven, or scroll to the utilisation table and find the transformer plants at 66–74%, or divide gross profit by revenue and see the margin go backwards, or open the contracts-in-progress table and find an HVDC converter station whose delivery date has moved five and a half years.

We also want to be honest about what this exercise cost us in error. We nearly published a one-year return of +43% for a stock that actually returned +262%, because an unadjusted price series broke at a stock split. We found and fixed it before publication. That is the entire reason we print our methods, our as-of dates and the things we could not verify: on a topic this heavily narrated, the discipline is the product.

Disclaimer: This article is for informational and educational purposes only and is not investment advice, a recommendation, or an offer to buy or sell any security. Figures are sourced from DART regulatory filings and Korea Exchange market data as of the dates stated and may be superseded by later filings. Market data is as of the settled close of 22 September 2026; prices, market capitalisations and valuation multiples change continuously. Currency conversions use ₩1,350.36/$1, the 22 September 2026 rate, and are indicative. Always conduct your own research and consult a licensed financial adviser before making any investment decision.