Leeno Industrial (058470): Inside the $468M Founder Sale

Every advanced chip made anywhere on earth gets pressed, at least once, against a spring-loaded metal pin so a tester can decide whether it works. A 677-person company in Busan makes about 30,000 varieties of that pin and the sockets that hold them, exports 84% of what it produces, and keeps roughly half of every won of revenue as operating profit. In April 2026 its 76-year-old founder filed a one-page notice saying he intended to sell 9.18% of it. What happened over the next seven weeks is a small masterclass in why the Korean filing and the English headline are not the same document.

🏢 What Is Leeno Industrial, and What Does It Actually Sell?

Leeno Industrial is a Busan manufacturer of the two consumable parts that sit between a semiconductor tester and the chip being tested: spring contact probes, which it brands LEENO PIN, and the IC test sockets those pins are assembled into. It also makes precision components for ultrasound probes. Nothing it sells ends up in a finished product a consumer sees; everything it sells wears out and gets replaced.

The company describes its own origin in the filing: it was founded in November 1978 as Leeno Industrial Co. (리노공업사), incorporated on December 20, 1996, and listed on KOSDAQ on December 18, 2001 (Source: DART annual report FY2025, rcpNo 20260318000182, filed Mar 18, 2026). Lee Chae-yoon, born August 1950, has been chief executive since incorporation — 29 years — and his current term runs to March 25, 2027. There is no English Wikipedia entry for the company; the founder has one, which is a fair summary of how little of this business is documented in English.

Two operating facts matter more than any description. First, the products are made to order: the filing says Leeno produces “about 30,000 varieties” and ships existing volume items within 10 days of receiving a purchase order, special orders in two to four weeks. Because of that, the company states it has no order backlog to report — “the customer’s purchase order is the order status.” Second, it runs every process in-house, from micro-machining through plating, from a single site in Busan’s Gangseo district.

Headcount at the end of 2025 was 677 (301 men, 376 women), with average tenure of about nine and a half years (Source: DART annual report FY2025, rcpNo 20260318000182). On our arithmetic that is roughly $408,800 of revenue and $194,200 of operating profit per employee for FY2025 — numbers you would expect from a software firm, not a metalworking one.

📊 Where Does Leeno’s Revenue Actually Come From?

Two-thirds of it comes from IC test sockets, and that share is rising fast — from 61.77% of revenue in FY2024 to 68.48% in the first half of 2026 — while the medical-device business shrinks as a proportion.

Separate-company (non-consolidated) revenue as filed. H1 2026 = the six months to June 30, 2026; FY = the twelve months to December 31. Percentages are the company’s own. Won amounts converted once at ₩1,345.99/$1, our own settled-price record for September 4, 2026, and that single rate is used for every dollar figure in this article. (Source: DART half-year report, rcpNo 20260814000445; annual report FY2025, rcpNo 20260318000182)

The geographic split is where it gets interesting, because the two segments point in opposite directions. In the first half of 2026 Leeno exported $151.1M of the $180.5M it sold — 83.74%, up from 79.81% in FY2025 and 77.17% in FY2024. But that average hides a split:

• Test sockets: 94.39% exported. Of $123.6M, only $6.9M was sold inside Korea.
• LEENO PIN: 79.97% exported.
• Medical components: 93.56% domestic. Of $13.8M, just $0.9M left the country.

The company itself says the semiconductor segment’s export-to-domestic ratio is “9:1”, which our arithmetic on the filed numbers confirms at 90.7%. The medical business is the mirror image, and the filing explains why by naming the only customer named anywhere in Leeno’s filings: “Since 2010 we have had a strategic partnership with Siemens, supplying components for ultrasound probes” (Source: DART half-year report, rcpNo 20260814000445, “Market conditions”). Siemens Healthineers runs its ultrasound operation partly out of Korea, which is the most plausible reason a German multinational’s purchases are booked as domestic sales — though the filing does not say so, and we are not going to assert it.

One caution before you read anything into the socket mix. Leeno discloses production volumes as well as revenue, so it is tempting to divide one by the other. Doing so gives roughly ₩1.28M of revenue per socket produced in FY2025 against ₩1.45M in H1 2026, and ₩1,339 per pin against ₩1,588. That looks like a 13–19% rise in realised price. It is our calculation, not the company’s, it uses units produced rather than units shipped, and “sockets” spans thousands of different designs — so treat it as a directional hint about mix, not an average selling price.

🗓️ How Did Leeno Get Here?

Leeno’s own filed history is unusually thin — it lists mostly awards — but the export-tower awards it lists are dated, third-party, and denominated in dollars, which turns them into a usable growth ladder.

Compiled from Leeno’s own filings: annual report FY2025 (rcpNo 20260318000182), half-year report (rcpNo 20260814000445), advance trading-plan report (rcpNo 20260424000538) and major-holding report (rcpNo 20260615000001).

Note the April 2025 line, because it will corrupt any comparison you make against an older English source. Leeno split its stock five ways in 2025. A 2024 article quoting a share price around ₩250,000, or earnings per share of ₩7,463, is not describing a different company — it is describing the same company before the split. We come back to this in the ticker section, because Leeno’s own annual report prints the pre-split and post-split figures side by side without restating the old ones.

🌍 Is Leeno Actually a Global Leader, or Just a Korean One?

Leeno’s filings state no market share and name no competitor anywhere — so anyone telling you it is “the global number one in pogo pins” is not quoting the company. We checked the full text of both the FY2025 annual report and the 2026 half-year report: the only occurrences of “market share” are a sales-strategy bullet expressing an intention to “expand market share” in the medical segment, and a line in the executive-compensation criteria. Neither is a share disclosure.

What is verifiable is scale and recognition, both dated and both conferred by third parties. The Korea International Trade Association awarded Leeno its $50M export tower in December 2016, $70M in 2017, $100M in 2021 and $200M in 2022 — a fourfold rise in exported value in six years. In November 2024 the Ministry of SMEs and Startups named it to the ninth cohort of Korea’s “Long-Lived Excellent Enterprise” list.

The intellectual-property table is the closest thing to a moat you can count. As of June 30, 2026 Leeno held 176 registered domestic patents and 153 registered overseas patents — 329 in total — with a further 19 and 127 pending respectively, plus 78 registered designs and 26 registered trademarks (Source: DART half-year report, rcpNo 20260814000445, IP holdings table). Research and development ran at 2.81% of revenue in the first half of 2026 ($5.1M), down from 3.52% in FY2025 and 3.61% in FY2024 — the ratio falls because revenue is growing faster than the R&D line, not because spending fell.

Its filings describe the barrier to entry in their own words: the business is “technology-intensive, takes a long time to accumulate know-how, and rests on product reliability.” That is a management assertion, not a measurement. The measurable version is the 10-day lead time on 30,000 part numbers, which is hard to fake and harder to copy.

📈 Is the Business Getting Better or Worse?

Better on every operating line — revenue up 27.25% and operating profit up 36.70% in the first half of 2026 — but the cash statement tells a different story from the income statement, and the gap is worth a paragraph of your time.

*FY2024 EPS is on the pre-split share base of 15,242,370 shares; Leeno did not restate it. On the post-split base it is ₩1,492.6. All other columns are post-split. “Q2 2026 alone” is the three-month column the half-year report prints separately from the cumulative one. (Sources: half-year report, rcpNo 20260814000445; annual report FY2025, rcpNo 20260318000182)

Quarter by quarter, six quarters in a row: 44.56% in Q1 2025, 47.50% in Q2 2025, 49.84% in Q3 2025, 47.61% in Q4 2025, 47.41% in Q1 2026, and 51.34% in Q2 2026 — the highest of the six and the only one above 50%. That is a wide, high plateau rather than a straight climb: the margin gave back more than two points in the fourth quarter of 2025 and did not exceed its Q3 2025 level again until Q2 2026. (Q3 2025 alone — revenue ₩96,841,827,642, operating profit ₩48,261,344,084 — is filed in the Q3 2025 quarterly report, rcpNo 20251113000285, which also gives nine-month cumulative revenue of ₩287,774,135,318 and operating profit of ₩136,643,461,445; the FY2025 annual report, rcpNo 20260318000182, gives full-year figures of ₩372,534,084,608 and ₩176,997,188,289.) Gross margin in the first half was 53.66%, and selling, general and administrative expense was just 3.93% of revenue. There is essentially no sales organisation to speak of: two teams reporting to the CEO.

One of those six quarters is ours rather than Leeno’s. The company publishes no fourth-quarter column, so we derive Q4 2025 by subtracting the nine-month cumulative figures above from the full year: revenue ₩84,759,949,290 and operating profit ₩40,353,726,844, a 47.61% margin. Both operands are filed; the subtraction is not.

Now the qualifications, and there are three.

First, a third of the pre-tax improvement came from below the operating line. Pre-tax profit rose ₩48.5bn ($36.1M) year on year in the first half; the increase in net non-operating income — ₩2,949,381,540 to ₩19,041,377,239 — accounted for ₩16.1bn ($12.0M), or 33.16%, of that. Break the non-operating line open and it is almost entirely currency: foreign-exchange gains and translation gains net of losses swung from minus $4.9M in H1 2025 to plus $5.8M in H1 2026, a $10.7M swing that alone equals 29.78% of the pre-tax increase. A company that exports 84% of its output and holds a quarter-billion dollars of financial assets will do that. It is real money; it is not operating improvement, and it can reverse.

Second, one line in that non-operating block is a black box. “Miscellaneous income” was ₩5,637,117,401 ($4.2M) in the first half, up from ₩3,429,592,750 ($2.5M) — larger than either foreign-exchange gain line on its own (₩4.12bn of transaction gains, ₩4.31bn of translation gains), larger than the securities gains, and the filing does not itemise it. We flag it because we cannot explain it, not because we think it is improper.

Third, and most concretely, operating cash flow fell while operating profit rose. Cash from operations was $50.9M in H1 2026 against $60.3M in H1 2025 — down 15.55% — while operating profit rose 36.70%. The reason is on the balance sheet: trade receivables went from $39.3M at December 31, 2025 to $75.5M at June 30, 2026, a 92.32% increase against 27.25% revenue growth. On our calculation that pushes days sales outstanding from about 52 days for FY2025 to about 76 days for the first half. Some of that is arithmetic — a business growing this fast carries more receivables — but 24 days is a lot of drift for a company whose customers are the best-capitalised firms in the industry.

🔍 Why Do Three Different Numbers Exist for the Founder’s Sale?

Because Korea’s insider pre-disclosure rule forces a seller to publish an estimate 30 days before the trade, and the executed price only appears afterwards in a Korean-language filing that no English outlet read. Here is the full sequence, filing by filing.

April 24, 2026. Lee Chae-yoon files an advance trading-plan report for specified securities by an officer or major shareholder — rcpNo 20260424000538. It states the purpose in eleven Korean words: “asset management through the sale of held shares.” It states the method (off-hours sale), the quantity (7,000,000 common shares, 9.18% of the company), and a settlement window running from May 26 to June 24, 2026 — 30 days. And it states a price: ₩123,300, with an explicit footnote saying this is “the KRX closing price on the day before submission (April 23, 2026: ₩123,300)” and that “the actual disposal price may differ.” Multiply out and you get ₩863,100,000,000 — $641.2M. That is a reference calculation, and the filing says so.

This report exists because of an amendment to Korea’s Financial Investment Services and Capital Markets Act that took effect on July 24, 2024. Officers and major shareholders of listed companies must now disclose planned trades of their own stock at least 30 days in advance, and the Financial Services Commission’s own English release confirms the design, including the permitted deviation: an insider may transact within 70–130% of the planned value. Leeno’s filing prints that band in a footnote, citing Article 173-3(3).

April 27, 2026. Seoul Economic Daily reports the plan in English, correctly quoting the filing’s ₩863.1bn and converting it to roughly $605 million. Leeno’s shares had closed at ₩124,400 on April 24 — the highest close in our own settled series, which begins March 31, 2026. The next session, Monday April 27, they closed at ₩109,800, down 11.74%, on volume of 3,215,459 shares, the heaviest day of 2026 to that point (Source: KRX settled quotations).

Then the market moved. By the day before execution the stock closed at ₩99,800, 19.06% below the reference price in the notice.

June 12, 2026. The trade is struck off-market. Two separate filings record the executed price and they agree exactly: the major-holding report (rcpNo 20260615000001) and the post-trade insider ownership report (rcpNo 20260616000258) both show 26,418,345 shares going to 19,418,345, a change of −7,000,000, at a disposal price of ₩90,000. The filings print the price and the share count but not the product; multiply them yourself and the sale is ₩630.0bn — $468.0M. Settlement was June 16.

June 15, 2026. The Elec reports the completed deal in English at “104,500 won per share” for “about 731.5 billion won.” Seven million shares at ₩104,500 is ₩731,500,000,000 to the won — and ₩104,500 was that day’s closing price, not the disposal price. The number in the filing is ₩90,000.

Converted once at ₩1,345.99/$1. Closing prices are settled KRX quotations for the dates shown.

Two things follow from ₩90,000 that do not follow from either of the other numbers.

The discount was larger than a normal block. Against the June 11 close of ₩99,800 — the last close before the contract date — ₩90,000 is a discount of 9.82%. Against the June 12 close of ₩104,500 it is 13.88%. Korean block deals typically clear at 5–10% below market.

The seller came within three percentage points of breaching his own filed plan. ₩630.0bn is 72.99% of the ₩863.1bn he told the market he expected to transact. The statutory floor is 70%. Had the stock fallen another 4% before he sold, the plan as filed would have had to be amended. This is the pre-disclosure rule doing exactly what it was designed to do — and costing the insider $173.2M relative to the price his own notice implied — while public shareholders had 49 days’ warning that 9.18% of the company was coming to market.

One more detail that puts the sale in scale. This was Lee Chae-yoon’s first major-holding report — the 5%-rule filing — since April 14, 2014; the previous-report line in rcpNo 20260615000001 is dated twelve years earlier, at 5,283,669 shares and 34.66%. In the intervening period the only movement in his position was the 2025 stock split, which multiplied those 5,283,669 shares into 26,418,345 without changing his percentage at all. On June 12 he sold 7,000,000 of them. On June 12 alone, 11,535,581 Leeno shares changed hands, against a previous 2026 daily record of 3,215,459; June’s total volume of 35,587,354 shares was 46.7% of the entire share count (Source: DART half-year report, rcpNo 20260814000445, monthly price and volume table).

One clarification on “first since 2014,” because it is the kind of claim a reader will test. The twelve-year gap is in the major-holding (5%-rule) series. Lee Chae-yoon did file a separate insider ownership report in the interval — rcpNo 20250507000264, dated April 25, 2025 — but that one records the stock split, a different filing type reporting no change in his economic stake.

He still owns 19,418,345 shares — 25.48%, worth $960.7M at the September 4, 2026 close — and remains chairman, chief executive and chair of the board. Nothing about control changed. What changed is that $468.0M of stock left the founder’s hands and entered the float, and the price it left at is recorded in one language only.

💸 Can Leeno Still Afford Its Own Dividend?

In the first half of 2026 it could not, on cash: free cash flow was $19.8M against $45.1M of dividends paid, a coverage ratio of 0.44× — the year before, the same measure was 1.05×. The gap was bridged by drawing down financial assets, which is exactly what a company with $333.1M of them and no debt should do. But the flip is worth understanding, because Leeno’s entire published Value-Up plan rests on it.

Leeno pays a single annual dividend, disbursed within a month of the March AGM, so the entire year’s payment falls in the first half. Free cash flow is operating cash flow less purchases of property, plant and equipment and intangibles, as filed. (Source: DART half-year report, rcpNo 20260814000445, statement of cash flows)

The capital expenditure is not maintenance. Leeno is building a new plant, and the money already committed is large relative to the company. On June 30, 2026 construction in progress stood at ₩157,665,495,427 ($117.1M) — 60.01% of net property, plant and equipment, 18.40% of total assets, and 3.4 times the ₩46,854,905,864 ($34.8M) net book value of every machine Leeno currently operates. None of it depreciates yet. None of it produces anything yet.

The programme, as filed: the board approved it on November 8, 2024 for an investment period running to November 10, 2026, with a budget of ₩97,182,000,000 ($72.2M) for construction and equipment, of which ₩78,553,778,109 ($58.4M) — 80.83% — had been spent by June 30. The land was bought separately from Korea Water Resources Corporation under a December 22, 2022 contract for a site in Busan EcoDelta City. The contract price was ₩75,480,820,998 ($56.1M), but Leeno paid the third through ninth instalments and the balance ahead of their agreed dates and earned an early-payment discount, so the amount actually paid was the discounted ₩74,513,878,908 ($55.4M) — ₩966,942,090 ($718k), or 1.28%, less. Payment completed on May 21, 2024, the day the corporation granted pre-completion consent to use the land; a small part of the discount had to be settled back on that date, because consent arrived before the original due dates on those instalments had passed. Together that is $127.6M, and the filing states plainly that it was “funded from the company’s own resources.”

One honest loose end: land paid plus construction spent is ₩153.07bn ($113.7M), while construction in progress reads ₩157.67bn ($117.1M) — a ₩4.60bn ($3.4M) difference the filing does not explain. Capitalised costs are the obvious candidate, but we are telling you the two numbers do not reconcile rather than quietly using whichever one suits.

Now put that beside the Value-Up plan Leeno filed on March 26, 2026 (rcpNo 20260326900771). It is a single page. Its entire content is two targets — “a year-end cash dividend of at least ₩800 per share” and “a payout ratio of at least 30%” — plus one sentence of plan: “we intend to strengthen profitability on the basis of the capacity expansion from the new plant, scheduled for completion in the second half of 2026, and thereby secure the funding source for shareholder returns (cash dividends).” There is no attached plan document, because Leeno qualifies as a “high-dividend company” under Article 104-27 of the Restriction of Special Taxation Act and is therefore exempt from filing one.

Read the two documents together and the logic is circular in an interesting way. The dividend is currently outrunning cash generation because the plant is consuming it. The plan says the plant will produce the profits that fund the dividend. Both may be true. But the “target” itself — ₩800 a share, 30% payout — is below what Leeno already does: it paid ₩800 for FY2025 at a 39.96% payout, and its stated internal policy in the annual report is to “maintain a payout ratio at the 30% level” with a three-year average of about 40%. As Value-Up commitments go, this one promises less than the status quo. For what the programme is supposed to achieve, see our guide to Korea’s Value-Up programme.

💰 What Do You Actually Pay for Leeno Today?

About 33 times last year’s earnings and 6.5 times book for a $3.77bn company that converts nearly half its revenue into operating profit — the cheapest multiple in Korea’s listed chip-test cluster, on the highest margin.

Price, share count and market capitalisation from KRX settled quotations for September 4, 2026. The trailing-twelve-month EPS stitches FY2025 (₩2,002) minus H1 2025 (₩928) plus H1 2026 (₩1,401); the weighted share bases of those periods differ slightly, so treat that one line as approximate. Equity is as filed at June 30, 2026 on a separate-company basis. “Cash and financial assets” is the sum of three filed lines at June 30, 2026: cash and cash equivalents ₩86,771,663,199 ($64.5M), other financial assets measured at amortised cost ₩340,041,996,800 ($252.6M), and other financial assets at fair value through profit or loss ₩21,482,908,704 ($16.0M) — ₩448,296,568,703, or $333.1M. The 52-week range is the highest and lowest settled closing price across the 243 KRX sessions from September 5, 2025 to September 4, 2026.

Against its listed peers — the same table we built for our ISC deep dive, on the same date and the same FX, so the two articles agree:

⚠️ Basis caveat, and it cannot be removed: ISC’s and TSE’s figures are consolidated; Leeno files separate-company statements only and reports no consolidated accounts, so its row is not a consolidated number and cannot be made into one. P/E is market capitalisation divided by FY2025 net income as each issuer reported it on DART, all at the September 4, 2026 close and ₩1,345.99/$1. Peer set selected by us. (Sources: FY2025 annual reports filed on DART by each issuer; the KRX market data portal for prices and share counts.)

Held to matched half-year periods, the ranking does not change:

Same basis caveat: Leeno separate, ISC and TSE consolidated. Leeno H1 2026: revenue ₩242,958,602,798, operating profit ₩120,814,329,893 (rcpNo 20260814000445).

Internationally, the comparable businesses are FormFactor in the United States and Technoprobe in Italy, both of which make probe cards and test interface hardware. We name them for orientation only: Leeno’s filings do not name competitors, and we have not put their numbers on the same basis as the Korean three, so we are not going to print them as if we had.

🏷️ Which Leeno Numbers Are You Actually Reading?

There is exactly one listed Leeno — KOSDAQ 058470 — with no ADR, no US listing and no preferred share class; the trap is not a look-alike ticker but the April 2025 five-for-one stock split, which makes every pre-2025 per-share figure incomparable unless you divide it by five.

The split took effect April 14, 2025: par value ₩500 → ₩100, shares 15,242,370 → 76,211,850, with authorised shares raised to 200,000,000 at the March 26, 2025 AGM. Leeno’s FY2025 annual report prints the pre-split and post-split per-share figures in adjacent columns and restates only the current period — the footnote reads “the 30th fiscal year figures reflect the stock split effective April 14, 2025.” (Sources: rcpNo 20260318000182; rcpNo 20260814000445)

If you want the mechanics of reading Korean tickers and converting quotations, our companion piece on Samsung and SK hynix share prices in USD covers the conversion traps in more depth.

🌏 Is Foreign Money Buying or Selling Leeno?

It depends entirely on the window you choose, and the windows disagree in sign — foreigners were net buyers over the last five sessions and over the 55 sessions since the block settled, but net sellers over the last twenty and the last thirty.

Method: each day’s net share balance multiplied by that day’s settled close, then summed — an approximation, not execution prices. Our own settled data series for 058470 begins on June 18, 2026, two days after the block settled, so “55 sessions” is the full history we hold rather than a chosen window; we say so rather than presenting it as a standard lookback. Foreign ownership was about 32% of shares outstanding as of September 4, 2026 (Source: Naver Finance). For how these flows work mechanically, see how foreign and institutional flows move Korean stocks and our weekly Foreign Flow Watch.

The 20-session column is the one to be careful with. A writer using only that window would tell you foreign investors are exiting Leeno. A writer using only the 5-session column would tell you they are accumulating. Both would be describing the same tape — and a 30-session window, which we also ran, is negative too (−$5.7M), so the 20-session reading is not a one-week artefact. What the full 55-session record shows is the more useful fact: since the founder’s block settled, foreign investors have bought $237.2M net of Leeno while domestic institutions have sold $272.2M net.

Be precise about what that does and does not say. The $237.2M is roughly half the $468.0M the founder sold, but it is not evidence that foreign investors bought the block — it is on-exchange buying in the eleven weeks after settlement, and the matching sales came from Korean institutions, not from Lee Chae-yoon. Who took the 7,000,000 shares on June 12 is not disclosed anywhere we can find, and we are not going to infer it from a flow series that starts six days later.

The register bears that out at the top end. The two Korean asset managers above 5% both shrank across the period: Samsung Asset Management from 5,714,876 shares (7.50%) to 4,891,973 (6.42%), and Mirae Asset Global Investments from 6,346,847 (8.33%) to 4,505,763 (5.91%) (Sources: the FY2025 annual report, citing their February 9 and March 3, 2026 filings, and the half-year report, citing their July 8 and August 3, 2026 filings — the dates are the filing dates, not a single common measurement date). Neither of them bought the block.

One more name crossed 5% and then left again, and it is worth reading rather than counting. J.P. Morgan Securities PLC filed a new major-holding report on July 10, 2026 showing 3,896,206 shares — 5.11% — as of July 7. Read the second page and neither entity crossed 5% on its own: PLC held 2,884,338 shares (3.78%) and its affiliate J.P. Morgan Securities LLC held 1,011,868 (1.33%). The obligation arises on the combined book, which is why the filing prints the aggregate next to the reporting entity’s name. We found the same shape a fortnight later at Techwing (089030), where J.P. Morgan crossed 5% on a 45,000-share stock borrow rather than a purchase — two names, one lesson: a name on the 5% register is not automatically a buyer (rcpNo 20260710000204), then filed again on July 29 showing 3,073,540 shares, or 4.03%, as of July 23 (rcpNo 20260729000050). It does not appear in the register above because it was below 5% again by June 30’s measurement basis and by the time the half-year report was written. It also is not the block buyer. The detailed change schedule in the July 29 filing is dominated by stock-lending mechanics — line after line of 차입 (borrow), 차입주식반환 (return of borrowed shares) naming Vanguard Group Inc (as agent) and the California Public Employees’ Retirement System, plus On-Lend and On-Lend Return entries — with only small on-exchange purchases and sales alongside. That is a prime-brokerage book crossing a disclosure threshold on borrowed stock, not a principal taking a 9.18% stake. The buyer of the founder’s block remains undisclosed.

🏛️ Who Controls Leeno, and What Do Minority Holders Get?

Lee Chae-yoon controls it with 25.48% after the June sale, chairs the board he sits on, and has 62,851 small shareholders holding 58.49% of the stock behind him — a governance structure with real strengths and one glaring succession hole.

Shareholdings as of June 30, 2026 except where a filing date is given. Small shareholders numbered 62,851 holding 44,576,433 shares (58.49%) as of December 31, 2025. (Sources: rcpNo 20260814000445; rcpNo 20260318000182; rcpNo 20260615000001)

What is genuinely good here, by Korean standards. There is no holding company, no cross-shareholding, no subsidiary web — Leeno has no consolidated accounts because it has nothing to consolidate. Related-party dealings are microscopic: sales to Leeno Precision of ₩220,341,856 ($164k) and purchases of ₩47,680,000 ($35k) in the first half, together about 0.11% of revenue. Total compensation for all directors and auditors in the half-year was ₩1,193,097,600 ($886k). And on March 26, 2026 the articles were amended to replace the standing auditor with a proper audit committee of at least three directors, two of them elected separately from the main director slate — plus a new clause reading, in full: “In performing their duties, directors shall protect the interests of all shareholders and shall treat the interests of all shareholders equitably.” That is the fiduciary language at the centre of Korea’s governance reform, written into a mid-cap’s articles. The board also resolved on February 19, 2026 to adopt electronic voting and electronic proxies, which is what actually matters to an overseas holder who cannot fly to Busan in March — though read the standing rule, not just the resolution: the half-year report says Leeno has adopted neither cumulative nor written voting and uses electronic voting and electronic proxy solicitation “where necessary, for a specific general meeting only, by resolution of the board.” It is a per-meeting switch the board can decline to flip, not a permanent facility. It has been used once, at the 2025 annual meeting.

What is not. The board is seven directors: four inside — the founder, the finance director, and two sales directors, one of them Lee Kyung-min, listed as a lineal descendant of the largest shareholder — and three independent. The founder chairs it. Insiders are a majority. And the same March 2026 amendment that added the equal-treatment clause also fixed the meeting format as “held only by shareholders attending in person at the place of the meeting” — the company opted out of virtual and hybrid general meetings in the same sitting at which it wrote in a duty to all shareholders.

On dividends, the predictability problem remains untouched. Leeno’s record date for the FY2025 dividend was December 31, 2025; the amount was not fixed until the March 26, 2026 AGM. Its annual report answers “X” — no — to both the question of whether the articles permit setting the record date after the dividend is determined, and the question of whether predictability is provided; the plan for improvement is recorded as “undecided.” In plain English, you had to own the stock on the last day of December to receive a dividend whose size you would not learn until the following March. That is a live component of the Korea Discount, and if you are wondering what you would actually receive net of tax, see our guide to Korean dividend withholding tax.

The dividend record itself is strong: 23 consecutive years of payment from FY2002 through FY2024, with FY2025 making 24, never an interim or quarterly payment, and a payout ratio of 41.05%, 40.20% and 39.96% over the last three years. The treasury holding is trivial at 0.41% but is finally going to be retired — the half-year report states that “treasury shares that continue to be held will be cancelled by September 2027 in accordance with the relevant laws, for the enhancement of shareholder value.” Leeno bought those shares in 2015 and has sat on them for eleven years.

Which leaves the succession hole. Lee Chae-yoon turned 76 in August 2026. The half-year report’s director table lists Lee Kyung-min — born June 1984, female, a sales director since 2022 — with her relationship to the largest shareholder given as 직계비속, lineal descendant. The filing does not say daughter, and we are not going to say it either; what it does say is that her shareholding is zero. The founder has just converted 9.18% of the company into $468.0M of cash rather than transferring it, and the stated purpose in the filing is “asset management.” We are not going to speculate about what that means. We are going to point out that it is the single most important unanswered question about this company, and that it is answered nowhere in the filings.

📚 Lingo Check

🎯 Why It Matters for K-Export Stars

Leeno is the cleanest illustration we have found of why we read the Korean filing rather than the English wire. Every material number in this article — the ₩90,000 disposal price, the 70–130% statutory band the seller nearly breached, the ₩157.7bn of idle construction, the 0.44× dividend cover, the equal-treatment clause added on the same day the company opted out of virtual shareholder meetings — is in a document that has no English version and, as far as we can find, no English summary.

It also sits precisely where our thesis says the operating leverage of Korea’s chip cycle actually shows up: not in the giants, whose margins are diluted by scale, but in the consumables layer beneath them, where a 677-person company in Busan can run a 51.34% quarterly operating margin because every advanced package needs a socket and every socket wears out. That layer now has two deep dives on this site — Leeno on metal pins, ISC on rubber sockets — and they are worth reading against each other, because the same test-socket market has produced one company at 47.5% margin and 33.4× earnings and another at 27.3% and 68.5×.

This company is one of ten in our K-Semiconductor Filing Map — a supply-chain map where every cell is marked by what the filing itself supports, and each one links to the receipt number it came from.

Disclaimer: This article is for informational and educational purposes only and does not constitute investment advice, a recommendation, or an offer to buy or sell any security. All figures are drawn from public filings on Korea’s DART system and from settled market data on the dates stated, and are accurate as of those dates; prices, ownership and financial figures change. Foreign-currency amounts are converted at a single stated rate and will differ from rates on other dates. Do your own research and consult a licensed financial adviser before making any investment decision.