Coway Stock: What I Am Paying For Ends on September 8

I opened a Korean regulatory filing on Coway stock, took two numbers out of it, and divided one by the other. The filing, dated July 10, 2026 (Fri), says Netmarble intends to spend 49,999,750,000 won, roughly $35.1 million, buying shares on the open market between August 10, 2026 (Mon) and September 8, 2026 (Tue). It also says the holding will move from 18,943,446 shares to 19,531,681 shares. The difference is 588,235 shares. Divide the money by the shares and you get 84,999.6 won, which is to say the filing was built on a reference price of 85,000 won, about $59.67.

On August 5, 2026 (Wed) the shares closed at 92,800 won, about $65.15. That is 9.18% above the reference price, by my calculation. Spend the same 49,999,750,000 won at 92,800 won and you buy 538,790 shares instead of 588,235. The resulting stake works out to 27.53%, short of the 27.60% written into the filing.

Those three numbers are the whole piece. Everything below explains why I think the third one matters more than the first two.

Contents15 min read

The disclosure arithmetic inside Coway stock

Coway Co., Ltd. trades on the KOSPI, the main board of the Korea Exchange in Seoul, under the code 021240. The KOSPI is the senior of Korea’s two domestic markets, the junior one being the KOSDAQ, and it holds most of the country’s large industrial and consumer names. Coway rents household appliances, mostly water purifiers, air purifiers and bidets, on monthly subscription contracts, and it sends technicians to service them. The company also sells outright in some overseas markets and runs a water treatment engineering subsidiary.

Two shareholders are buying the stock right now for opposite reasons. Netmarble, the Korean mobile game publisher that acquired control in 2019, is buying to entrench. Align Partners, a Korean activist fund, is buying to press. Dailian reproduced the Netmarble filing in full: 18,943,446 shares and 26.77% before, 19,531,681 shares and 27.60% after. Against 70,767,241 shares outstanding those percentages reproduce exactly, which is a useful side benefit. It tells me the share count my data vendor is carrying is the same one the filing uses.

Why 0.07 points is worth writing down

The gap between 27.60% and 27.53% is seven hundredths of a percentage point. In governance terms it changes nothing. I wrote it down anyway because of what produces it. When the two sides compete for shares, the price rises, and the risen price shrinks what each side’s money can buy. The competition eats its own target. If the price keeps climbing across the thirty day window, the shortfall widens instead of closing.

I have not seen this calculation anywhere in the Korean coverage, which is mostly framed around the idea that the rivalry supports the price. That framing is correct as far as it goes. I simply want the arithmetic on both sides of it.

The other buyer, and the company itself

Align Partners held roughly 3% of the company in early 2025 and 6.21% in 2026. S Journal, writing on August 3, 2026 (Mon), described the accumulation as a push to widen influence at shareholder meetings, and reported that the company plans to cancel 50 billion won, about $35.1 million, of treasury stock within the year. NSP News on July 28, 2026 (Tue) set out the activist demands as raising total shareholder return from 20% to 40%, introducing quarterly dividends, and cancelling treasury shares.

So the float is shrinking from three directions at once. The controlling shareholder buys, the activist buys, the company retires. I have no argument with the view that this supports the price. My argument is only that one of those three has a printed end date and the other two do not.

What Coway stock is built on when the filings are set aside

Reading this company only through the stake contest would be a mistake. First quarter 2026 consolidated revenue was 1,329.7 billion won, about $933 million, up 13.2% year on year. Operating profit was 250.9 billion won, about $176 million, up 18.8%. Net profit was 182.0 billion won, about $128 million, up 31.1%. News1 covered the May 8, 2026 (Fri) release in the Korean press.

The split matters more than the total. Domestic revenue was 742.8 billion won, up 9.5%. Overseas subsidiaries together contributed 537.0 billion won, up 20.2%, and inside that figure Malaysia alone was 406.2 billion won, up 23.5%. Thailand was 55.4 billion won, up 29.3%. Indonesia was 12.6 billion won, up 14.7%. The United States subsidiary was 57.5 billion won and it shrank 4.1%.

Malaysia therefore supplied about 76% of overseas revenue in the quarter, on my arithmetic. Anyone describing this as broad global growth is compressing four very different markets into one word. Southeast Asia carried the quarter. North America went backwards.

The second quarter is still an estimate

At the time I am writing, the second quarter of 2026 has not been filed. Last year’s second quarter landed on August 8, 2025, so the release is close. Until it arrives there are only forecasts. Hana Securities analyst Park Jong dae published estimates of 1,429.2 billion won of revenue, about $1.00 billion, and 273.3 billion won of operating profit, about $192 million, growth of 14% and 13% respectively, above a market consensus of 265.2 billion won. Herald Business relayed those figures on July 9, 2026, along with domestic rental account additions of 190,000 in the first quarter and an estimated 230,000 in the second.

I stopped short of building a time series out of the account numbers. Insight Korea carried a 2024 forecast from Hanwha Investment and Securities analyst Lee Jin hyup putting net account additions at 300,000, 370,000 and 400,000 for 2024, 2025 and 2026. Domestic quarterly additions in 2026 already approach that annual figure within the first half. I could not establish that the two counts cover the same scope, so I use the direction and leave the magnitudes alone.

The engineering subsidiary is a similar case. Coway Entech won water treatment work at an SK Hynix semiconductor plant in the United States. News1 put the award at 70 billion won, about $49 million. The Guru, on July 15, 2026, described it as roughly 80 billion won of incremental revenue expected over two years. One of those is a contract value and the other is a revenue expectation, so they may well both be right. I record both and adopt neither.

The year the dividend fell and Coway stock did not

The 2025 fiscal year dividend was 1,940 won a share, about $1.36. The 2024 dividend had been 2,630 won. That is a cut of 26.2%. Total cash distributed came to 137.3 billion won, about $96 million, which was 51.8 billion won and 27.4% below the prior year, and the payout ratio fell from 33.4% to 22%. Reporter Kim Hong gun of the Korean shareholder press outlet S Econ set out those numbers on February 6, 2026 (Fri). In the same year revenue rose 15.2%, operating profit 10.5% and net profit 9.2%. Earnings up, dividend down.

The company’s own value up filing explains where the money went. The February 6, 2026 (Fri) corporate value enhancement progress disclosure records 137.3 billion won of cash dividends alongside 110 billion won, about $77 million, of treasury share purchases, and states that roughly 1.9 million treasury shares bought in 2017 and 2024 were cancelled in full. Total shareholder return is to stay at a 40% posture through 2027. Those 1.9 million shares represent about 2.61% of the pre cancellation share count, on my arithmetic.

Changing the channel triggered a tax rule

The total held. The channel changed. And the channel it changed to failed a tax test. Under the Korean regime a company whose payout ratio is at least 25% and whose dividend rises at least 10% year on year lets its shareholders elect separate taxation on dividend income. A 22% payout ratio does not clear that bar. The company appears to have registered the point, because the same filing says that from 2026 cash dividends will be prioritised in order to qualify for separate taxation, with a payout ratio of 25% or more and dividend growth of at least 10%.

That is why I do not treat the 2025 cut as a retreat from shareholders. The aggregate stayed inside the 40% posture, the mix moved toward buybacks, the mix produced a worse tax outcome for holders, and the following year the company reversed direction. A policy the company set for itself was corrected by the tax code. I find that a more interesting failure mode than a simple stinginess story, and it is a different problem from the one I looked at in a Korean industrial trading at a heavy governance discount despite record profits, where the discount attached to who controls the company, not to how the cash leaves it.

One quoted figure that did not reconcile

Reporter Yang Jin won, writing in the Korean press on April 2, 2026, put Netmarble’s take from the 2025 dividend at about 35.9 billion won, roughly $25 million. Divide 35.9 billion won by 1,940 won and you get 18,505,155 shares, which is 26.15% of the register. The July filing says 26.77%. The two do not match, and the reason is almost certainly that Netmarble bought roughly 40 billion won of stock in May and June of 2026, between the dividend record date and the filing. So 26.15% describes the position at the record date and 26.77% describes it in July. Two accurate numbers, two different dates, and no basis for putting them on the same line.

The subscription premium Coway stock does not receive

This company does not sell hardware so much as lend it and bill monthly. Markets are supposed to pay up for that. I have repeated that idea for years. This is the first time I have actually counted which companies carry the premium, and the count did not go the way I assumed.

The two American rows come from stockanalysis.com screens and each row carries its own timestamp. A.O. Smith is quoted at $60.73 as of the August 3, 2026 close, 16:00 US Eastern time. Cintas is quoted at $203.65 as of the August 4, 2026 close at the same hour. The price to sales and net margin cells are mine, obtained by dividing market capitalisation and net income by trailing revenue from those same screens, so they carry a calculation label instead of a quotation label.

Read across the rows. Net margin at Coway is 12.44% and at A.O. Smith 13.17%, essentially the same business quality by that one measure, while the multiples are 10.84 times and 16.92 times. Cintas, whose revenue model is the closer analogue, trades at 41.48 times. Coway sits at roughly a quarter of that.

I want to be careful here. Cross border multiple comparisons are crude. Korean listed companies carry a persistent discount that has many causes, Cintas earns a materially higher net margin, and none of these three competes with the others for a single customer. The point I take away is narrower: whatever premium the subscription model commands, it is attached to companies other than this one. That is a fact about where the money sits, not a prediction that it will move. Something similar came up when I looked at a Korean holding company whose listed stakes were worth more than the parent: a gap can persist for years without anything closing it.

Coway stock at a glance, August 5, 2026 close

Figures come from Kiwoom Securities data and reflect the August 5, 2026 (Wed) close as checked at the time of writing. This piece may publish later, so the numbers can differ from live quotes. The Korean won is the reference currency throughout and dollar figures are approximate, converted at 1,424.5 won per dollar, the Seoul market close on the same date as reported by Seoul Finance. I dropped several vendor fields because they fail their own internal consistency checks. Total equity of 2,914.9 billion won conflicts with book value per share times shares outstanding, which gives 3,664.4 billion won, and dividing net profit by the stated equity gives 21.18% against a quoted return on equity of 18.1%. With no two of the three confirming each other I removed the whole equity family and kept only the book value figure, whose basis the P/B reproduction settles. Cash flow and EBITDA fields were dropped for the same reason.

The case against my read

The dividend demand collides with the business model. Analyst commentary published by the Korean outlet Press9 on March 1, 2025 by Na Han ik reported that Align had argued for a payout ratio of 90%, funded partly by taking net debt to twice EBITDA, and called that a request to stop growing. The supporting figures are the interesting part. Through the first three quarters of 2024 net profit was 461.3 billion won while operating cash flow was 277.5 billion won, about 60% of it, and finance lease receivables grew by 714.7 billion won over the same span. A rental company converts profit into receivables as it installs accounts. Paying out more means either installing less or borrowing.

The margin thinned. Reversing the growth rates gives a 2024 operating margin of 18.45% against 17.70% in 2025, a decline of 0.75 percentage points in a year when revenue grew 15%. Growth of that kind deserves a question about its quality.

North America went backwards. 57.5 billion won in the first quarter, down 4.1%, while the Southeast Asian units compounded.

Price to book of 1.79 is the one screening test this company fails. On the seven metric checklist I keep, it scores six of seven, or 86 points, and price to book is the single miss. This is not a cheap stock against its equity. It is cheap against its earnings, which is a different claim.

The support has a printed expiry. The Netmarble window closes on September 8, 2026 (Tue). The 50 billion won cancellation is scoped only as within the year. Where the price settles once that flow stops is genuinely unknown.

Where I stand on Coway stock

The name sits on my buy candidate list and I did not send an order at the August 5 close. Not because the business disappoints and not because the price is extended. I did not buy because I cannot separate what I would be paying for. Business value and contest value are inside one quotation, and one of them expires on a date I can read. I decided to look at the price after that flow is gone. This is a wait for something to end, which is a different posture from waiting for something to happen.

The sell side numbers belong here too. According to The Guru, UBS moved its valuation from 110,000 won to 125,000 won, JP Morgan from 120,000 won to 130,000 won, and Citi from 93,000 won to 105,000 won. Against the August 5 close those imply 34.7%, 40.1% and 13.1% of headroom respectively, all my own calculations. Domestically, Smart Economy on August 4, 2026 (Tue) relayed the Hana Securities view that the shares change hands at about 7.8 times twelve month forward earnings.

What 7.8 times actually requires

That multiple should not be copied without unpacking it. At 92,800 won a share, 7.8 times implies earnings per share of 11,897 won, and multiplied by the share count that implies net profit of 841.9 billion won, about $591 million. Against 2025 net profit of 617.5 billion won that is growth of 36.3%, all recovered by my own division. Now extend the first quarter net profit growth rate of 31.1% across the full year instead: net profit of 809.5 billion won, earnings per share of 11,440 won, a forward multiple of 8.11 times. The two land close together, which tells me the brokerage figure sits on the first quarter trend line and not on something more heroic. What the article does not say is which closing price it divided by, so I treat 7.8 times as a number carrying an unstated assumption that the first quarter pace holds to year end.

My two conditions and my breaking point

Two conditions have to hold together before I buy. First, after September 8, 2026 (Tue) the price returns toward the 85,000 won area, the zone the shares occupied before the contest intensified. Second, domestic rental account additions hold their first half pace. If only the first arrives, the business may simply have cooled. If only the second arrives, I am paying the support price in full.

My thesis breaks if the purchase window closes and the price holds around 92,800 won anyway. That would mean the support was never mostly about the stake contest, that the business was setting the price all along, and that my patience cost me a higher entry. The condition is deliberately written against my own position, which is how I think these should be written.

Notes for readers outside Korea

There is a structural problem with this particular trade for anyone reading in English, and it is not the usual one about brokerage access. The fact this piece turns on lives in a Korean language regulatory filing. Netmarble’s purchase declaration went to the Korean electronic disclosure system on July 10, 2026 (Fri) and was reported the same evening in Korean. English coverage of Korean mid cap corporate actions typically arrives days later, in summary, and often without the share counts that make the arithmetic possible. Anyone tracking whether the buying is running ahead of or behind schedule during the window will be reading Korean filings to do it.

The irony sits in the ownership table. Foreign investors hold 54.71% of this company, a majority, and three of the four valuation revisions I could find came from foreign houses. More than half the register belongs to people who mostly cannot read the primary document on the day it appears.

The practical points are ordinary by comparison. I could not confirm a US listed depositary receipt for this name, so direct exposure means an account with Korea Exchange access that settles in won. Broad Korea funds hold it only as a small weight. Korean dividends are paid once a year in cash and subject to withholding at source before you see them, and the payment converts at whatever the exchange rate happens to be on the payment date, not the date you read this. At about $65 a share, a position sized in whole shares is lumpy for a small account.

Two numbers I will not resolve until then

My notes carry two prices next to this company. The first is 85,000 won, the reference price the controlling shareholder built into its own filing and the level the shares occupied before the contest heated up. If the price comes back there after September 8, 2026 (Tue), I will take that as something close to the business price.

The second is 92,800 won, which is today’s price. If that level survives the disappearance of the purchase flow, I split the two forces wrongly, and the answer then is to discard the framework, not to refine the sums.

Either way the information arrives on the same date, and not before it. Until then this name sits on my list with no order behind it.

Related reading

• Kangwon Land Stock, Two Profit Numbers Pointing Opposite Ways