Kakao Won the SM Entertainment Bidding War. A Court Is Still Deciding If That Was Legal.

Two acquirers wanted the same K-pop label. One bid ₩120,000 a share and asked shareholders to sell. The other spent ₩240 billion buying shares on the open market at prices above that bid, for the specific effect of making sure nobody had to. The second company won the company. Its founder then spent a year under indictment for the trades that won it, was acquitted, and is now waiting on an appeal.

That is the SM Entertainment fight in one paragraph, and it is worth a founder's attention for a reason that has nothing to do with K-pop. Kakao's founder Kim Beom-su did not get indicted for outbidding a rival. He got indicted for how the bidding worked, in a jurisdiction where the line between "won the auction" and "rigged the auction" runs through a statute that asks about intent rather than outcome, and where a court can look at the identical set of trades twice and reach different answers. The Seoul Southern District Court's first answer, in October 2025, was not guilty. Prosecutors appealed. The second answer is due in October 2026, roughly six weeks after this sentence gets published.

Nothing about the tender offers explains why Kakao and HYBE were fighting over SM Entertainment in the first place. The reason sits in a contract most SM shareholders had never heard of.

Lee Soo-man founded SM in 1995 and, for most of the company's public life, took roughly 6% of its revenue every year through a private company he owned outright, called Like Planning. It paid him for producing, a job he had by then mostly stopped doing, a career stage most people call retirement and SM's filings called an ongoing services agreement. SM had never paid a dividend until 2021.⁴ The producing contract, once you extended its logic to its stated terms, would have paid Lee royalties on the company's back catalog until 2092.⁵

Align Partners, an activist fund, built a position of about 1% in SM and used it to argue that the arrangement was value extraction wearing a job title.⁶ In October 2022, SM's board terminated the contract. In January 2023, Align filed litigation against Lee and seven former board members.⁶

A 1% shareholder ending a related-party contract is the kind of governance story that usually stays a governance story. This one didn't, because SM's professional management, led by co-CEO Lee Sung-soo (no relation to the founder), needed a way to keep Lee Soo-man from simply buying his way back into control now that his fee had been cut off. On February 3, 2023, SM agreed to issue Kakao roughly 9% of the company through new shares and convertible bonds, at a discount, to seal a "strategic partnership." Lee Soo-man read it correctly as a dilution aimed at him and filed for an injunction five days later.⁷

That is the piece worth sitting with before the tender offers start. The war wasn't outside capital discovering an undervalued asset. It was incumbent management picking a friendly shareholder to dilute the founder who had just lost his toll booth, and the founder finding a much larger friendly shareholder of his own. HYBE didn't show up because SM was cheap. It showed up because Lee Soo-man needed someone with ₩422.8 billion who would take his side.

Here is the first structural fact that shapes everything downstream, and it's worth stating plainly because most founders assume takeover defenses travel the same way everywhere. They don't.

A US board facing a hostile threat can adopt a shareholder rights plan, the poison pill, a mechanism Delaware courts blessed decades ago and boards can generally adopt without a shareholder vote. Korea's Commercial Act contains nothing equivalent. What a Korean board can do instead is issue new shares or convertible bonds to a friendly third party under Article 418(2), but only if that issuance serves the company's genuine management purpose, such as raising financing it actually needs. Korean courts have read that requirement strictly since at least the Hanjin KAL case in 2020, when a private equity fund enjoined a defensive issuance for failing exactly that test.⁸

SM's placement to Kakao ran into the same wall. On March 3, 2023, the Seoul Eastern District Court blocked it, finding SM had no urgent need for the financing and that the deal would mainly serve to entrench the side that proposed it, at the expense of the founder's and other shareholders' stakes.¹⁰ SM cancelled the issuance to comply.

Put the two courts next to each other and you get the actual rule Korean boards operate under: you may pick a favorite shareholder to defend yourself, but only if you can also explain, with a straight face, why you needed the money. "We needed a friend" has never been an accepted answer, in 2020 or in 2023. Delaware would have let SM's board explain itself in a proxy statement, months later, to shareholders who had already been diluted. Seoul made it explain itself to a judge, first, before a single share changed hands. Same instinct, entirely different amount of homework required to act on it.

Once the injunction landed, Kakao had exactly one instrument left: buy shares itself, from anyone willing to sell them, at whatever price it took. That is not a workaround. In a country with no pill and, as of 2026, still no mandatory bid rule on the books, buying shares in the open market is close to the only durable way to acquire control of a public company.¹⁶ Everything that follows is what that instrument looks like when two well-funded acquirers use it against each other at the same time.

HYBE's plan was conventional by American or European standards. Buy the founder's 14.8% for cash. Then run a public tender for another 25% at a fixed price, aiming to cross 40% and take control outright.¹²

The mechanical problem with a tender offer is one that reads as obvious only after you've watched it fail: it works exclusively when the tender price beats the market price. HYBE offered ₩120,000. Through the two-week offer window, SM's stock traded above that figure and closed the period at ₩127,600.⁹ Any shareholder holding SM stock could get a better price by doing nothing and selling on the exchange the next morning. Rational shareholders did approximately that. HYBE's tender drew almost no acceptances, and the company that had just spent ₩422.8 billion to become SM's largest shareholder controlled a fraction of what its plan required.

A tender offer, in other words, is not a purchase. It's a dare that says "the market is wrong about this stock, and I'm betting real money that you'll agree with me before this stops." The market's response to HYBE was to keep trading above the dare, which meant somebody was buying at prices HYBE hadn't offered. Prosecutors would later argue that "somebody" wasn't the market being efficient. It was Kakao, on purpose.

Between mid- and late February 2023, Kakao and Kakao Entertainment made 553 separate purchases of SM stock, spending roughly ₩240 billion, at prices above HYBE's ₩120,000 tender.³ Prosecutors alleged this wasn't ordinary accumulation ahead of a later bid. It was designed to keep the market price pinned above HYBE's offer for the specific purpose of making the tender fail, a tactic Korean enforcement has a name for: closing-price management, buying concentrated at the end of the trading session to hold the print where you want it. The charge sits under Article 176(3) of Korea's Capital Markets Act, which bars trading "for the purpose of fixing or stabilizing" a security's price.¹⁸

Notice the word doing all the legal work there: purpose. Every large, real purchase moves price. A pension fund building a position moves price. A strategic acquirer accumulating shares before making its own tender offer, which is precisely what Kakao also did, moves price. None of that is manipulation on its own, in Korea or anywhere else that regulates markets, because if it were, every serious buyer would be a felon. What separates a defensible accumulation from an indictable one is evidence that the purchases were never about owning the stock at all, just about denying someone else a price.

That evidence, structurally, lives in exactly one place: internal communications. Not the trades. The trades looked the same either way, 553 purchases at rising prices during a rival's tender. What decides the case is whether a chat log or a call recording has someone stating the sabotage plainly, on the record, the one thing nobody involved in a corporate maneuver is ever supposed to do, and the Seoul Southern District Court's first-instance ruling in October 2025 found the prosecution hadn't produced enough of that. Large-scale buying that happens to move the price during a competing tender offer, the court held, cannot be manipulation merely because it had that effect; the state hadn't proven intent to artificially distort the price rather than genuinely compete for the stock.²⁰ Kim Beom-su, former Kakao Entertainment investment chief Bae Jae-hyun, two former Kakao co-CEOs, and the corporate entities themselves were all acquitted.²⁰

Prosecutors were not persuaded by their own loss. On appeal, they argue the trial court simply didn't grapple with the messages and recordings they submitted.¹⁹ The appellate court heard arguments through mid-2026 and is expected to rule in October.²¹ Whatever it decides will not be a ruling about whether Kakao bought a lot of stock in February 2023. Everyone agrees it did. It will be a ruling about what was in someone's head while doing it, which is a strange thing to litigate for two and a half years over a K-pop label, and an entirely normal thing to litigate under securities law in a country that, like most of them, criminalizes intent rather than volume.

Kakao's actual winning move, once the private placement was dead, was to stop trying to be clever and just outbid. On March 7, 2023, it announced a ₩1.25 trillion tender offer, ₩150,000 a share for up to 35% of SM, a 25% premium to HYBE's number.¹¹ Five days later HYBE said the price had run past any range it considered rational and withdrew.¹² By March 24, HYBE had agreed to do the one thing that made the whole exercise pay for itself: tender its own 14.8% stake into Kakao's ₩150,000 offer, a roughly $87 million gain on shares it had bought five weeks earlier.¹⁴¹⁵

Read the sequence again and notice who actually profited from losing. HYBE never ran SM Entertainment for a single day. It identified a control fight worth entering, bought a blocking-sized position, forced a rival to pay up to make it go away, and exited with a nine-figure gain in under two months. Losing a takeover battle and making $87 million doing it is not a consolation prize. For a strategic buyer without the balance sheet or the appetite to actually win, it may be the better outcome on offer, and HYBE's board would have been negligent to have preferred the alternative.

Kakao, for its part, closed the tender with combined ownership of 39.87% between Kakao Corp and Kakao Entertainment, becoming SM's controlling shareholder.¹⁷ It did not get 35% at a discount. It got roughly 40% at whatever price it took to make sure nobody else could get there first, plus the legal exposure that came from how it kept the price up while doing it.

The premium didn't hold, which is the detail every party to a bidding war would rather you forget once the press release goes out. Two years later, in May 2025, HYBE sold what by then was a 9.38% SM stake to Tencent Music Entertainment in a block trade at ₩110,000 a share, ₩243 billion total, making the Chinese streaming company SM's second-largest shareholder behind Kakao.²²²³ That price is 27% below the ₩150,000 Kakao paid at the peak of the fight to make its rival go away. A control premium prices the moment someone needs the fight to end. It says very little about what the stock is worth once it doesn't have to.

Step back from SM Entertainment specifically, because the transferable lesson isn't about K-pop labels. It's about what happens when a jurisdiction gives you exactly one tool for winning a public-company fight and then polices how you use that tool after the fact.

SM Entertainment sits at the intersection of three structural facts about Korean capital markets, and all three were necessary to produce a fight that looked like this one:

No dual-class stock. Article 369 of the Commercial Act makes one share, one vote mandatory for listed companies.¹³ Neither HYBE nor Kakao could buy votes cheaply through a Control Wedge the way a founder can in the United States. Every percentage point of control cost a percentage point of real economic ownership, paid in cash, at whatever the market demanded that week.

No mandatory bid rule. The United Kingdom's Takeover Code requires anyone crossing 30% ownership to offer all remaining shareholders the same price. Korea has debated an equivalent rule for years and, as of mid-2026, it remains a bill before the National Assembly rather than law.¹⁶ Without it, nothing obliged Kakao to offer minority shareholders equal treatment once it had control. It only had to win the specific shares it was chasing, and the rest of the register found out what its shares were worth by watching the news like everyone else.

No poison pill, and a strict judicial test for the substitute. As covered above, the only defensive tool available to a Korean board is a friendly issuance that has to clear a genuine-business-need bar, one that has failed in court before and failed again here.

Take any one of those three away and the SM fight becomes a different story. Add a mandatory bid rule and Kakao's 40% stake would have triggered an obligation to buy out everyone else at the same price, changing the economics of the whole contest. Allow dual-class stock and Lee Soo-man could have kept votes while selling economics, instead of needing to find a buyer for his entire block. Allow an easier pill and SM's board might never have needed a court's permission to pick a side.

None of that makes Korea's framework worse than Delaware's or London's. It makes it different in ways that determine exactly where the fight will actually happen, and founders who assume their home jurisdiction's playbook travels are the ones who find out the hard way that it doesn't. The playbook that travels is the one where you check, before you commit capital or accept an offer, which of the three levers above exists where your target is actually listed.

If you sit on the board of a company that could become the rope in a two-bidder tug of war, or you're the acquirer sizing one up, work through this before signing anything:

1. Find the mandatory bid threshold, if one exists. If your jurisdiction has one, a crossing acquirer owes every shareholder the same price, which changes what "winning" costs. If it doesn't, as in Korea through 2026, control goes to whoever can hold the highest price the longest, and minority shareholders get whatever price they can find on the exchange that week.
2. Compare the tender price to the trading range before you evaluate it, not after. A tender priced below where the stock is actually trading is not a real offer. It's a floor being tested. HYBE's ₩120,000 bid against a ₩127,600 close told you the outcome before the offer period even closed.
3. Read the defensive-issuance standard, not the board's press release. "Strategic partnership" and "financing need" are marketing language for the same instrument. Ask what the courts in that jurisdiction actually require to uphold a friendly placement against a challenge, and assume a large enough aggrieved shareholder will test it.
4. If you're accumulating shares during a live tender fight you didn't launch, document your investment thesis contemporaneously, and mean it. The line between winning an auction and being prosecuted for it runs through your own internal messages, not your trade blotter. Two sets of 553 identical purchases can end in an acquittal or a conviction depending entirely on what somebody wrote down at the time.
5. Price the loser's exit alongside the winner's prize. Before you assume you need to win a control fight to make it worth entering, model what a strategic block position is worth to you if the other side simply pays you to leave. HYBE's board planned for control and got a very good trade instead. Plan for both outcomes from the start, and you won't need the second one to feel like a consolation.

None of this is legal advice for a specific deal. Takeover defenses, mandatory bid thresholds, market-manipulation standards, and criminal intent requirements differ by jurisdiction, move over time, and in Korea's case are actively the subject of pending legislation and a live appeal. Get counsel licensed where your target trades before you buy a single share on the strength of this article.

A Korean trial court said no in October 2025, finding the prosecution hadn't proven the purchases were made with the specific intent to fix or artificially stabilize SM's price, as opposed to a genuine attempt to acquire and win a competing bid.²⁰ Prosecutors disagree and have appealed, arguing the court didn't adequately weigh the messages and calls they submitted as evidence of intent.¹⁹ A ruling is expected in October 2026. Until then, the honest answer is that one court has said not guilty and a second court is deciding whether to agree, which is a normal, if uncomfortable, place for a securities-fraud case to sit for years, because intent is exactly the kind of question that takes courts a long time to answer twice.

It could have. HYBE said publicly that the price had exceeded what it considered a "fair range," a decision to protect its own balance sheet rather than a legal constraint.¹² Folding turned out to be the more profitable choice: tendering its stake into Kakao's higher offer produced an $87 million gain in roughly six weeks, a better risk-adjusted outcome than running SM Entertainment would likely have been for an acquirer with its own separate business to run.¹⁵ A bidding war doesn't require both sides to want to win outright. It only requires both sides to be willing to be paid to stop.

SM Entertainment's shareholders never got to vote on any of this. The producing contract ended in a boardroom. The private placement died in a courtroom. The two tender offers were decided in the open market, share by share, by whoever had the larger checkbook and the higher tolerance for a later phone call from the Financial Supervisory Service. Control changed hands twice in six weeks, and the artists whose contracts were the entire point of the fight found out where they worked from the same press releases everyone else did.

The appeal ruling lands in October 2026. If it goes against Kakao, the lesson for the next set of acquirers won't be "don't buy stock during a rival's tender offer." Every serious bidder in every market does that. The lesson will be narrower and more uncomfortable: write down why you're buying, before a prosecutor has to ask you what you meant.