Why Kakao's Spinoff Could Reshape How the World Values Platform Companies

A 17.4 Trillion Won Mystery

Kakao is worth less than half what analysts say it should be. That is the raw number driving everything. According to sum-of-the-parts estimates from domestic and international brokerages — including reports from Shinhan Securities, KB Securities, and Jefferies — the Kakao Group’s latent value sits at 34.2 trillion won. Its three-month average market capitalization is 16.8 trillion won. The difference — 17.4 trillion won — is not a rounding error. It is a structural blind spot that has persisted despite repeated corporate promises to unlock value.

The split that Kakao announced on August 23 is an attempt to close that gap. But the structure of the split matters as much as the numbers. What Kakao has chosen — a human spinoff, or inhyeok bunbal in Korean corporate law — is unusual, politically deliberate, and worth watching closely as a potential precedent for the entire Korean corporate sector.

Not What You Expect From a Chaebol Spinoff

In a typical Korean spinoff, the parent company retains 100% of the new entity’s shares and later lists the subsidiary separately. This has become known in Korea as jjogae — literally “tearing apart” — and it has drawn sharp criticism from institutional investors and minority shareholder advocates alike. The fundamental flaw is blunt: minority shareholders of the parent get no direct stake in the spun-off business. If that subsidiary surges in value, the parent’s shareholders do not share in it proportionally. When Kakao previously discussed similar moves over the past two years, regulators flagged the problem. The Financial Services Commission has been tightening rules around duplicate listings that do not adequately protect ordinary shareholders, and the Korea Exchange has signaled it will scrutinize such arrangements more rigorously in listing applications.

Kakao’s human spinoff sidesteps that trap entirely. Instead of the parent holding the new companies’ shares, the newly created entities issue stock directly to existing Kakao shareholders in proportion to their current holdings. Kakao shareholders will receive shares in both KakaoAI — valued at 36% of net asset book value — and KakaoX — carrying the remaining 64% — according to their existing stake. No one gets left behind. No one gets diluted without consent. The mechanics are straightforward, but the political signal is complex: the board is choosing a structure that makes it visibly difficult for insiders to extract private benefit from the separation.

This is not merely procedural. It is a signal that Kakao’s board is calculating its reputation cost with unusual precision. In a market where investor activism is growing — and where retail holders control a significant portion of voting power — protecting the optics of fairness has become itself a corporate strategy. The choice also reflects the personal dynamics of founder-CEO Kim Beam-su, who has faced recurring accusations of entrenchment and who now needs the transaction to succeed on both financial and credibility grounds.

The Two Engines

KakaoAI takes the messaging and artificial intelligence side — essentially the KakaoTalk platform augmented by what the company calls its “agentive AI” ecosystem, branded as Carrana. The Carrana platform has been rolling out to KakaoTalk’s 68 million Korean monthly active users, and the company says it is already integrating generative AI capabilities across customer service, content recommendation, and productivity tools. KakaoAI is targeting 1 trillion won in AI revenue by 2030, with projections of more than 20 million daily AI users and total revenue above 6 trillion won — roughly a 20% compound annual growth rate. The valuation logic is that a messaging platform with deep AI integration trades more like a software company than a traditional telecom-adjacent conglomerate, and software companies command materially higher multiples in current markets.

KakaoX absorbs the vertical businesses: techfin through its Daum Financial Services subsidiary, content production, mobility operations including the Drive X division, and the global investment portfolio that includes stakes in overseas technology firms. KakaoX will come into existence with roughly 2.3 trillion won in cash from the sale of Twohand’s stake and other liquidation moves, plus the capital tied up in mobility and entertainment operations. Total investable resources: about 6.4 trillion won. That war chest is substantial enough to fund autonomous driving development, expand the financial services footprint, and make strategic acquisitions without depending on the parent company’s balance sheet.

The arithmetic is striking because the logic is simple: a messaging platform with deep AI integration does not look like a mobility and payments company. Investors price them differently. Bundling them together has compressed Kakao’s multiple, creating a conglomerate discount that penalizes the high-growth AI business by associating it with slower-growing, capital-intensive operations. Unbundling them lets each story trade on its own merits.

The Bigger Picture: Korean Corporate Reform

This is the first major test of whether a top-tier Korean tech group can restructure without triggering the old chaebol accusations — that insiders are extracting value while ordinary shareholders get the short end. The stakes extend far beyond Kakao. Samsung, Hyundai, and SK Group have all faced pressure to unlock shareholder value through structural reform, and none have attempted a human spinoff at this scale. Kakao is effectively volunteering to be the proof of concept, betting that a successful model will create pressure on peers to follow — or risk falling further behind on valuation.

There is also a timing element that cannot be overlooked. South Korea’s government has signaled that chaebol governance reform will remain a policy priority through 2027, with the Ministry of Economy and Finance and the Financial Services Commission jointly pushing for greater transparency and shareholder-friendly restructuring. A successful human spinoff that demonstrably closes the valuation gap without harming minority shareholders would give reformers a concrete reference point and strengthen their hand against entrenched business groups resistant to change. A failure — if the market still punishes both entities or if either company struggles to operate independently — would set the governance reform agenda back by years and hand ammunition to skeptics who argue that structural tinkering cannot overcome Korea’s deeper cultural and ownership problems.

The second-order effects are already visible. Legal firms are fielding inquiries from other Korean conglomerates about the mechanics of inhyeok bunbal. Asset managers with large Korean equity positions are watching closely, knowing that if the model works, it could trigger a wave of similar transactions across the KOSPI. Competitors like Naver are likely reassessing whether their own bundled structures create similar hidden value — or hidden risk if investors begin demanding the same transparency.

What Happens Next

The split itself does not create value. It creates the conditions for the market to price things correctly. Whether that happens depends on execution in the quarters and years ahead. Kakao has said it will buy back and retire 3 trillion won in shares this year as a goodwill gesture — a form of shareholder return that ties the reorganization to immediate financial benefit and helps offset any short-term volatility from the separation. The buyback is also a signal that management believes the market is undervaluing the combined business even before the split takes effect.

The real question is whether KakaoAI can deliver on its AI revenue targets fast enough to convince investors the platform side is worth the premium brokerages assign it. KakaoTalk’s user base is massive but concentrated in Korea, and the company faces stiff competition from Naver’s Line and Google’s messaging products. If Carrana adoption stalls or monetization proves harder than expected, the AI story loses its anchor and the new entity could trade at a discount rather than a premium. Meanwhile, KakaoX must deploy its 6.4 trillion won war chest in ways that generate returns exceeding what the market currently implies from Kakao’s bundled valuation. That means the mobility business needs to reach profitability faster, the financial services division needs to find growth beyond incumbent positions, and the investment portfolio needs to be managed with discipline rather than sprawl.

If both succeed, the 17.4 trillion won gap shrinks — potentially dramatically, as multiple expansion compounds with earnings growth in each independently priced entity. If either stumbles, the new structures merely provide a cleaner view of underperformance rather than a reset, and the valuation discount simply gets disaggregated rather than eliminated.

For investors outside Korea, the broader takeaway is simpler than the corporate law details. Platform companies worldwide — from Line in Japan to Tencent in China to Meta’s various divisions — face the same problem: markets struggle to price diversified tech giants fairly when different parts of the business deserve fundamentally different multiples. Kakao is attempting a structural solution rather than just hoping for a re-rating, which is either clever or desperate, depending on how the market responds in the quarters ahead. The world will be watching to see if a human spinoff can deliver on its promise, because the answer will determine whether Korean corporates follow or flounder — and whether the conglomerate discount that has weighed on Asian platform stocks for decades finally begins to break.