SK Hynix just dropped a 40 trillion won bomb. Stock buyback. Not a drill.
In the DeFi winter, we didn't see this kind of capital return. Protocols burned tokens, sure. But real companies? They hoarded. SK Hynix is different. They're calling their own cycle. t saying.
Context: SK Hynix is the world's leading HBM memory manufacturer. HBM3E? They're the sole supplier for Nvidia's H200 and B100. That's a monopoly in a gold rush. Their cash flow has exploded. The market cap? Around 130 trillion won. This buyback represents 30% of market cap over three years. That's aggressive. Every crash is just a story that hasn't been written yet. This buyback writes a new chapter.
Core: The math is brutal. SK Hynix's free cash flow for 2024 is estimated at 12 trillion won. By 2025, that could double to 24 trillion. Their capital expenditure peaked in 2023 at 17 trillion. Now they're reaping the rewards. The buyback is funded by operational cash, not debt. Their net cash position is improving. I didn't believe it at first. I checked the balance sheet. The numbers are real.
Let me break down the mechanics. A 40 trillion won buyback over three years means roughly 13 trillion per year. That's about 50% of their projected 2025 FCF. The rest goes to growth and dividends. They're committing to a minimum 50% payout ratio of adjusted net profit. That's higher than any DRAM competitor. Micron pays 20%. Samsung pays 25%. SK Hynix is leading the capital return race.
But here's the contrarian angle. The market is celebrating. Citi just upgraded to buy with a 310,000 won target. That's a 30% upside from current levels. But I see a trap. The buyback is a signal of confidence, but it's also a sign that the company expects no better use for the cash. If they had a 50% CAGR growth opportunity, they'd invest. They're not. They're returning capital. That means the growth phase is maturing. Smart money often sells when companies announce large buybacks at peak earnings. Look at history. In 2021, when Nvidia announced a $10 billion buyback, the stock peaked three months later. Not saying it's the same. But patterns repeat.
Contrarian: The real risk is not the buyback. It's the HBM competition. Samsung is ramping HBM3E production. Micron is winning orders from AMD. SK Hynix's monopoly is under threat. Their premium pricing power will erode. The buyback masks that reality. Retail investors see the headline and buy. Institutional investors see the dilution and sell. Every crash is just a story that hasn't been written yet. This one's about Samsung's technology roadmap.
Based on my audit experience, I've seen this play out in crypto. When a DeFi protocol announces a massive token buyback, the token price pumps, then dumps as insiders distribute. SK Hynix is different because it's a real company with real earnings. But the psychology is the same. The market is pricing in peak FCF. If HBM margins compress from 60% to 40%, the buyback becomes unsustainable. The stock could correct 30%.
Takeaway: The key levels to watch are 250,000 won and 300,000 won. If SK Hynix breaks above 280,000 on the buyback news, it's a short-term momentum trade. But if it fails to hold 250,000, the support is broken. I'm not buying here. I'm waiting for the pullback. The buyback is a floor, not a catalyst. The real catalyst is the October earnings call. If they guide lower HBM market share, the buyback won't save the stock. In the DeFi winter, we didn't learn this lesson. I did. I lost 40% in 2018 chasing buybacks. Not this time.
So, is SK Hynix a buy? The answer is: it depends on your time horizon. For the next six months, the buyback provides a safety net. For the next two years, the HBM cycle is peaking. I'd rather wait for the next generation of memories. When the hype dies, the real value emerges. t saying.
