The 30-year U.S. Treasury yield has climbed to its highest level since 2007 while U.S. government debt hovers around $40 trillion. This combination is lifting long-term discount rates and widening perceived risk premia across equity markets, directly challenging valuations that rely on cash flows far into the future.
High-duration growth and AI equities sit at the center of that pressure. Market leaders such as NVIDIA Corp. (NVDA), Microsoft Corp. (MSFT), Alphabet Inc. (GOOGL), and Advanced Micro Devices (AMD) trade on elevated multiples that embed aggressive assumptions about long-run AI adoption and profitability.
Historical episodes in 1999-2000, 2013, and 2021-2022 show that when long-dated Treasury yields reset sharply higher and remain elevated, richly valued growth cohorts have tended to underperform for quarters or years. In those periods, valuation compression often persisted even where the underlying secular technology thesis ultimately proved directionally correct.
With long-term yields now at multi-decade highs, the market is actively repricing the cost of capital applied to AI infrastructure and software platforms. That repricing does not require an outright reversal in AI demand to weigh on share prices; it only requires investors to demand higher returns on long-dated, narrative-heavy cash-flow streams compared with shorter-duration or more cyclical equities.
Terminology
- 01Discount rates: Interest rates used to convert future cash flows into today’s present value.
- 02Risk premia: Extra return investors demand for holding riskier assets over safer benchmarks.
- 03Cost of capital: Required return that investors expect for providing financing to a company.