The long-end rate scenario confronting US equities is no longer hypothetical. On October 6, the 10-year Treasury yield stood at 5.27% and the 30-year yield at 5.64%, according to the US Treasury’s daily yield-curve data.
That creates a sharper Q4 test for the S&P 500’s AI leadership. The question is not simply whether cloud and AI revenue can continue expanding. It is whether the companies carrying that leadership can turn immense spending on computing infrastructure into returns fast enough to justify their valuations when investors can obtain materially higher yields from long-dated government debt.
There is meaningful fundamental support. Earnings expectations remain strong and large technology companies have already delivered on cloud and AI-related revenue. Yet higher long-term yields raise the discount rate applied to future cash flows, an issue that bears especially heavily on richly valued businesses whose investment cases depend on growth extending far into the future.
A 5.27% 10-year Treasury resets the valuation test
Equity valuations do not need an earnings recession to come under pressure. A rise in the risk-free rate can reduce the price investors are willing to pay for a given stream of expected profits, even when those profits remain intact. That is the central arithmetic challenge posed by a 5%+ 10-year yield.
In its March 2026 Earnings Insight report, FactSet put the S&P 500’s forward 12-month price-to-earnings ratio at 19.9. That matched its five-year average, but it was above the 10-year average of 18.9. The distinction matters: the index was not at an unprecedented valuation by the shorter comparison, but it also was not priced at a long-run discount that could readily absorb a higher discount rate.
A return toward the 10-year valuation norm would amount to multiple compression unless earnings estimates rose enough to offset it. This does not establish that such compression will happen, nor does the March reading describe the index’s valuation on October 6. It does show the starting vulnerability identified in the available consensus data: the market had less valuation room if long-end yields stayed high.
AI leaders are central to that sensitivity because their share prices reflect expectations of sustained growth and future monetisation. A 5.27% 10-year yield therefore changes the standard these companies must meet. Delivering growth is helpful; delivering growth that exceeds already demanding expectations becomes more important.
Fed projections and the long-yield problem
The Federal Reserve’s September projections offer little support for assuming rapid and lasting relief at the long end of the curve. The median forecast called for real GDP growth of 2.3% in 2026 and 2.4% in 2027, while core PCE inflation was projected at 3.4% and 2.5%, respectively. Those figures point to an economy with continuing growth but inflation still above the Fed’s target.
The same Federal Reserve projections showed a median federal-funds-rate forecast of 4.1% at year-end 2026 and 4.1% in 2027. That is below 5%, but remains above the policymakers’ 3.2% longer-run median. In other words, a lower policy rate and low long-term borrowing costs are not interchangeable outcomes.
Long-dated yields reflect expectations that extend beyond the next policy decision. With projected inflation above target and policy expected to remain restrictive relative to its longer-run setting, investors have reason to distinguish prospective short-rate easing from a decisive fall in long-term discount rates. For equity markets, that distinction is material. An AI-led index rally can coexist with rate cuts, yet still face pressure if the 10-year yield does not follow them lower.
This is also why the debate should not be reduced to whether the Fed cuts. Q4 equity performance will depend on the interaction of earnings delivery, valuation and the yield investors demand over a much longer horizon.
Earnings growth can cushion the index
Consensus earnings estimates provide the strongest counterweight to the rate concern. FactSet’s March forecast called for S&P 500 earnings growth of 19.0% in Q4 2026 and 17.1% for full-year 2026. If achieved, that pace would give the index a substantial fundamental cushion against some valuation compression.
The AI thesis also has evidence beyond projections. S&P Global Market Intelligence, in its Q1 2026 earnings review, said the Magnificent Seven broadly exceeded expectations on cloud and AI-related revenue. The Information Technology sector delivered 33% year-over-year operating-EPS growth.
Those results matter because they demonstrate that the market’s AI leadership has been accompanied by operating earnings growth, rather than solely a narrative about a distant technology opportunity. They also help explain why a high-yield environment need not automatically break the S&P 500: genuine earnings expansion can support prices even when valuation multiples are constrained.
But the same evidence concentrates the burden. A small group of large companies must keep producing enough cloud and AI-related growth to validate their elevated strategic importance to the index. As valuations become harder to defend, quarterly earnings are likely to be judged not only on whether revenue rises, but on whether the resulting profits and cash flows show that AI investment is becoming economically productive.
$495 billion of hyperscaler capex is an execution test
The investment scale raises the difficulty of that task. Alphabet, Amazon and Microsoft projected approximately $495 billion in 2026 capital expenditure, up 61% from 2025, according to S&P Global Market Intelligence. The same report said analysts could not yet clearly link the spending to appreciable returns.
That gap is the critical Q4 vulnerability in the AI leadership case. Capex can build capacity, improve products and defend market position, but it is not itself proof of returns. At nearly half a trillion dollars across three companies, the spending shifts investor attention from the size of the AI opportunity to the efficiency and timing of monetisation.
Higher yields make that scrutiny more consequential. When discount rates are lower, investors may be more willing to value distant benefits from infrastructure spending. When the 10-year Treasury is above 5%, the market is more likely to demand evidence that revenue gains can produce durable margins, cash flow and returns on investment rather than merely support a larger capital base.
Competition could further narrow the room for error. S&P Global Ratings identified intensifying competition from open-weight models and Chinese technology providers across multiple layers of AI. That does not negate the Q1 earnings strength or determine the future winners. It does challenge the assumption that today’s leaders will preserve pricing power and excess returns indefinitely.
For the S&P 500, AI leadership can survive a 5%+ Treasury market if earnings growth stays strong and investment begins to produce demonstrable returns. The more difficult proposition is that leadership can remain insulated from higher yields while capital spending accelerates and competitive conditions make those returns less certain.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.
