Goldman Sachs: U.S. Stock Returns Hit Historical Extremes, but Market Breadth Continues to Narrow
nashnova research
Goldman's latest report shows S&P 500 returns in 2026 sit at the far-right tail of the distribution since 1928 — yet the rally is narrowing to fewer stocks, valuations are near 20-year highs, and the easy phase of re-rating may already be over.
After gains this extreme, is there real fundamental support?
S&P 500 year-to-date returns land in the extreme-right tail of the historical distribution going back to 1928 — this kind of performance is exceptionally rare.
Goldman stresses it is not a hollow rally: global earnings revisions have been persistently revised upward, unlike most years when estimates drift lower.
This means → the returns are extreme, but they rest on actual earnings momentum, not pure sentiment.
Why does the rally feel fragile?
The US market's gains are increasingly concentrated in a handful of stocks, with breadth steadily narrowing; European markets show the same pattern.
At the same time, 12-month forward P/E ratios — valuations based on the next year's expected earnings — are near 20-year highs across most regions.
In plain terms = the market is rising, but fewer stocks are doing the lifting and prices are getting expensive — structurally, it is becoming brittle.
Big Tech's bill vs. the US government's — which is scarier?
Annual capex by hyperscale cloud companies — spending on data centers and chips — is expected to exceed $1 trillion next year, and its share of operating cash flow keeps climbing.
Meanwhile, US federal debt just crossed $40 trillion last week; rising long-end rates are pushing interest costs higher as a share of GDP.
This means → two "burn-rate lines" are steepening in parallel — AI capex on the corporate side, sovereign debt interest on the government side. Either one losing control would reprice the market.
How long can European earnings hold up?
Goldman forecasts STOXX 600 EPS growth of 15% in 2026, but expects a sharp slowdown to 5% in 2027.
This reflects a key warning: today's high earnings growth rate cannot be extrapolated — the deceleration next year is two-thirds.
What does the market need from here to keep climbing?
Goldman states explicitly: none of the above constitutes a standalone sell signal on its own — but "the easy phase of re-rating is probably behind us."
From here, the market needs three conditions to hold simultaneously: earnings keep delivering, AI capex stays productive (the money spent earns returns), and long-end rates remain orderly — all three, not two of three.
Put simply = it is not a call to sell; it is a shift from "gains come easy" to "every step needs verification." Goldman's core advice: watch these three lines and see which one cracks first.
市场有风险，内容仅供研究参考，不构成投资建议。