When Safe Money Pays Nearly 5%, the Premium You Get for Owning Stocks Gets Very Thin
Robert Shiller's excess CAPE yield — the extra return stocks have historically offered over safe bonds — just fell below 1%, roughly a quarter of its average since 1950. Here's what that does and doesn't mean.
By Sean Anees Saifi · Capital Wealth · Published Friday, September 11, 2026 · Source: The Wall Street Journal, September 11, 2026 edition (Heard on the Street)
Key Points
Robert Shiller's excess CAPE yield — stocks' earnings yield less a real bond yield — just fell below 1%.
That is roughly a quarter of the measure's average since 1950, a level historically followed by weak 10-year returns.
The 10-year Treasury reached 4.975% by Friday's close and the 30-year 5.354%, the highest since 2023.
Multi-decade-high yields in France and Britain suggest this is not only a story about American AI growth.
One of the largest budget deficits ever run during an expansion is helping supercharge the growth those yields are measured against.
Below 1%
Shiller's excess CAPE yield
~4x
Its average level since 1950
4.975%
10-year Treasury, Friday's close
5.354%
30-year Treasury, Friday's close
Comparing what safe bonds pay against what stocks earn is one of the oldest valuation exercises there is. It describes the price of future returns, not their timing.
In one line: As safe yields climb, the extra return investors are paid for owning stocks shrinks — a statement about valuation over a decade, not a call on next month.
Markets don't actually say anything. People say things and then put the words in the market's mouth, which is quieter than owning the opinion outright. Bonds come closest to an exception, because a yield is arithmetic rather than mood. And the arithmetic has moved a long way this year.
The premium, in plain English
Strip the jargon and the idea is simple. Buy a Treasury and you are promised a known payment on a known date. Buy stocks and you own a claim on earnings that might grow, might shrink, and won't tell you which in advance. The difference between what stocks earn and what safe bonds pay is your compensation for accepting that uncertainty. Shiller's version, the excess CAPE yield, nets a real bond yield against a cyclically adjusted price-to-earnings ratio, which averages a decade of profits so that one exceptional year can't flatter the picture.
That measure just dropped below 1%. Since 1950 it has averaged around four times that level. The paper went to press Thursday night with the 10-year Treasury at 4.943%; by Friday's close it was 4.975% and the 30-year 5.354%, the highest since 2023. Readings this thin have historically been followed by poor S&P 500 returns over the next decade — and historically is carrying real weight in that sentence. These are relationships drawn from a handful of non-overlapping decades, not forecasts, and they have been early before by years at a stretch.
A valuation question, not a market call
What the measure can't tell you is what happens next month, or next year. It describes the price being paid today for a decade of future earnings, which makes the honest response a planning response rather than a trading one. Two caveats sit on the other side of it. The growth those yields are measured against is being supercharged by one of the largest deficits ever run during an expansion, with corporate borrowing pouring disproportionately into AI infrastructure — productivity gains could justify all of it, and that if keeps getting larger. Meanwhile France and Britain are printing multi-decade-high yields without being AI powerhouses at all, which points at debt as much as at growth.
None of this argues for abandoning stocks, and anyone who tells you a single valuation ratio times the market has a chart to sell you. It argues for knowing what you're being paid to take the risk, and for checking that the next decade's plan doesn't quietly assume the last one repeats. Savings rate, time horizon and how much risk the plan actually requires — that's the conversation this number belongs in.
What It Means For Your Portfolio
Hold — a valuation reading, not a signal to act today
A thinner equity premium doesn't tell you what the market does next; it tells you the price of future earnings has gone up while the risk-free alternative finally pays something real.
General planning principles, not advice for anyone in particular: valuation measures describe the price of expected returns, not their timing, and historical relationships are not forecasts. A thinner premium for owning stocks is a reason to revisit savings rate, time horizon and how much risk a plan genuinely requires — it is rarely a reason to make a sudden change.
In the book, the bias is cautious: defensive on duration, neutral on stocks, confidence medium. No long-duration bonds are being added and nothing new is being bought, because the September letter's conditions — a core print of 0.1% or less, a Fed hold, and vol-of-vol back under 90 — were not met after Friday's report.
