Shiller PE (CAPE)
The S&P 500 priced against ten years of inflation-adjusted earnings.
41.4×
Expensive
As of 2 Oct 2026 · Daily (monthly history)
Higher than 97% of months since 1871. Comparable to Mar 1999.
History
What the Shiller PE is
The Cyclically Adjusted Price-to-Earnings ratio — CAPE, or the Shiller PE — divides the S&P 500's price by the average of its last ten years of earnings, with both sides adjusted for inflation. Robert Shiller and John Campbell introduced it in 1988 to solve an obvious problem with the ordinary P/E: earnings collapse in recessions, so a trailing P/E goes to its most alarming reading exactly when stocks are cheapest.
Averaging a decade smooths a full business cycle into the denominator. What comes out is a valuation number you can compare across eras — 1929, 1982, 2000, today — without the cycle doing the talking.
How it is built
Ten years of reported S&P 500 earnings are each restated into today's dollars using CPI, then averaged. The index price, also in today's dollars, is divided by that average. Shiller's own series runs monthly back to January 1871, which is why this gauge can quote percentile ranks against a century and a half rather than a few decades.
DeepValues persists the monthly history and refreshes the current reading daily, so the headline number tracks the market between month-ends while the distribution behind it stays the canonical long series.
Source: Robert Shiller CAPE series · series page
How to read it
The bands come from the long-run distribution itself: the mean sits near 17×, one standard deviation up near 24×, two near 32×. So under 10× is deeply undervalued territory (1920, 1932, 1982), 10–17× undervalued, 17–24× fair, 24–32× elevated, and above 32× expensive by any historical standard — a reading only 1929, 1999–2000 and the 2021 peak have reached.
CAPE's genuine predictive content is about long horizons. The relationship between starting CAPE and subsequent 10-year real returns is one of the more durable regularities in equity data: high starting multiples have historically been followed by lower decade-ahead returns. That is a distribution, not a forecast.
Bands
- Deeply Undervaluedbelow 10.0×
- Undervalued10.0× – 17.0×
- Fair Value17.0× – 24.0×
- Elevated24.0× – 32.0×
- Expensive32.0× and above
What it does not tell you
The accounting underneath the ten-year window has changed. FAS 142 (2001) ended goodwill amortisation and mark-to-market write-down rules changed how sharply recessions hit reported earnings — both of which mechanically raise the modern CAPE relative to the pre-1990s series. Comparing today's 38× to 1929's 32× is not quite comparing like with like.
The ten-year window also carries its own baggage. For most of 2018 the denominator still contained the 2008–09 earnings collapse, which flattered the ratio downward; when that window rolled off, CAPE "fell" without anything changing about price. And like every valuation measure, it says nothing about the next twelve months.
The other gauges
- Buffett IndicatorTotal US stock market value divided by the size of the US economy.
- CBOE VIXWhat the options market is charging for the next thirty days of risk.
- AAII SentimentHow bullish individual investors say they are — the classic contrarian read.
- 10-Year Treasury YieldThe discount rate underneath every valuation on the platform.
- Fed Balance SheetThe liquidity tide — how much of today's price is money rather than business.
Published for information only, not investment advice. Readings refresh hourly; the as-of date above is the date of the underlying observation.