Statistics · 2026 · Stocks
Shiller CAPE Ratio Statistics (2026)
The Shiller CAPE ratio was 40.9 in September 2026, the S&P 500 price divided by its 10-year average of inflation-adjusted earnings. That is 2.3× the 17.8 average and higher than in 98.9% of the 1,749 months since January 1881. Only 19 months were higher, 18 of them in the dot-com boom. The record is 44.19 in December 1999.
Key takeaways
- The CAPE averaged 40.9 in September 2026, against a 17.8 mean and a 16.6 median since January 1881.
- Today’s reading is higher than 98.9% of 1,749 monthly values; 19 months were higher.
- The CAPE peaked at 44.19 in December 1999 and bottomed at 4.78 in December 1920.
- Stocks bought in the top fifth of CAPE readings returned 3.6% a year after inflation over the next 10 years, against 10.8% in the bottom fifth.
- The correlation between starting CAPE and the next 10 years’ real return is −0.49: an association, not a forecast.
- The regular P/E was 26.0 in September 2026, below the CAPE because the two use different earnings bases.
What is the Shiller CAPE ratio today?
The Shiller CAPE ratio averaged 40.9 in September 2026, the latest complete month. CAPE stands for cyclically adjusted price-to-earnings: the S&P 500 price divided by its average inflation-adjusted earnings over the previous 10 years. It is the 20th-highest of 1,749 months since January 1881, so only 19 months were higher.
1. The CAPE averaged 40.9 in September 2026
multpl.com defines it as the price divided by the average of the past ten years of inflation-adjusted earnings. The monthly figure is an average of the month’s prices, not a month-end close. multpl’s October 2026 reading was 41.07 after the October 1 close, one trading day (WealthyBud data · 1,749 months · September 2026).
2. 98.9% of months since January 1881 had a CAPE at or below 40.9
Percentile method: the share of the 1,749 monthly values from January 1881 to September 2026 that are at or below September 2026’s 40.90. The 19 higher months fall in March 1999 to September 1999 (7 months) and November 1999 to September 2000 (11 months) and August 2026 (1 month) (WealthyBud data · 1,749 months · September 2026).
3. 136 months topped 30, and 26 topped 40
That is 7.8% and 1.5% of the 1,749 months. The 26 readings above 40 fall in January 1999 to September 2000 (21) and May 2026 to September 2026 (5) (WealthyBud data · 1,749 months · September 2026).
For company-level valuation, see our stock valuation statistics.
What is the historical average CAPE?
The average CAPE is 17.8 and the median is 16.6 across the 1,749 months from January 1881 through September 2026. Since January 1950 the mean rises to 20.8 and the median to 20.2. The September 2026 reading of 40.9 is 2.3× the long-run mean and 2.0× the post-1950 mean.
4. Mean 17.8 and median 16.6, January 1881 to September 2026; since 1950, 20.8 and 20.2
The window starts in January 1881, the first month with ten full years of data. multpl.com publishes values from February 1871 and showed a 17.42 mean on October 1, which matches WealthyBud’s calculation over that longer window. Since January 1950 (921 months) the mean is 20.8 (WealthyBud data · 1,749 months · September 2026).
When was the CAPE highest and lowest?
The CAPE was highest at 44.19 in December 1999 and lowest at 4.78 in December 1920, a 9.2-fold range across the 1,749 months since January 1881. The September 2026 reading of 40.9 is 7.4% below the record and 8.6× the low.
5. Record high: 44.19 in December 1999
The next-highest month was July 1999 at 43.83, 0.36 points lower (WealthyBud data · 1,749 months · September 2026).
6. Record low: 4.78 in December 1920
The next-lowest month was August 1920 at 5.02; the lowest readings cluster in 1920–1921 (WealthyBud data · 1,749 months · September 2026).
The four highest peaks, one per run of months above 30:
- December 1999 peaked at 44.19, inside a run of 51 months above 30 (June 1997 to August 2001).
- August 2026 peaked at 41.13, inside a run of 35 months above 30 (November 2023 to September 2026).
- November 2021 peaked at 38.58, inside a run of 22 months above 30 (August 2020 to May 2022).
- January 2018 peaked at 33.31, inside a run of 15 months above 30 (September 2017 to November 2018).
Does a high CAPE predict lower returns?
High CAPE readings have gone with lower 10-year returns. Stocks bought in the highest fifth of readings earned a mean 3.6% a year after inflation over the next decade, against 10.8% for the lowest fifth. The link is an association in overlapping windows, not a forecast or proof of cause.
7. Highest-CAPE fifth: 3.6% a year; lowest fifth: 10.8%
Start months from January 1881 to June 2016 (1,626) are sorted by CAPE and split into fifths. Return is the annualized real total return (dividends reinvested) over the next 120 months, ending by June 2026; ties split by date (WealthyBud data · 1,626 start months · June 2016).
8. Correlation between starting CAPE and the next 10 years’ real return: −0.49
Pearson r over 1,626 start months, January 1881 to June 2016; since January 1950 (798 months) it is −0.55. Windows overlap, so the sample holds about 14 independent decades, not 1,626 (WealthyBud data · 1,626 start months · June 2016).
9. 90 of 326 top-fifth windows lost money after inflation; 0 of 325 bottom-fifth windows did
Top-fifth annualized real returns ranged from −5.9% to 12.8%, so some high-CAPE starts still did well. The bottom fifth ranged from 1.2% to 20.0% (WealthyBud data · 1,626 start months · June 2016).
10. Starts at a CAPE of 35 or more: −2.6% a year over 34 start months, all in 1998–2001
These 34 start months form one episode. Results ranged from −5.9% to 1.1% a year. No decade-long window yet exists for a start at today’s 40.9 (WealthyBud data · 1,626 start months · June 2016).
11. Campbell and Shiller: valuation ratios mostly forecast stock prices, not earnings or dividends
Their NBER working paper, an update of a 1998 article, found that price–earnings and dividend–price ratios “do poorly” at forecasting dividend, earnings or productivity growth. They “appear to be useful primarily in forecasting future stock price changes” (NBER Working Paper 8221, 2001).
See P/E ratios for 103 large-cap stocks
| Starting CAPE group | CAPE range | Mean real return | Median real return | Windows below zero | Months in sample |
|---|---|---|---|---|---|
| Lowest fifth | 4.8 to 11.1 | 10.8% | 10.6% | 0 of 325 | 325 |
| Fifth 2 of 5 | 11.1 to 14.6 | 7.2% | 7.2% | 36 of 325 | 325 |
| Fifth 3 of 5 | 14.6 to 17.5 | 6.4% | 6.5% | 36 of 325 | 325 |
| Fifth 4 of 5 | 17.5 to 21.3 | 5.5% | 5.6% | 31 of 325 | 325 |
| Highest fifth | 21.3 to 44.2 | 3.6% | 4.5% | 90 of 326 | 326 |
The table does not forecast the next decade. Earnings definitions and payout policies have changed, and Shiller’s data page offers a total-return CAPE variant for that reason.
How does CAPE compare with the regular P/E ratio?
The CAPE divides price by 10 years of average real earnings, while the regular P/E divides price by trailing 12-month as-reported earnings. In September 2026 the CAPE was 40.9 and the P/E was 26.0. The CAPE is usually higher, with a median CAPE-to-P/E ratio of 1.11 against 1.57 now.
12. The regular P/E was 26.04 in September 2026, against a CAPE of 40.90
multpl describes its P/E as “based on trailing twelve month ‘as reported’ earnings” (multpl P/E page). multpl flags September 2026’s P/E as an estimate from the latest reported earnings. Both series are monthly averages for the same months. The P/E is higher than 92.6% of the 1,749 months (mean 16.6, median 15.5); the CAPE is higher than 98.9% (WealthyBud data · 1,749 months · September 2026).
13. The P/E peaked at 123.73 in May 2009, when the CAPE was only 16.0
A one-year P/E rises when trailing earnings fall; the ten-year average smooths a single bad year. The P/E exceeded the CAPE in 597 of 1,749 months (34.1%). Their correlation is 0.53 (WealthyBud data · 1,749 months · September 2026).
| Month | Shiller CAPE (10-year average real earnings) | Regular P/E (trailing 12-month earnings) | CAPE ÷ P/E |
|---|---|---|---|
| September 2026 (latest complete month) | 40.90 | 26.04 | 1.57× |
| December 1999 (CAPE high) | 44.19 | 29.66 | 1.49× |
| May 2009 (P/E high) | 16.00 | 123.73 | 0.13× |
How does CAPE compare with bond yields?
The CAPE earnings yield, 1 divided by the CAPE, was 2.44% in September 2026, on inflation-adjusted earnings. The 10-year Treasury yielded 4.99%, a nominal figure. Stocks offered 2.54 percentage points less than the Treasury, mixing real and nominal bases, which is a rough gauge, not a like-for-like spread.
14. CAPE earnings yield 2.44% vs a 4.99% 10-year Treasury yield, September 2026
The yield is the monthly average of 21 daily FRED DGS10 observations (nominal). The earnings yield is real. The gap is −2.54 points (FRED DGS10) (WealthyBud data · 777 months · September 2026).
15. The earnings yield fell below the Treasury yield in 437 of 777 months since January 1962
That is 56.2% of months, with a mean gap of −0.26 points; only 9.8% had a gap as low as today’s. The widest shortfall was −4.38 points in January 2000; the widest surplus was 4.69 in March 2009 (WealthyBud data · 777 months · September 2026).
16. For 10-year real returns, the earnings yield correlated at 0.46 and the Treasury gap at 0.24
Pearson r across 654 overlapping start months, January 1962 to June 2016. The gap correlation is weaker than the earnings yield alone in this sample. See interest rates and stocks for more, and the Buffett indicator for another valuation gauge (WealthyBud data · 654 start months · June 2016).
What this means for investors
Treat the CAPE as context, not a timer. A 40.9 reading is high by history, yet 236 of 326 top-fifth decades still gained after inflation, and results ranged from −5.9% to 12.8% a year. The data show an association, not a schedule. See stock market returns for long-run averages.
Use more than one gauge. The P/E (26.0), the Treasury gap and the Buffett indicator each measure value differently. See company-level multiples in the stocks hub.
Stick to your plan. Diversified, low-cost funds such as SPY let you stay invested through valuation swings. Set risk by your horizon, not by one ratio.
More Stocks statistics
Buffett Indicator Statistics (2026)
The Buffett indicator, U.S. corporate equity value as a share of GDP, from Federal Reserve and BEA data since 1947 Q4.
View ResearchStock Valuation Statistics (2026)
Price-to-earnings and price-to-sales ratios across 103 large-cap U.S. stocks, computed from SEC fundamentals and delayed market prices.
View ResearchInterest Rates and Stocks Statistics (2026)
How Treasury yields and the federal funds rate line up with stock valuations and returns, from Federal Reserve data.
View HubStock Fundamentals & Investor Scores
103 large-cap companies ranked by SEC-filed fundamentals.
View