By Mark Hulbert
The stock-market-valuation indicator with the best forecasting record is predicting that the S&P 500 Index’s total return over the next decade won’t even keep up with inflation.
Inflation will be 1.4% annualized between now and 2031, the latest forecast of the Federal Reserve Bank of Cleveland’s inflation model . And the benchmark index’s nominal 10-year total return will be only 0.7% annualized, according to the indicator.
Welcome to this month’s update of eight valuation indicators with good long-term records — a regular feature in this space. In the table at the end of this column I review their current status.
Careful readers will note that this month I am deleting one of the indicators that I had included in previous columns, and in its place I’m adding this new one with the unequaled long-term record.
This additional indicator was introduced in 2013 by the anonymous author of the Philosophical Economics blog, who dubbed it “ The Single Greatest Predictor of Future Stock Market Returns .” It is based on the average investor’s equity allocation: Of the total amount investors have allocated to equities, fixed-income and cash, how much is allocated to the stock market?
This is a contrarian indicator, with higher equity allocations a bad sign for the stock market’s subsequent long-term return.
According to the most recent Federal Reserve data, this indicator stands at 45.2%, which is 10 percentage points higher than the historical average. As you can see from the table below, this latest reading is higher than 95% of all other quarterly readings since 2000 — and higher than 94% of all readings since 1950.
To appreciate how good a track record this indicator has, consider the r-squared of a simple regression model that uses it to forecast the S&P 500’s /zigman2/quotes/210599714/realtime SPX -1.31% dividend- and inflation-adjusted return over the subsequent decade. It is an almost unbelievably high 70.1%, far higher than that of any of the other valuation indicators I feature in the table below.
That’s really saying something, since these other indicators were themselves chosen because they have some of the highest r-squareds in the stock market forecasting business.
The reason I had not included this indicator in previous monthly updates is not that it didn’t have a good track record. I excluded it because it is updated only quarterly and with a long time lag. The most recent Federal Reserve data, for example, is updated through Sept. 30.
The indicator that I have dropped from my monthly valuation table is a close relative to the much-better-known Cyclically-Adjusted Price Earnings ratio, or CAPE — which continues to appear in the table. The variant that I have dropped is the Cyclically-Adjusted Total Return Price Earnings ratio, or TR-CAPE, which differs from the CAPE in that dividends are included in its calculation. Though no doubt superior in a theoretical sense, I am dropping it because its track record is so close to that of its better-known relative, and its 10-year forecast almost identical.
For a full description of how each of the indicators in the table below is calculated, please refer to my Oct. 30 column .
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at email@example.com .