Are Stocks Undervalued Yet? – WSJ

The recent bounce in the S&P 500—up 5.3% since May 19—could be just a passing bear-market rally, or it could be a new bull-market leg.

But if it proves to be the latter, it is all but certain that factors other than undervaluation are helping drive stocks higher.

The eight valuation indicators that have proved best at predicting 10-year returns, inflation-adjusted, for the stock market are a subject I have covered before. And while it is possible that other valuation models exist that are just as good at predicting bull markets, I haven’t discovered any.

On balance, these eight indicators at the mid-May low stood at more than twice the average valuation of the bear-market bottoms seen in the past 50 years. And, seen in comparison with all monthly readings of the past 50 years, the average of the eight measurements was in the 88th percentile.

CAPE fear

Let’s start by looking at the cyclically adjusted price/earnings ratio, or CAPE, made famous by Yale University finance professor (and Nobel laureate)

Robert Shiller.

It is similar to the traditional P/E ratio, except that the denominator is based on 10-year average inflation-adjusted earnings instead of trailing one-year earnings. As with the traditional P/E, the higher the CAPE ratio, the more overvalued the market.

On May 19, the CAPE ratio stood at 30.4. That is more than double the average CAPE ratio at all bear-market bottoms since 1900, according to an analysis by my firm, Hulbert Ratings, of bear markets included in a calendar maintained by Ned Davis Research. While some might think comparisons from so long ago aren’t relevant under current conditions, a comparison with more-recent decades yields a similar conclusion. The average CAPE ratio at bear-market bottoms over the past 50 years, for example, is still 17.0.

Another perspective is gained by comparing the CAPE’s reading at the May low with all monthly readings over the past 50 years. Even at the recent low, the CAPE was in the 95th percentile of all results—more overvalued than 95% of all the other months over the past 50 years.

This message of extreme overvaluation isn’t easily dismissed, since the CAPE ratio has an impressive record predicting the stock market’s 10-year return. You can see that when looking at a statistic known as the R-squared, which ranges from 0% to 100% and measures the degree to which one data series explains or predicts another. When measured over the past 50 years, according to my firm’s analysis, the CAPE’s R-squared is 52%, which is very significant at the 95% confidence level that statisticians often use when determining if a correlation is genuine.

Many nevertheless reject the CAPE for various reasons. Some argue that the ratio needs to be adjusted to take into account today’s interest rates which, though higher than they were a year ago, are still low by historical standards. Others contend that accounting changes make earnings calculations from previous decades incomparable with today’s.

Still, the same bearish signals are being sent out by the other seven indicators that my firm’s research has found to have impressive stock-market forecasting abilities, and those indicators are based on different criteria.

Valuations at stock market’s mid-May low

Percentile of reading at stock market’s mid-May low, relative to distribution of monthly readings over the past 50 years

Most overvalued reading of past 50 years

Most undervalued reading of past 50 years

Average investor equity allocation

Percentile of reading at stock market’s mid-May low, relative to distribution of monthly readings over the past 50 years

Most overvalued reading of past 50 years

Most undervalued reading of past 50 years

Average investor equity allocation

Most overvalued reading of past 50 years

Percentile of reading at stock market’s mid-May low, relative to distribution of monthly readings over the past 50 years

Most undervalued reading of past 50 years

Average investor equity allocation

Percentile of reading at stock market’s mid-May low, relative to distribution of monthly readings over the past 50 years

Most undervalued reading of past 50 years

Most overvalued reading of past 50 years

Average investor equity allocation

Percentile of reading at stock market’s mid-May low, relative to distribution of monthly readings over the past 50 years

Most undervalued reading of past 50 years

Most overvalued reading of past 50 years

Average investor equity allocation

Here are those seven indicators, listed from high to low in terms of their accuracy over the past 50 years at predicting the stock market’s subsequent 10-year return, and showing by how much each indicates that the stock market remains overvalued:

• Average investor equity allocation. This is calculated as the percentage of the average investor’s financial assets—equities, debt and cash—that is allocated to stocks. The Federal Reserve releases this data quarterly, and even then with a time lag, so there is no way of knowing where it stood on the day of the mid-May market low. But at the end of last year it was 68% higher than the average of the past 50 years’ bear-market bottoms, and at the 99th percentile of the 50-year distribution.

• Price-to-book ratio. This is the ratio of the S&P 500 to per-share book value, which is a measure of net worth. At the mid-May low, this indicator was 95% higher than it was at the past bear-market bottoms, and was in the 90th percentile of the distribution.

• Buffett Indicator. This is the ratio of the stock market’s total market capitalization to GDP. It is named for

Berkshire Hathaway

CEO

Warren Buffett

because, two decades ago, he said that the indicator is “probably the best single measure of where [stock market] valuations stand at any given moment.” At May’s market low, the Buffett Indicator was 145% higher than at the average of past bear-market lows, and at the 95th percentile of the historical distribution.

• Price-to-sales ratio. This is the ratio of the S&P 500 to per-share sales. At the mid-May low it was 162% higher than at the past 50 years’ bear-market bottoms, and at the 94th percentile of the historical distribution.

• Q ratio. This indicator is based on research conducted by the late

James Tobin,

the 1981 Nobel laureate in economics. It is the ratio of market value to the replacement cost of assets. At the mid-May low it was 142% higher than at the bottoms of the past 50 years’ bear markets, and in the 94th percentile of the historical distribution.

• Dividend yield. This is the ratio of dividends per share to the S&P 500’s level. It suggests that the stock market is 121% overvalued compared with the past 50 years’ bear-market lows, and in the 87th percentile of the 50-year distribution.

• P/E ratio. This is perhaps the most widely followed of valuation indicators, calculated by dividing the S&P 500 by component companies’ trailing 12 months’ earnings per share. It currently is 16% above its average level at the lows of the past 50 years’ bear markets, and at the 58th percentile of the distribution of monthly readings. (This average excludes the bear-market low in March 2009, when U.S. corporations on balance were barely profitable and the P/E ratio artificially skyrocketed to near infinity.)

Short-term inscrutable

It is worth emphasizing that these valuation indicators’ impressive track records are based on their abilities to forecast the stock market’s subsequent 10-year return. They are much less useful when predicting the stock market’s shorter-term gyrations. So it wouldn’t be inconsistent with the message of these indicators for the stock market to mount a powerful short-term rally.

Assuming the future is like the past, however, the path of least resistance for the stock market is to decline. Short-term rallies notwithstanding, odds are good that the stock market on balance will produce a below-average return over the next decade.

Mr. Hulbert is a columnist whose Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at [email protected].