For many years now a relatively large contingent of analysts, investors and journalists has been convinced the stock market was in a bubble because the "Shiller P-E" ratio was just too high. Back on 8/12/2013, in our Monday Morning Outlook, we made our case that the Shiller model was too pessimistic. Now that looks like a pretty good call.
In the past four years, the bull market in stocks has continued, with the S&P 500 generating a total return – including both share price increases and reinvested dividends – of almost 60%, or more than 12% per year.
The response from many in the market is to say "OK, maybe a Shiller P-E ratio of 23.4 (from 8/2013) wasn't too high. But now it's 30.5, and the bull market really is doomed."
As we said four years ago, Yale University economics professor Robert Shiller has made some great calls. In his 2000 book, "Irrational Exuberance," Shiller argued that a 10-year average of corporate earnings smooths out the ups and downs of the business cycle. Then, using this "cyclically-adjusted" level of earnings and comparing it to stock prices he claimed to generate a better version of the P-E ratio. It was an all-time high in 1999-2000, a clear signal of a "bubble" in stocks.
Several years later, Shiller warned investors about a housing bubble: Another prescient call. Although we don't see eye to eye with Shiller about how the economy works, no one should doubt his intelligence or sincerity.
But that doesn't make him infallible. We still think the Shiller P-E is painting an overly pessimistic view of the stock market. One basic flaw is that by using a ten-year time horizon for earnings, the Shiller P-E includes a massive drop in earnings from 2008-09 generated by awful political management of the economy and a mis-use of mark-to-market accounting.
But even if you take the Shiller P-E at face value, the situation isn't as bad as some claim.
Let's go back to December 1999, when the Shiller P-E peaked at 44.20, the highest level on record. That P-E corresponds to an earnings yield of 2.26% (100 divided by the Shiller P-E). At the time, 10-year Treasury Notes were yielding 6.28%. In that environment, investing in Treasury securities was clearly the better deal. The following ten years proved that proposition, with the S&P generating an annualized total return of -0.9%.
Next look at May 2007, when the Shiller P-E peaked at 27.55 prior to the last recession. That translates into an earnings yield of 3.63%. At the time, the 10-year Treasury yield was 4.75%.
At first glance it looks like bonds were more attractive. But that doesn't factor-in the potential of corporate earnings growth. And, in the end, despite the deepest recession since the Great Depression, earnings growth helped generate a higher return for stocks than for bonds. In the ten years after May 2007, the S&P generated a total return of 6.9% per year, beating the 4.75% on a 10-year Treasury Note.
Where does this leave us today? A Shiller P-E of 30.5 corresponds to an earnings yield of 3.28%, which is already more generous than the yield on any Treasury security.
The 10-year Note yield would have to be 7.3% today for the stock market to be as bad a deal relative to bonds as it was back in 1999. (Take today's earnings yield and add the gap between the Treasury yield and earnings yield in 1999, or 3.28% + (6.28% - 2.26%) = 7.30%)
For stocks to be as "bad" a deal as they were in May 2007, the 10-year Note yield would have to be about 4.4% today, which remember, even then didn't mean bonds beat stocks.
Moreover, all these comparisons will look even better once the 2008-09 atypical earnings swoon drops out of the ten-year earnings window.
This economic expansion is not over. Earnings have, and will, continue to grow. And, as we said four years ago, those who swear by the Shiller P-E are going to be underinvested in stocks. This bull market has a long way to go.
Brian S. Wesbury - Chief Economist
Robert Stein, CFA – Deputy Chief Economist
Click here for PDF version