So and so says the market is overvalued. Should we get out?
Gauging whether the market is under or overvalued is almost an obsession with some of the financial press.  It is completely understandable to me that people might get the impression that high valuations portend market doom.  Ah, if only it were that simple.

Valuation is not a good timing tool. Turns out, there really isn’t any good timing indicator when we apply statistical rigor to the data.

Vanguard did a study in 2012, “Forecasting stock returns: what signals matter and what do they say now,” that examined a variety of market valuation measures and found all them to be almost entirely useless in predicting market returns in the coming year.  Some, like trailing 12 months P/E ratios or Shiller’s CAPE, gave some clue as to returns 10 years out, but they explained only about 40 percent of the returns of the following 10 years.

A couple of years ago, I heard a speaker at a conference cite the Shiller CAPE ratio and declare the markets 20 percent undervalued. The next day, I heard another economist cite the exact same statistic and declare the market 30 percent overvalued.  Difference of opinion is normal and this is a significant reason why trying to time markets is folly. You can't be a buyer unless you find a seller and you can't be a seller unless you find a buyer.

What about the times when markets are “clearly” over or undervalued?

Even points of valuation extremes are not particularly useful as timing mechanisms. By most measures, stocks had become expensive on a historical basis in early 1996. If you took that as a sell signal, you would have missed out on the four most profitable consecutive years in the history of the U.S. markets. Since the crisis of 2008, many people have pointed to the historically high prices of U.S. government securities and recommended abandoning these holdings.

Rather than viewing valuation as a timing tool, I think looking at valuation and adjusting expectations for the next decade can be a useful exercise. Historically, when market valuation was high, future returns were below average more often than not and when valuations were low, future returns were better than average more often than not.