FORTUNE — If you believe what you read in The New York Times, it’s time to sell stocks.
On Tuesday, the Times ran a pair of pieces saying the market is overvalued. Once again, it’s bubble time. The first piece, written by David Leonhardt, was mostly about a stock market valuation gauge invented by Noble-prize winning economist Robert Shiller.
But Leonhardt failed to mention that there has been debate about the usefulness of Shiller’s market measure, and he appears to have gotten some of his math wrong.
MORE: The BIG risks in the Nasdaq
The article says you should be wary of buying stocks right now because, on average, they are trading at 25 times earnings according to the Shiller P/E, which looks at 10 years of earnings, rather than the typical previous 12 months measure. Leonhardt says when the Shiller measure gets that high, stocks tend to “disappoint,” dropping on average 12% over the following five years.
But that isn’t correct. Going back to 1900, the market crossed 25 on the Shiller P/E on five occasions, and, on average, stock prices have risen nearly 4.7% during the following five years. Leonhardt factors in inflation, but that only gets you down to a loss of 2.6%, not 12%, and inflation is currently much lower than it has been in the past.
What’s more, Leonhardt’s calculations ignore dividends, which are significant. Factor those in and the average five-year return for stocks following a Shiller P/E of 25 is 12.5%. Add in inflation, and it’s more like 20%. And that’s just an average. Sometimes the market has done much better, like following late 1995, when stocks rose 130% after the Shiller P/E crossed into the current danger zone.
I don’t know where Leonhardt got his numbers. Mine come from Jeff Rubin of Birinyi Associates, who has, admittedly, been skeptical of the Shiller P/E measurement for a while. (“Let me get out my Shiller file,” he says when I call him.)
MORE: Three ways to reignite the job market
Rubin’s main beef with the measure is that it almost always says stocks are overvalued. A little over five years ago, back in March 2009, Shiller said, with his gauge at 13, that stocks were still expensive. That was a bad call. March 2009 was the bottom. The S&P 500 is up about 180% since that point, far higher than the Shiller P/E would have predicted. “If the Shiller P/E never told us it was time to buy, how can it be saying it’s time to sell now?” asks Rubin.
That being said, investors should temper their expectations about what the market can do from these levels. I have written about the 4% market. And certainly, technology stocks in particular look expensive right now.
And a 12.5% total return over the next five years isn’t great. But it’s still better than what you are likely to get investing in bonds or gold or real estate. But the implication of Leonhardt’s piece is that it is time for stock investors to run for the hills. You’d likely be better off staying put.