More on the P/E Ratio

Barry Ritholtz was kind enough to link to my post about the P/E ratio of the S&P 500 hitting a 10-year low. I have to confess that I wasn’t trying to make a bullish call for stocks. I was simply pointing out that P/E ratios are at a 10-year low. I think it’s fascinating is that the bull market has seen its P/E ratio shrink over time. I believe that’s unprecedented. (Barry included a counter post from Scott Frew.)
But I want to use this opportunity to make a few important points. The first is that ALL financial measurements are flawed. Some are worse than others. This doesn’t mean we ignore them, but we do have to recognize each one’s limitations.
Of course, I was the one who raised P/E ratios in the first place so let’s zero is on that one. (Don’t get me wrong: I love the P/E ratio. I love despite the flaws). The first problem is that the P/E ratio mixes two types of data. One number—the price—is a fixed point number, it lives in a specific point in time. Earnings, however, is a rate. It tells how much money was made between two specific points. It’s as if we’re comparing dollars to miles-per-hour.
This isn’t wholly unkosher, but we do have to look out for pitfalls. For example, P/E ratios often rise at the beginning of a bull market because prices tend to anticipate earnings. (Not only that, prices can even influence earnings.)
The P/E ratio is also highly dependent on pay out ratios, meaning how much of its profits a company pays to its shareholders (the owners) in the form of dividends. When companies pay out a larger share of their profits as dividends, earnings multiples should be lower. According to Professor Robert Shiller’s latest data, only about one-third of corporate profits are paid out as dividends. That’s very near the lowest percentage in all his data, which stretches back more than 130 years. It’s far lower than 1929 or 1987.
P/E ratios also influenced by long-term interest rates. This is for two reasons. The first is that borrowing costs are a major business expense, so if the cost of renting money falls, companies will be more profitable. The second, and more important reason, is that bonds compete against stocks for investors’ dollars. So higher bond yields means that stocks have to adjust and offer investors more “bang for the buck,” hence lower earnings multiples. The reverse is true for lower rates. It’s a never-ending battle between bonds and stocks. Periodically, gold likes to jump in the battle, too.
Twenty-five years ago, long-term Treasury yields climbed over 15%, and it was common to see P/E ratios in the high-single-digits. While bond yields have crept up recently, we’re still in a period of low long-term yield relative to the last thirty years.
And finally, the P/E ratio is a fine general gauge of market sentiment, but it’s still far from perfect. As Barry said, “cheap stocks can get cheaper.” That’s very true. Also, expensive stocks can get even more expensive. Much more. I’ll give you a good example. According to Dr. Shiller’s data (his P/E ratio data uses trailing earnings for the last 10 years), the P/E ratio reached 26 in February 1996. That was highest reading since October 1929.
But if you sold out, you would have been kicking yourself. Four years after October 1929, the S&P 500 was down by two-thirds. But four years after February 1996, the S&P had doubled. Thanks for nothing Mr. P/E Ratio!
As long-time readers know, I’m a perma-bull. This doesn’t mean I’m always bullish. It just means that I avoid timing the market. When you realize how many different variables affect the market stock, it’s simply overwhelming. The moral of the story is to look at all financial data, but don’t be a slave to any.

Posted by on June 5th, 2006 at 6:11 am


The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.