Ben Levisohn of the WSJ wrote two articles over the past week on the P/E ratio. One was titled “The Decline of the P/E Ratio” and the other “Is it Time to Scrap the Fusty Old P/E Ratio.” The gist of both is that due to economic uncertainty, a macro focus driving trading, HFT and etf’s, the P/E ratio is less relevant. While it’s easy to dismiss when it compresses, I believe the P/E ratio will always be a useful measure of a stock’s value, especially when viewed over not just the past 12 months, as future earnings forecasts are always a guess, but over a longer period of time. What is happening now, and leading to articles such as these, is the exact opposite of what occurred in the late ’90s, moving from multiple expansion to multiple compression. Against a LT P/E ratio of 15x, the P/E of the S&P 500 got to 31 in Mar ’00 and all we are now is back to the LT average of 15x (14.5 to be exact). If one believes that something that overshoots eventually undershoots its historical average, the average fall should continue.
The average P/E ratio in the ‘90s to the March ’00 peak was 21.6x and is running at a still relative high of 19x over the past 10 years since. A move down to the 666 in the S&P 500 back in early ’09 achieved a level that was clearly cheap at 10x earnings but we don’t have to do that again to get cheap once more. A flat market with rising earnings can do the same. The point of all this is that don’t dismiss the P/E ratio just because one doesn’t like its direction. Also, 14.5x earnings is not cheap, just cheaper as the pendulum usually swings to the other extreme and 31x was quite historic on the high end with levels in ’74 and ’80 falling to 7x. Yes, some like to argue that low interest rates make the market cheap relative to earnings but its just one of many factors and interest rates are currently artificially low because of the Fed and considering where rates have come from over the past 20 years, there isn’t much room left for them to fall.