Thursday, July 16, 2009

The Ongoing Perils of PE Ratios

One of the interesting things about writing a book is that you get feedback from lots of folks about what they liked and disliked the most about your writing. Recently, someone told me they read and re-read the chapter on P/E (price-to-earnings) ratios (The Incredible, Amazing P/E Ratio) several times and still “didn’t get it.” It’s not surprising that someone didn’t get it, because clearly the huge majority of investors don’t “get it” either. If you are a proponent of value investing, and one of the basic tenants of your portfolio construction is to be wary when the market is expensive, then understanding this basic valuation measure is very important. Unfortunately, as I say in the book, it’s also an ongoing mystery where bulls and bears will look at the same data and reach vastly different conclusions.

At Pinnacle, we look at a wide variety of valuation measures involving earnings and price. Last week we routinely updated our valuation data and came up with the following P/E measures: (Note: we actually look at more than a dozen earnings-based indicators.)

The current range of P/E’s is fairly wide from a low of 10.83 based on a variation of John Hussman’s price-to-peak earnings methodology, to a high of 18.47 based on a ten-year average of S&P earnings. Obviously different investors using different methodologies will reach different conclusions from this data. Even investors who get the same numbers will reach different conclusions. For example, the 10-Yr normalized GAAP (Generally Accepted Accounting Principals) P/E ratio looks expensive when comparing to the median P/E of around 15 going back to the late 1800’s, but if you exclude the Great Depression and only compare to post WW-I data the median rises to about 17 and the current 18 figure doesn’t look too bad.

The bottom line: Buy and Hold investors who are looking for a simplistic, quantitative solution to fundamental value investing will be hard pressed to find it. It turns out that finding value is as much “art” as it is “science.”