Thursday, August 25, 2011

Is the Market Expensive or Cheap?

While I remain constructive on the outlook for stocks - especially dividend-payers - I am aware that some of my valuation metrics might be challenged.

For example, I have pointed out that the price/earnings (p/e) multiple of the S&P 500 is attractive relative to historical averages.

In addition, the dividend yield on the S&P is nearly equal to the 10 year Treasury note, which is a situation that has not existed in the capital markets for nearly 50 years.

On both measures, then, I say that it is not a bad time to be buying.

The Lex Column in this morning's Financial Times takes issue with using either P/E ratios or dividend yields as a guide to stock purchase decisions.

In the spirit of keeping Random Glenings a democratic site, I thought I would mention their comments.

First, as to P/E ratios:

... In the U.S., for example, investors are hearing that the market is trading on a forward p/e of about 10 times, while the long term average is about 15. Forget that analysts always inflate future earnings. The big flaw with this approach is that current or near-future earnings are very unlikely to represent an equilibrium return from stocks.

Oh, so this means that we shouldn't be buying here?

Unfortunately, it's not that simple, says the Financial Times:

Valuation mirages can work the other way, too. As Smithers & Co notes, the trailing p/e for the U.S. market in 1933 was 50 percent above average, and the forward p/e looked high too. The following three decades, however, turned out to be fantastic for stocks. Indeed, looking at investment periods of up to 30 years, beginning in 1871, American companies actually ended up being expensive in hindsight a third of the time, in spite of looking cheap on a p/e basis beforehand.

As to dividends, the FT piece that it is not exactly fair to compare dividend yields to bonds, since dividends are paid at the discretion of management while coupon payments are legal obligations.

Whoever said this was easy?