From the January 2009 issue of Wealth Manager Web • Subscribe!

January 1, 2009

MEASURING THE IMMEASURABLE

In an interview on November 21, legendary investor Jeremy Grantham, chairman and co-founder of GMO, declared that, "It's a nicely cheap market, but bear in mind that it can be spectacularly cheap if it wants to be." Tempering his enthusiasm a bit further, he added that the market has been expensive 10 out of the last 13 or 14 years, and current conditions qualified as "quite ordinary" on the cheapness scale.

On that day, the S&P closed a hair above 800, after which the largest one-week rally (on a percentage basis) ensued. Was Grantham's call stunningly prescient, and was the market truly undervalued at that point?

The most common tool for assessing overall market valuations is the price-to-earnings, or P/E ratio. Basic P/E analysis indicated that the market's fair value was substantially above 800. Wharton professor Jeremy Siegel proclaimed the market "dirt cheap" in late October when the S&P stood at 968. Siegel argued that the average historical P/E was about 15, compared to a value of just over 10 at the time.

But as has become the common practice on Wall Street over the last several years, Siegel used operating earnings, as opposed to GAAP earnings, in his calculations. Operating earnings strip away the supposed "one-time charges" that corporations incur. Unfortunately, these one-time charges have a habit of reappearing on an all too frequent basis, leading many to question the predictive ability of operating earnings.

Nearly a half century ago, Graham and Dodd argued that using a single year's earnings in the P/E calculation placed too much emphasis on the current year's earnings, and that earnings should be averaged over an entire business cycle. Yale economist Robert Shiller has been an outspoken advocate of this methodology and has shown that P/E data, when using a moving average of the last 10 years' earnings data, can be very predictive of market performance.

When viewed through the 10-year lens of earnings history, the market no longer looks that cheap.

Vitaliy Katsenelson, a fund manager with Denver-based Investment Management Associates and author of Active Value Investing, provided the following data:

Using 10-year data, the average P/E over the last century is 18.3. Katsenelson's data from mid-October, as depicted in the chart above, shows the P/E at 20. By the time of Grantham's call on Nov. 21, the P/E had declined to about 16--below the historical average. But the data show that bull markets typically start from P/E valuations of 13 or less. Katsenelson believes the market is in a "range bound" stage, characterized by a cycle of bull and bear markets resulting in a lack of overall performance.

Grantham is also a proponent of the use of historical earnings in P/E analysis, and it is likely that the P/E of 16 gave rise to his call of a "nicely" but not "spectacularly" cheap market. Another clue that he is not endorsing the start of the next bull market can be found in his comment that he had just increased the equity allocation in his sister's account--to a mere 30%.

P/E analysis is not the only tool for assessing market valuation. In "Advisor Perspectives," we have written about firms that value each company in the S&P 500 by projecting their free cash flow and then discounting to determine a present value. The Applied Finance Group, a Chicago-based investment advisor, operates such a model, and it was producing a fair value of the S&P of just under 1,000 as of late November. These models eliminate the oversimplification inherent in P/E analysis, but they introduce a host of new assumptions, such as the method for projecting cash flows and the selection of a discount rate. Because of its simplicity, therefore, it is much easier to understand and critique a P/E model.

Determining the market's fair value is truly an exercise in measuring the immeasurable. However, as an imprecise science, the 10-year historical earnings P/E ratio has proven to have predictive value.

The wild card in this analysis is the course of future earnings announcements.

With an accelerating global recession and housing prices still on the decline, corporate earnings are highly likely to undershoot expectations, producing a negative psychology that will overwhelm any calculated attempt to determine the market bottom.

Robert Huebscher, rhuebscher@advisorperspectives.com, is CEO of "Advisor Perspectives" (www.advisorperspectives.com), a free online database and weekly newsletter for wealth managers and financial advisors.

Reprints Discuss this story
This is where the comments go.