Looking at the chart below, you might conclude that P/E ratios today, by recent historical standards, are not that high.  Although they spiked during the dot com era, they have clearly subsided to levels that were typical for the previous 20 years, and are not much more than the values since 1936 (as far back as S&P provides a time series).

Source:  Standard and Poor’s

These P/E ratios are calculated by the common practice of dividing the stock price by the prior 12 months earnings.  Leonhardt makes the case that this is the wrong way to look at P/E ratios.  Going back to classic 1934 textbook, “Security Analysis,” by Graham and Dodd (whose pupils included Warren Buffett), he notes that those authors argued the earnings (the denominator in the P/E equation) should not be the most recent 12 months, but should look back “not less than five years, preferably seven or ten years.”

Graham and Dodd believed P/E is valuable tool in stock analysis, but a year or less of earnings data does not necessarily reflect whether a business can generate earnings on a sustainable long term basis.  A single year’s data can be more reflective of the overall economy than of a company’s future prospects, and could be “deeply misleading” when it comes to predicting the company’s long term prospects.

Robert Shiller, an economist at Yale and author of Irrational Exuberance, and John Campbell, an economist at Harvard, have resurrected Graham and Dodd’s theories, by calculating P/E ratios based on ten years of historical earnings data.  Their data, as it appeared in the Times article, is presented below:

In contrast to the previous picture, today’s P/E ratios are remarkably higher than their historical averages.  Calculated in this manner, today’s P/E’s are in the mid 20’s, as compared to values of 16-18 using a single year of earnings data.  With the exception of the dot com era and the market run up prior to the Great Depression, P/E ratios are higher than they have been in the past 130 years.

Leonhardt notes corporate profits have soared in the last few years, citing statistics from the Commerce Department which show inflation adjusted corporate profits increasing over 30% from 2003 to 2006.  Yet, despite this boom in profits, P/E ratios have remained fairly low. 

Interpreting the Data

Leonhardt offers two ways of interpreting this data.  The optimistic view asserts globalization, automation, and other productivity increases have fundamentally changed the underlying valuation model.  The past ten years loses relevance, because it reflects a time when businesses were operating less efficiently.  The Graham-Dodd P/E model no longer matters and investors can look forward to a bull market.

The conservative view is that the boom in corporate profits is non sustainable, and today’s market valuations have not completely shaken off their inflated dot com multiples.  Tweedy Browne, in their September 30, 2007 letter to investors, noted that “It is not at all clear that corporate earnings can continue to aggressively advance in the face of a tripling of oil prices, an accelerating housing crisis, and higher financing costs.  The bottom line: corporate earnings are fairly volatile over time and Graham’s methodology accounts for that.  Modern day P/E calculations that focus on recent and projected earnings do not.”  [Ed. Note:  Next week’s issue of Advisor Perspectives features an interview with Tweedy Browne’s managing partners.]  Leonhardt’s final observation in his piece is many investors believe gains in productivity have led the market into a “new era,” but investments that depend on a “new era” usually do not do well.  The stock market will continue to be a good investment for the long term, but over the next few years the Graham and Dodd P/E data suggests that there will be more risk “in a market that is relearning the meaning of the word.”

Our Analysis

We concur with Leonhardt’s view that today’s markets embrace an added degree of risk, but we believe it is a mistake to place too much emphasis on the P/E methodology used.  In contrast with the dot com era, the current bull market is the result of increased corporate earnings, not the result of inflated P/E ratios.  Corporate earnings are clearly illustrated in the quarterly EPS data for the S&P 500, which shows the upward trend since 2002:

Source:  Standard and Poor’s

The run up in corporate earnings began after the end of the dot com era.  If Leonhardt had looked at five or seven years of earnings data (the lower end of what Graham and Dodd advocated), the effect would have been even more dramatic.   Moving average analysis is very sensitive to time period used, and choices other than the ten years used by Shiller and Campbell can produce strikingly different results.

Although the housing market and credit crises loom as large now as they did when Leonhardt wrote his piece, there are still many reasons to be optimistic about the continued trend in profitability.  One of these is the continued decline of the dollar, which benefits businesses dependent on foreign markets.  Approximately 45% of the earnings from the companies in the S&P index come from overseas markets.  Manufacturing industries are reaping the benefits of this trend.

Globally, P/E data is lower than it has been over the last ten years:

Investors that are broadly diversified internationally can be less worried about inflated P/E ratios.

Even if profits subside there will still be an upward trend in the ten year P/E data.  It is not an issue, as Leonhardt states, of whether the market “may not have fully worked off” their dot com valuations.  The issues are whether current levels of profitability are sustainable, and whether current P/E ratios of 16-18 will continue to hold.  These two issues are interrelated.  In fact, it is significant that traditional market P/Es have remained constant, despite the increase in profitability.  Perhaps this is the lesson the market learned from the dot com era.

Display article as PDF for printing.

Would you like to send this article to a friend?