Valuation Is in the Eye of the Stockholder

"Fast, cheap and out of control" isn't just the name of a recent documentary. It's also a pretty good description of the stock market. But just how cheap is the subject of a heated debate on Wall Street.

The bulls, of course, contend the market is dirt cheap after a 28-month shellacking. Nonsense, say bears, who note price-to-earnings ratios remain much higher than the historic average of around 14, no matter how they're measured.

History seems to be on the side of the skeptics, as the table that follows suggests. The bottom line is that trailing P/Es for the S&P 500 are significantly higher today than at any of a number of significant market bottoms in the 20th century.

Treasury yields -- to which the relative value of stocks is compared -- are currently lower now than in several more recent episodes. But today's yields aren't nearly as far below the average of past bottoms as current P/Es are above their comparable average.

The data used for the table below were culled from the Web site of Robert Shiller, Yale economics professor and author of Irrational Exuberance. The historic Treasury-note yields come from the Federal Reserve.

Past as Prologue: P/E Ratios and Bond Yields at Past Market Troughs
Date	S&P 500 P/E (trailing 12-months)	Treasury Yields
Dec. 1920	8.52	NA
July 1932	9.35	3.2% (long-term govt. securities)
March 1938	10.62	2.64% (long-term govt. securities)
April 1942	7.93	2.44% (long-term govt. securities)
May 1962	18.32	3.87% (10-year note)
Sept. 1966	14.12	5.18% (10-year note)
June 1970	13.69	7.84% (10-year note)
Dec. 1974	7.54	7.43% (10-year note)
July 1982	7.83	13.95% (10-year note)
Oct. 1990	14.21	8.72% (10-year note)
Sept. 1998	24.2	4.81% (10-year note)
Avg.	13.63	5.96%
Current	34.5*	4.58% (10-year note)
Sources: Robert Shiller for P/Es except *, from Barron's Federal Reserve for yields.

The relevant dates come from a recent report by GNI Capital, a relatively small ($20 million) hedge fund in New York, which sought to compare past market junctures to the current environment.

Most prior market lows occurred during economic recessions, GNI found, which isn't surprising, given the stock market generally bottoms ahead of an upturn in earnings as investors discount the future recovery.

Notably, we're not in a recession currently, meaning either it's different this time or maybe the stock market's recent swoon is telling us what's coming for the economy, as discussed last week.

The Great Debate

The above data sparked a torrent of emails when I posted them yesterday in RealMoney.com's Columnist Conversation. Several readers suggested comparing current P/Es with historical levels is a fruitless exercise, given the differences not just in bond yields but in inflation rates, dividend payout levels, and general market action. Additionally, the composition of the S&P 500 has changed dramatically over the years. Then there's the overriding question of how, exactly, we're supposed to be measuring earnings these days.

Earlier this week, Richard Berntstein, chief strategist at Merrill Lynch, suggested the "variability [and] unpredictability" of earnings is "more important when determining the current market's valuation than are interest rates and inflation."

In addition to the cycles for the economy and corporate earnings, variables in the current equation include the debate over reported vs. "core" earnings, the potential wave of restatements when the Aug. 14 sign-off date arrives, and whether options should be treated as an expense.

For those reasons, Brad Ruderman, managing partner at RCM Partners, a Los Angeles-based hedge fund with about $100 million under management, agreed the "What's the 'E'?" question is the crucial one facing investors today.

Additionally, he noted that past P/E troughs coincided with "one-off" events such as bad policy decisions in the 1930s and the oil embargo of the 1970s. Thus, comparing current P/Es to historic levels doesn't necessarily assist investors today.

"One can always make an academic argument for both sides," of the bull/bear debate, Ruderman said. "What serves us well is buying undervalued businesses after extreme selloffs and selling rallies."

RCM has recently reined in short positions previously focused in consumer names such as Kohl's , Wal-Mart and Tiffany's .

The fund has been buying media companies such as Disney , Viacom (VIA Quote) and Rainbow Media Group .

Ruderman, who declined to discuss the fund's performance, also maintains the bullish stance on cable stocks detailed here in May, despite the recent drubbing the group has taken. "They've been clobbered by guilt by association with Adelphia and EBIDTA but we've been buying more as they're down," he said. "We think cable is one area where you have pricing power."

GNI Capital's Allen Gillespie, who authored the firm's report, reached a conclusion similar to Ruderman's about interpreting current P/Es vs. past bottoms.

The Nasdaq Composite and the Dow Jones Utility Index have gone through bear markets akin to the Dow Industrials in 1929-32, Gillespie said. "But that doesn't mean the whole list will do that."

Gillespie argued names such as J.P. Morgan Chase , Citigroup (C Quote), and Duke Energy are trading at attractive levels based on P/E and/or book value.

"These are names that have a lot of fleas but unless you think they're bankruptcies, it's harder to stay short," he said.

GNI, which was up 15.9% year to date through Monday, has been trading in and out of those names lately but had no positions as of Tuesday's close.

"I think we're in a situation where earnings have quit declining, which makes the real bearish case much tougher," he said. "That doesn't mean we're going to have a great bull market, but if earnings continue to lift, the big bearish bet has to be taken off the table."

Continued improvements in corporate profitability is, of course, the big "if."

More Notes and Notables

Finally, the historic P/Es cited in the table above conflict with the message of money manager Ken Fisher's recent column in Forbes, in which he argued: "Correctly calculated, the market's highest P/Es ever were in 1920, 1932 and 1982."

Fisher, however, didn't specify what "correctly calculated" means or what the specific P/Es were in those eras.

The notion that P/Es would be extremely high at market bottoms makes sense when you consider earnings were at a nadir at those junctures. "The market's P/E usually is higher at a bear market bottom than its prior peak, because as a rule, earnings disappear faster than stock prices," Fisher wrote.

However, he also wrote: "History shows there is simply nothing about P/E levels, cut any way you want, to help predict market tops or bottoms, or market levels several years out. That we believe otherwise is mythology."

That supports the conclusions of sources queried for this article and seemingly undermines the point of Fisher's, as well as his recently adopted bullish stance.

The fund manager's office did not return multiple phone calls seeking comment on Tuesday.