Is Alan Addled?

"Greenspan model" indicates stocks today are overvalued by about 18%


Fed Chairman Alan Greenspan hasn't said much about the stock market this year, but his favorite valuation model is just about screaming a sell signal. The so-called Greenspan model was brought to our attention last summer by Edward Yardeni, economist at Deutsche Morgan Grenfell, who found it buried in the back pages of a Fed report. The model's very presence in such a report was noteworthy because the Fed officials normally don't tip their hand about their views on the stock market. The model surfaced at a particularly interesting time: Stocks were near a high point, and the Greenspan model indicated that the market was about 20% higher than it should have been.

That turned out to be a pretty good call. By October, stocks had fallen as much as 15% from their summer high point. By year-end, of course, the Dow had recovered to around 7900, but it still remained about 5% below its peak for the year.

Now that the Dow has climbed above 8600, Greenspan's model is again flashing a warning signal. To be exact, the Greenspan model now indicates that stocks are 18% overvalued.

"The market is as overvalued now as it was before it took a dive in October," Yardeni avers.

Over the past 18 years, the Fed's model has done a pretty good job of indicating the fair value of the S&P 500 (top chart). Since early last year, the model has shown stocks to be overvalued (bottom). Are we due for another decline like the one we had last October?

None of this means that Greenspan is planning to make any dire pronouncements. Indeed, he may have learned his lesson back in December 1996 when he pointed to the stock market's "irrational exuberance." Since then, of course, the Dow has risen more than 2,000 points. So far this year, about the only thing Greenspan has uttered about stocks is a warning that any very rapid drop in stock prices could be a "virulently negative force in the economy."

Though Greenspan didn't say it, a pretty nasty fall is suggested by his model. And as the accompanying graph shows, the model has done a pretty good job of predicting the path of the S&P 500 over the past 18 years. The point of most extreme overvaluation came just before the crash in 1987, when stocks were 32% overvalued.

Lately, the model has accurately indicated that stocks were undervalued from 1993 through 1995, and, oddly enough, it indicated that the market was fairly valued in late 1996 when Greenspan made his "irrational exuberance" comment. The model suggests that in 1997 stocks moved into overvalued territory, where they have remained.

The Fed's model arrives at its conclusions by comparing the yield on the 10-year Treasury note to the price-to-earnings ratio of the S&P 500 based on expected operating earnings in the coming 12 months. To put stocks and bonds on the same footing, the model uses the "earnings yield" on stocks, which is the inverse of the P/E ratio. So while the yield on the 10-year Treasury is now 5.60%, the earnings yield on the S&P 500, based on a P/E ratio of 21, is 4.75%.

In essence, the Fed's model asks, Why would anyone buy stocks with a 4.75% earnings return, when they could get a bond with a 5.60% yield?

The Fed's model suggests the S&P should be trading around 900, well under its current level of 1070.

"The market is as richly valued now as it was last summer -- before the Asian crisis," Yardeni observes. "Stocks aren't cheap." Yardeni says stocks look even worse than the Fed model indicates because he believes profits on the S&P Index will be hard-pressed to hit the current estimate of $50.78, which comes from IBES International, a unit of Primark. "I think we'll be lucky to do $48. If you use $48, the market is about 25% overvalued," Yardeni says.

Last year, operating profits for the companies in the S&P hit about $45.50. To reach $50.78 for this year, profit growth would have be 11%. Given that earnings are expected to rise less than 5% in the current quarter, an 11% gain for the entire year may be a tall order.

Joe Abbott, research director at IBES, points out that his firm's monthly calculation of expected S&P earnings for the coming 12 months dipped in February to $50.78 from $50.98, marking one of the few monthly declines in the current bull market. Given that Wall Street profit estimates have been cut lately for such big companies as Intel, Compaq and Motorola, the IBES 12-month estimate could drop farther in March.

In that light, Yardeni's skittishness is easy to understand. After being bullish on stocks for much of the 1980s and 1990s, Yardeni has turned cautious lately. "I do think the days are numbered for this bull market," he says, adding that he expects trouble between now and the end of next year, partly because of the huge cost of fixing the Year 2000 computer problem. Yardeni points out that the market can remain overvalued for quite some time, and that an overvalued situation doesn't have to be corrected by a fall in stocks. Profit estimates can rise or interest rates can fall, making stocks more attractive.

Stocks vs. Bonds

Critics of the Fed model and similar ones used on Wall Street say the models don't give enough credit to stocks. "I don't pay a lot of attention to that kind of stuff," notes Peter Canelo, domestic strategist at Morgan Stanley Dean Witter. His view: "You're not getting paid enough to be in bonds right now."

Canelo looks at several measures that suggest stocks look good vis-a-vis bonds. One is the yield on the 30-year Treasury bond relative to the 1.5% dividend yield on the S&P. That gap, now 4.4 percentage points, is low by standards of the past 20 years, meaning that bonds don't offer that much extra yield. He says this margin has ranged from 3.5% to 9.5% since the early 1980s. A narrow gap favors stocks. Before the 1987 crash, the spread was over seven points.

Another indicator favoring stocks is the projected return on equities relative to bonds. Canelo figures that stocks can generate an annual return of 9.5% over the long haul, based on dividends of 1.5% and estimated annual corporate profit growth of 8%. By this measure, stocks will return about 3.5 percentage points more than 30-year Treasuries yielding around 6%. "Whenever the premium has been over three percentage points, stocks generally have outperformed bonds by double digits in the next 12 months," Canelo says.

Maybe Canelo's right. But somehow remembrance of last October's tremors seems to tip the scale in favor of Greenspan's model.


  1. This article talks about a equity valuation model used by the Federal Reserve that suggests that stocks are overvalued. What are some of the difficulties in building such a model? What advantages might the Fed have over other investors in building such a model?
  2. This article mentions that the Fed model has done a fairly good job of predicting market upturns and downturns over the last 18 years. Is this sufficient evidence that the model works? Why or why not?
  3. The model usedby Morgan Stanley suggests that stocks are not overvalued. What is the underlying basis of the Morgan Stanley model and what is it assuming?