Publish date:

Price-to-Earnings Ratios Are Contracting, but How Far, How Fast?

Some analysts foresee a period of more mundane returns as multiples compress further.

It's useful to track price-to-earnings ratios. They give a good read on how much investors are willing to pay for every dollar of earnings per share. And they are great indicators of value, or lack thereof, in the market, and of investor psychology. At the moment, P/E's are compressing, and that's a bit worrying.

If we get much unexpected bad news -- a medium-hard landing as opposed to a soft one -- those P/E's will likely shrink a good deal more. And bad things happen to stock prices when P/E's (the ratio of a stock's price to earnings per share) shrivel. It is bad enough when earnings estimates decline. Add in a decline in the multiple investors will pay for those earnings, and you have a double whammy to the downside.

One thing that bugs me is that, even as multiples on the S&P have come down slightly since the highs of last year and earlier this year, they remain so darn far above the historical average. Take a look at the following charts put together for me by Joe Kalinowski, equity strategist at

I/B/E/S International.

S&P 500 P/E Remains High by Historical Standards

Tech-Stock P/E's Are Still Flying

Multiples on the S&P 500 have taken a dip from the highs reached in late 1999 and early 2000. But that's all it is, a dip. (I'm talking here about P/E's based on 12-month rolling forecasts, by the way.) The small decline in the S&P's P/E shouldn't come as any big surprise. A bunch of new, high P/E companies -- mainly tech -- have replaced lower-tech, lower P/E stocks. (Think

JDS Uniphase




(RAD) - Get Rite Aid Corporation Report

.) The S&P's P/E has also held up pretty well because analysts have boosted their earnings estimates a bit for many companies in the S&P, even as the economy has showed signs of slowing.

Kalinowski isn't especially worried. For this year, he's still looking for 18% earnings growth. For 2001, he expects 13% to 15% profit growth, which he thinks is pretty good. "I am very bullish on the market," he says. "Earnings growth will be very favorable and acceleration of earnings is the fundamental driver of stock prices."

TheStreet Recommends

He notes that the so-called PEG ratio -- the P/E multiple divided by the earnings growth rate -- has come down lately. The lower the P/E relative to the earnings estimate, the better. True, but the PEG ratio has been a lot lower than it is today. It is about 1.3 now, down from its peak of about 1.5 earlier in the year. And that is still a far cry from a PEG of .9 back in 1991, when this bull market started.

I remain skeptical. I wonder whether third and fourth quarter earnings this year will be received so well when they come in. And if they miss estimates, watch out below.

I am also concerned that when you look at the universe of stocks wider than the S&P, earnings multiples have compressed more. Here, I am relying on research done by the folks at

Morgan Stanley Dean Witter

, in particular analyst Hernando Cortina, who has studied P/E compressions on the

MSCI World Index

(excluding Japan).

Cortina writes, "We are increasingly of the view that 2000 may mark the start of a period of more mundane equity returns as multiples compress -- an investment landscape where asset allocation and sector/stock selection become more than a one-decision exercise."

He notes that, when the current bull phase really kicked into gear in the second half of the 1990s, the market's overall P/E was only 16 times trailing earnings. That popped to a record 30 times in December 1999 and at the end of July stood at 26 times earnings. Compare that with current levels, and we've seen a 14% decline in the P/E on the MCSI index.

Cortina takes no comfort from that. "Despite its recent decline," he writes, "the multiple remains in near-record territory and is almost double its 30-year average of 15 times. The compression of more than four multiple points ... since the end of 1999 is the largest P/E decline since the October 1995 trough. Only twice over the last 30 years has the market multiple avoided significant further contraction after this amount of compression."

The other two times were less than stellar periods for stock investors. From April 1972 to August 1974, the market P/E went from 20.3 to 6.8. The MCSI world index fell 43%. In the second instance, January 1994 to October 1995, P/E's went from 25.2 to 16.3, and the index returned 7%.

Morgan Stanley's Cortina asks the question in his report: "If compression were to continue, would it be benign, driven by rising earnings, or a harsher price-driven contraction?" He then answers his own question. "We think at the present it is difficult to identify factors significant enough to warrant a major global multiple derating

i.e., a big drop."

Knowing how sensitive sell-side analysts are about speaking badly of the market, I pay attention to cautionary words like this. You should too.