>From John Hussman
Market Outlook, January 1999
1999 will be one of the worst years of the century for the stock market, beginning at valuation extremes, plunging as a recession unfolds, and ending in pre-millennium panic. The bulk of market returns over the past 16 years have been driven not by earnings growth but by an expanding price/earnings ratio. The S&P 500 Index now trades at 31 times peak earnings, compared to a historical average of 12 times peak earnings, and multiples no higher than 20 at major pre-crash tops such as 1929, 1972 and 1987. From 1995 to May 1998, the market enjoyed uniform trends which allowed previous valuation extremes to be breached. But that uniformity was lost in May. Since then, despite marginal new S&P highs, the average stock and mutual fund has lagged T-bills. Weighted equally, the average stock lost ground in 1998. Such unfavorable valuation and divergent market action has preceded every major crash. But, hey, never mind. Sure, the S&P 500 Index is at the most abominably extreme valuations ever, but remember, every company in the S&P is capable of setting up an internet site. If you compare their valuations to those of other internet stocks, the S&P 500 are cheap! If you don't realize we're kidding, you're about to lose half of your money.
"Why is this market so unshakable? It must be because 1) so many people have so much surplus money to invest; 2) that hundreds of thousands of newcomers have watched others make money in the market, and are determined to do likewise; 3) stocks are scarce in relation to total demand; 4) the specter of inflation makes common stocks widely sought; and 5) corporate earnings are rising. This roaring market shows no signs of slackening its pace."
- Ira Cobleigh, Happiness is a Stock that Doubles in a Year, 1968
The bear market that immediately followed took stocks down by -30% over the next 2 years, and -40% over the next 6 years. Buy-and-hold investors lagged risk-free Treasury bills for the next 18 years, finally outpacing T-bills in early 1986.
"Look at the group of 1967 doublers. Most of these companies were less
than five years old! Traditional and time-tested criteria for stock
selection on the basis of established earnings, substantial book value, and dividend payments would have put you out in left field in any appraisal of these computer equities. A more psychedelic approach was required, and a new spectrum of price/earnings multiples!".
Psychedelic. Same book. Over the next few years, those computer-related companies lost nearly 80% of their value. Did somebody say "internet"?
About twice a week, I visit one of several large bookstores in our area. In two long shelves of business and investment books, I noticed that only one book marked with a "30% off" sticker. That book was John
Rothchild's The Bear Book, which by the way, is an excellent historical primer (even for bulls), and highly recommended even at full price.
In the current market environment, investors have formed their
expectations of future returns by measuring past returns from
trough-to-peak. Over the long term, of course, the best way to estimate the returns from any investment approach is to measure from peak-to-peak or trough-to-trough. But when markets begin at extreme overvaluations, even these comparisons can be overly optimistic. What investors are totally unprepared for, of course, is how badly the market performs when measured from peak-to-anything else, or anything-else-to-trough.
Rothchild offers a few reminders.
"The Dow moved into record territory at 734.91 in 1961, and sold for
exactly the same price in 1980, 19 years later". "The Dow bottomed at 566.05 in 1960. It sunk to 577.60 in 1974, gaining all of eleven points in 24 years".
Speaking of peak-to-peak movements, I recently read Jeremy Siegel's
analysis of the Nifty Fifty in his book Stocks for the Long Run. He
writes "The average price-to-earnings ratio of these stocks was 41.9 in 1972, more than double that of the S&P 500 Index". Today's "Nifty Fifty" still includes many of those issues, including Coca-Cola, Merck, Philip Morris, Pfizer, Gillette, and Johnson & Johnson, among others. Today, the largest stocks trade at an average P/E of 49, so one would expect that the Nifty Fifty would have performed slightly better than the S&P 500 for the period from 1972 - 1998. This is indeed the case. As would be expected, whether you measure from trough-to-trough, peak-to-peak, or even hypervaluation to hypervaluation, an investor will earn returns approaching the historical average.
What is stunning about Siegel's analysis is that he goes on to conclude that the Nifty Fifty weren't overvalued in 1972. Evidently, it doesn't matter that these stocks suffered a staggering plunge in the 1973-74 bear market, and underperformed the market for the 24 years from 1972-1996, at times by a dramatic margin. Siegel explains away this trip to hell, saying "the Nifty Fifty were in fact properly valued at the peak, but a loss of confidence by investors sent them to dramatically undervalued levels".
Here is a better interpretation. The Nifty Fifty were steeply overvalued in 1972, but they are more steeply overvalued today. Measured from peak-to-peak, the returns on the Nifty Fifty don't look so bad. But measured from peak-to-anything- else, these stocks have badly underperformed. If you thought the performance of the Nifty Fifty was bad during the 24 years from 1972 to 1996, just wait. As an old Danish proverb says "Bad is never good until worse happens".
Mathematically, if the P/E when you buy is the same as the P/E when you sell, you will, by definition, earn a capital gain equal to the growth rate of earnings. Your total return will be the earnings growth rate plus the dividend yield. Today, the average P/E on the largest stocks in the S&P is 49. If your sole interest is in matching the long term performance of the market (however low or high), and you believe that at some point in the distant future, the P/E on large stocks will still be 49, and you will never sell until the P/E reaches that level again, then go ahead and buy the largest S&P stocks here. But you'd better be willing to hold on to a poorly performing portfolio for decades and decades, because if the P/E drops below 49 in the interim, you'll be saddled with a portfolio which is underperforming the average stock.
Investors are in a market which demands further price increases when
their primary engines (rising P/E ratios and falling dividend yields)
are exhausted beyond belief. But with memories faded and valuations
ignored, buy-and-hold has become gospel. At 31 times record earnings, no less. Robert Prechter quotes a 1917 book called One-Way Pockets: "The public is always trading oriented in the early stages of a bull market, and 'investment' oriented at the top". In an otherwise interesting set of articles in Investors Daily, William O'Neil offers the following advice for growth fund investors (I am not making this up): "It's easy: You never, ever sell a domestic, diversified growth stock mutual fund. You hold it until you die". He left out that the cause of death will probably be from a heart attack as growth funds plunge in value during the coming bear market.
In our view, the truly fascinating thing about the current internet
stock craze is the misunderstanding about where the profits will come
from. We are certain that the internet will continue to grow at
fantastic rates. Yet we are equally certain that the true winners will
not be the "dot-com"s, but rather, the "dot-net"s. By this we mean that the winners will be those that provide access, not content.
The problem with internet sites is that there are no barriers to entry. As soon as anybody develops a business model which shows a profit, they will invite heavy, unbridled competition. As we have noted before, when internet sites become profitable, the market share of each company will move inversely with growth of the internet, which is another way of saying that revenue growth will be dramatically less than envisioned, not to mention earnings growth.
The best way to capture the growth of the internet is to buy those
companies that have proprietary technology or barriers to entry, and
primarily those that provide access and infrastructure rather than
content. The other requirement is that the valuation of the stock must
not fully discount the future growth. There are only a handful of stocks in this category that satisfy our own criteria. (These stocks are listed on the "Master Portfolios" page of the newsletter).
We can't believe the desperate attempts being made to justify the
valuations of current market leaders. As our longer term subscribers
know, we discovered a little networking company back in 1991, and our
recommendation was featured in Personal Investor Magazine under the
heading "A True Growth Stock". The company had a P/E of 35, but we
expected revenues and earnings to grow rapidly enough. That little
company was called Cisco Systems, which promptly multiplied in value 7
times over before we sold it 2 years later. While we arguably sold too
soon, our replacements kept pace, and it was only in the past year's
blowoff that Cisco began to outpace our Diversified Portfolio
selections.
Well, last month, an analyst at Donaldson, Lufkin & Jenrette recommended the stock (current P/E near 100 on mature earnings), saying "On the face of it, we certainly agree the stock is in uncharted valuation territory, but as an internet stock, Cisco is cheap!". Oh, Brother. We're the first to say that Cisco is a great company which we would like to own again, but at current prices, look out below.
The bottom line - there is a wide gap between the stocks that the herd
thinks are attractive and the ones that really are. We expect our stock selection to continue its exceptional record, but we are not so glib to hold anything without a solid hedge. Even a decline to last summer's low would recover all of our hedging costs since 1995, while our stocks would give up only a fraction of their gains since then. The end result, of course, is that by the end of the coming market cycle, whenever that occurs, we will be able to look at both our stock selections as well as our put options and see net gains in both, earned with consistently low risk of overall loss. That, for us, is the real objective.
|
