It's All About the Earnings
Wednesday January 16, 11:50 am Eastern Time
SmartMoney.com - Ahead of the Curve
By Donald Luskin
ONE MONTH AGO I suggested in this column that it was time to make a short-term tactical asset-allocation trade - to sell technology stocks and buy long-term government bonds. Now with more than a week of the new year under our belt, the case for making this trade is even stronger.
Thirty-year Treasurys continue to yield over 5.5% in an environment of no inflation risk whatsoever (if anything, I've argued that the risk is on the side of deflation). That puts real bond yields (the nominal yield adjusted for inflation) at probably the highest levels ever.
At the same time, technology stocks are in what has to be called a ``bear-market bubble.'' It doesn't feel as wild and crazy as the bull-market bubble in early 2000 (and it isn't nearly as much fun). And it doesn't seem possible, with the prices of technology stocks and of the Nasdaq having fallen to mere shadows of their former glory. But the fact is technology stocks' price/earnings ratios, based on next year's earnings forecasts, have snuck up to levels that nearly rival the insanity of the great market top in March 2000.
It's all a question of earnings. If technology companies can deliver spectacular earnings growth this year, then today's sky-high P/E ratios will be justified. But even if that happens, investors who buy technology stocks at these prices may not make any money ? and almost certainly not enough money to justify the risk they are taking. Why? Because spectacular earnings growth is already built into stock prices. So if it actually happens, then stock prices might not even budge. But if it doesn't happen, then it's going to be Code Blue in the boiler room ? there will be a lot of disappointed investors wondering how they managed to get fooled yet again, and dumping their stocks helter skelter.
I believe that the tone for the coming earnings season has already been set by two key company presentations this week. And based on these early clues, technology companies just aren't going to make a down payment in the first quarter on the earnings surprise they are going to have to produce to justify today's prices.
Cisco (NASDAQ:CSCO - news ) Chief Executive John Chambers spoke Tuesday at Salomon Smith Barney's annual Entertainment, Media and Telecommunications Conference. He's still talking about ``order linearity,'' which just means that Cisco is going to make its quarter, and may even beat forecasts by a little. And he's talking about ``dramatic'' market-share gains ? but nothing more than what I've suggested here that we should expect during a downturn from No. 1 players like Cisco, and nothing that doesn't just come out of the hide of rivals like Juniper Networks (NASDAQ:JNPR - news).
And Chambers said that he's going to go back to his long-standing policy of not talking about Cisco's quarterly performance in the midst of a quarter ? investors will just have to wait for the quarterly earnings announcement. True, that's a signal that the worst is behind us. Over the past year Cisco and every other technology company has been in the emergency room hooked up to life-signs monitors ? Dr. Chambers has had to report frequently on his patient's condition to its worried friends and relatives. Now the monitors can come off, and the patient can leave the ER...but for where?
Not the golf course. This patient is simply heading for intensive care. Chambers admitted that while customers' business had stabilized, they were still buying Cisco gear conservatively. Gone, gone, gone are the bold promises of 30% to 50% revenue growth that Chambers was still clinging to less than a year ago. Now the best he can manage is, ``I'd like to say I knew where the economy is going...but there's still limited visibility.''
So the message is: It's stopped getting worse, but it's never going to be as good as it used to be.
Earlier this week AOL Time Warner (NYSE:AOL - news ) ? you remember them, the former ``Internet stock'' that is now a ``media stock'' ? also formally abandoned its long-standing claims of 1990s-style revenue growth. In an analyst call on Monday, departing CEO Gerald Levin lowered the bar for his successor Richard Parsons by trimming revenue-growth guidance by at least half, and at the same time warning of what might be the largest write-down of goodwill in history. Its 2002 business plan assumes no growth for the U.S. economy, and no improvement whatsoever in advertising revenues.
And yet, at the same time as these two leading New Economy companies make these glum forecasts, tech stocks are carrying nearly their highest multiples in history. What do investors think they know that the CEOs of America's leading companies don't know?
Perhaps some overoptimism is normal. For instance, we know that Wall Street securities analysts are perennially overoptimistic. The chart below shows the amount by which the actual earnings of the S&P Information Technology sector have fallen short of rosy analysts' forecasts since 1985. On average, actual earnings have trailed the forecasts by about 10%.
The analysts have been especially optimistic this last year. At the beginning of 2001, they were forecasting $88.3 billion in annual earnings for the sector. The reality was $32 billion, a shortfall of 60%.
Now the analyst consensus is forecasting $39.2 billion for 2002 ? a 22% increase over 2001. But that estimate of $39.2 billion is the E in the P/E ratio that's nearly the highest in history. So that means that stocks are priced far more optimistically than would be justified by even the perennially overoptimistic analysts.
How much more optimistically? There's a simple calculation that will tell us.
We start with a concept I discussed in my column on Dec. 12 ? the ``yield gap'' that exists between the earnings of stocks and the interest payments from long-term Treasury bonds. If stocks have a P/E of 20, that means that earnings are 5% of stock prices. If long-term bonds are yielding 5% at the same time, that would make the yield gap zero.
Since 1985, the yield gap between 30-year Treasury bonds and the S&P Information Technology sector has been negative 0.9%. That means that the uncertain earnings for tech stocks over a typical year have been a bit less than the guaranteed income from bonds. Investors are willing to accept less in a given year from stocks because they expect earnings to grow over time, while the income payments from bonds are locked in.
The yield gap gets more negative when either interest rates go up (meaning that bonds offer stiffer competition for stocks), or when P/E ratios go up (meaning that a dollar invested in stocks represents less in earnings). Today the yield gap is at negative 3.4% ? a level that has only rarely been seen in modern history. It's slightly more negative than the level seen just before the crash of October 1987, and almost as negative as the level seen in March 2000, just before the onset of the tech wreck.
Using the concept of the yield gap, we can now precisely measure just how optimistic stock prices really are about earnings growth. That's because we can precisely calculate the level of earnings that it would take to return the yawningly negative yield gap of -3.4% to its historical average of -0.9%.
The answer will shock you. For the yield gap to return to its historical norm, the S&P Information Technology sector would have to earn $81.8 billion in 2002. That isn't quite as much as the $88.3 billion that was being forecast a year ago ? but it's more than the record-breaking $78.4 billion that was earned in 2000...and it's a lot more than the $32.0 billion earned in 2001 ? 156% more, to be precise.
It's also a lot more than the $39.2 billion that analysts are forecasting for 2002. In fact, if through some miracle of a super-duper-hyper-ultra-``V'' recovery it were actually achieved, it would represent an upside surprise of 108%, more than double the largest upside surprise there has ever been. Here's the same chart with that surprise entered in.
So what we have on our hands right now is a severe case of manic-depressive split personality in the markets. On the one hand, the CEOs are telling us that while the worst is over, we shouldn't expect very much in the way of earnings growth. On the other hand, the analysts are telling us that tech stocks are nevertheless going to deliver solid 22% earnings growth. On the other other hand, stocks themselves are priced to tell us that earnings will grow by 156%.
So anything less than 156% is going to be a disappointment relative to today's pricing. And that means that to make any money from tech stocks today (other than by shorting them) earnings growth will have to exceed 156%.
Now do you see what I mean about those nice, sane, 30-year Treasury bonds yielding a juicy 5.5%?
biz.yahoo.com
Donald Luskin is chief investment officer of Trend Macrolytics, an economics consulting firm serving institutional investors. You may contact him at don@trendmacro.com. One of his model portfolios has a long position in AOL Time Warner. |
