Related Articles
10 signals to watch in a pivotal 3rd quarter
Ignore the arm-waving. An IT rebound's just a mirage
What correction? I'm going long
Related Resources
Check MSN Moneys economic calendar
What are the latest upgrades and downgrades?
Whats going on in the markets?
Company Focus
Recent articles:
• A critical leading indicator: Who's hiring?, 7/2/2003
• 3 Best Buy rivals look like better buys, 6/25/2003
• The natural-gas rally hasn't passed you by, 6/18/2003
More...
| | Company Focus
A stock-picking system thats working -- again
You don't have to trust analysts to make money off their earnings estimates. Simply watch their revisions for a tip-off to changing market expectations.
By Michael Brush
Somewhat rusty and a little worse for the wear, a once-lucrative money machine from the 1990s is back up and running and worth using again -- even if its only spitting out $5 bills compared to the twenties it cranked out in its glory days.
This money machine is a stock-selection system that tracks earnings-estimate revisions to find and ride improving trends in the outlook for corporate profits on Wall Street.
It may seem like a dumb idea these days to pick stocks by using the earnings forecasts of Wall Street analysts -- that discredited lot whose reputations went down in flames with the bursting of the market bubble in 2000. But the strategy reaps rewards in spite of the taint of scandal that hangs over the Street by cleverly putting the caution and paranoia of Wall Street to work for you.
Want some proof the strategy is up and running again? Try this:
Crushing the broad market
A theoretical portfolio of stocks selected with the system for this column by Mitch Zacks, of Zacks Investment Research, crushes the market. For 2003 through May, it gained 32.7% before trading costs compared with a 14% gain in the S&P 500 ($INX) in the same time frame. This is one of the better years so far for earnings-estimate revisions, says Zacks. It is the fourth-best year since 1994.
Zacks created the portfolio by selecting the 300 stocks with the best earnings-estimate revisions each month, out of the largest 3,000 stocks. Trading costs would have knocked those returns down to around 26%, estimates Zacks. In reality, it would be difficult for a money manager with a large portfolio to match those returns, because many of the favored stocks are so small the manager would move the price up before he could get a sizable position. But it's easier for individual investors.
Last year, the strategy beat the S&P 500 by only around five percentage points before trading costs. Back in 1999 -- or one of the better years for the strategy -- the system posted gains of 62%, 40 percentage points better than the S&P 500.
Zacks' focus-list stocks with good estimate revisions now include: Marvel Enterprises (MVL, news, msgs), which houses a library of comic-book characters including Spider-Man and The Incredible Hulk; Fisher Scientific (FSH, news, msgs) and Omnicare (OCR, news, msgs) in health-care research services; and Doral Financial (DRL, news, msgs) in banking.
What the quants are seeing
In a universe of the biggest thousand or so stocks tracked by quantitative analysts at Prudential Equity Group, the 20% getting the best upward estimate revisions advanced 24% by the end of June, compared to 17% for the overall group.
Quants, or managers who rely on numeric measures to pick stocks, are increasing the use of estimate revisions and earnings surprises, which are also back as a good way to predict stock-price moves. Louis Navellier of Navellier Associates, for example, has upped the weighting of revisions and surprises in his models to 11% to 16%. That compares with lows in the mid-single digits last year when those tactics worked less.
Estimate revisions are producing gains again for a simple reason. Boiled down to its basics, the system is a way to help investors latch on to economic trends. But when the economy is adrift or at an inflection point -- say, because it's stuck in the transition phase between recession and growth theres no trend for the strategy to pick up. Thats one reason the system worked so poorly last year.
Rising profit forecasts
Now that the economy seems to be growing again (or is about to start growing again), managers are putting money in stocks where analysts are increasing profit forecasts, knowing that those increases are likely to be followed by more positive numbers bumps, which will drive stock prices even higher.
Their reasoning: Positive business trends at companies tend to continue, at least in the near term. A better answer, however, emerges from the psychological tendency of Wall Street analysts to err on the side of caution.
To understand, lets say youre a Wall Street analyst. Some of your research makes you think a company will earn $1 more per share next year than you previously thought. You rush to increase your estimates by $1 ahead of the competition, right?
No. First, you know that some random event -- an unforeseen economic downturn, a new competitor -- just might wipe out that extra earnings potential. So you reason its better to increase your estimates by a portion of that dollar for now, say a quarter, and then wait for confirmation that the story is playing out as planned.
Second, you've learned it's much worse for your career to go out on a limb and be wrong than to play it safe and mess up along with everyone else. The second way is a lot easier to explain to your boss, not to mention your clients.
For all of these reasons, investors know that when they see analysts up their estimates by a dime or a quarter, in all likelihood those analysts really think theres more upside. The bottom line: You can be pretty sure that positive revisions will be followed by more positive revisions.
Now if everyone knows this -- and a lot of people do -- then all that good news behind most numbers bumps should get priced in to stocks right away, rendering the system useless, right? Not really. Its true that greater use of the system now means it will probably never be as effective as it was at its peak last decade.
A lagging effect
But even if a lot of people have caught on to how it works, investors still dont fully price the implications of an upward estimate into the market immediately.
The first issue is that not everyone will agree that the information is accurate, says Ed Keon, a quant with Prudential who has used earnings estimate revisions for years. "We all tend to be slow to change our beliefs, so when we get a piece of unexpectedly good news, our first tendency is to reject it. Some people will want more evidence.
With the stock market, if you wait for something to be proved beyond a reasonable doubt, it's too late. Besides, even if money managers are fairly certain that a revision foreshadows more good news, they typically build a position in a stock slowly, simply because that's how they do business. For all these reasons, it can take a while for estimate revisions to be fully priced into the market.
What about the Wall Street scandals and reports revealing how many analysts simply invented forecasts for stocks to help generate investment-banking business? The analysts and brokers still work under the same roof, despite the reforms. So doesnt that mean theres little value in Wall Street estimates because analysts still have an incentive to hype stocks?
Nope. It doesnt really matter what the level of forecasts are, says Keon. It is just a matter of trying to be on the right side of changing expectations.
But this doesnt mean you can buy just any stock getting good upward estimate revisions. Many send you down the wrong path if you aren't careful.
Next week: A closer look at the details of how to use earnings estimate revisions, where to find them on the Internet, what investment advisors use them, and what stocks come up the strongest using this system.
|
