Your Next Great Stock: How to Screen the Market for Tomorrow's Top Performers

Chapter 18. The Rising Expectations Screen

Never get on an empty New York City subway car when all the other cars are full. The empty car isn't a sign of your good fortune. It's a sign that either the air conditioning is broken or that someone has done something unpleasant enough to make everyone leave.

That's an example of advice that doesn't expire. You can use it today or a year from now. Not so, analyst recommendations. Those are best used fresh, or not at all.

I've noted throughout the book that analyst buy recommendations generally aren't worth following. Yet some studies suggest otherwise. One found that during the 10 years ended 1996, stocks with strong buy ratings outperformed those with strong sell ratings by 13 percentage points a year.

Look through the stack of contradictory findings on the subject, and two common points emerge. First, sell recommendations are better predictors of poor stock performance than buy recommendations are of good performance. That makes sense. Analysts are loath to advise investors to sell stocks. Most firms issue anywhere from 3 to 15 times as many buy recommendations as sell recommendations. A cynical investor would blame that on conflict of interest. Recall that investment firms try to sell services to the same companies their analysts rate. There's a benign explanation, though. Analysts look for stocks worth buying because that's what account holders at their firms are interested in.

So there's probably money to be made in following sell recommendations ...