MAKING SENSE -- November 21, 2012 at 11:04 AM ET
Why Do Analysts' Estimates Matter?
Photo by Andrew Rich via Getty Images.
Paul Solman answers questions from the NewsHour audience on business and economic news here on his Making Sen$e page. Here is Wednesday's query:
Tony Merlo: If a stock earns $1 a share for a quarter while analysts predict 98 cents, the stock typically goes up because it beats the estimate. But if the analysts estimate $1.02 a share it typically goes down because it fell short. Why should the stock be punished because the analysts predict incorrectly? They don't have access to internal documents so their estimates are only guesses. This doesn't make sense to me.
Paul Solman: The perfect week to answer this question, given the plunge in Hewlett-Packard's share price.
You're right, Tony -- about analysts' estimates, that is. They cannot be based on "inside information" that would be "material" to the stock's price. But it doesn't mean the estimates aren't better informed than yours or mine would be. While I'm reporting for the NewsHour and tending this page, analysts are spending their days learning all they can about company prospects and coming up with a best guess as to quarterly earnings. So their estimates amount to very good guesses.
But why does an earnings estimate matter so much, you might ask. Because the price of a stock is entirely a function of its expected future earnings.
What is a company? An organization dedicated to the objective of "earnings," aka profits. That's why investors put their money behind one company as opposed to another: the higher the earnings, the better the return.
On any given day, the market price of a stock is the best collective guess of investors as to what those earnings will be in the future. The best indication -- short of inside information -- is how much the company is earning right now. The best estimate of current earnings comes from the analysts.
Now imagine the day of a company's earnings announcement. Just this week, analysts were expecting Hewlett-Packard, for example, to announce quarter-of-the-year earnings of 85 cents per share -- total earnings divided by the total number of shares. But then H-P announced that a software company named Autonomy which it had bought for $10 billion was a purchase fraught with fraud and the company was deducting $5 billion from earnings to account for the loss. Never mind the arithmetic, but trust me: quarterly earnings now dropped to around 35 cents.
So what happened to the stock? It promptly fell 12 percent.
Why? Because the earnings were so much lower than analysts estimated and, therefore, than investors expected.