<<expected 2005 earnings for the "big 9" biotechs will be almost double 2002 earnings>>
The problem here, is:
1. foreward earnings forecasts are random numbers. Neither you nor I have any idea which blockbuster will be the next one to dissolve heart valves. Or which drug put on the market targetted at a small niche, gets used to cure diabetes or depression. That is, nobody has a decent track record of accurate forecasts, and so current forecasts should be used as a gauge of current sentiment, and nothing else; i.e., if the forecasts are rosy, that's a sell signal.
2. even if those forecasts were accurate, there is zero correlation between biotech earnings and stock prices. The industry just isn't mature enough, yet, for that. You can use trailing, 12M foreward, 5-year forecasts, whatever, the picture is the same.
3. even if there was a correlation, and the last-5-years of data showed it, that time period is full of events on the thin edge of the bell-shaped-curve, so most of the data points should be thrown out.
4. even if there was correlation, and even if the recent (5Y) past predicted the future, correlation isn't causation. Until this industry gets mature enough that cash flow supports research/development/marketting costs, profits probably won't be helpful in predicting stock prices. |
