We need forecasts — but don't count on them

Electronic News, Nov 19, 2001 by Bernard Levine

Business forecasts for semiconductor makers and the rest of the electronics industry are useful and necessary, just don't take them as gospel. Be cautious in evaluating any predictions. Projections of future sales growth are vital to making plans, but no one can really be sure what will happen next week, much less next year.

Next year's sales for the entire electronics industry, and all businesses, will be drastically impacted by the results of the war against terrorism, both in the United States and abroad. Yet nobody today can possibly know what those results will be, so there is no way to factor them into the equations companies are now making for 2002.

The Semiconductor Industry Association (SIA) coming under fire for its forecast of 6 percent growth for the chip industry in 2002 and a return to more normal growth in 2003 was a recent headline in Electronic News, with critics calling the SIA too optimistic. Sure, the SIA is on the optimistic side, but what would you expect from an industry trade association? Maybe some analysts are being too pessimistic right now, although they would probably counter that they are just being realistic. However, coming business could have an upward surprise. We've had so much bad news in recent months, perhaps we are due for some good fortune.

At this time next year, the state of the world, and the electronics industry, could be as uncertain and troubled as it is today. Or we all could be emerging victorious from a critical challenge with business well on the road to recovery.

Either road is possible, or it could turn out somewhere in between, although the latest world news as I write this has provided more encouragement. Something may be said for taking the optimistic approach now and seeing the bottle as half full. Sometimes perceptions do help shape reality-just don't bet the farm on it.

Economists' forecasts don't always come true, explained JoAnne Feeney, assistant professor of economics at the University of Albany, because it is impossible to know all the coming factors that might prove crucial; such as changing technologies and consumer preferences. "Unless you can predict the future, risk is something we are stuck with," she said at the recent Albany Symposium on Global Semiconductor Issues. She added it would often be "more useful if the forecasters indicated their degree of confidence in the forecast, their margin of error."

I have long been in the business of making forecasts and projections for this newspaper and its annual supplements, and our paper's forecasts are as good as anyone's, maybe better. In other words, they are amazingly accurate many times, but wide of the mark other times.

Many of you probably do forecasting for your own businesses. With the time rapidly approaching to get next year's plans ready, remember that forecasts are always something of a crapshoot, and more so amid today's political uncertainties. Coming plans must rely on the best projections available, but always be cognizant that final sales might come in at the top or bottom of the forecast range, or maybe even outside of it. Make your decisions as best you can, but keep plan B in mind just in case.