March 19, 2004 – Saratoga, CA – Bookings of U.S.-based semiconductor equipment manufacturers were $1,305.5M for February 2004 and are at their highest level in three years, reported by Semiconductor Equipment and Materials International (SEMI) yesterday. The first two months of 2004 exhibit a 69% growth in comparison to the same period in 2003, leading to the question on everyone’s minds these days: When will it come to an end?

The common denominator between 2000 and 1995 was the overly optimistic forecasts perpetuated by industry experts as well as forecasting pundits. In 1995, the much-touted slogan “No more Silicon Cycles” was waved; in 2000 it was promised that the boom would extend “at least until the end of 2001”, or even better, until the end of 2002. Going back in time, we see the same sources predicting in 1984 that 1985 would be an excellent year. Currently, the word is that the industry can’t build capacity fast enough to bring upon itself another recession. Or, at least it won’t be able to do so until the end of 2005.

Since statements such as ‘boom-to-bust IC cycles have come to an end’ are surfacing, is history repeating itself now?

“Another recession will take place; the semiconductor industry cannot avoid it,” said Rosa Luis, Director of Marketing and Sales for Advanced Forecasting. “No matter how many safeguards are put in place because of the last recession, this industry has yet to learn to be proactive in preventing the recession’s catalysts instead of being reactionary.”

“The most important factor required for accurately timing the next downturn is to know the Underlying Demand facing the industry and how capacity will be brought online to match that demand,” she added.

If past behavior has some indication about the future, one can expect industry decision-makers to overshoot Underlying Demand. The reason is their inherent tendency to extrapolate the most recent period’s situation into the next. I.e., they assume the boom continues even though Underlying Demand already changed direction. As Underlying Demand dips, decisions are made in the opposite direction, perpetuating the over-heating situation.

“We do not believe the industry has significantly overcome its inability to control double and triple-bookings,” said Moshe Handelsman, President and founder of Advanced Forecasting. “Since this is the main cause for buildup of over-capacity, the risk of repeating the excesses of 2000 remains intact.”

Luis revealed that “Advanced Forecasting’s prediction of Underlying Demand is strong through the end of 2004. Actual Shipments are running in tandem, thus reducing a potential threat of over-heating. However, as the industry witness increasing lead-times, shortages, growth rates becoming inflated, and companies boost their quarterly guidance, over-heating will surely follow.”

Founded in 1987, Advanced Forecasting is a leader in forecasting demand for semiconductors, semiconductor equipment and materials industries. Advanced Forecasting uses a purely quantitative forecasting model, a unique and highly sought-after viewpoint that is never modified retroactively, thus appealing to high-level executives and industry leaders. It provides the industry's most accurate market forecasts and has acquired, in its seventeen years, a user base of more than 400 companies worldwide.

Back to Press Releases

Back to Top