We might gain an edge in perspective if we target selected "blue-collar" power applications, not the products themselves, that historically pick up in hard times. That might include automotive, basic industrial tasks and certain alternative energy schemes. We'll know that both the power and general economies are picking up when the number or percentage of products targeting those areas go below a pre-determined point.

Looking at selected "bread-and-butter" applications is another assault on traditional analysis, and it's already done to some degree, but I have yet to see it fully applied using a top-down approach. It amounts to keeping track of power products that imply basic need and functionality, not futuristic bells and whistles. Yet more people seem interested in raw IC or module production.

Indeed, I get calls every week from people asking for traditional market data, which indicates that the masses equate chip numbers with financial success. It ain't necessarily so. And never has been.

Major surprises aside, why markets go up and down is often a mystery, even to the experts. Basically, though, a stock's worth often boils down to how its owner feels at the end of a given day. It's one reason "the market" tends to drop on Fridays in good economic times (profit taking), and stabilizes at week's end in bad times, allowing the stockholder to go home feeling somewhat reassured. A stock's price likely always will be a strong function of a person's beliefs.

That's a scary proposition. It's been that way for a long time, but there should be a better solution. Looking at selected power applications first, not the products they imply, or raw numbers, may be a unique way to more accurately gauge both the power and general economies.