Editor’s note:
Intel CFO Stacy Smith is a featured speaker at the CFO Rising West conference September 21-23, 2009 in Las Vegas. For more information, click here.

Also, this article has been updated to correct an erroneous reference to Intel’s next generation of processing technology. The company is building 32-nanometer processors, not 32-nanosecond processors.

Like jittery individual investors who sell low after stock prices tumble, companies often put off capital investments when the economy tanks in hopes of preserving cash. But for those that have cash available, spending on the future while the market is soft can be a way to cultivate a long-term competitive edge.

That was part of the rationale behind Intel Corp.’s decision earlier this year to commit $7 billion over two years to retool three existing plants to build new 32-nanometer processors. The factories won’t be ready until 2011, but when they are, says Intel CFO Stacy Smith, the company will be positioned to produce the products at a lower cost and with higher performance levels than companies that have pulled back their spending. (News that Intel was investing in U.S. manufacturing plants earned the company a lot of positive press early this year, at a time when many others were announcing layoffs.)

The investment underscores the capital-intensive nature of the semiconductor business, notes Smith, who will offer a presentation on investing for growth at the CFO Rising West 2009 conference, to be held September 21–23 in Las Vegas. “We have found over the years that investing in big capital projects [can work well] countercyclically, as we go into these downturns,” he told CFO.com.

Among chipmakers, of course, the need for high capital investment remains, even when there isn’t a down cycle. Intel and its competitors are driven by Moore’s Law, the fact — identified in 1965 by company co-founder Gordon Moore — that the number of transistors on integrated circuits doubles about every two years. For the past 50 years, Moore’s Law has continued to hold true with remarkable consistency, bringing great advances in computing but also great capital-expense demands for companies such as Intel. Smith says Intel is already conducting R&D not only for the generation of products to come roughly two years from now but also for the generation after that.

Intel CFO Stacy Smith: bound by Moore’s Law on capital spending

Still, says Smith, the principle that investing during weak economic climates produces a disproportionate competitive advantage should apply to other industries as well. He allows, though, that “you have to be careful — history is littered with companies that went bankrupt by blindly following that rule.”

The “art,” he says, is to make investments in new products while simultaneously tightening other areas. Fortunately for Smith and Intel, however, cutting costs during the recession has been less challenging than it might otherwise have been, thanks in part to what he calls “just luck.”

Back in 2006, even though the economy was booming, Intel decided it was too management-heavy and slow in executing on business plans. It began an ongoing head-count reduction strategy, driven by a focus on design and manufacturing efficiencies, that has pared the work force from a high of 103,000 to less than 80,000 today.

But a large majority of the cuts came before last year’s meltdown of the economy, according to Smith. In January the company announced that 5,000 to 6,000 jobs would be eliminated or relocated after closing or stopping production at five plants, two of them in the United States. Two weeks later, however, Intel announced that its investment in retooling the three other plants would preserve about 7,000 jobs. “We have downsized significantly less in the heat of the recession compared to many companies, because we had done a lot of it prior to that,” says Smith. “It was very beneficial to have the luxury of time to attack the underlying inefficiencies, which is one of the things that has enabled us to continue making investments.”

Meanwhile, Intel, like many companies, is seeing the skies over the economy starting to clear. It recently issued guidance forecasting specific results for the third quarter, after having not done so for the previous two quarters. Given the market volatility, Smith says, doing so would have amounted to “making up a number” and risking its credibility. Instead, the company gave qualitative forecasts, discussing market trends, inventory trends, and business plans.

Because Intel resides at the end of a supply chain, its business for any particular period depends as much on the amount of inventory in the chain as current end-user demand. In late 2008 and early 2009, that inventory level was so high that orders for Intel products fell way more than computer sales did, Smith notes.

Now, he says, the supply chain has regained enough equilibrium to warrant a specific forecast, although it’s on the conservative side. Third-quarter revenue is expected to be $8.5 billion, plus or minus $400 million. That’s a somewhat wider range than Intel typically has estimated in the past, and the top-line figure is at the low end of the company’s typical seasonal variance between the second and third quarters.

In fact, the improvement actually began in the second quarter, when Intel sales fell 15% from the year-earlier period to $8.02 billion but roared past analysts’ expectations of $7.28 billion, according to Thomson Reuters. Profit was 18 cents per share, excluding a charge for a $1.4 billion antitrust fine by the European Commission, compared with the 8 cents analysts had expected.