Semiconductor Jujitsu

"Reversal of Fortune? The Recovery of the U.S. Semiconductor Industry" by Jeffrey T. Macher, David C. Mowery, and David A. Hodges, in California Management Review (Fall 1998), Univ. of California, S549 Haas School of Business #1900, Berkeley, Calif. 94720–1900.

During the 1980s, the woes of the U.S. semiconductor industry became a symbol of America’s alarming competitive plunge. In 1989, the Massachusetts Institute of Technology’s Commission on Industrial Productivity, reflecting widespread expert sentiment, issued a report saying the industry was too "fragmented." Yet since then, semiconductor makers have made a dramatic recovery—assisted, ironically, by that very "weakness."

In the United States—in contrast to Japan and Western Europe—the semiconductor industry consists of numerous, relatively small firms, from industry leader Intel to Micron and other, more specialized companies. The U.S. firms dominated the world market until the mid-1980s, when Japanese producers, concentrating on the dynamic random access memory (DRAM) devices that supply computer memory power, surged into the lead, observe Macher, Mowery, and Hodges, a doctoral student, professor of business and public policy, and emeritus professor of engineering, respectively, at the University of California, Berkeley.

State-aided Japanese giants such as Hitachi and Toshiba enjoyed an access to capital that U.S. firms lacked, and in the fierce race to develop higher-capacity memory devices and sell them cheaply, they drove many of their American competitors out of the DRAM business by 1985. Analysts who regarded DRAM production as an indispensable "technology driver" for semiconductor manufacturing painted a gloomy picture of the U.S. semiconductor industry’s future. And indeed, the U.S. global market share fell from almost 62 percent in 1980 to a low of 37 percent in 1989.

But the industry’s "fragmented" nature, a handicap in DRAM competition, turned out to be an asset. U.S. semiconductor makers were able with relative agility to shift their focus to higher-end products that played to American strengths in innovation, the authors say. "U.S. firms have reoriented their strategies . . . to concentrate on logic and microcomponent products, where foreign competition was less intense," and they could exploit their ties to computer software developers in coming up with new products.

That strategic repositioning, together with improvements in the quality of the products and in the manufacturing process itself, helped the U.S. industry regain global leadership in semiconductors by 1993, the authors say. By 1997, U.S. chip makers such as Intel and Texas Instruments controlled more than 50 percent of the world market, while Japanese firms, now facing DRAM competition from South Korea and Taiwan, saw their market share slip to 29 percent.

But the American "reversal of fortune" may not be permanent, the authors warn. Although worries about industry fragmentation proved groundless, there is reason for concern about the state of basic scientific research, which fuels commercial advantage in technology industries. Today, Bell Labs and the other huge corporate labs that did much of the fundamental research underlying semiconductor technology concentrate on short-range corporate goals, while the leading semiconductor firms themselves focus their research on new-product development. Hitachi and other major overseas competitors, the authors note, still do "considerable" long-range research.