What does it mean that the U.S. government had to step in last week to help the semiconductor manufacturers of the Silicon Valley? It means that we have to give up one of our comforting illusions about the toughness of global competition.
For years we have taken a peculiarly American pride in the way U.S. entrepreneurial companies have secured technological leadership in the crucial business of electronic microchips. Where our automobile and steel giants had failed against foreign competition and gone running to the government for protection, the innovators of the Silicon Valley--actually the Santa Clara Valley north of San Jose--were self-reliant and successful.
Firms such as Intel Corp., led by its engineer-founders Gordon Moore and Andrew Grove, or Advanced Micro Devices, founded by the flamboyant former salesman Jerry Sanders, competed and won against Japanese companies many times their size. Against what we perceived to be an empire-like Japan Inc., we felt good about relying on Luke Skywalker.
But something has been going wrong with the script. Last year about this time the normally ebullient Andy Grove was angrily complaining to anyone who would listen that Japan's Hitachi Ltd. was cutting prices severely on an electronic device that Intel had just begun producing at its new plant in Albuquerque, N.M. As a result, Intel's product--an erasable programmable read-only memory chip, or EPROM in the jargon of the trade--was rendered unprofitable from the start.
Grove and other semiconductor makers cried "unfair" to the U.S. government, and last week they got action. The Commerce Department ruled that Japanese firms were selling EPROMs and other electronic products in the United States for less than their price in Japan. The United States calls this "dumping" and punishes it by attaching a tariff to the price of the offending import.
That tariff, along with the effect on Japanese competition of the stronger yen and weaker dollar, will benefit U.S. semiconductor makers by allowing them at least some profit on sales of the microchips in question.
But it will be only a breather. The basic memory chips the Japanese were dumping are in severe oversupply worldwide. Even U.S. government edicts have little long-term effect against falling prices for commodity products. And in the more advanced semiconductors, such as Intel's microprocessors, U.S. firms face increasing competition from far larger Japanese companies that have closed whatever technological gap existed years ago.
Against Intel's or National Semiconductor's $1 billion-plus in sales, for example, Hitachi does more than $20 billion in annual turnover, NEC Corp. $7.5 billion and Fujitsu Ltd. $5 billion. As much as overall size, the ability to support one business with profits in another can be telling in competition. To wit: NEC, a world power in telecommunications equipment, has become the world's largest producer of semiconductors.
We should not overlook the fact, for all the xenophobic talk of Japan Inc., that Japanese business has its visionary, entrepreneurial leaders, too. NEC, led since 1964 by Koji Kobayashi, is a case in point. Kobayashi determined that, if NEC was to play a leading role in computers and communications, it had to build a presence in the computer's principal component, the semiconductor. So NEC worked at making semiconductors for 13 years before it showed a profit. "In business," Kobayashi has written, "you have to look ahead."
That is not to slight in any way the Silicon Valley entrepreneurs, who have been protean investors in research and development and in new plants. Intel, for example, invests 14% of its sales in R&D; Advanced Micro Devices invests 18%. But there's no getting away from the fact that it is harder for Intel to keep up that level of investment in good years and bad than far larger and richer NEC.
On top of that, capital is cheaper to the Japanese firm because national policy keeps interest rates low--as U.S. policy used to do. The United States hasn't been keeping interest rates low over the last decade--nor has it been winning in international competition.
So where does that leave the U.S. semiconductor industry? Embattled but able. Our two largest producers, Motorola and Texas Instruments--each with $5 billion-plus in sales--have other businesses to see them through the tough cycles in semiconductors.
As for the Silicon Valley, increasingly it sees a safer future in making semi-custom microchips that its customers can imprint with final product designs. Such specialization is thought to be safe from the torturing competition in commodity products. But semiconductor analyst Richard Billy, of the Gartner Group research firm, points out that Japanese companies are the emerging leaders in semi-custom chips, too.
Ah, the Silicon Valley, such a wonder while it lasted--like Camelot.