But in the mid-’70s, TI decided to move more heavily into the consumer electronics business, with calculators and wristwatches. In 1978 the company picked Chang to run the consumer business. He was proud of Speak & Spell, the breakthrough speech-synthesis toy that came out that year, which used the world’s first single-chip speech synthesizer [see “25 Microchips That Shook the World,” IEEE Spectrum, May 2009]. But overall, he says, the move was a mistake, for the company and for him personally.
Mark Shepherd, then chairman and CEO of TI, agreed with the prevailing wisdom at the time that a good manager could manage anything. In this case, Chang says, “I think he was wrong. I found the consumer business to be very different. The customer set—completely different. The market—completely different. And what you need to get ahead in that business is different, too. In the semiconductor business, it’s just technology and costs; in consumer, technology helps, but it’s also the appeal to consumers, which is a nebulous thing.”
In 1983, with TI’s consumer business struggling and Chang sidelined as head of “quality and people effectiveness,” he knew his path at TI no longer led to the executive suite. He decided to leave.
He had no real plans, but he was confident he’d be a hot job prospect. And so he was; after his departure hit the news, his phone started ringing, he says, and didn’t seem to stop for days. Eventually, he had two offers he seriously considered—one from a venture capital firm and one from General Instrument, an East Coast manufacturer of semiconductors and cable set-top boxes.
He joined GI in 1984 as president and chief operating officer and moved to New York City; he’d once dreamed of living there and walking to work. He moved to an apartment on Fifth Avenue, just a few blocks away from his office in the General Motors building. And he had, he thought, free rein to build up the company’s R&D operation.
He didn’t. The company, he says, wasn’t really serious about R&D, preferring to buy businesses rather than grow them internally. And the pleasure of walking to work paled quickly in the rain and winter slush. Meanwhile, he had separated from his then wife, Christine.
He left GI after just a year. Once again, he quit without a plan. “After these two setbacks, at TI and at GI,” he says, “I did not think that my aspiration to be the CEO of a major U.S. company was in the cards.” He contemplated going to a venture capital firm, but, he says, “I didn’t think I would be very good at it.”
Then a government official in Taiwan, K.T. Li, contacted him. Chang had met Li several years previously, when TI was negotiating to build an assembly plant in Taiwan. Li was looking for someone to manage the decade-old Industrial Technology Research Institute (ITRI), turning it into the Bell Labs of Taiwan.
Just about everybody he knew told him not to do it. His soon-to-be ex-wife thought he’d lost his mind for even seriously considering the offer. But Chang, thinking he had little to lose, decided to give it a try. He took over as president of ITRI in 1985 and immediately began shaking things up. For starters, he broke the “iron rice bowl,” challenging the assumption that once you got a job in a government organization, you’d be set for life. Instead, Chang instituted a review process that put the lowest 2 percent of performers on probation, with the possibility of dismissal looming if they didn’t improve. He also began transitioning ITRI toward partial funding by industrial contracts, not just by government subsidy. Not surprisingly, these changes didn’t go over well with the staff. Hate mail began to pour in to Taiwanese legislators, complaining about Chang and his policies. "Back 25 years ago," he says, "they considered me a foreigner who suddenly became their boss. They were scared of me."
“He’s known not to tolerate fools,” says Chenming Hu, a professor emeritus at the University of California, Berkeley, and former chief technology officer for TSMC. "And that engenders both respect and fear."
The situation at ITRI had only just started to settle down when Li came back with another proposal. The government, Li told him, wanted Taiwan to have a semiconductor company and thought that if anyone could get it off the ground, Chang could.
Chang knew it wouldn’t be easy. TI was still the dominant company in the semiconductor business, with Intel rising fast and National Semiconductor Corp. and Advanced Micro Devices also strong players. Just to survive among these giants would be a challenge.
And he would be starting at a huge technical disadvantage. In 1975, ITRI had licensed from RCA Corp. a 7-micrometer semiconductor processing technology that at the time was already out of date. By 1985, research at ITRI had advanced to a 3-µm process; however, the rest of the industry had moved forward even faster—about two generations ahead, with advanced products being manufactured in 1.5-µm technology.
But Chang said yes anyway. “It’s like in the movie The Godfather,” he says, “an offer you can’t refuse,” explaining that to say no would have tagged him as someone completely without ambition and would likely have dimmed his future in Taiwan.
Taking on the semiconductor behemoths head-to-head was out of the question. Chang considered Taiwan’s strengths and weaknesses as well as his own. After weeks of contemplation he came up with what he calls the “pure play” foundry model.
In the mid-’80s there were approximately 50 companies in the world that were what we now call fabless semiconductor companies. The special-purpose chips they designed were fabricated by big semiconductor companies like Fujitsu, IBM, NEC, TI, or Toshiba. Those big firms drove tough bargains, often insisting that the design be transferred as part of the contract; if a product proved successful, the big company could then come out with competing chips under its own label. And the smaller firms were always second-class citizens; their chips ran only when the dominant companies had excess capacity.
But, Chang thought, what if these small design firms could contract with a manufacturer that didn’t make any of its own chips—meaning that it wouldn’t compete with smaller firms or bump them to the back of the line? And he realized that this pure-play foundry would mean that Taiwan’s weaknesses in design and marketing wouldn’t matter, while its traditional strengths in manufacturing would give it an edge.
TSMC opened for business in February 1987 with $220 million in capital—half from the government, half raised from outside investors. Its first customers were big companies like Intel, Motorola, and TI, which were happy to hand over to TSMC the manufacture of products that used out-of-date technology but were still in some demand. That way, the companies wouldn’t have to take up their own valuable fab capacity making these chips and would face little harm to their reputation or overall business if TSMC somehow failed to deliver.
Soon, Chang says, start-up companies, which would live or die depending on TSMC’s manufacturing runs, signed on. Early customers like Broadcom, Marvell, Nvidia, and Qualcomm, Chang says, “started with us when they were small,” and being able to tap into TSMC’s manufacturing prowess was a big reason for their success.
“Not having us would have slowed down innovation in the industry,” Chang maintains, pointing out that those little companies would likely have had to invest in manufacturing capacity instead of R&D or share their intellectual property with a TI or an IBM.