The Taiwanese are wryly amused, rather than worried, about the U.S. government’s attempt to supersede Taiwan’s multi-generational semiconductor project.
Hsinchu, Taiwan – The glitzy capital city of Taipei disappeared into the haze of a grey and wet Tuesday morning as we sped uneventfully into Taiwan’s interior, far from the tourists and foreign dignitaries who descend in their masses on Taipei. A bleak industrial landscape punctuated by the occasional residential high-rise rose from the jungle along the way, the tropical flora being all that saved these structures from an unflattering comparison to the Khrushchevka aesthetic they appeared to mimic. The emergence of the sleek, contemporary complex that compose the Hsinchu Science Park couldn’t have struck a more desirable contrast with their surroundings.
Established in 1985, Hsinchu and other industrial, biomedical, and science parks like it form an archipelago of innovation across the length of Taiwan’s populous West coast. Hsinchu occupies a special place insofar as it played a prominent role in incubating what has become one of the island nation’s foremost economic advantages over its competitors: the semiconductor industry. Chips, as those without the time for polysyllabic expressions describe them, were the sine qua non of the “smart” consumer electronics industry well before the pandemic demonstrated the frailty of the supply chains that provision them to gadget makers worldwide.
When the globe severed the interconnectivity that typified global trade since the fall of the Berlin Wall in 2020, the industrialized world’s polities received a rude awakening about the degree to which outsourcing this industry to countries with a comparative advantage in the semiconductor market had become a liability. Countries like Japan, Germany, and the United States (often working with their Taiwanese partners) are crash-coursing their own domestic semiconductor industries. America’s experiment in this realm — a top-down, centrally planned industrial project Congress called the CHIPS and Science Act — has so far failed to meet expectations.
“Until now, the dominant belief has been that increasing chip manufacturing capacity was simply a matter of money,” the Financial Times reported in August. But it turns out that simply decreeing a new domestic semiconductor industry into existence hasn’t gotten the job done. The industry requires about 160,000 new engineers and technicians every year, but it has so far drawn only about 1,500. “For chip technicians,” the FT continued, “that figure is even lower with just about 1,000 new technicians joining each year. In the next five years, the demand for these workers is forecast to reach 75,000.”
That sluggish growth explains my Taiwanese interlocutors’ confidence when I asked whether they were apprehensive that their overseas customers might one day undercut their advantage in the semiconductor sector. Sure, there are some promising facilities rising from the Arizona desert, HSP’s associate researcher Scott Huang conceded. But they do not, and probably cannot, allocate capital efficiently enough to bring the enormous cost associated with semiconductor research, development, production, and distribution down to competitive rates in the medium or even long term. It’s one thing to develop domestic capacity and repatriate industries to satisfy the military’s demand for chips, but consumer electronics are another story. With the utmost confidence, Huang dismissed the notion that America’s experiment with industrial policy would bear fruit anytime soon.
Those remarks dovetail with the assurances to which we were privy from the researchers at the Chung-Hua Institution for Economic Research. Taiwan presently provides the world with roughly 80 percent of the semiconductors in use today. That share is expected to decline by about 5 percent in as many years — hardly the stuff of nightmares. Over 60 percent of the world’s chip supply is made in Taiwan. Of the rest that are made abroad, the vast majority are made with the support and investment of Taiwanese industrial interests. It’s a truly global enterprise. Inputs like design, wafer fabrication, testing, materials, equipment, and photomasking are undertaken by dozens of multinational firms. Cooperation in research, development, and manufacturing employs hundreds of thousands of highly skilled workers on every habitable continent. That level of success is what you might expect from an industry that was born in the early 1980s, and the wry smile you encounter, when the Taiwanese are asked if the U.S. can scuttle this multi-generational project virtually overnight, is reflective of a well-founded self-confidence.
Taiwan’s commercial dominance in this crucial sector isn’t something that the Chinese Communist Party can simply requisition. The brainpower and experience necessary to create this thriving industry, to say nothing of the market incentives that made it a viable investment in the first place, will not take root in a Marxian social compact. But nor can Tawain’s success be legislated into existence by its jealous partners in the free world. Like so many external threats, the Taiwanese don’t seem all that worried about the CHIPS Act. But unlike foreign military threats, when it comes to semiconductors, the Taiwanese have a point.