By Don Clark and Ana Swanson | The New York Times
In December 2022, Taiwan Semiconductor Manufacturing Company, the key maker of the world’s most cutting-edge chips, said it planned to spend $40 billion in Arizona on its first major U.S. hub for semiconductor production.
The much ballyhooed project in Phoenix — with two new factories, including one with more advanced technology — became a symbol of President Biden’s quest to spur more domestic production of chips, the slices of silicon that help all manner of devices make calculations and store data.
Then last summer, TSMC pushed back initial manufacturing at its first Arizona factory to 2025 from this year, saying local workers lacked expertise in installing some sophisticated equipment. Last month, the company said the second plant wouldn’t produce chips until 2027 or 2028, rather than 2026, citing uncertainty about tech choices and federal funding.
Progress at the Arizona site partly depends on “how much incentives that the U.S. government can provide,” Mark Liu, TSMC’s chairman, said in an investor call.
TSMC is just one of several chip makers running into obstacles with their U.S. expansion plans. Intel, Microchip Technology and others have also adjusted their production schedules, as a sales slump in many kinds of chips pressures the companies to manage their spending on new infrastructure. New chip factories are hugely complex, involving thousands of construction workers, long construction timelines and billions of dollars of machinery.
