Post

1/ Dan Wang’s NYT essay on TSMC's Arizona plant argues that TSMC is doing for America what Apple did for China. But the analogy rests on two assumptions and it's not clear that either one of them is holding. www.nytimes.com/2026/10/23/o...

2/ The first assumption is about supplier spillover. Apple’s impact on the Chinese economy ran through thousands of domestic component, tooling and assembly firms pulled up to Apple's standard — Foxconn first, then a whole ecosystem.

3/ TSMC Arizona’s equipment comes from ASML, Applied Materials, Tokyo Electron. Those vendors basically supply every advanced fab on earth. But what actually gets localized around Phoenix, and over what timeline, is still open.

4/ The second assumption is about labor mobility. Spillover in China happened because engineers left and founded or joined competitors. That’s basically the Shenzhen story.

5/ But, TSMC keeps R&D concentrated in Taiwan, and Taipei codified an “N-1” rule barring export of the leading-edge node. Arizona is a generation behind Hsinchu as a matter of law.

6/ The absorption piece is also not fully settled. Diffusion in China worked because engineers left for domestic rivals. In American, the domestic rival is Intel, and the Intel Foundry may not be in shape to absorb much.

7/ Rotation makes both difficult. Engineers come on contracts of up to three years, the first cohort is already heading home. The question is whether some of what they know stays behind when they leave.

8/ Of course none of this means Arizona fails. But presence alone doesn’t guarantee learning. A fab can run well for years without the skill behind it reaching anyone who doesn’t work inside it, or spilling out into the broader economy.

10/ Allied industrial investment depends on engineers being able to enter and stay. Enforcement volatility and unpredictability can undo a project much faster than subsidies can attract a new one.