[zenmacro]

Two thoughts.

1)

> Globalization is Dead and No One is Listening

We're talking about it on HN page 1. And I hear a lot of talk among the macro and policy people I follow about friendshoring.

2)

The author makes an argument that the fact that TSMC had to send US engineers to Taiwan for training was a sign of a talent shortage. But earlier the author said this:

> TSMC is arguably the one company that most epitomizes all the forces of globalization – free trade, hyper specialization...

Hyperspecialization to me implies that you need to go to HQ to understand how everything works and it will take a while to understand it all. So the earlier argument about a shortage of talent in the US seems tenuous to me.

[hn_throwaway_99]

> Globalization is Dead and No One is Listening

Yeah, from that perspective the author is just wrong. I posted this the last time HN discussed this article, https://news.ycombinator.com/item?id=33957643, but despite the author saying "No American, or any Western media outlet for that matter, bothered to cover this speech", I first read about it on an article on the front page of the NY Times website.

[svnt]

The author seems to have missed that TSMC previously failed to transplant TSMC culture into the USA. In nearly thirty years they probably realized they need more than just semiconductor expertise transfer — they need a transfusion of Taiwanese culture.

One way to do that is to send an entire employee base from Taiwan, but then how do you maintain or integrate in the US? Splitting the employee base in half seems more likely to succeed, probably with process/engineering roles predominantly Taiwanese and more external-facing roles skewing American.

[AlbertCory]

There are lots of Japanese and Chinese companies with substantial presences in the US. TSMC probably realizes it's a learning process and they need to get on with it.

[svnt]

My experience with those (and German companies) is that it tends to be using the US employees as an innovation hub, which in practice means we get the projects that are more likely to fail, because that is less culturally embarrassing here.

This approach can’t work with advanced semiconductor manufacturing for a number of reasons but primarily the same factors that drove Intel’s adoption of “Copy Exactly!”

[guessbest]

Globalism isn't dead. It has just diminished slightly from its height in 2007-9 after the global economic collapse, which should be called a double dip recession first in 2007 and again in 2013. It has gone from 60 percent of the world's trade to 55%. Not exactly the end.

https://www.weforum.org/agenda/2019/01/how-globalization-4-0...