pull down to refresh

Much has been written about tariffs and what Trump hopes to achieve with them. One of the things oft discussed is bringing manufacturing back. A tiny bit of research suggests that this will in no way happen for anything sophisticated on any medium-term timescale.
This article was a nice and simple overview of why, including from some perspectives you don't often hear:
Chinese workers work longer hours more happily and they’re physically faster with their hands; they can do things that American labor can’t. It’s years of accumulated skill, but it’s also a culture that is oriented around hard work and education that the United States no longer has.
Sadly, what I describe above are not theoretical situations. These are things that I have experienced or seen with my own eyes. It’s fixable, but the American workforce needs great improvement in order to compete with the world’s, even with tariffs.
The story could be different in twenty years, of course. Anything could happen in two decades. But the idea that we'll gut through a couple hard years and suddenly have an industrial base again, or even the seedlings of one, strikes me as bonkers.
So bonkers that even Trump, for all his many and obvious limitations, must understand it. And is doing all this anyway. Which is also a good thing to understand.
There's a very important distinction, that many people have raised, between manufacturing and manufacturing jobs.
Not only can manufacturing easily return, it's not clear that it ever left. Whether American workers will be competitive, compared to robots and foreign workers, is an entirely different issue.
reply
Point taken re: jobs vs manufacturing as an industry. However:
Not only can manufacturing easily return, it's not clear that it ever left.
I guess this is where talking in generalities is unhelpful. It's clear that certain crucial kinds of manufacturing did in fact leave -- I'm thinking of almost everything pertaining to the semiconductor industry, to pick a current very important darling. Building capacity to do normal semi manufacturing at scale, let alone the frontier-level work that TSMC is doing, is something that did leave, and would take a long time to come back.
You seem to have a different point of view. Care to expand on it?
reply
I'm not sure I have a different point of view on that. The types of manufacturing have changed and people can attach their value judgements to whether that's better or worse.
Where manufacturing takes place has also changed, which is what caused the Rust Belt. Not all of that is due to foreign competition, though. Lots of manufacturing moved to the southern states, which had more attractive policy bundles for manufacturers.
Could semiconductor manufacturing be re-shored in a reasonable timeframe? I don't know. I don't understand why Taiwan's comparative advantage in this is so strong that it's worth perennially saber rattling towards China and making outlandish security guarantees.
Knowing as little about the industry as I do, it doesn't seem like something America would be particularly ill-equipped to do.
reply
The US is definitely ill-equipped to manufacture semiconductors. TSMC is the powerhouse it is because of decades of investment and coordination across Taiwan and its society. This is not just in the form of financial support from the government. We're talking a university feeder system that trains TSMC's highly-skilled workforce, monetary policy that favors exports, localized brownouts and water shortages in nearby communities to keep the lights on at the factory. It's not just about manufacturing, it's a culture. Not to mention making semiconductors is incredibly difficult, especially at the size and scale that TSMC does it.
reply
Interesting. You could think of a bunch of those as in-kind subsidies.
I can see why it would be unacceptably costly for someone to try outcompeting them head on.
Eventually, there will likely be a technological advancement that makes another production process profitable. Then, being so heavily invested in the current technology will be a disadvantage.
reply
If you wait long enough, everything ends.
A useful question is how long you're willing to wait, and what you're willing to endure in the interim.
Could semiconductor manufacturing be re-shored in a reasonable timeframe? I don't know. I don't understand why Taiwan's comparative advantage in this is so strong that it's worth perennially saber rattling towards China and making outlandish security guarantees.
Knowing as little about the industry as I do, it doesn't seem like something America would be particularly ill-equipped to do.
It just Takes A Long Time to build complicated shit in the real world.
You're talking about literally the most sophisticated and technologically ornate industry on the planet, with the highest tech inputs, the highest expertise (including all the tacit knowledge that comes with it) swimming upstream against the forces of globalization that have encouraged all of that expertise and infra to accrue elsewhere.
If you want it to re-shore all that for security reasons (which seems like a no-brainer, and has for a while) then you need to do something to countervail all those forces and all that comparative advantage. But you can't just snap your fingers and expect it to happen tomorrow.
reply
In broad strokes, I appreciate all those factors.
Industrial centers do move, though. There's some margin that would cause a shift, but I don't know how large it is.
I mentioned in another comment that the most efficient way to address the national security concern is to require the government to buy stuff that was made in America. That would allow the industries to establish themselves domestically and evolve at least semi-organically. Maybe some even hit a tipping point in the global landscape and become economically viable.
reply
Yeah, they should use those tools. Back in the day, DoD procurement worked kind of like that, and the stuff they bought was egregiously expensive, just bonkers expensive. They eventually rescinded the rule because it wasn't viable in the face of competition for all the reasons that Libertarians love competition. We'd be flying planes made out of calculators and Speak and Spells.
Chip Wars talks about some of this; may be worth a re-read. That guy should write a sequel, that's for sure.
I don't understand why Taiwan's comparative advantage in this is so strong that it's worth perennially saber rattling towards China and making outlandish security guarantees.
It's simple. From The Usual Suspects:
They realized that to be in power, you didn't need guns or money or even numbers. You just needed the will, to do what the other guy wouldn't.
Tariffs won't help ya. Don't lure TSMC in with subsidies or "less tariffs". These are all self-owns. If you want to build semiconductors, build semiconductors.

JUST DO IT.

reply
What can "JUST DO IT" mean, other than state-run enterprise, then?
That's the worst self-own of all. If Taiwan is dead-set on producing way below cost (which I take to be the implication), then we should just be happy to accept the artificially inexpensive inputs.
reply
Huh? Why does the state decide what "we" wanna do?
I meant it more like:
Current scenario: I as a citizen of the mighty USA think that the USA should manufacture semiconductors and thus I am voting for Donald Trump because he will make it so.
Desired scenario: I as a citizen of the mighty USA think that the USA should manufacture semiconductors and thus I am building a semiconductor factory and I shall elevate our might.
You can increase the tariffs enough... that eventually you'll get 'all the manufacturing back' when it's cheaper to make goods domestically.
But at what cost?
reply
We will see later but the tariff affect all us
reply
Make Manufacturing Great Again!
reply