I have a sense that manufacturing is not coming back until there is a real Cold War 2.0 (or Hot War 3.0 which hopefully never realizes). I mean a real one, like the NATO-WP one when there was not a lot of trading between the two blocs.
1. Manufacturing takes time and money to setup. This is a software forum where changes are made at lighting speed. Manufacturing requires capital, years, and will. In software it might take a day to make an obvious change. Manufacturing hopefully 6 months. Working with physical systems takes time.
2. Manufacturing does not mean jobs. It will be automated with a fifth or less required to setup and maintain. Some technical jobs will come back, but nothing like what has been said.
The tariffs initially put in place will have the same level of effects as a war. This is At Covid levels. The effects take time, inventories are getting sold. You are responding to the news not the effects give it a few weeks for company CEOs to get through the meetings that their employees are pushing up the chains.
Outside of all of that. The biggest issue is consistency of action which this admin squirellllll.
Trump has 4 years if he survives that long. He's an old man. The trade war can't possibly last long enough for manufacturing to be forced to restart domestically. Even if this was his first term and he got a second, by the time the second rolls around everyone would just hope to wait it out until the Democrats retook the White House and they could import everything from China again. But now no one can even make the argument and be taken seriously that we need this, because if Trump's doing trade wars then it must be a bad idea.
This investment thesis is about to be tested at scale.
The whole government has been bathing in the globalization alcohol for so many years that there is a huge interest group rooted for it to go forever. You can't do much without removing this group.
There is no "legal", "orderly" retreat from the current situation IMHO. It's going to be messy. The issue is whether Trump is doing the job or he is just faking it for whatever the reasons.
.. so what's the point of this again?
Once the purge is done in the public and private sectors (yeah you can imagine that such interest groups exist everywhere), Trump and his allies need to install competitive people to create and execute policies to "bring manufacturing back".
Failing either of the two (purging and reinstalling competitive policy makers) fails the job.
No one in the United States feels rich, when we look around, we don't see wealth or prosperity. We suspect, though we'd feel silly to say it out loud, that if anyone ever busted into Fort Knox and looked in the vaults, those would be empty of the gold that it was once famous for.
>What's the US supposed to look like after this purge?
I imagine we'll look like what we really have been for a long while, instead of this illusion that everyone has of us.
It will get poorer economically in upcoming years thanks to gov moves, the wheels have been set in motion. Maybe dollar will tank so that my first sentence won't be valid anymore. Not sure it will be balanced with rest above though.
But Trump can only impact public sector at the moment, right?
Proof of the uncertainty is that Wall Street clearly thinks there's a good chance he'll back off. That's why the market still has room to keep falling, rather than having already fallen as far as it will for a while. It's got a lot of room left. All that room is people betting he's not going to keep this up.
Exactly. I think a lot of people who say they "want manufacturing to come back to the USA" have a picture in their imagination of a 1950s factory employing hundreds of middle class people standing along an assembly line picking up parts with their hands and working on them, or turning bolts on a car as it goes down the line. Then the old steam whistle blows, and all those workers exit the building and drive back to their middle class 1950s homes.
This world does not exist anymore and is never coming back. If manufacturing comes back, it's going to be heavily automated plants with a lean staff of robot technicians keeping the line going. Manufacturing may come back to the USA, but those Fred Flintstone jobs are not part of the picture.
The idea that one can be functionally illiterate yet afford a home, a car and a family by spending 40hr and no more riveting spring hangers onto Chevrolets, or whatever, is a legacy from a blip in time during which the US had a bear monopoly on state of the art for the time manufacturing.
Those men who worked comparable factory jobs in the 1870s through the 1930s had wives and kids doing piece work at home. And in the 1970s their wives all went back to work again, just outside the home.
And a stable political system. If I were in the manufacturing industry, I wouldn't make big bets on moving manufacturing back to the states unless I had some real guarantees that these tariffs weren't going to go away in the next week.
It is genuinely a hard problem. Even 50 years ago or so, simply having two hands and reasonable health was a gateway to a decent-paying job. These jobs have kind of evaporated, and aren't going back. That's a bit part of the nostalgia for manufacturing, I would imagine.
In an ideal world, you would ramp up free education, churn out more skilled labour. But for many reasons, this isn't happening and isn't likely to happen soon. So what should societies do with all this "unskilled surplus labour"?
UK has sort of given up, and since 1970s whole regions of England rely heavily on continued social transfers. I.e. the bankers and the oil rigs pay for the budget, the budget pays for former miners. But that's increasingly unsustainable, not to mention massively unattractive to everyone involved.
US seems to (maybe?) follow a note darwinian approach. But this is, it seems, also not working. Bringing back manufacturing, as you say, won't fix it either. It feels like there is a whole class of people without access to good jobs or education to reskill.
Other than spending a lot of money on this, like Switzerland, I don't really know how to fix it, it is genuinely a hard problem in a liberal country. Would be keen to hear suggestions.
First you invite industry to reshore via subsidies and preferential access to government contracts. If necessary, the government must directly invest in new firms. (They already do this in a very small way with In-Q-Tel and others, so it's not totally beyond the pale. For a time there was even a US Army VC firm.) If you talk to a Chinese factory owner or mine boss, many of them will tell you that they got their start with a >$2M direct investment from their government.
Second you gradually tighten the screws on foreign finished products, not industrial inputs like metals, plastics, ores, etc.
Third you streamline export paperwork requirements and relax things like ITAR.
Then, when that's all humming along and the factories are working, you can launch blanket tariffs to protect your nascent industries, if need be. But you must exempt necessary industrial inputs from tariffs.
What's happening now is completely backwards/inverted and it's going to lead to total chaos.
Also, in most foreign countries, court cases are cheap. (They could literally be 100x cheaper to litigate than they would be in the US.)
If you're talking about in a US context specifically, then if absolutely necessary the federal government could either buy or nationalize the patents in dispute, just as all states nationalize land from time to time under eminent domain. Zvi recently wrote about this in the context of the state buying the Ozempic patent: https://thezvi.wordpress.com/2025/01/21/sleep-diet-exercise-...
> [The US Government] should buy out the patents to such drugs.
> This solves the consumption side. It removes the deadweight loss triangle from lost consumption. It removes the hardship of those who struggle to pay, as we can then allow generic competition to do its thing and charge near marginal cost. It would be super popular. It uses government’s low financing costs to provide locked-in up front cold hard cash to Novo Nordisk, presumably the best way to get them and others to invest the maximum in more R&D.
> There are lots of obvious gains here, for on the order of $100 billion. Cut the check.
That's not true at all. US companies register patents in all foreign jurisdictions as well. It wasn't legal for Chinese companies to copy all that US IP, they were just able to do it because to stop them, our diplomats would have basically had to persuade the Chinese government not to develop.
The difficulty lies in political will and execution. Either fails then the whole scheme fails.