Chips are like agriculture: It does not matter that the home grown product can only compete with government support, you never want to be in a position where you cannot provide for your country’s minimal needs purely from local sources.
It depends what the minimal needs are, but basically the US has this for chips already. Yes the US only makes older chips, but in fact those are the ones that really matter, that you can’t do without. The ones that are in everything.
However making everything domestically just isn’t really viable for any country any more. The scale of our technological civilization and the diversity of goods and materials it depends on is more that any one continent can support, let alone one country.
It would be possible to collapse that down to one continent if absolutely necessary, but it would be incredibly economically painful and the US would need to give up on a lot of non-essentials and other priorities to devote resources to duplicating capacity that already exists elsewhere.
Buying rocks and exporting chips is fine. Of course it's not that extreme but the sentiment is there. It's about expertise capital and concentrating that domestically.
I think it is not about one country doing to all. But technology transfer should happen within like minded countries and within a smaller circle. So may be two continents max.
This is a completely meaningless statement. It sounds good, but contains no actual argument. why are chips like agriculture? What makes that statement not apply to literally every product category that exists?
Agriculture is self-evident, we need food on a daily basis to live. I don’t need three square microprocessors a day.
Farming equipment needs microchips, as does practically any modern device, vehicle, machine, etc. If you don't have domestic microchip production then pretty much everything else you make will have foreign dependencies that could be embargoed.
Of course we'd think that, we're on the Internet, discussing it. The people who aren't, aren't here to rebut that. I met this woman who lives part-time in Romania, and was saying that they don't have smartphones still, which is why she has to spend her time in Romania, because she can't just have this particular subset of elderly people in Romania get information from her off of Instagram.
Just like you don’t want to wait many months to tool up to make food if there is conflict…
You don’t want to wait many years to tool up to make chips that go in weapons, either.
Else, you risk being structurally defeated very early, and this leads to others concluding that they can take you on. It undermines other preparations.
Yes, we understand that, but it’s not a good comparison.
Producing chips and agricultural goods are fundamentally different, starting with limits like climate, seasonality, and land. At most, you could compare it to mining minerals for chip production
>It does not matter that the home grown product can only compete with government support,
Even that isn't a given, because unless you have amassed a certain amount of technocratic and governmental competency chances are it can't compete even with government support and you just produce crony dysfunctional companies.
And of course there's economic trade offs. If you're politically ordering your economy to make chips, it doesn't make something else, and whatever it was making and trading for chips it was better at, and so you get fewer chips, that's comparative advantage. Industrial policy (and tariffs) do not increase aggregate production, they reduce it. And given that the circle of items you "can't do without" seems to be a bit of a moving target these days, at some point you're actually more brittle because you've replaced large chunks of the market with state production.
People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing. But cronyism is a policy and structural failure, it's usually because the incentives for the different parties involved encourage it to happen. Institutions can be designed carefully if policymakers actually want to do it.
USPS is a great example of an organization that's managed to largely avoid this. Whenever you mention that people crawl out of the woodwork to complain about the 7 different times they lost their package, but their logistics at scale is still unmatched by the private sector, while also not completely negating the value of private sector alternatives (which so often is argued would happen if the government actively started doing anything new).
It is established economic theory not just some opinion .
Constraints in form of regulations, subsidies, tariffs by government intervention will produce inefficiencies.
This is the core economic philosophy behind both the modern centre right[2] i.e. neoliberal democrats (80s-today) and the right wing republican conservatives (pre MAGA -2016).[1]
Even communists would agree this is correct, it is understood in different levels since Adam Smith invisible hand, from far left to centre left just posit that different polices that benefit the people not the economy are worth the cost .
That is actually true for all other groups except anarchists and libertarians all agree government intervention is needed in some form, they just disagree on how .
——-
Postal services anywhere in the world are inefficient by design. They are governed by universal service obligation principles not efficiency same with telecom providers and other utilities.
The question then becomes how much inefficiency is acceptable given the objectives . There are no right answers, you could have competing private couriers who dump unprofitable low density routes to USPS while serving profitable ones themselves (UPS, fedEX, Amazon ) or you could have gigantic postal service which is sole delivery provider socialist style, there are going to be some problems either way.
It is just matter of preference which set of trade offs we are okay with .
—-
[1] Post MAGA it is a populist party economically there is no fixed ideology to be characterized
[2] the centre right label is economic and technical to contrast say the socialist left party of FDR/New Deal 1930-70s, not meant to offend.
But that's exactly the thing, where we care about an industrial capacity for its strategic reasons rather than because we're trying to maximize efficiency of revenue extraction. Pure market efficiency is great if you're trying to maximize profits, but that's not necessarily the goal here because the benefits are indirect, just as they are with universal postal service (or universal healthcare, etc).
>People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing.
No, it's the opposite. It's a conclusion taken from empirical evidence looking at success rates in the real world. As the US has engaged on its most recent bout of industrial policy, industrial activity has declined, not gone up. Again, this is expected. If you through protectionism make chips and steel in the US inefficiently, everyone in the US using those products as inputs suffers.
There's some limited cases like developing countries engaging in catch up growth , but there's virtually no evidence for effectiveness of these policies in cutting edge technologies, which isn't surprising because they by definition tend to rely on supply chains and knowledge sourced from all across the globe, and to this protectionism is particularly disruptive. And even in developing nations most of the time practices like ISI (import substitution industrialization) fail, it's devastated Latin America in the 1950s to 1980s.
I don't know what USPS proves in this case because it's sustained by a government monopoly. Obviously you could if you wanted privatize mail delivery.
How? What if we just decide we will never go to war with Mexico or Canada and get comfortable with the idea of importing from our allies? There is no serious future risk from doing that
100 years ago the British Empire ruled the world, now it's a small island you don't hear much about...
The US is only about 250 years old, I'd be cautious about predicting the future
You're off by about 150 years. 100 years ago was 1925, the British has already largely collapsed losing the US, Canada, Australia. 100 years isn't forever but it's a long time
And we haven't had any serious threats from Canada since 1812. I think the most reasonable estimate is 100-200 years
~1920 is the peak of the British Empire in term of territory, anyways, the details are meaningless, what matters is that things move fast and just because you're at the top of your game right now doesn't mean you'll be in the same position in 100 years
I could also take the example of world wars, in France ww1 was deemed "la der des ders", which meant "the very last war" or "the war to end all wars", well 20 years later we were at it again
Or simply look at China, you don't even have to go back 100 years in the past to see drastic changes.
That's only the case if you include Canada and Australia, which were functionally independent at that point.
I'm not claiming nobody will invade France or Taiwan in the next 100 years, I'm claiming that the US is special. We haven't been invaded since 1812 and haven't really been attacked since 1941. It's reasonable to predict we won't be invaded or go to war with our neighbors for 100 years since it hasn't happened for 213 years
No, based on the history of conflict we can say that the more time that passes with neighbors not invading one another, the less likely they will in the future
If we include the idea that either one of them is allied to a major power at war with the US over a hundred year horizon, right now that looks pretty likely, and arguably is one of the things the current US admin are trying to stop before it becomes inevitable.
The US admin is very clearly pushing us in the opposite direction. You believe that Trump's actions make war with Canada less likely? What's the mechanistic explanation?
So the argument is that the neighbors will never ally and be involved in a war that puts US on the opposite side? What is the argument for the sovereign neighbors to always be neutral or on US side come what may?
It's an extension of a peace through strength philosophy. If you lose your critical sovereign capabilities you become weak and vulnerable. You no longer get to decide who you do and don't "never go to war with".
Yes I understand the reasoning, it's just obviously wrong. This doesn't happen and hasn't happened in hundreds of years. This is not why people get invaded, otherwise Switzerland would have been invaded many times over the last century
This kind of answers your own question: the reality is you are only a reliable US ally [1] if you can hold them by the balls TSMC style. Given that countries go to great lengths to develop and maintain such dependencies. Canada's current weakness is at least in part because it has failed to do so.
[1] Edit to add: This was/is poorly worded - I mean that the US will only guarantee that you remain an ally while they are in some sense dependent on you, and while doing so they may work to break that dependence, which you may interpret as them trying to abandon you.
You are claiming that Canada will stop exporting to the US in the future? Populism does the opposite, they would likely stop importing. There's almost no risk to us that they would not export chips
My claim is Canada has no unique product that it offers that the US is sufficiently conspicuously dependent on for it to guarantee any real sense of true independence, and so it finds itself subject to the whims of a foreign state.
Right now there is a very large Canadian boycott of US products, services and tourism. I also had to explain to a client this week that because of the import tariffs on Chinese goods to the US the US assembled products are now no longer competitive with alternatives. The fact you were seemingly unaware of this kind of demonstrates the level of effect of it.
The Nintendo Wii SoC was actually fabbed in Canada and exported, but that facility has changed into something slightly different because the whole east coast/hudson river valley fab world went sideways a while ago.
> We import tons of food and energy from them and have no alternative on time scales or 10 years
More importantly for Canadians that food or energy has no alternative competing market to sell into. Consequently the Canadians are totally dependent on the US market to even set the price of it. This applies to many other sectors as well.
Canada is currently having a huge desperate push to export to non US markets because of the levels of uncertainty that have been created. And I say this as someone not totally dismissive of the US position, but they need to do a far better job of bringing their allies into the tent with them.
Russia has been totally isolated from global market and they are still producing hundreds of cruise missiles a month just fine. America could easily figure it out long enough to survive a few yrs of conflict. Even stockpiling 10yr old chips would be good enough 95% of military industry. Then emergency investment and smuggling will cover the rest.
Meanwhile cutting off your markets and wasting hundreds of billions on a long term bet with a small probability that another global war will happen is pretty dumb financial thinking.
Tariffs and corporate welfare will actively make a country poorer and create unproductive zombie markets while raising taxes on everyone. Not to mention diverting budgets and new revenue away from actual national security investment.
Up until relatively recently we were the SOTA and #1 semiconductor exporter. When people talked about the "american manufacturing sector" a significant portion of it was actually that.
Those foundries didn't go away, they're still manufacturing with the same capabilities they used to (and they're much cheaper now since they're competing with the better ones in Taiwan.)
It's good to hear TI hasn't given up on high performance SOCs as it was beginning to look like they had. But most of this stuff is still here. Freescale and many other American companies are still making the same (better even) chips they always have which is more than enough for cruise missiles (more than enough for decent PDAs and smartphones really) even without "stockpiling."
Yes and those Russian semiconductors aren't sophisticated or high end at all which supports my point. The military isn't making cellphones, they make missiles and fly jets with decade old computer chips. A few years of war with China is not going to magically eliminate all computer chips. The global market will still exist in some form and wars are far more motivating than a few government grants.
Not to mention China (and/or Taiwan) is still going to want to sell to someone to survive, and those countries can smuggle them into America - just like Russia does for it's drone industry and oil. America is much more capable in that regards with NATO and it's huge purchasing power.
I still think TI and Apple should be investing in foundries and domestically. It should just make sense as a business otherwise it's going to be a very expensive embarrassment.
The transfer of wealth from tax payers to tech giants has never been better.
Enormous state subsidies.
For doing something the free market would do if it made economic sense.
How many bailouts is Intel on now?
The idea of forcing a merger between Intel and TMSC at what could be called
be barrel of the gun, would at least make it (indirectly) Taiwan that would
have to take over paying for Intel.
I'm a bit biased because my home state is Ohio and they have it in their constitution that the state can't have a stake in any private company and can't even lend credit to any private company. And this amendment was written in blood, in the early 19th century the state nearly bankrupted itself investing in and taking stake in private companies.
* The state can't risk taxpayer money on ventures that might not pay off or lose them
money. How the state "gets around this" is by issuing zero recourse loans. The advantage is that when economic development money is handed out there's not an asset on the balance sheet. It's treated like it was spent. The value the state gets from spending the money has to be independently worth it for taxpayer without considering financial returns.
* It eliminates a whole category of conflicts of interests where the government will get squeamish regulating or punishing bad behavior because it would hurt the taxpayers' investment.
* It also eliminates vectors for corruption as well as the negative effects of the government having direct influence over specific businesses. No backdoor regulations from the state's ownership stake that don't go through the legislature.
So I'm very heavily in the camp that government shouldn't ever be allowed to have stake in any private company. The line between government and private enterprise should be the wall this admin likes to talk about. I certainly didn't expect it would be republicans I would be trying to convince that state ownership of business is a bad thing.
> So I'm very heavily in the camp that government shouldn't ever be allowed to have stake in any private company. The line between government and private enterprise should be the wall this admin likes to talk about.
completely agree, private sector should be private and public be public, and the two should never co-mingle (and that includes politicians owning stocks and bonds etc)... right now there is so much corruption (sorry, "investment") it boggles the mind
> I certainly didn't expect it would be republicans I would be trying to convince that state ownership of business is a bad thing.
because they are competing with china and they are wiping the floor with the competition, so they feel they need to do the same i assume
America is the last place that is short of capital for industry investments where it requires gov taxes going to it, they have a huge domestic financial market and tons of foreign investment (but those require legitimate plans, not national security woo woo). This is just propping up weak megacorp industry like they did with Boeing, instead of fostering real progress.
Sure.. calculators and MSP430s for remote power meters are keeping TI from closing up shop, but TI doesn't have the capitalization structure to bring up a fab for the types of chips people say they want. I mean sure... If you want to make 28nm chips, they're fine, and you can do a lot with 1 and 2 GHz parts, but... We keep saying we want to make the chips in the states that they're making in Shenzhen and Taipei... And honestly, a $1.6B grant from daddy warbucks may not be enough to prevent TI from taking the money and dropping out of the program in a few years.
And this comes from a place of love... My family's been invested in GSI for almost 100 years
> calculators and MSP430s for remote power meters are keeping TI from closing up shop
calculators have consistently been a minor percentage of TI’s business (~5% of profits per source below). I doubt MSP430s in particular amount to a huge percentage either
Depends on which line of DACs. And calculators are an almost irrelevant amount of TIs revenue. They don’t report it individually, but it’s categorized under the miscellaneous “other” bucket which is only 6% of their business and includes DLP and “other charges” related to M&A. $947 million with all those other things means you’re talking about probably 100-300 million in revenue. There’s other businesses within TI that do more revenue than that by themselves.
I’m not saying that at all. I’m saying it’s nearly meaningless in the grand scheme of their overall revenue. They make roughly 50% margin on their $17 billion in revenue right now. If somehow calculators cost nothing (100% margin) it wouldn’t make a meaningful difference on their overall margin.
Saying "TI makes money on calculators" is a pretty misleading statement outside of any other context. Its a tiny part of their profits and revenues. It's like suggesting McDonalds is an ice cream shop. Sure it's on the menu and they make a profit on it but it's a small side business after selling burgers and fries.
I’m really confused how you got that from the comment you’re replying to, and why you’re continuing to defend misinformation you’re spreading in the original comment. you implied TI primarily makes money from calculators and MSP430s. this is easily provably false
the person above made an analogy —- they didn’t claim TI loses money per calculator
Microcontrollers and calculators are a small part of TI's revenues. Most (>70%) of their revenues come from analog devices like amplifiers, DC-DC converters, ADC/DACs, and things like that.
They make important chips and many top of the line products of their segments but they're not things like server grade CPUs or GPUs.
Yes. They make most of their money from low margin parts. That's not as good a story as you might think it is. Though... making money is certainly better than not making money. And yes, they have a decent mixed signal story.
But... everyone seems to think TI will be competing with TSMC's and Samsung's small-node parts. And they probably could, but they would need to a) build a fab that can make 5 or 3nm parts and b) build a sales channel for new parts. I was alive in the 2000s so remember TI doing an exceptionally poor job of step b.
Their analog division has >50% margin, a good bit more margin than their MSP430's and graphing calculators. That's not far off from TSMC's overall margin.
It's a better story than your misleading statements acting like TI only makes calculators and old microprocessors and flat out inaccurate ones about profit margins.
I only saw their OPM broken out by division. OPM was around 37-38% in 2025Q2. Do you have numbers for NPM broken out by division? But yes, if they could get volumes like analog or mixed signal with margins like "other" or "embedded" that would be pretty awesome.
as I understand it (from outside), such calculators are part of american curriculum and is pretty much demanded on SAT and similar school-independent testing
Everyone talks about how the CHIPS, IRA, etc. purposely invested more in red states, but they don't talk about the mechanisms that made that happen.
I would be curious
Chips are like agriculture: It does not matter that the home grown product can only compete with government support, you never want to be in a position where you cannot provide for your country’s minimal needs purely from local sources.
It depends what the minimal needs are, but basically the US has this for chips already. Yes the US only makes older chips, but in fact those are the ones that really matter, that you can’t do without. The ones that are in everything.
However making everything domestically just isn’t really viable for any country any more. The scale of our technological civilization and the diversity of goods and materials it depends on is more that any one continent can support, let alone one country.
It would be possible to collapse that down to one continent if absolutely necessary, but it would be incredibly economically painful and the US would need to give up on a lot of non-essentials and other priorities to devote resources to duplicating capacity that already exists elsewhere.
Buying rocks and exporting chips is fine. Of course it's not that extreme but the sentiment is there. It's about expertise capital and concentrating that domestically.
But at the expense of what? We’re not lopping off the rock farmer end of the economy to redirect efforts toward chips.
I think it is not about one country doing to all. But technology transfer should happen within like minded countries and within a smaller circle. So may be two continents max.
This is a completely meaningless statement. It sounds good, but contains no actual argument. why are chips like agriculture? What makes that statement not apply to literally every product category that exists?
Agriculture is self-evident, we need food on a daily basis to live. I don’t need three square microprocessors a day.
Farming equipment needs microchips, as does practically any modern device, vehicle, machine, etc. If you don't have domestic microchip production then pretty much everything else you make will have foreign dependencies that could be embargoed.
The Internet is made of microprocessors. Maybe we don't need the internet to live, but it seems pretty important these days!
Of course we'd think that, we're on the Internet, discussing it. The people who aren't, aren't here to rebut that. I met this woman who lives part-time in Romania, and was saying that they don't have smartphones still, which is why she has to spend her time in Romania, because she can't just have this particular subset of elderly people in Romania get information from her off of Instagram.
Just like you don’t want to wait many months to tool up to make food if there is conflict…
You don’t want to wait many years to tool up to make chips that go in weapons, either.
Else, you risk being structurally defeated very early, and this leads to others concluding that they can take you on. It undermines other preparations.
Yes, we understand that, but it’s not a good comparison.
Producing chips and agricultural goods are fundamentally different, starting with limits like climate, seasonality, and land. At most, you could compare it to mining minerals for chip production
This is true. Also true for oil production.
>It does not matter that the home grown product can only compete with government support,
Even that isn't a given, because unless you have amassed a certain amount of technocratic and governmental competency chances are it can't compete even with government support and you just produce crony dysfunctional companies.
And of course there's economic trade offs. If you're politically ordering your economy to make chips, it doesn't make something else, and whatever it was making and trading for chips it was better at, and so you get fewer chips, that's comparative advantage. Industrial policy (and tariffs) do not increase aggregate production, they reduce it. And given that the circle of items you "can't do without" seems to be a bit of a moving target these days, at some point you're actually more brittle because you've replaced large chunks of the market with state production.
People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing. But cronyism is a policy and structural failure, it's usually because the incentives for the different parties involved encourage it to happen. Institutions can be designed carefully if policymakers actually want to do it.
USPS is a great example of an organization that's managed to largely avoid this. Whenever you mention that people crawl out of the woodwork to complain about the 7 different times they lost their package, but their logistics at scale is still unmatched by the private sector, while also not completely negating the value of private sector alternatives (which so often is argued would happen if the government actively started doing anything new).
It is established economic theory not just some opinion . Constraints in form of regulations, subsidies, tariffs by government intervention will produce inefficiencies.
This is the core economic philosophy behind both the modern centre right[2] i.e. neoliberal democrats (80s-today) and the right wing republican conservatives (pre MAGA -2016).[1]
Even communists would agree this is correct, it is understood in different levels since Adam Smith invisible hand, from far left to centre left just posit that different polices that benefit the people not the economy are worth the cost .
That is actually true for all other groups except anarchists and libertarians all agree government intervention is needed in some form, they just disagree on how .
——-
Postal services anywhere in the world are inefficient by design. They are governed by universal service obligation principles not efficiency same with telecom providers and other utilities.
The question then becomes how much inefficiency is acceptable given the objectives . There are no right answers, you could have competing private couriers who dump unprofitable low density routes to USPS while serving profitable ones themselves (UPS, fedEX, Amazon ) or you could have gigantic postal service which is sole delivery provider socialist style, there are going to be some problems either way.
It is just matter of preference which set of trade offs we are okay with .
—-
[1] Post MAGA it is a populist party economically there is no fixed ideology to be characterized
[2] the centre right label is economic and technical to contrast say the socialist left party of FDR/New Deal 1930-70s, not meant to offend.
But that's exactly the thing, where we care about an industrial capacity for its strategic reasons rather than because we're trying to maximize efficiency of revenue extraction. Pure market efficiency is great if you're trying to maximize profits, but that's not necessarily the goal here because the benefits are indirect, just as they are with universal postal service (or universal healthcare, etc).
>People take it as an assumption that cronyism will always happen if the government invests too strongly/consistently in a certain thing.
No, it's the opposite. It's a conclusion taken from empirical evidence looking at success rates in the real world. As the US has engaged on its most recent bout of industrial policy, industrial activity has declined, not gone up. Again, this is expected. If you through protectionism make chips and steel in the US inefficiently, everyone in the US using those products as inputs suffers.
There's some limited cases like developing countries engaging in catch up growth , but there's virtually no evidence for effectiveness of these policies in cutting edge technologies, which isn't surprising because they by definition tend to rely on supply chains and knowledge sourced from all across the globe, and to this protectionism is particularly disruptive. And even in developing nations most of the time practices like ISI (import substitution industrialization) fail, it's devastated Latin America in the 1950s to 1980s.
I don't know what USPS proves in this case because it's sustained by a government monopoly. Obviously you could if you wanted privatize mail delivery.
How? What if we just decide we will never go to war with Mexico or Canada and get comfortable with the idea of importing from our allies? There is no serious future risk from doing that
> if we just decide we will never go to war with...
That's hubris. Although the US does indulge elective wars, one does not always get to choose with whom one will war.
Okay so make a prediction. What is the probability that Canada or Mexico will declare war on the United States in the next 100 years?
100 years? Only a fool would attempt that. 100 years from now Uzbekistan could be fighting Brazil in orbit around Venus.
100 years ago the British Empire ruled the world, now it's a small island you don't hear much about... The US is only about 250 years old, I'd be cautious about predicting the future
You're off by about 150 years. 100 years ago was 1925, the British has already largely collapsed losing the US, Canada, Australia. 100 years isn't forever but it's a long time
And we haven't had any serious threats from Canada since 1812. I think the most reasonable estimate is 100-200 years
~1920 is the peak of the British Empire in term of territory, anyways, the details are meaningless, what matters is that things move fast and just because you're at the top of your game right now doesn't mean you'll be in the same position in 100 years
I could also take the example of world wars, in France ww1 was deemed "la der des ders", which meant "the very last war" or "the war to end all wars", well 20 years later we were at it again
Or simply look at China, you don't even have to go back 100 years in the past to see drastic changes.
That's only the case if you include Canada and Australia, which were functionally independent at that point.
I'm not claiming nobody will invade France or Taiwan in the next 100 years, I'm claiming that the US is special. We haven't been invaded since 1812 and haven't really been attacked since 1941. It's reasonable to predict we won't be invaded or go to war with our neighbors for 100 years since it hasn't happened for 213 years
This strikes me as hubris in the extreme.
My own death has not yet been a problem for me, but I can safely assume it will be.
Because everyone dies. If everyone lived to be 1000 you'd be wrong to worry about dying in 100 years
It sounds like you’re making the assumption that things will remain static because the alternative is unfathomable to even consider.
Surely based on history the odds of a conflict between neighbouring countries increases with time passed.
No, based on the history of conflict we can say that the more time that passes with neighbors not invading one another, the less likely they will in the future
Mexican warlords (which we colloquially call "cartels") are fighting a small war with the US right now and have been for many years.
What does that mean? No they aren't? When is the last time cartels attacked the American military?
Definitely non-zero. If you were Mexican or Canadian you would not take the bet on even 20 years right now so who would bet on 100?
If we include the idea that either one of them is allied to a major power at war with the US over a hundred year horizon, right now that looks pretty likely, and arguably is one of the things the current US admin are trying to stop before it becomes inevitable.
The US admin is very clearly pushing us in the opposite direction. You believe that Trump's actions make war with Canada less likely? What's the mechanistic explanation?
Didn't the US join 2 world wars it did not start (or was involved in starting) in the last ~100 years?
Not against our neighbors
So the argument is that the neighbors will never ally and be involved in a war that puts US on the opposite side? What is the argument for the sovereign neighbors to always be neutral or on US side come what may?
Yes, that's right. This has been the case for over 200 years so I think it's reasonable it will continue to be true for at least another 100
The fact that they're much smaller than the US and right there. Both would have their key cities flattened within 30 days.
> go to war
It won't be war. It'll be one-sided trade deals [1,2], and a slow erosion of economic and political sovereignty, culminating in a puppet state.
[1] https://en.wikipedia.org/wiki/Canada-China_Promotion_and_Rec...
[2] https://www.cbc.ca/news/canada/fipa-agreement-with-china-wha...
One sided trade deals in which they continue exporting to us and import less
There is plenty of risk that our neighbors stop importing and almost no risk they stop exporting
It's an extension of a peace through strength philosophy. If you lose your critical sovereign capabilities you become weak and vulnerable. You no longer get to decide who you do and don't "never go to war with".
Yes I understand the reasoning, it's just obviously wrong. This doesn't happen and hasn't happened in hundreds of years. This is not why people get invaded, otherwise Switzerland would have been invaded many times over the last century
This kind of answers your own question: the reality is you are only a reliable US ally [1] if you can hold them by the balls TSMC style. Given that countries go to great lengths to develop and maintain such dependencies. Canada's current weakness is at least in part because it has failed to do so.
[1] Edit to add: This was/is poorly worded - I mean that the US will only guarantee that you remain an ally while they are in some sense dependent on you, and while doing so they may work to break that dependence, which you may interpret as them trying to abandon you.
You are claiming that Canada will stop exporting to the US in the future? Populism does the opposite, they would likely stop importing. There's almost no risk to us that they would not export chips
My claim is Canada has no unique product that it offers that the US is sufficiently conspicuously dependent on for it to guarantee any real sense of true independence, and so it finds itself subject to the whims of a foreign state.
Right now there is a very large Canadian boycott of US products, services and tourism. I also had to explain to a client this week that because of the import tariffs on Chinese goods to the US the US assembled products are now no longer competitive with alternatives. The fact you were seemingly unaware of this kind of demonstrates the level of effect of it.
You are arguing my point. Canada may stop importing from us but will never stop exporting. There are no incentives for that and never will be.
We import tons of food and energy from them and have no alternative on time scales or 10 years
If we imported chips from Canada, that supply chain would be safe for at least 50 years, probably hundreds
The Nintendo Wii SoC was actually fabbed in Canada and exported, but that facility has changed into something slightly different because the whole east coast/hudson river valley fab world went sideways a while ago.
> We import tons of food and energy from them and have no alternative on time scales or 10 years
More importantly for Canadians that food or energy has no alternative competing market to sell into. Consequently the Canadians are totally dependent on the US market to even set the price of it. This applies to many other sectors as well.
Canada is currently having a huge desperate push to export to non US markets because of the levels of uncertainty that have been created. And I say this as someone not totally dismissive of the US position, but they need to do a far better job of bringing their allies into the tent with them.
Russia has been totally isolated from global market and they are still producing hundreds of cruise missiles a month just fine. America could easily figure it out long enough to survive a few yrs of conflict. Even stockpiling 10yr old chips would be good enough 95% of military industry. Then emergency investment and smuggling will cover the rest.
Meanwhile cutting off your markets and wasting hundreds of billions on a long term bet with a small probability that another global war will happen is pretty dumb financial thinking.
Tariffs and corporate welfare will actively make a country poorer and create unproductive zombie markets while raising taxes on everyone. Not to mention diverting budgets and new revenue away from actual national security investment.
Up until relatively recently we were the SOTA and #1 semiconductor exporter. When people talked about the "american manufacturing sector" a significant portion of it was actually that.
Those foundries didn't go away, they're still manufacturing with the same capabilities they used to (and they're much cheaper now since they're competing with the better ones in Taiwan.)
It's good to hear TI hasn't given up on high performance SOCs as it was beginning to look like they had. But most of this stuff is still here. Freescale and many other American companies are still making the same (better even) chips they always have which is more than enough for cruise missiles (more than enough for decent PDAs and smartphones really) even without "stockpiling."
Yes and those Russian semiconductors aren't sophisticated or high end at all which supports my point. The military isn't making cellphones, they make missiles and fly jets with decade old computer chips. A few years of war with China is not going to magically eliminate all computer chips. The global market will still exist in some form and wars are far more motivating than a few government grants.
Not to mention China (and/or Taiwan) is still going to want to sell to someone to survive, and those countries can smuggle them into America - just like Russia does for it's drone industry and oil. America is much more capable in that regards with NATO and it's huge purchasing power.
I still think TI and Apple should be investing in foundries and domestically. It should just make sense as a business otherwise it's going to be a very expensive embarrassment.
The transfer of wealth from tax payers to tech giants has never been better. Enormous state subsidies. For doing something the free market would do if it made economic sense.
How many bailouts is Intel on now?
The idea of forcing a merger between Intel and TMSC at what could be called be barrel of the gun, would at least make it (indirectly) Taiwan that would have to take over paying for Intel.
>For doing something the free market would do if it made economic sense.
It didn't make sense.
Bailout?
Merging Intel and TSMC?
Sorry I am not following.
Desperate measures to save Intel: US reportedly forcing TSMC to buy 49% stake in Intel to secure tariff relief for Taiwan
https://news.ycombinator.com/item?id=44801486
Is fabric here a mistranslation of "fabrica"? Correct translation would be "factory."
Probably it should be “fabrication plant”.
Chip manufacturing factories are traditionally called fabs, short for that.
Title reads very weird in American English, I'd expect "factory" instead of "fabric" (even though in Dutch, I'm bilingual, "fabriek" means factory).
The actual title of the article used the word "project" not "fabric"
Any reason to use "fabric" here?
“As for federal support, TI got $1.6 billion of CHIPS Act funding, and a whopping 35% investment tax credit from Trump’s big bill passed in July.”
So how much ownership is the US gov gonna get in this one?
TI is a going concern, no bailout necessary
Intel is not getting any money that legislation (law) had not already allocated to them. The transfer of shares to the Gov is just a shakedown.
Just? You mean they should have received the stupidly large sum of money without anything for the tax payer?
Yes? That's what grants are. The government is buying a domestic chip industry with that money.
Why shouldn’t they get some equity in return for giving money to a for profit company?
I’m not a fan, but it’s been going on for a very long time.
It’s not all that different from development tax incentives which are very common.
I think the grants had some requirements to build up domestic production.
I'm a bit biased because my home state is Ohio and they have it in their constitution that the state can't have a stake in any private company and can't even lend credit to any private company. And this amendment was written in blood, in the early 19th century the state nearly bankrupted itself investing in and taking stake in private companies.
* The state can't risk taxpayer money on ventures that might not pay off or lose them money. How the state "gets around this" is by issuing zero recourse loans. The advantage is that when economic development money is handed out there's not an asset on the balance sheet. It's treated like it was spent. The value the state gets from spending the money has to be independently worth it for taxpayer without considering financial returns.
* It eliminates a whole category of conflicts of interests where the government will get squeamish regulating or punishing bad behavior because it would hurt the taxpayers' investment.
* It also eliminates vectors for corruption as well as the negative effects of the government having direct influence over specific businesses. No backdoor regulations from the state's ownership stake that don't go through the legislature.
So I'm very heavily in the camp that government shouldn't ever be allowed to have stake in any private company. The line between government and private enterprise should be the wall this admin likes to talk about. I certainly didn't expect it would be republicans I would be trying to convince that state ownership of business is a bad thing.
They should be doing neither.
America is the last place that is short of capital for industry investments where it requires gov taxes going to it, they have a huge domestic financial market and tons of foreign investment (but those require legitimate plans, not national security woo woo). This is just propping up weak megacorp industry like they did with Boeing, instead of fostering real progress.
Sure.. calculators and MSP430s for remote power meters are keeping TI from closing up shop, but TI doesn't have the capitalization structure to bring up a fab for the types of chips people say they want. I mean sure... If you want to make 28nm chips, they're fine, and you can do a lot with 1 and 2 GHz parts, but... We keep saying we want to make the chips in the states that they're making in Shenzhen and Taipei... And honestly, a $1.6B grant from daddy warbucks may not be enough to prevent TI from taking the money and dropping out of the program in a few years.
And this comes from a place of love... My family's been invested in GSI for almost 100 years
> calculators and MSP430s for remote power meters are keeping TI from closing up shop
calculators have consistently been a minor percentage of TI’s business (~5% of profits per source below). I doubt MSP430s in particular amount to a huge percentage either
one random source: https://www.meta-calculator.com/blog/ti-graphing-calculator-... (this is pretty easy info to find)
I can assure you that TI's margins on calculators and MSP430s are much higher than their margins on DAC*s.
Depends on which line of DACs. And calculators are an almost irrelevant amount of TIs revenue. They don’t report it individually, but it’s categorized under the miscellaneous “other” bucket which is only 6% of their business and includes DLP and “other charges” related to M&A. $947 million with all those other things means you’re talking about probably 100-300 million in revenue. There’s other businesses within TI that do more revenue than that by themselves.
Saying "TI makes money on calculators" does not mean "TI does not make money on DLPs."
Also... revenue, profit and margin are all different things.
I’m not saying that at all. I’m saying it’s nearly meaningless in the grand scheme of their overall revenue. They make roughly 50% margin on their $17 billion in revenue right now. If somehow calculators cost nothing (100% margin) it wouldn’t make a meaningful difference on their overall margin.
Saying "TI makes money on calculators" is a pretty misleading statement outside of any other context. Its a tiny part of their profits and revenues. It's like suggesting McDonalds is an ice cream shop. Sure it's on the menu and they make a profit on it but it's a small side business after selling burgers and fries.
So TI is losing money on every calculator they sell? News to me.
I’m really confused how you got that from the comment you’re replying to, and why you’re continuing to defend misinformation you’re spreading in the original comment. you implied TI primarily makes money from calculators and MSP430s. this is easily provably false
the person above made an analogy —- they didn’t claim TI loses money per calculator
Microcontrollers and calculators are a small part of TI's revenues. Most (>70%) of their revenues come from analog devices like amplifiers, DC-DC converters, ADC/DACs, and things like that.
They make important chips and many top of the line products of their segments but they're not things like server grade CPUs or GPUs.
Yes. They make most of their money from low margin parts. That's not as good a story as you might think it is. Though... making money is certainly better than not making money. And yes, they have a decent mixed signal story.
But... everyone seems to think TI will be competing with TSMC's and Samsung's small-node parts. And they probably could, but they would need to a) build a fab that can make 5 or 3nm parts and b) build a sales channel for new parts. I was alive in the 2000s so remember TI doing an exceptionally poor job of step b.
Their analog division has >50% margin, a good bit more margin than their MSP430's and graphing calculators. That's not far off from TSMC's overall margin.
It's a better story than your misleading statements acting like TI only makes calculators and old microprocessors and flat out inaccurate ones about profit margins.
I only saw their OPM broken out by division. OPM was around 37-38% in 2025Q2. Do you have numbers for NPM broken out by division? But yes, if they could get volumes like analog or mixed signal with margins like "other" or "embedded" that would be pretty awesome.
What does GSI refer to? Googling did not lead to any obvious results.
The original name of Texas Instruments (Geophysical Service Inc.).
TI's older name: Geophysical Service Inc.
will this fix xkcd 768 problem, tho?
Do teenagers even use those calculators anymore?
as I understand it (from outside), such calculators are part of american curriculum and is pretty much demanded on SAT and similar school-independent testing