I think you underestimate the dependency of all modern economy on semiconductor industry.
I honestly think there is 100% chance that if China invaded Taiwan today, the US would declare war and send troops to defend Taiwan. In contrast, I think there is >0% chance that the US would not declare war on Russia if they did a (tactical) nuclear attack in Ukraine.
I wish I could remember the origin of this. But a quip on this idea I had read was along the lines of "If China invaded Taiwan and Kansas, the US would send troops to Taiwan first". Whether you want to call Taiwan an ally, a protectorate, vassal or whatever your political standpoint would dictate, the US is very protective of Taiwan.
But you underestimate our ability to downsize, turn around and build them locally on a 5-year horizon, and our ability to backtrack tensions anyway if we re reaching a nuclear point.
We will give Ukraine to Russia if we can save Paris.
> We will give Ukraine to Russia if we can save Paris.
Only that you just taught the already aggressive ruling elite of a huge country with an abundance of resources who don't care about anyone including their own except that they need them for work and for the fighting that threatening use of nukes gets them anything they want. Moldova next - it's not EU or NATO, already very low risk for Russia, if they can get there. Which was (is) a stated goal for the current war, to get the entire south of Ukraine to take away their sea ports and to get to Moldova.
They'll try the Baltic states next. Not a full invasion, just lots of little aggressive actions. Even previously they did murders in the EU, financing of radical parties out to undermine current EU country governments, supported by propaganda. I don't know how much it actually influenced US elections, but I think it's save to say they at least tried.
Giving them Ukraine will be massive. They will also have lots more of the oil and gas reserves under Ukraine and around the Krim. They will also get tens of millions of new citizens, lessening the problems of a shrinking number of people available inside Russia significantly. There also are significant parts of former USSR production in Ukraine, which will all go to Russia. They will also own even more of the prime agricultural lands of Eastern Europe, which at least so far seems to suffer less than Western Europe (look at the heat maps of this summer) under climate change so it may become even more valuable than it already is. The land is some prime real estate - unlike Siberia, Ukraine is much better, you can't look at the map and think "it does not add all that much to Russia" because the value of Ukraine lands is much higher.
I have no idea how you get this idea. Giving up Ukraine is really, really massive in its long term consequences, greatly strengthening Russia directly as well as showing them that the means they use actually work. This would be a gigantic loss for the West.
This is a good comment, but rather than tens of millions of new citizens, they would get tens of millions of new insurgents. Nearly the whole population of Ukraine is involved in the war effort in some way, and it would be impossible to break this completely. The only thing that could be given to Russia with the conquest of Ukraine is the option to turn into Afghanistan instead of North Korea.
> they would get tens of millions of new insurgents.
I doubt it. Most people will be passive and will just live their lives. They will get a few for sure, but they won't be able to do all that much. It's not like the ruling elite cares if there's an occasional killing, after all, they already use that method themselves, see the list of Russian businessmen and manager deaths.
Not everyone would be involved in directly fighting, but there are intelligence networks, supply networks, opportunities for discreet sabotage and falsifying critical data, and many other ways that people can support a resistance movement that would continue even in a fully occupied Ukraine.
In 1922 Russians have won the war against Ukrainian People's Republic.
After 20 years, vast majority of Ukrainians who opposed the occupation were dead, scared to death, refugees in other countries, or in forced labor camps in Siberia and other remote places of USSR.
This is a terrible truth, but the situation has changed in this war because everyone is contributing in organized volunteer centers. The amount of coordination is orders of magnitude greater than was possible a hundred years ago. I don’t think such a large society has ever been so fully engaged in a single purpose in history. Because the strategy is so new, it would be a mistake to assume that old tactics will work against it in the same way they have in the past. Nothing is guaranteed, but I wouldn’t bet against the Ukrainian people even in the worst circumstances.
I think there is little chance Russia could successfully occupy and hold Ukraine. Look at the US's utter failure to do the same in Afghanistan, despite vastly more resources and some significant popular support.
No you massively overestimate the importance of economics or semiconductor industry. Whatever will happen to Taiwan has nothing to do with Chips.
For China it's about nationalism, for US it's about protecting allies/upholding treaties and protecting democracy from the strongest authoritarian regime. Chips are not important. After all chances are high they might be destroyed even in a successful defense of Taiwan.
As a Ukranian American I wish we had and were doing more for Ukraine but it's not about chips or economics. Ukraine had only recently grown closer to the US. The US has promised to defend Taiwan for a long time (well sort of, arguably the US does keep some strategic ambiguity about this which might let it wiggle out)
I don't know where you got the idea that the US will not go to war to protect the US economy, national security, or military capacity, but you are severely misunderstanding the situation. That is why the military exists. We spend the amount we do on the military to support American economic dominance and define the rules on how world trade happens. The reason we are not doing more for Ukraine is because they are not very important to the US, outside of being a buffer against Russia.
Ideas like "protecting democracy" are used to sell citizens on wars they don't care about. The full destruction of TSMC is likely preferred over a Chinese dictated world technological economy. The truth is, if one side has TSMC chips and the other doesn't, what we're talking about may necessitate a total war.
You're overestimating the value of semiconductor industry in this hypothetical.
TSMC chips aren't critical, considering that ASML can also deliver the same equipment to US... and cut off China from their equipment. (or sabotage in critical cases)
There's a myriad reasons, why US would probably would send in military to protect Taiwan. But it's not going to be "just because TSMC"
The US is really not _that_ interested in protecting democracy from authoritarian regimes. If we were, we’d have boots on the ground in many African states.
While upholding treaties is vitally important, I think you’re underestimating the importance of chips(a rare occurrence on HN!).
Wars are generally fought over resources rather than ideas, and pretending that US is defending Taiwan to defend democracy instead of defending its strategic interests (access to vital resources — chips) is misguided.
> For China it's about nationalism, for US it's about protecting allies/upholding treaties and protecting democracy from the strongest authoritarian regime.
I doubt this is true for China (I very much suspect economic concerns trump any other concerns for them as well), but I am quite convinced you are wrong about the USA - one of the biggest supporters of non-democratic regimes in the world. There is little in US history to suggest they have any preference for a democratic regime over a subservient autocratic one. They are also extremely clearly uncaring of international treaties.
And make no mistake: the USA is coordinating Ukraine's defense because it sees it as a good chance to weaken Russia, not out of some deep care for the people of Ukraine.
It is not just about semiconductors. It is about pride. Taiwan shows another way for China that is not the PRC just as Hong Kong did. Taiwan is not just a separatist state, it is a successful separatist state. The better Taiwan does the worse the PRC looks. Semiconductors are just another gut punch. Why can't mainland China do what tiny Taiwan has done?
If Taiwan is taken off the board for semiconductors, or if all semiconductors have to go through China, that means the entire US military (and all western militaries) are dependent on a geopolitical rival. To allow that would be nothing short of giving away the game.
This is what I don't get. I thought military and aerospace used decades old cpu designs on decades old fab technology for the radiation hardness? Not saying they couldn't benefit from an upgrade, but it's not like Lockheed is putting Nvidia GPUs in fighter jets, right? It doesn't seem like a deal breaker to use a 14nm node compared to a '5nm' node (or whatever is the latest TSMC process). Seems like a weird line to draw in the sand to me. Frankly, seems like the only applications which are make or break on EUV lithography are all gaming related. Am I that off base here? Certainly prices would rise, and critical supply chains would have to be remade, and there would be no more iPhones, but seems like we'd get by just fine for a few years before catching back up and then likely surpassing Taiwan. And it's not like the real brains behind TSMC would willingly help a Chinese controlled takeover, and add on a new layer of corruption and bureaucracy, and in a few years TSMC is irrelevant anyways. And don't forget, TSMC is reliant on ASML, who certainly wouldn't be shipping any more EUV lithography equipment to a Chinese controlled Taiwan.
I think that's true for some key components of the military, but for example, anything involving AI systems necessitate cutting edge chips. You can probably look to the effect chip shortages had on car manufacturers to see how much of this likely works. Cars manufacturing wasn't entirely halted in most cases, instead they had downgraded functionality. Less automatic windows, more window cranks. The totality of every military system isn't based off high end chips, but some of it certainly is.
There are advanced Fabs in Israel, Intel and TSMC are both building next gen Fabs in Arizona. It's not like Twain is the only place with advanced chip fabs(though yes the vast majority of the capacity is there and IIRC thats where their latest process nodes are but I don't think the military is dependent on those absolute cutting edge process nodes)
Taiwan is also the center of a lot of specialist equipment (not the big swiss-watch ASML machines, but stuff like wafer transports, cleanroom gear, etc.) and consumables (chemicals, bunny suits, etc.). I don't know how extensive and fragile that ecosystem is but I live in eastern Michigan and it seems like every small to midsize town has some tiny automotive supplier that is somehow still in business even while the big plants have moved on. I suspect this is largely because they have no competitors and it is specialist work that doesn't really scale (meaning there is only so much of this work to go around, even if global auto sales 10x) so no one is really motivated to compete either. The result is that the GM and Ford plants leave the country, but still rely on a relatively small set of expertise and tooling that only exist in the rust belt. Presumably, a thorough nuking of the Midwest would (in addition to lots of other unpleasant effects, like the death and famine millions, perhaps billions) at least require the pause of the majority of auto manufacturing around the globe. Assuming people still wanted cars after such an event, it could be recovered, but it would take time.
I got a little carried away with the Michigan analogy, but IMO it doesn't go far enough: Taiwan is far more integral to the global semiconductor industry than Detroit is to the global auto industry.
I agree that this isn't an indefinite problem, but the current state of affairs is that TSMC is irreplaceable on the world's fabrication scene. The military definitely has some need for cutting edge process nodes, though I can't say how much. This is intentionally vague, but it seems safe to say there are a number of high end missile guidance systems, a variety of AI implementations, and drone systems that likely all have some dependence on high end chips. Not to mention the economic reliance on TSMC, Apple alone is ~7% of the S&P500.
I think if we fast forward 5-10 years, Taiwan will not be this much of an absolute, but it will take years for these new fabs to come online. Until there are viable alternatives, Taiwan is a massive risk.
Semiconductors alone aren't even remotely enough. We have enough supply to wait out and build out other semiconductor fabs, to go to WW3 just over TSMC.
