TSMC to build second Japan chip factory | Hacker News Reader

TSMC to build second Japan chip factory

finance.yahoo.com

634 pointsytch466 comments

I think a while back TSMC finally understood that building a factory in the US is just not feasible, so their backup is to just transition to Japan long term if Taiwan's situation doesn't pan out. During the Pandemic for example, when Japan noticed that their supply chain is too dependent on China, and that during an emergency they too are subject to export controls, even for their own factories, they immediately acted to bring manufacturing of giants such as Iris Ohyama back to Japan. Contrast that to the US and Europe who keep talking about these things, but don't actually execute(although the US at least tries to throw money at the problem).

TSMC lost the Chinese market, because their government went along headfirst with US trade war policy(similar to what Japan did in 1986, but worse in fact). South Korean officials on the other hand lobbied heavily to get long term exemptions, which allowed them to turn around their profit situation.

The thing is, TSMC floundered in America because they had to compensate American workers.

American workers aren't cheap.

These brutal Asian work cultures can only exist in Asia.

Americans and Europeans compete through ingenuity and intellect economically, not through a meat grinder of hard work. It's much different.

America's plan to revitalize their industry cannot be contingent on the graces of foreign countries.

China is grinding out their own domestic chip manufacturing even though it's far behind the technological sophistication of Taiwan. In the long run, China's strategy is vastly superior.

It's a bit embarrassing.

By the stats, USA is the brutal one: https://data.oecd.org/chart/7kW3.

While that does have Japan, the link lacks China and many other countries. Check out this list from Wikipedia[0]

[0] https://en.wikipedia.org/wiki/List_of_countries_by_average_a...

Could be a case where US workers work those hours officially, but Japan workers work a lot on unpaid overtime in addition to those hours.

Per source: “ The data are intended for comparisons of trends over time; they are unsuitable for comparisons of the level of average annual hours of work for a given year, because of differences in their sources and method of calculation.” https://data.oecd.org/emp/hours-worked.htm

From my experience working at TSMC, although the number of work hours are higher it definitely is not brain dead the way you are sketching out to be. They are solving really hard problems with really short deadlines. I also don't see how a factory can work any other way. Every second the machine is down you lose millions, literally.

There have been former workers who claimed that there were unsafe working conditions there, and that they were called lazy by higher-ups:

https://www.theguardian.com/business/2023/aug/28/phoenix-mic...

I don't see how this connects to what I said.

My understanding is that the US factory is underway and going to start producing chips next year and that they have plans to construct a second factory already. I’ve been seeing this negative narrative a lot around the US factory, but I’m curious if there’s any evidence that they’ve actually stopped progress. I feel like that would be a huge political loss for Biden and the CHIPS act at this point.

TSMC fab construction site in the US is a popular target for drone video and such videos are regularly uploaded to YouTube. I watch them, and there is no evidence of stopping.

My understanding is that TSMC is happy with construction, and their main worry is CHIPS act. No CHIPS act fund is actually distributed to advanced fabs yet because US government has so many conditions.

It’s going AFAIK, but had a bunch of mishaps and talent/labour shortage, so it’s going slower. I’m curious how big its output is going to be, compared to the one that’s supposed to start producing chips in 2024 in Japan.

Just a bit disheartening timeline wise. It took Japanese ~3 years (2021-2024) from the announcement to production, versus ~5+ (2020-2025 TBD) for North American factory. I hope we figure out the logistics and have it easier for the second factory though!

There would be an easy fix: pressure Tim Cook to increase the US-manufactured content of iPhones. Trump actually did a version of this with some success; it was the version of the Mac Pro nobody wanted, because Trump has the reverse Midas touch, but previous presidents could have done more here.

Have the Intel plans concerning Germany changed?

https://www.intel.com/content/www/us/en/corporate-responsibi...

"This is the first phase of Intel’s plans to invest as much as 80 billion euros in the European Union over the next decade along the entire semiconductor value chain—from R&D to manufacturing and advanced packaging."

Germany earmarked $22B for chipmakers support.

Though experts think the EU needs $500B.

https://www.electronicsweekly.com/news/business/eu-chip-goal...

| TSMC lost the Chinese market, because their government went along headfirst with US trade war policy(similar to what Japan did in 1986, but worse in fact).

Taiwan is definitely caught between a rock and a hard place. They watched what happened in Hong Kong and no longer is interested in rejoining Xi’s China. So now they are threatened with invasion unless they rejoin China. They are reliant on the US to prevent that’s from happening. So given the choices, it’s not surprising that they are choosing the US to continue existing.

It’s crazy to me that you haven’t see more divestments in China from Taiwanese companies.

China and Taiwan use the same language so China is still very attractive for oversea expansion for Taiwanese companies.

Ukraine and Russia also mostly speak the same language. Doesn’t mean much if the great leader decides to make a move, actually it is another reason to make a move.

About 1/3 of Ukrainians know Russian.

Ukrainian and Russian are mutually intelligible languages.

https://en.wikipedia.org/wiki/Mutual_intelligibility#Slavic

Whereas different dialects of Mandarin may not be mutually intelligible

> Many varieties of Mandarin, such as those of the Southwest (including Sichuanese) and the Lower Yangtze, are not mutually intelligible with the standard language

https://en.wikipedia.org/wiki/Mandarin_Chinese

The Mandarin used in Taiwan and the Mandarin used in China are both standard Chinese and mutually intelligible.

IIRC Cantonese and Mandarin are the two big language groups that are not mutually intelligible.

But Cantonese isn’t spoken in Taiwan except a handful of Hong Kong immigrants. What’s your point?

But they do speak Taiwanese in Taiwan which you wouldn’t understand if you only spoke mandarin.

> they do speak Taiwanese

Do you mean Hakka and/or Hokkien?

Hokkien which is primarily spoken across the strait in Fujian.

Hakka is primarily spoken in Guangdong, Fujian, and Jiangxi.

Xi himself was the Party head of Fujian for most of his career and Xi's father was the Party head of Guangdong when he was rehabilitated in the Deng Xiaoping era.

This is why most manufacturing in China ended up coastal Southern China - most Chinese Taiwanese trace their heritage to there barely 2-5 generations ago at most.

The younger generation (post-1989) in Taiwan speaks and understands Mandarin.

This whole thread has a bunch of answers which are confusing the topic.

The issue is why would Taiwanese businesses care about the China market? Aside from the fact that the China market is massive, there is a simple answer: Taiwan and China have the same business language, and that is Standard Chinese aka Mandarin.

Yes, lots of Taiwanese people also speak other Sinitic languages that are not Mandarin, and are not mutually intelligible with it. And lots of Chinese people also speak other Sinitic languages that are not Mandarin and are not mutually intelligible with it. And even some variants of Mandarin itself are not mutually intelligible. But - outside of Cantonese in HK and Macau - none of those languages are used as the primary business language anywhere in either China or Taiwan, so it's an interesting side note but doesn't change the point.

All that said, aside from the Chinese market being massive, and the common language being convenient, there is a much bigger elephant in the room that explains why Taiwanese companies might not have a fun time doing business in China: politics.

It doesn't matter how much money Taiwanese companies might want to make if the CCP can threaten to turn off the spigot any time they want to influence Taiwanese politics, which unfortunately nowadays appears to be all the time. Sure, it's leaving a lot of money on the table, but doing business with Japan or the US or other countries that aren't run as a single party dictatorship whose leadership has a stated platform of dismantling your own government might be a less risky option.

> none of those languages are used as the primary business language anywhere in either China or Taiwan

我同意。

I was just trying to dig into what OP meant by "Taiwanese" as a language.

It's always going to be Mandarin for anything commercial.

That said, you can't deny the benefit the Hakka and Hokkien diaspora provided to China's manufacturing capacity - it was diaspora Chinese from Thailand (CP Group was the first foreign private company to incorporate in China), Malaysia, Singapore, Indonesia, and Taiwan had on PRC's catchup.

In Taiwan it’s called Taiwanese. It’s similar to Hokkien which is why it’s often referred to as Taiwanese Hokkien. But it’s not 1:1. And people outside of Taipei will assume you speak it and understand it.

> Xi himself was the Party head of Fujian for most of his career and Xi's father was the Party head of Guangdong when he was rehabilitated in the Deng Xiaoping era.

This is because the CCP does not trust southerners (people from Fujian/guangdong especially) and will send northerners in to manage/rule those provinces. China is still afraid of Fujian revolting and joining Taiwan.

Yea no.

If that was the case, then Hakka wouldn't be so overrepresented within the CCP.

Deng Xiaoping himself was Hakka, as were most of the the core CCP leadership until recently [0]

Hakka is the primary language spoken in TW after Mandarin btw.

If you're a fellow Bay Area native, maybe visit Fremont or Richmond District in SF sometime and actually learn about us Asian subcommunities.

[0] - https://www.jstor.org/stable/654189

Deng Xiaoping was born/raised in sichuan, only his ancestors were kejia. I’m sure if you look at anyone in china’s family tree long enough, you’ll find a lot of mixing, which is about as weird as a German having ancestry from Italy or France. It’s even weirder, as the paternal side was originally from Sichuan moving to GD during Ming and back to Sichuan during Qing.

> Deng Xiaoping himself was Hakka, as were most of the the core CCP leadership until recently [0]

I’m going to not believe your source given how much wrong you’ve gotten already. But even if true, Chinese identity politics are more attuned to region than ethnicity. Southerners also were trusted because the KMT was heavily southern biased while the communists were the opposite. This is why the capital is in BJ and not NJ.

> Hakka is the primary language spoken in TW after Mandarin btw.

No no no. It is definitely Hokkien, a dialect of Min, which is the base dialect for Fujian.

> If you're a fellow Bay Area native, maybe visit Fremont or Richmond District in SF sometime and actually learn about us Asian subcommunities

Is that some kind of Asian American thing? You can learn plenty about Chinese ethnicities in China, why bother doing it in the states? Southern ethnicities are also over represented in overseas Chinese communities given the propensity of people from GD or FJ to go abroad, and then cultures kind of diverge a bit (another reason the CCP doesn’t trust GD/FJ).

> Deng Xiaoping was born/raised in sichuan

Guang'an, the town he was born, was and is Hakka speaking, after the Hakka-Punti wars were lost by the Hakka.

> Chinese identity politics are more attuned to region than ethnicity

I can agree with that with the post 1970 generation. Deng Xiaoping and his ilk were from before that era, during the Long March.

> It is definitely Hokkien

Yes, Hokkien is prominent in TW but so is Hakka. Around 10% Hakka speaking based on the last census.

> Is that some kind of Asian American thing

Taiwanese (and HK) American specifically. Iykyk I guess. Hence why I equally smell shit from your response.

My perspective is primarily mainland where I lived for 9 years, and primarily interacting with mainlander coworkers (eg from Fujian, talking politics was ok).

you're correct. My taiwan colleagues visit HK, they request "no cantonese please"

Outside of Taipei, a lot of people speak Taiwanese. (While they also speak mandarin if you don’t know Taiwanese you can only understand a bit of what people say)

Nearly every Ukrainian understands Russian, but many Russians would only understand the gist of what Ukrainians are saying, because the languages only share about 60% of their vocabulary.

A lot of common, everyday words differ in Ukrainian and or arise from different roots (e.g Polish).

Oh no, only 60%? Surely that's plenty for a conversation, no?

Now I wonder if this number, provided it is a real one, went up or down during the last 30 years. I would bet on lower but it's only a gut feeling.

> Oh no, only 60%? Surely that's plenty for a conversation, no?

It's plenty for communication, not plenty for a conversation.

(It's also a mirror for colonialism, by the way, where the occupied speak the language of the occupier, but the occupiers can't be arsed to learn the language of the occupied.)

It's not even necessarily enough for communication. With the Pareto curve on word commonality it's really quick to get high percentages of vocabulary. But it's the words you don't know on a sentence that are usually the important ones.

You are right. 60% is nothing, especially given the political situation

can't even talk about chips and language without being attacked by the crew

• https://news.ycombinator.com/item?id=39295035

• https://news.ycombinator.com/item?id=39294267

• https://german.stackexchange.com/a/53783

It's a linguistic topic. Good luck trying to talk with Germans as an English speaker.

Ukrainian shares 84% of vocabulary with Belarusian, 70% with Polish, 66% with Slovak.

English and German share 60% of vocabulary.

Except English, German, and Dutch are not mutually intelligible.

Although as an English-speaking native who has studied German, Dutch often maddeningly looks like it should make sense, but it doesn't.

> Whereas different dialects of Mandarin may not be mutually intelligible

Except for some slang words which nobody would use in business anyway, Sichuanese is largely intelligible to native Mandarin Chinese speakers if spoken slowly and repeated a couple times. People from Sichuan can also speak standard Mandarin. The written language is identical.

As for Taiwan and China it is even less an issue. The very few words that are different may be the source for some humor sometimes but that's it. It's kind of like how British people say "lift" and Americans say "elevator". If you're not brain-dead you'll figure it out pretty quickly and maybe crack a joke or two about it. When you see a sign that says "lift" you don't panic and say that it's not intelligible, you can make some sense of the word.

It's a non-issue in practical terms.

I was born in Kiev and spoke Russian at home. Can barely understand Ukrainian unless it's spoken slowly by a native Russian speaker. I can get the gist of what Zelensky is saying in an interview but can pretty much never understand native Ukrainian speakers. I think there's also a gradient of dialects and accents West to East, so I'm sure you can find some Ukrainian villager I would understand better but in general they're not mutually intelligible (to me).

Wow. Did you attend Russian language-only schools when growing up? Is this common in your generation?

Moved to the US as a kid (by way of Estonia and then Italy), went to a year of Russian language "transitional" school here. So don't have any insight on the Ukrainian school experience.

How close is mandarin to Cantonese?

About as close as English and Swedish.

I have warm memories as a kid going with my Mom to the daily market, and watching people communicating by furiously writing words in their hands, in addition to the simplified tradespeak between the language groups. It's an interesting thing, having both a shared writing system and mutually unintelligible spoken language!

Maybe it should be more recognized that what the quote "a language is a dialect with an army" means is that borders of nations don't coincide with borders for languages, or put more simply, it has such two meanings that, there are languages that are realistically just weird accents on one another, and one "language" that are realistically two or more.

I have some confidence with dialects of my primary language(not Chinese) within ~150mi of where I am; beyond that, mutual intelligibility with local dialects isn't guaranteed. Yet, those dialects are rarely considered(including by speakers) to be separate from the standard. They're just local accents. That aren't even intelligible to city dwellers.

To me, cantonese may as well be a separate language from Beijing Mandarin.

I mean it is?

"Cantonese (traditional Chinese: 廣東話; simplified Chinese: 广东话; Jyutping: Gwong2 dung1 waa2; Cantonese Yale: Gwóngdùng wá) is a language within the Chinese (Sinitic) branch of the Sino-Tibetan languages"

https://en.wikipedia.org/wiki/Cantonese

I simply left room for error.

Around 1/3 are native speakers but the number of know Russian is significantly higher as it’s the formal business language the way English is in a lot of the world. Hard to find exact numbers, but according to Wikipedia a 2008 gallop poll had 80% of Ukrainians claim to prefer Russian as the language of business.

Ahh TIL! Thanks.

Well, this is pretty much dated. You would be surprised to learn how the tables started to turn starting from 2014 and finished turning today.

Is this an anecdote of your experience or?

11 years does not seem like long enough time for a language to first start declining and then end as being the primary formal language.

It’s not “declining” but rather being actively replaced and rejected by the population. When your nation suffers brutal aggression perpetrated by the neighbor - it makes it no longer fashionable to speak the language of the aggressor. The fact that Russia also denies that Ukraine and Ukrainians are even a real nation and culture distinct from Russia fuels the sentiment too.

Organically, no, 11 years is not long enough.

But you may recall that in 2014, a few political directives regarding culture and language use have been made by the Rada, and then a few political decisions were made in the Kremlin, and then everything turned to shit (To put it simply).

It's easy to do a lot in 11 years when you start banning foreign-language media, stop using a language for government services, stop teaching it, etc, etc.

Still, that 80% hasn’t died off so those people still know the language

February 2014 Moscow occupied Crimea. 12.04.2014 Moscow occupied Slovyansk.

Name "few political directives" in 2014 before Moscow invasion. Ukraine actions are direct response to Moscow aggression. People don't want to be occupied by Moscow like Donetsk, Luhansk. Life is awful there, million fled from occupation. That's why changes were supported by majority of Ukrainians.

Still occupants language was learned in schools, media could use it though eventually quotas set to use Ukrainian too. And officials continued using it.

Ukraine policies fought discrimination of Ukrainian in Ukraine. Discrimination that stems from centuries of occupation by Moscow. In 2016 state stated at least 60% TV should be on Ukrainian. Only in 2017 education in schools was switched from occupants language to Ukrainian. Since 2019 Ukrainian should be used in services unless requested otherwise by customer. People switch to Ukrainian voluntarily, state provides means.

Ukraine is a democratic state, check out Euromaidan. Stop pretending like changes is anything but result of Moscow agression.

There was a bit of a coup against one of the branches of government on Feb 2014, it's odd that it's missing from your timeline, given that it kind of precipitated everything else that followed.

But perhaps that's how you think democracies work - when you don't like the government, you bring your friends to wear funny hats and storm the capitol, and get a new one... Should Americans do that the next time an unpopular politician ends up heading the executive? It certainly speeds up the transfer of power, even if it drops the 'peaceful' aspect of it...

Euromaidan was response to violent dispersal of protesters. Government escalated, eventually killed hundred of citizens. Do you claim Americans would do nothing if killed in hundreds? No persecution, approved by "unpopular politician", passed laws on dictatorship (16.01.2014).

Moscow invaded Ukraine (Crimea) 20.02.2014. Yanukovych fled 21.02.2014. Occupation does not just "happen", it was staged. Ukrainians felt that as betrayal, seen as occupants population cheared in support. That hurts, breaks cultural ties. In a few months Moscow invaded east of Ukraine while spreading lies. Lies obvious for Ukraine citizens, believed by occupants population.

> Do you claim Americans would do nothing if killed in hundreds?

They'd blame the people who died. At least, that's how Kent State went down (And the students there weren't even trying to overthrow the government).

There's a process for peaceful transfer of power. Some countries have good processes for this, some have bad ones, some are in between. As far as I'm aware, though, no country has a process of 'Enough people storm the capitol' for determining when that happens.

When you don't follow the permitted process, this compromises a democracy's legitimacy. Now, obviously the coup was only carried out against the executive, not the legislature, so the resulting government was partially legitimate - at least, the legislature remained representative of the public (And the issue was resolved in the subsequent election).

But that aside, just because the coup only finished on the 21st, and the invasion happened on the 20th, doesn't mean that the weeks of the revolution leading up to it weren't intimately related to the start of the war.

Laws on dictatorship passed 16.01.2014, copied from Belarus, Russian Federation https://en.wikipedia.org/wiki/Anti-protest_laws_in_Ukraine

Peaceful transfer of power is not possible in Belarus or Russian Federation. Ukrainians have no guns, democracy is not stable, judiciary and special forces are not independent, media influenced by state and oligarchs. Euromaidan saved Ukraine from Belarus fate.

Moscow invasion staged not in preceding weeks but in years. Putin revealed intentions in 2007, occupied Georgia in 2008.

"Name "few political directives" in 2014 before Moscow invasion."

"On February 23, 2014, the second day after the flight of Viktor Yanukovich, while in a parliamentary session, a deputy from the Batkivshchyna party, Vyacheslav Kyrylenko, moved to include in the agenda a bill to repeal the 2012 law "On the principles of the state language policy". <...> The bill would have made Ukrainian the sole state language at all levels. <...> However, the move to repeal the 2012 law "On the principles of the state language policy" provoked negative reactions in Crimea and in some regions of Southern and Eastern Ukraine. It became one of the topics of the protests against the new government approved by the parliament after the flight of Viktor Yanukovich." [0]

And more generally: Ukrainization Post-1991: Independent Ukraine. [1]

"check out Euromaidan."

“The mob, whatever it is, has no legitimacy before the sovereign people expressing themselves through its elected representatives.” [2]

[0] https://en.wikipedia.org/wiki/Language_policy_in_Ukraine#Att...

[1] https://en.wikipedia.org/wiki/Ukrainization#Post-1991:_indep...

[2] https://www.nytimes.com/2023/03/22/world/europe/macron-pensi...

Something about having your country invaded and missiles fired at your cities tends to change perceptions of the culture initiating said aggression.

But they do not make you forget a language is the point.

I thought we were talking about the 80% who "preferred" it as a language of business. Surely that... has dropped like a stone.

That and the way people respond to polls now is gonna change.

In any case I think going forward you'll see English's fortunes rise in Ukraine, and probably Polish as well.

The parent comment of this thread by @Waterluvian is whether they understand the Russian language.

not anecdotal, I speak both languages. The trend now is to reject everything russian even though you do understand it, no way around this. And yes, the “kitchen language” for many ukrainians, especially east part, remained russian. However, on public or outside ppl try their best to speak Ukrainian. The younger generation will be more like the one in the baltic counties or Georgia. Understand russian but rather speak their native language.

anecdotal means your personal experience.

Based on the recent (September 2023) social study:

Which is your native language? 2012: 57% Ukrainian, 42% Russian; 2023: 82% Ukrainian, 16% Russian

Which language do you speak at home? 2023: 63% Ukrainian, 27% mixed, 7% Russian

Coincides with my andecotal experience. Source: https://ratinggroup.ua/en/research/ukraine/soc_olog_chne_dos...

I’m sure it is, but terrible relations doesn’t make people forget a language and we’re only speaking to the number who know the language. I’ve no doubt a generation from now that number will be a lot lower if things continue on this path.

In reality it’s certainly over 90% (probably closer to 99%)

The only Ukrainians I've that that didn't speak Russian are 2nd gen immigrants outside of Ukraine. Often hilarious because their parents often barely speak Ukrainian themselves.

"83% of Ukrainians responding to a 2008 Gallup poll preferred to use Russian instead of Ukrainian to take the survey." [0]

[0] https://web.archive.org/web/20100518073110/http://www.gallup...

Most of their construction material comes from China and most of Taiwanese exports go to China. Taiwan will never compete on the global market for most of its products. This current policy is economic suicide, and anyone not blinded by ideology knows it.

Not a single US president pushed for a FTA with Taiwan. So far every single one has opposed it, while on the back bullying Taiwan to open their markets to dump cheap American pork into it, completely against the will of the Taiwanese people by the way.

The current result is that Taiwanese wages have stagnated for decades and as a result save for a select few that go to Japan and the US, a lot of other people look for opportunities in Mainland China.

What made Taiwan so successful in the past is what makes Dubai and Singapore so successful now. Open trade with everyone, and easy business opportunities. Both things that it no longer engages in.

BTW the big joke nobody really seems to know is that the pro independence party has never submitted a bid for independence, while the pro China party has.

Why people keep believing that "open trade" solves everything and makes "easy business opportunities"?

You can have open trade but also subsidy a portion of your economy to make that "open trade" not so open anymore. I'm not against it, just saying that a single specific decision makes Dubai and Singapore success is a reductionism that helps in nothing the discussion. Let take OPEC in matters, would you say they follow a "free trade" ideology?

Many wrong things with comments. Your pro-china KMT bias is showing

> Taiwan will never compete on the global market for most of its products.

the aggregate brand value of Taiwan's 25 largest brands totaled US$13.84 billion in 2023, a 5 percent rise from 2022 and marking the fourth consecutive year of the value surpassing the US$10 billion mark. https://focustaiwan.tw/business/202312020012

> The current result is that Taiwanese wages have stagnated for decades

The average annual salary for full-time employees in Taiwan reached an eight-year high of NT$694,000 (US$22,242) this year (2023). https://focustaiwan.tw/business/202311290017

Taiwan to surpass Japan in GDP per capita this year (2023): JCER https://asia.nikkei.com/Economy/Taiwan-to-surpass-Japan-in-G...

> The average annual salary for full-time employees in Taiwan reached an eight-year high of NT$694,000 (US$22,242) this year (2023).

I am the complete opposite of a CCP shill, but that is by definition stagnation.

The 2015-16 recession was brutal in Taiwan [0]. It basically was a lost decade.

South Korea and Japan ate Taiwan's market in the upper bracket, and China ate Taiwan's market in the lower bracket.

[0] - https://www.cnbc.com/2015/10/30/taiwan-gdp-falls-for-first-t...

Not really seeing a "stagnation" using a graph, in the last 6 years

https://www.ceicdata.com/en/indicator/taiwan/annual-househol...

13k in 2017, 17k in 2022, 30% increase. and 22k in 2023, 30% increase.

the stagnation from 2011 to 2017 is due to the pro-China president from 2008 to 2016, which fueled market and investment losses going to China. That reversed in 2016 with the Pro-Taiwan president.

When you actually show the rate of change from e.g. 2014-2022, it's about 3.5%. Inflation being about let's say 2% over those years. Although the fact that it's reported in USD probably matters as well to really understand the economy. Anyways, the average salary in Taiwan across the entire workforce was just under NT$41,000 per month (median being surprisingly close to that figure), which is comparable to many cities in China. Also you can't compare 2008-2016, the US caused GFC caused a lot of issues. 2015-2022 household income also corresponds with how China was doing in terms of exports, since China was rapidly expanding exports around that time showing probably there's a pretty strong tie/correlation between the two regions.

The fact that Taiwan GDP per capita is close to surpassing Japan's, shows how poorly Japan has been doing despite its stock market.

I tend not to trust CEIC - they have had issues converting data over time periods, which is critical for the NTW as it has been extremely volatile over the past decade.

For now, let's use Monthly Household Income sourced from the Taiwanese DGBAS and in NTW [0]

In 2016 it was NTW 84000 but by 2022 it was NTW 99000, which isn't a significant change, especially factoring the craziness the NTW has had since 2016.

> stagnation from 2011 to 2017 is due to the pro-China president ... That reversed in 2016 with the Pro-Taiwan president

It wasn't a DPP vs KMT issue. China had a stock market crash in 2015-16 [1], and Taiwanese companies were heavily exposed, as China is Taiwan's largest trading partner by a longshot. On top of that Taiwanese companies were already facing the brunt of the collapse of the CSSTA [2].

[0] - https://tradingeconomics.com/taiwan/disposable-personal-inco...

[1] - https://en.wikipedia.org/wiki/2015%E2%80%932016_Chinese_stoc...

[2] - https://en.wikipedia.org/wiki/Cross-Strait_Service_Trade_Agr...

Taiwanese companies run most of the major hightech manufacturing in China. For example, Foxconn is a taiwanese company. Most of my dealings with these types of manufactures has been in discussions of how to circumvent tariffs, not in really exiting manufacturing in the mainland.

When I look at what occurring, I think it has more to do with their cheapest labor force is in China. So the scale and profit is hard to leave.

The Chinese labor force has been reaching parity recently. Hence pushes into Vietnam & India for manufacturing.

> Hence pushes into Vietnam & India for manufacturing.

Doesn't help Taiwanese companies long term.

They had an advantage in Mainland China being Chinese speaking. Already in India, Tata is becoming the Indian version of Foxconn for most manufacturers and VinGroup or Korean Chaebols like Lotte (SK has a FTA with Vietnam) the Vietnamese version of Foxconn in Vietnam.

While South Korean and Japanese companies were actively de-risking in China by returning to ASEAN+India in the early-mid 2010s, Taiwanese companies only began doing this in the late 2010s after the 2015-16 recession.

It’s more than that: most of the Taiwanese companies are owned/controlled by mainlanders who still have a lot more economic power than native Taiwanese, even if the latter has more political power these days.

The Republic of China (aka Taiwan) had never any interests in effectively and finally surrendering to the communists on the mainland.

The "threat of invasion" has been the case since 1949. The communists stopped and did not 'finish the job' because it was hard, because of a deal with the Nationalists, because they got distracted in Korea, whatever other plausible reasons. This has been the situation since.

Interestingly this does not mean that the people, especially pro-KMT, are necessarily 'pro-US'. Those people are rooting for China but not for the communists and see the US as a necessary 'evil', so to speak.

There is an enormous amount of skepticism towards the US TSMC deal in Taiwan, in the sense that Tsai Ing-wen “sold out” Taiwanese IP and top engineers due to political pressure.

And in fairness, the US does not have a strong track record wrt its overseas military shenanigans actually helping locals, to put it lightly. A lot of people in Taiwan are anti-CCP, but at the same time, not pro-US because they see the US as an untrustworthy or at least unreliable military ally.

> And in fairness, the US does not have a strong track record wrt its overseas military shenanigans actually helping locals, to put it lightly.

Furthermore the looming US presidential election is making people nervous even in countries that don't depend strongly on the US...

OTOH it seems fairly safe to build a TSMC factory in the US and loan us some engineers. I mean it isn’t as if we’re actually going to make the long term investments in the education of our people required to steal TSMC’s secrets.

> There is an enormous amount of skepticism towards the US TSMC deal in Taiwan, in the sense that Tsai Ing-wen “sold out” Taiwanese IP and top engineers due to political pressure.

I followed that story in the media for months. I never saw this sentiment. Can you share some sources? I couldn't find anything from mainstream media.

This is one of the most articulate explanations of KMT. You hit it on the nail that KMT isn’t so much pro-Communists as it is pro China.

But, How would you separate one from the other (China from government)?

Sounds like a strawman for them

The same way you can be pro-Korea and anti-communist or, in the past, pro-Germany and anti-communist...

'China' does not mean the People's Republic of China, which is the communist state that occupies mainland China (from KMT's point of view). Taiwan is China, too.

The majority of Taiwanese people do not even consider themselves partially Chinese and none of the major political parties have any interest in political unification with China.

Some people in Taiwan might wish the people of China well because they have family there, but this is no different to how members of the Chinese diaspora around the rest of the world feel about the country.

The pro-China political parties in Taiwan are primarily right-wing parties, which is to say they are much more interested in the Chinese market than in Chinese politics.

There's an issue in how these polls are being conducted and it's hard to tell what's happening especially with Western articles that don't fully give all information or give the polling questions. The word "Chinese" have many different ways of stating it in Chinese. So when they ask in polls e.g. are you "Chinese", it really depends on which word they use. They can use 中国人, which does mean Chinese, but it also has a much stronger political connotation related to the mainland. So most people in Taiwan will say no they aren't 中国人, since they have their own government. However if you were ask, e.g. are you 华人 also a word for "Chinese", etc, they will more likely say yes. After all the official country's name is 中華民國. Mainland Chinese people will also say they are 华人 too.

Kind of like how russian has русские (russkiye) which means ethnic Russians and россияне (rossiyane), which are Russian citizens.

Making the 華人 distinction is like asking white Americans if they consider themselves ethnically European or Irish or Italian or whatever. Just as people in the US with European heritage may have an interest in what is happening in the nation of their ancestors, people in Taiwan with Chinese heritage may have an interest in theirs. But in neither of these cases do the majority of people see the nation of their ancestors as the country that they call home.

The context of these surveys in Taiwan is trying to determine if Taiwanese people see their own country as a different or perhaps more legitimate version of China, and the contemporary answer - unequivocally - is no. The only people pushing the myth that most Taiwanese people see themselves as citizens of China is the ruling party of China.

I don't think 华 is necessarily the same as ethnicity as it's interchangeable with the concept of "China" itself. 中華民國 translates into Republic of China the official country's name, of which Taiwan is mere province of. It's the same "hua". I think most Chinese historians recognize that China as a concept has often been "split" in the past, with different governments each stating their legitimacy. It happened e.g. during the Three Kingdoms period, the North and Southern dynasties period. It happened during the Song dynasty, etc., etc. It's happened an awful lot for long periods of time. To Westerners the Taiwan and China situation seems a bit odd, but not really in the context of Chinese history and the Chinese mindset. Regardless each government still saw themselves as the "legit" Chinese government with all that entails, e.g. mandate of heaven, etc..

If we are talking about ethnicity I believe that the large majority would say they are 漢人, which is more equivalent of an America saying he's Irish, Italian, etc. Huaren is more equivalent to saying you're Chinese, culturally, ethnically, etc... 华裔 is basically saying you have Chinese blood, which may be also different conceptionally.

中国人 used to not have such a huge political connotation to mean only PRC people but was more interchangeable with 中华, 华人, etc., but that's now not the case, and the polls reflect that change in mindset.

I don't disagree that there are many different overlapping terms and identities related to the concept of "China" and the people who have an ancestral connection to it. However, I think it is incorrect to blend all of these together and come out with the conclusion that at the end of the day there is only One China and One Chinese People and these concepts are immutable, inescapable and eternal.

The CCP and other historical leaders have used these concepts to try lay claim to every land that a Chinese person ever walked across, every sea they ever sailed and every fen they ever earned, but that has to be understood as the imperialism it is. Inside the empire it behooves the subjects to not publicly challenge that narrative, but outside of it people have the freedom to define themselves as they see fit.

Polling shows that the vast majority of Taiwanese people today do not consider themselves Chinese citizens (中國人), nor do they support Chinese unification (中國統一). If you put a Taiwanese person on the spot and ask them "well, in the context of the last five thousand years, could there ever be a scenario in which the island of Formosa is ruled by the same leaders as the region of East Asia along the Yangtze, Yellow and Pearl rivers?" they might say yes, because in those last five thousand years there were a couple of hundred when that was indeed the case. But that doesn't mean that they see themselves right now as having any deeper connection to the PRC than the Chinese diaspora in Singapore, Malaysia, Thailand, US, Canada or anywhere else.

I'm pretty sure at this point we are just talking pass each other. Taiwanese people don't consider themselves 中国人, because zhongguoren is now considered to mean a PRC citizen (although historically this was not necessarily the meaning). This is of course true, because Taiwan has their own government that runs pretty much independently from the mainland government. However, I don't think this means they don't see themselves as "Chinese", but simply that POLITICALLY they see themselves as not citizens of the PRC, but not necessarily that they aren't "Chinese".

And I think you're quite wrong that people in Taiwan don't have a deeper connection to Chinese in China. Diaspora in Singapore, Malaysia, Thailand, etc., have largely mixed with the indigenous cultures in those areas. Most of the later generations often speak a broken form of Chinese, or Singlish, etc., and if they do speak Chinese, perhaps they speak a Chinese dialect (brokenly) and probably practically illiterate in Chinese. In Taiwan this is simply not the case, by and large they all speak Mandarin and by and large all share the same history, know the same legends, pray to basically the same deities, can watch and read each others films/media, still uses the same flag as the one that represented China 80-100 years ago, often have close enough relatives still in China that aren't e.g. 5th removed since Taiwan's population double in size post '49, etc. Although there's a difference in traditional and simplified characters, it's not enough that someone who is fluent in one or the other and spends maybe a few months trying to learn the other would have much difficulty. Not to mention I believe most people say that if they already know traditional characters, it's easier to learn simplified, with simplified characters often being the running or grass script form of the traditional character. Moreover because there's actual communication between mainland China and Taiwan, Chinese spoken in both areas won't drift like it did between N.K. and S.K.

What makes you "pretty sure"? Clearly not the polling, which as I have pointed out does not support your view of the situation. Do you have personal experience that provides anecdotal evidence contrary to the polling?

I am not Chinese, but I lived in China for several years and now I live in Taiwan. Anecdotally, I have not met or spoken to a single Taiwanese person who sees China as their true and native home. Many Taiwanese people have never visited China and don't have any close family living there. Of course, some do, but even those have spent far less time in China than a migrant worker like me, and are less in touch with the contemporary culture of the country.

Sure, Taiwanese people still have a lot in common with Chinese people. Nobody disputes that. But Australians, New Zealanders, Americans, Canadians and Brits all watch the same shows and listen to the same music and eat the same food and speak the same language and pray to the same god, and yet nobody would suggest that they are a monolithic people, destined to inevitably unite under a single flag. That's not how emigration works. People emigrate and then they find a new identity, and that's just as true of Chinese people as every other people.

Taiwan as a country is in an unfortunate situation of having been under a military dictatorship for 40 years, and by the time the first democratic elections came about, the die was already cast. The constitution cannot be changed, the flag cannot be changed, the official name of the country cannot be changed - any of these things are likely to trigger a Chinese invasion, because of some politicking that happened between China and the US while the people of Taiwan were still under the jackboot. This status quo is not liked by anyone, but most Taiwanese people have made peace with it. Three decades ago. No candidates in the recent election stood on a platform of uniting with China. Nobody cares about that. The idea that Taiwanese people are all secretly holding on to a dream that they could one day be part of a Greater China is a fantasy held mostly by people who live in present-day China.

That's historically and factually illiterate on many levels but it would be too tedious and likely fruitless to attempt to develop.

Just pointing out, New Zealand, Canada, Australia, UK all have the same monarch. And perhaps the situation between them and e.g. the US, with its five eyes program, is actually closer to something that could occur between China/Taiwan.

On this, we agree. Perhaps some day the governments of China and Taiwan could find themselves congenial enough to have a "special relationship" that builds on the shared history of the people. I think that best describes the hopes of the pan-Blue parties and their supporters in Taiwan today. Unfortunately, the current Chinese government have staked out such an extreme position on Taiwan that it would be hard for them to walk it back and accept a sovereign but friendly neighbor without losing face. I think this mixed messaging is a big challenge for pan-Blue parties who want to convince voters of their vision on cross-Strait issues.

And usually in the mainland the whole of China is referred as 中华.

For other readers, please note that 华 is the simplied form of 華. Outside of mainland China in Southeast Asia, most ethnic Chinese will refer to themselves as 華人/华人 -- roughly "Chinese descent". Most Chinese language speakers (and readers) in the region understand 中国人 to mean citizen of mainland China (and much more rarely Taiwan).

That's a great way to put it. In the US we describe things as some "freedom vs communism" conflict. I kind of had that point of view until I went to school in Hong Kong, long ago. I found that in general the Chinese in HK had more sympathy or patriotism for China then Europe (particularly England!) or the US.

They may not have wanted to be part of the Chinese mainland government as it was, but they were very supportive of a strong China. They felt very aggrieved about how Western Powers had treated China in the past.

I think the problem is-- thinking about my friends in Hong Kong-- is that it is hard to find a third way and that may not be stable in the long run.

Regardless of your observed relative sympathy towards China versus UK/US, how do you explain the multi-year protests against the proposed security law? Maybe the sympathy isn't so strong after all. Mostly, people didn't want to lose their freedom. They could clearly see the security law was the "beginning of the end" for their democratic freedoms, include free speech, right to protest, and voting for their choice of leaders (excluding CEO).

He explained.

Hongkong is a part of China that was seized by foreigners for 150 years. Inhabitants are Chinese in China. Just look at HK movies that often depict Chinese' struggle against Europeans with the 'white guys' usually the bad guys.

But that obviously does not mean that they would necessarily support the communist party or agree to relinquish their (very new) rights.

This is the same everywhere. People may oppose the government but it does not imply that they are not belonging to their country. Obviously.

It's quite extraordinary to see the depth of the anti-China narrative. Only them can be forcibly invaded and still be the bad guys when they peacefully recover their territory.

That is the story if you read western media but it is important to remember that there is a massive world outside the west where that story is turned on its head: Here the US is a frail aging empire that just cannot handle the peaceful rise of China, and the idea of it no being the top dog, and thus lashes out in an increasing desperate manner.

Speaking as someone who now lives in Singapore, this is not the universal view outside the west and the rise of China is perhaps not warlike in the invading territory sense but people are wary of them.

Sorry - just to be clear - just as "Xi is an evil dictator" is common but not universal point of view in the west; the "US is a declining and aggressive empire" is a common but not universal point of view in the rest of the world.

But that is its own form of exaggerated propaganda from the opposing side. In reality, America is playing on geopolitical easy mode and is essentially too big to fail.

- It has vast natural resources, some like its oil surplus are resources that even China needs to import.

- It has a vast, diverse economy that includes high representation in high-complexity high-value products like software, airplanes, and, yes, semiconductors (Intel manufactures 75% of their products in the US).

- It is highly developed and highly educated with particular strength in higher education

- It is essentially impossible to physically attack

- It produces a crazy amount of food

- Its multinational corporations own a large amount of foreign assets, with a banking system that is entangled with the rest of the world

- It is close allies with basically everyone that matters except China and Russia, and arguably China is more dependent on the US than the US depends on China.

- NAFTA countries are a huge strength to the US. The US now buys more from Mexico than China.

- The world’s second largest military and navy are basically non-existent, and the US military is unmatched in logistics.

The US empire can fall apart without the US mainland being invaded or even threathened. Just like the UK empire was intact after the second world war but over the next three decades completely fell apart.

Simplified empire is about ever increasing conquest and exploitation. At one point the conquests and military upkeep becomes more expensive than the spoils and the whole process starts going in reverse.

Of course I am not saying that this is what is happening right now; maybe Russia and China are falling apart and we are entering another unipolar moment.

The difference is that the US is not an empire. It's an economic hegemony (if you're looking at it from a relative power perspective).

Its power stems from sovereign, contiguous territory with resources, or a network of global treaties both economic and military.

In contrast, Britain was a globe-spanning empire, which disintegrated with national independence movements.

The equivalent would be if California and Texas decided to split from the United States.

And as the quip goes... the next few largest economies, including China, are pretty incentivized to keep the global economy rules as-is, because they benefit.

De-dollarization is happening, but is going to be a slow shift, with an uncertain outcome.

I think the word empire is too loaded to be useful. And mechanisms of conquest and exploitation are different. But the core mechanism is the same.

And I think in two hundred years when history is written by someone who is dispassionate about it, the US empire will be seen a continuation of the UK and Dutch empires that was before it.

IMHO, the Dutch arc is definitely a more sound comparison. Albeit, as you say, with the military force of the British.

The key difference with the post-WWII economic world order was that the US generally tried to ensure that everyone else got theirs.

Granted, the US got more, but there was insight that a global economy that benefited all was more stable than a system that left powerful economies outside, with an incentive to topple it.

What happened in HK left a sour taste in the mouth of a lot of people. I’m all for China rising but all I see is a country still closed to the world.

> US is a frail aging empire that just cannot handle the peaceful rise of China

Not seeing that by the numbers

1.) The heavy market losses in 2024 come hot on the heels of a bruising run last year, when the CSI 300 index, comprising 300 major stocks listed in Shanghai and Shenzhen, fell more than 11%. By contrast, the United States’ benchmark S&P 500 index climbed 24% in 2023, while Europe’s grew almost 13%. Japan’s Nikkei 225 soared 28% last year and is still going strong, notching gains of nearly 10% so far this month. https://www.cnn.com/2024/01/22/business/china-stock-market-f...

2.) China suffers from deflation, while the rest of the world combats inflation. Not only does deflation signal a stagnating economy, it can lead to high unemployment, unaffordable debt repayment, and dismal outcomes for businesses. In the worst cases, deflation can lead an economy into a recession, or even a depression. https://www.wsj.com/world/china/deflation-worries-deepen-in-...

3.) Crushing debt. Going back further, China accounts for over half of the entire world’s total debt-to-GDP increases since 2008. https://www.geopoliticalmonitor.com/backgrounder-china-econo... https://www.bloomberg.com/news/newsletters/2024-01-06/bloomb...

4.) China’s youth unemployment rate hit consecutive record highs in recent months. From April to June, the jobless rate for 16- to 24-year-olds reached 20.4%, 20.8% and 21.3% respectively. https://www.cnn.com/2023/08/14/economy/china-economy-july-sl.... For reference, G7 countries is at 10%, US is at 8% https://data.oecd.org/unemp/youth-unemployment-rate.htm

Take a moment to consider what you just said. It’s just that - a story you are getting from the media. If these stories change from source to source maybe they’re not an authority you should appeal to.

Yep.

Taiwan is basically "captured" by:

1) Being too close and too dependent on China mainland

2) Enjoying a huge amount of surplus with China (it overweights everything else combined, and almost doubled)

IMO, there is really no point for China to invade Taiwan, even if Taiwan declares independence. The best move of China, when and if that happens, is to simply grab Jinmen and Mazu (the two small islands close to Xiamen), and then start a economic debacle of some sort. They don't even have to put up a physical debacle -- the only thing they need to do is to remove that surplus by removing all economic preferential policies. The best move for Taiwan, is always to be on the brink of independence without actually getting into it. After all, it is independent in all other ways and it's fine as long as US is strong enough.

Now the real point is: Can China put military equipment in Taiwan to break through the so-called first island chain? I think it's a Yes given enough time. I guess that's why NATO has been busily working on the second chain.

> there is really no point for China to invade Taiwan

Which is why, historically, it didn’t. But Xi isn’t rational: he’s a dictator.

It wasn’t rational for him to force his way onto Hong Kong; the territory would automatically become China in a few decades. Same for his corruption/loyalty purges and decimation of China’s tech sector. Irrational, short-term sacrifices of China’s long-term potential for his short-term politics.

> It wasn’t rational for him to force his way onto Hong Kong

When that happened, Hong Kong was involved in ongoing riots and people were proclaiming (Hong Kong) independence on the streets for months. The legislature was stormed in an event that was not unlike Jan 6 Capitol Hill in the US (we sure did feel the awkward resemblance of the two events...)

Everyone can have their views on whether suppressing the riots was morally justified, but Xi had to do something to assert the CCP's sovereignty over Hong Kong. It was the only rational move unless they were prepared to give up Hong Kong.

> the territory would automatically become China in a few decades

Not sure which place you're talking about. Hong Kong is already part of China at least since the 1997 handover. And Taiwan will never become under CCP rule if status quo is preserved. Don't think there's any mechanism for "automatically" taking over Taiwan.

I don't know how familiar you are with the history, but long-term thinking and patience was essentially the reason China was able to gain back control Hong Kong from the Brits, pretty much on China's terms. They took advantage of the 99-year lease on the New Territories, waited until the Brits started developing on the leased (as opposed to ceded) land, and caught them with their pants down when they realized they had no feasible plan to partially hand back the territory when the lease expired.

Hong Kong has always been a template (in the CCP's mind at least) for taking back Taiwan, I don't know how you'd come to the conclusion that they would forego long term potential for short term gains. It's like you're talking as if they already invaded Taiwan and you're shaking your head over the stupidity of it. (No, it hasn't happened yet.)

The threat of course always seem real enough. It's all a game of chicken. The one who seems most crazy wins. Back then during the 1970s, China also threatened to invade Hong Kong if the Brits didn't hand it back to them. They want you to think the leadership is irrational. Makes them harder to predict.

China was forcing HK to accept China laws and HK didn’t want it. They protested.

Don’t try and and make it sound like HK were just randomly protesting nothing.

HK is part of China.

And it was part of 1 country 2 systems.

Which proved there is no such thing. Because the CCP still has their foot in HK and making sure there is no 2 systems.

> Hong Kong was involved in ongoing riots

Because Xi ham-fistedly forced through a badly-written security law [1] and intervened in its politics [2]. It’s like stupid Tiannamen Square.

> long-term thinking and patience was essentially the reason China was able to gain back control Hong Kong from the Brits

China was rational under the CCP. It’s myopic under Xi.

> don't know how you'd come to the conclusion that they would forego long term potential for short term gains

They already did.

Hong Kong’s mess was avoidable. Most Taiwanese identifying as Taiwanese and not Chinese was avoidable [3]. The trade wars, and China being reëclipsed in manufacturing by U.S. + Japan + Deutschland, were avoidable [4]. The bank losses on overseas loans were avoidable [5]. The corruption in the rocket forces, which guarantees no Taiwan action until China’s military strength is past its relative nadir, was avoidable [6].

All of these missteps would not have happened under proper CCP leadership. They are expressions of Xi’s personal hubris. China sacrificed the immortality of its state for the favour of its mortal dictator; it’s possibly the luckiest geopolitical stroke for America after Russia’s invasion of Ukraine.

> during the 1970s, China also threatened to invade Hong Kong if the Brits didn't hand it back to them. They want you to think the leadership is irrational

But they didn’t. That’s rational.

[1] https://en.m.wikipedia.org/wiki/2019_Hong_Kong_extradition_b...

[2] https://www.bbc.com/news/world-asia-china-56264117

[3] https://www.pewresearch.org/short-reads/2024/01/16/most-peop...

[4] https://www.safeguardglobal.com/resources/top-10-manufacturi...

[5] https://www.ft.com/content/da01c562-ad29-4c34-ae5e-a0aafddd3...

[6] https://www.reuters.com/world/china/chinas-military-rocket-f...

> Because Xi ham-fistedly forced through a badly-written security law [1] and intervened in its politics [2]. It’s like stupid Tiannamen Square.

[1] has no mention of Xi. [2] is in 2021, way after the protests/riots.

I have reliable sources (not mass media, especially not western mass media) telling me that the security law was championed by the Hong Kong government (Carrie Lam in particular). The Chinese Central People's Government was not actively involved until the whole thing spiraled out of control.

I won't/can't tell you how involved I was in the events, but Trust Me Bro ™ I'm not just reading the news and coming to the conclusions that I wrote.

> have reliable sources (not mass media, especially not western mass media) telling me that the security law was championed by the Hong Kong government (Carrie Lam in particular)

Sure. This was in the open. That doesn't preclude Beijing's feet on the scales, which from everything I saw, in the open and with Trust Me Bros, what clear as day.

There were better ways to handle Hong Kong. Xi handled in a way that made sense for him. It cost China dearly in the medium term, and may prove to have been a geostrategic blunder in the long term.

> That doesn't preclude Beijing's feet on the scales, which from everything I saw, in the open

This is what I'm telling you didn't happen. My sources are not public, and you're free not to believe it.

I know you love the CCP and try your best to defend them, doesn’t change the fact that one of the primary reasons for the protests was the security law which came from Beijing.

> championed by the Hong Kong government

There was a significant divide within LegCo (local parliment). So no, it was not "championed". For other readers: The head of state (CEO) is hand selected by CCP. (Yes, I know about the elitist voting system to select a CEO, but the choice needs to be pre-approved by CCP.) Saying the person who was pre-selected by CCP is "championing" a proposed law by CCP isn't saying much. That was her job.

The appointment of Carrie Lam was a mistake and I have no idea why the CCP did it. If the original claim is that CCP/Xi was irrational for appointing somebody who had a past record in somehow inciting mass protests (she was responsible for the reform bills in 2014 that sparked protests on similar scale), then yea sure I'll take that.

What I was responding to was the claim that "Xi ham-fistedly forced through a badly-written security law". This really isn't the case, and if your argument is that "well, everything done by the government is attributed to Xi personally, I rest my case.", I guess you're right then. Whatever you please.

> But Xi isn’t rational: he’s a dictator.

This is a non-sequitor. There's no reason, a priori, to believe a dictator is any less, or more rational than a popular idiot (and we've elected quite a few of those).

(You also don't stay a dictator for very long by being a madman or an idiot, whereas an elected official usually at least gets to finish their term.)

Leaders serve the interests of those that have a say in whether or not they stay in power—in democratic countries with wide enfranchisement, that’s the population in general, in other systems it might be a combination of the military, party, international supporters, and the populace.

I think what people mean when they say a country “acts rationally” or doesn’t is that decisions are made that vaguely make sense in when analyzed by applying the rationalism framework to ”the country,” usually used as a shorthand for the populace. But of course, this would only make sense if the leaders decide that the best way to stay in power is to serve the interest of the populace. Which isn’t the case in a non-democratic country.

Rationalism is a framework in international relations, and it makes sense that some terms-of-art will sneak out into informal English. Unfortunately, in English, “irrational” is also, basically, a fancy way of saying somebody makes stupid decisions. And lots of countries that are, broadly, antagonistic toward the US are not representative democracies*. So it seems to have basically morphed into a way of saying that my country’s adversaries are stupid. An unfortunate end for an otherwise interesting term.

* not to say that we haven’t been willing to morally compromise ourselves and ally up with dictators or overthrow democracies.

> no reason, a priori, to believe a dictator is any less, or more rational than a popular idiot (and we've elected quite a few of those)

Dictators are less constrained and longer serving. The latter is particularly damning, since it means the need to save face prevents course corrections.

I'm concerned that Xi might be more concerned with leaving a legacy than with practical goals when it comes to Taiwan. The civil war was never fully won and resolving that seems important to him. I hope it's just posturing.

On the feasibility of an invasion: there was a very good episode of the Sinica podcast on wargaming this: https://thechinaproject.com/2023/08/17/wargaming-a-taiwan-in...

Very scary for everyone and especially the US. Think several lost carriers within the first 24-36 hrs

Edit: important to note that at the end of the episode they discuss that the wargaming exercise assumed < 100 dedicated amphibian landing boats, but Chinese officials have previously said that they'd use their merchant marine which would mean we are talking about thousand of ships! The whole thing is unfathomable and clearly unprecedented since WWII

I don't think anyone is going to come to Taiwan's aid if a major invasion comes up. There will be material and intelligence support, for sure, but sending troops? Not so sure.

I agree that Xi probably wants some legacy coined to him, but he is only 70 and looks pretty healthy (as far as I see) so I guess he can still wait it out.

I still don't believe there is going to be an invasion of Taiwan happening in the next few years, but if it does, I think it's going to be a lot bigger than just Taiwan. From this perspective, it is scary.

Declaring closed straits to Chinese military surface vessels on the Taiwan-Philippines and Taiwan-Japan gaps, along with resupply of Taiwan from the east, seems a reasonable response for the US (and possibly Japan).

China's calculus changes if they have to strike Japanese and Philippine sovereign territory (even if it's hosting US forces) in order to accomplish their goals.

And it's unclear if even Xi is willing to send as many ships and bodies as it would take to the bottom of Taiwanese straits to get an invasion force across. (Which then faces a guerrilla insurgency in mountainous terrain)

If Xi sees Taiwan as a gateway to Pacific hegemony, maybe that math is worth it...

But asking a populace to support that in the face of economic weakness and high youth unemployment is a tall order. The COVID lockdown riots demonstrated that even China's social control is a fine line.

Many leaders have discovered that war against an external enemy is a powerful unifier, at least in the short to medium term. I would never assume that Xi won’t eventually take advantage of that.

Honestly, Russia invading Ukraine, the global economic response to Russia, and China seeing all of that... probably saved Taiwan.

China doesn't have the natural resource base within its borders (they're digging up their iron ore reserves faster than any country, and they have to import oil) that Russia does.

Consequently, economic sanctions slow their economy down a lot faster than Russia's.

War may be an internal unifier... but unemployment, poverty, and scarcity is a rapid internal disintegrator.

>"unemployment, poverty, and scarcity is a rapid internal disintegrator."

Looking at "disintegrated" Cuba...

China is in no shape to invade, being in a great depression https://www.forbes.com/sites/miltonezrati/2024/01/22/chinas-..., up to its eyeballs in debt, and having horrible demographics. And having horrible tax base.

For reference, when Germany and Japan was expanding, they had great, young demographics, and thus very good tax base. Japan had very low debt in 1920 when they started the wars https://www.rieti.go.jp/en/columns/s15_0004.html. Germany had suspended the gold standard and financed the war by borrowing, something China cannot do today.

Another way to think about this is, instead of all the possible money China made that it could have spent on war, instead it decided to build half-finished tofu dreg buildings, then the CCP elites took those dirty money out of China and into western economies :). Even the elites knew China was no match for the combined wealth/forces of western countries.

Also, the Chinese army in general is probably non-functional. Too much corruption, case in point: Corrupt Chinese Officials Filled Missiles With Water, Report Says https://www.newsweek.com/china-missiles-rocket-fuel-corrupt-...

That is piss-poor journalism. Most military rockets are solid fueled, for one.

Perun covers the topic of Chinese military readiness with excellent sourcing and great detail, as always:

https://youtu.be/vhI_tTEE2ZQ

What state was Russia in before the Ukraine invasion? Putin and Xi aren't leading democracies, they can do whatever they want. Their countries are basically their property. Neither will go hungry if their country has a famine, etc

I agree China is probably not in the best shape regarding fighting, but for a different reason: bureaucracies don't like wars. They are hugely unpredictive, and more importantly, wars bring generals into the front row, so if you are not a general you are going to stand back, lose power and receive order from somewhere.

For example, usually the party provincial secretary (most powerful man in that province) concurrently serves as the first secretary of the provincial military region party committee, but he doesn't have real power in the military. When war comes, it's the group of people who are more experienced with military (the generals) going to make the call, and everyone else is not going to be happy.

Think Politburo and the Central Military Commission. When war comes it's the CMC who is going to make the call. Some people take seats in both but some not. They all stand in the Central Committee of the Chinese Communist Party -- but power distribution is going to shift when things move.

That's why I think the biggest resistance comes from the party itself, not depression. Depression, historically, usually leads to wars, not from them.

But you also need consider that the Chinese demographics are basically only going to get worse from here. Their dependeny ratio is going to get really bad, like it could be 1.5 retirement age people per 1 working age person.

Are they gonna try and fight a war while the average soldier has more then one aging parent back on the mainland?

Who would do the domestic production to support the effort?

If China ever wants to do it they need to do it in the next couple decades

https://www.pewresearch.org/short-reads/2022/12/05/key-facts...

I actually don't think they want a war. They are preparing for a war, that's for sure, but to start one, with NATO? That's not a good idea. Time is China's friend at least for the next decade as you mentioned.

Where is NATO in this picture? There is no alliance trigger with a Taiwan invasion.

There isn't in the case of Ukraine either, but NATO has mostly stepped up anyway. Sure we are not putting troops on the ground (yet???) but NATO has provided a lot of support.

Countries like South Korea, Japan, Philippians, and Australia are likely to get scared if China does too much and send some form of help.

> Where is NATO in this picture?

Nowhere. The relevant alliances are AUKUS [1] and the Quad [2].

[1] https://en.m.wikipedia.org/wiki/AUKUS

[2] https://en.m.wikipedia.org/wiki/Quadrilateral_Security_Dialo...

> Being too close and too dependent on China mainland

This is changing quickly.

New investments in China by Taiwanese companies declined 10.4 percent year-on-year in the first quarter of the year (2023) to US$758 million...That follows an almost 14 percent decrease in such investment last year.https://www.taipeitimes.com/News/front/archives/2023/04/21/2...

Exodus of Taiwanese businesses from China: push and pull factors amidst trade and tech tensions https://english.cw.com.tw/article/article.action?id=3573

We are going to see when the 2024 report is out. AFAIK the huge trade surplus is still there.

> It’s crazy to me that you haven’t see more divestments in China from Taiwanese companies.

There is a ton of divestment, what do you mean?

According to a survey conducted by the Center for Strategic and International Studies Trustee Chair in Chinese Business and Economics, over a quarter of surveyed Taiwanese firms with operations in China had already moved some of their production or sourcing out of China, while another third were considering doing so in the near-term. https://english.cw.com.tw/article/article.action?id=3573

Apple Aims to Make a Quarter of the World’s iPhones in India. Supplier Foxconn plans to build more factories and give India a production role once limited mostly to China https://www.wsj.com/tech/apple-aims-to-make-a-quarter-of-the...

> New investments in China by Taiwanese companies declined 10.4 percent year-on-year in the first quarter of the year (2023) to US$758 million...That follows an almost 14 percent decrease in such investment last year

How's total investment going? If the flow is still greater than zero, then presumably the stock is still increasing... It's a funny sort of 'divestment' where you have more invested (and at risk) every year.

Investing less every year is how you eventually get to investing nothing and finally to reducing total footprint every year in China.

It makes sense that this is a slow process when you consider the massive scale and complexity of the supply chains involved.

A decrease of prior investment level is divestment, just not happening all at once. If you want to see that kind of activity:

Foreign investors have snatched back nearly 90% (!!!) of the money they put into Chinese stocks this year (2023) https://markets.businessinsider.com/news/stocks/china-econom...

> Foreign investors have snatched back nearly 90% (!!!) of the money they put into Chinese stocks this year (2023)

So, they left in China 10% of the money that they sent to China? I'm still not seeing how this equates to a net decrease... 10% > 0%.

> no longer is interested in rejoining Xi’s China.

I am not clear about this statement.

1. When were they ever an actual part of China (less so Xi's china?) 2. When were they ever interested in giving up their sovereignty?

There is actually a ton of reshoring going on in the EU. They are just relatively quiet about it. Here’s a database that only goes up to 2018

https://reshoring.eurofound.europa.eu/reshoring-cases

I wonder if this is relating now to remote work as well?

Been seeing a lot more "Remote only in EU/UK" type roles, and when I ask they're explicitly not catering to other timezone-aligned regions.

Good question but I don’t know to be honest.

I was hiring that way in the previous company I worked for. Before we decided that, we had a few good candidates from UA and several African countries because initially we thought we'd limit only by the time zones.

In the end, we decided for EU/UK because of law compatibility, ease of enforcement (in case we would have to deal with some serious problems) and ease of gathering together from time to time.

Every major country is trying to build new fabs as fast as they can. ASML (maker of the highest resolution lithography machines) has a 400% increase in orders for their equiopment. This isn't a decision of "Japan instead of US", the US is building as many fabs as they can, so is Japan, so is Europe

Seems like everyone is chasing the high end fans, but what about basic components and PCBs? The supply chain is more than latest node chips!

Have you noticed a shortage of low-end components?

Raspberry pis were chronically in short supply for the last 3 years and were scalped to $80-100, to the point where x86 microPCs were cheaper.

That has nothing to do with the broader ecosystem and has all to do with RPi’s weird relationship with Broadcom.

Do you have a source? I’ve seen a lot of unsubstantiated rumors about the source of the raspberry pi shortage.

Wasn't that just because everyone wanted the RPis themselves, not because of shortages of their components?

Raspberry Pi claimed it was a shortage issue. I think if the issue was an increase in demand they would have wanted to say so publicly as it would make them look good.

There were plenty of component shortages in very recent memory (eg, FPGAs), but those were all down to the pandemic.

> building a factory in the US is just not feasible

Please educate me. why?

I wonder what he's thinking when he said that.

Not OP, but: TSMC has tried before. The workforce is not educated properly and the workplace cultures are vastly different. In this case, the US workers were used to stronger labor protections than their Taiwanese counterparts.

The US has multiple fabs and has multiple more being built right now. This is just the propaganda of the elite class who sold off our industrial base and you’re repeating it verbatim

Every year it becomes harder to justify hiring a Westerner from a business perspective. America in 50 years will look like Argentina, full of mediocre workers that demand empire era wages. If we wanna change that, we need to work on developing global monopolies and crushing our enemies. Won't happen though, we'll just wither away wondering why our economy is wasting away.

I’ll take this bet and see you in 50 years. The US has surged ahead of the rest of the world in recent years and it’s only just starting to put itself first again.

How? Our industries are being hollowed out. More and more engineering jobs will go to China, Taiwan, Ukraine, Poland, etc. where they are paid half of a westnern's salary and perform nearly as well if not better. This trend will only continue until the U.S is cut out of the equation entirely. Sure we're doing better than Canada, but Canada is the prime example of a country in decline. They won't need to wait 50 years to be Argentina. Same with many other Western countries.

Why didn't this happen in Germany and Japan? Both were born as manuf'ing giants and remain as giants. Compared to many neighboring countries, their labour costs are very high. Yet, they continue to manuf a huge amount of good for domestic consumption and international export. And both countries have very strong labour laws. To an American, it appears almost impossible to fire people in Japan and Germany.

This is exactly right. Onshoring fabs back to the US is part of a long term political and economic strategic plan to counter China called The Clean Network / The "5G trifecta" — TSMC's new fab in Arizona will be the largest onshoring in American history.

https://en.wikipedia.org/wiki/The_Clean_Network#The_%225G_tr...

https://keithkrach.com/article/tsmc-12b-chip-plant-in-arizon...

> This is just the propaganda of the elite class who sold off our industrial base and you’re repeating it verbatim

They’re socializing Americans to get used to a future where their kids have to go to the Middle East and China in search of upwards mobility. (Of course those societies will never be as accommodating of Americans as America has been of Chinese and middle Easterners.)

> a future where their kids have to go to the Middle East

Where? UAE? Israel? Anywhere else? Both are still far less prosperous for the middle class than the US.

I'm unaware of how automatable fabs are. If the workers are high-cost then the machines need to do more or the government needs to subsidize production.

One of them is about five miles from me in Phoenix and it's going poorly. My read is that there are some legitimate labor concerns, some mismanagement, but also a lot of special interest strong arming in things like not bringing in enough Taiwanese workers.

Translation: the business didn’t want to pay enough

I have a friend in Taiwan who works as an engineer for an LED manufacturer. He makes about 2K USD a month. I don't think anyone would even clean toilets for that much in the US. US salaries are just not globally competitive.

And yet salaries in the US are sustained. To me it looks like the issue is that while we know how to start companies and have VC capital, we don't know how to outsource well (even with all the local immigrants)

> He makes about 2K USD a month. I don't think anyone would even clean toilets for that much in the US.

Do you mean to say that's low pay or high pay compared to the US?

In the US, 2k USD a month would barely be enough to rent a small apartment, let alone pay for utilities and groceries. You'd be left homeless or starving.

It's low pay. That's $12 an hour.

The majority of Americans of all races and genders earn above $15 an hour [0]

Taiwan's average wage (so skewed upwards) was ~$22k a year in 2023 [1]. That was an 8 year high btw - wages have been much lower.

Lots of White Collar Taiwanese would move to Mainland China for that reason - they'd earn similar if not higher salaries in Mainland China AND not pay income tax.

Basically, OP's point is that companies don't optimize for wages alone (and I can attest to that having hired abroad, and helped move the operations of a former employer to Israel+India from the US).

Even TSMC's founder admitted that:

On a podcast hosted by the Brookings Institution last year, Chang lamented what he called a lack of “manufacturing talents” in the United States, owing to generations of ambitious Americans flocking to finance and internet companies instead. (“I don’t really think it’s a bad thing for the United States, actually,” he said, “but it’s a bad thing for trying to do semiconductor manufacturing in the U.S.”) [2]

[0] - https://nationalequityatlas.org/indicators/Wages_15-hr

[1] - https://focustaiwan.tw/business/202311290017

[2] - https://www.politico.com/news/2023/02/14/taiwan-tech-king-pe...

About your last paragraph, how does Intel and Global Foundries (IBM and AMD) do it well? It sounds like moaning from a senior business person wants easy mode. This is a new step in TSMC's history: expanding manuf'ing overseas. I am curious how the new TSMC plant Japan will do.

> how does Intel and Global Foundries (IBM and AMD) do it well

They don't execute as well as TSMC or Samsung, but they're able to do it largely because they're too big to fail and they Defense related subsidizes (eg. Both Intel and GlobalFoundaries got $3Bil from the DoD for manufacturing Secure Enclave chips along with the CHIPS money).

Also, you don't need to be leading edge for most defense applications. i7 processors tend to be the norm for plenty of Western defense applications and don't need a sub-7nm type process that Intel/Samsung/TSMC are competing over.

The issue is companies like Samsung and TSMC would get a large amount of state subsidies, while the US only started getting back into that game in 2022-23.

TSMC began building the Chandler plant before they got the subsidizes needed to make it even more worthwhile.

> curious how the new TSMC plant Japan will do.

Probably pretty decent. Japanese and Taiwanese work culture is basically the same so they won't have to deal with labor unions or getting complaints about overwork.

I remember hearing a lot of the American workers hired by TSMC Chandler ended up leaving for Intel Chandler for that very reason.

Basically, TSMC wanted to replicate the exact kind of work and management culture that exists in Taiwan (eg. Long hours, dictatorial managers, power politics, relatively low wages, little to no stock compensation, etc)

I have no knowledge of this field, but my naive question would be, wouldn't building such advanced products involve so much more automation relative to number of human workers, that the salary of workers doesn't affect the cost that much?

> the salary of workers doesn't affect the cost that much

It doesn't and that's why Intel still has foundaries in Oregon and Arizona.

The difference is TSMC's leadership doesn't want to play ball with American work culture and wants to keep pushing the 996 mentality (yes, even Taiwan has an extreme overwork and underpay problem).

The Foundary space is a very low margin industry. There's a reason why the only companies left are TSMC, Samsung, Intel, and GlobalFoundaries.

While the TSMC plant in Chandler has been plagued with bad press, the Intel plant right next door has been expanding with almost no hiccups.

Or maybe US workers are worth their high salary, and that is why the high US salaries are sustained.

I think it's more that US living requires the high salary. Living expenses quickly cripple anyone who doesn't have a high salary

So money flow to people who need it, and the greater the need, the more forcefully the money flows, eh?

Unfortunately the vast majority of people who need money don't get as much as they need. :(

If they paid what US workers expect, the chips would cost so much that nobody would buy them.

This is the effort from the 90s to which I am referring:

https://www.oregonlive.com/silicon-forest/2022/04/tsmcs-morr...

“We still have about a thousand workers in that factory, and that factory, they cost us about 50% more than Taiwan costs,” Chang said.

50% doesn't seem that much, but it really needs to be compared to production. Geographical risk spread will always cost you.

I was recently chatting with someone in that industry but not at TSMC. It's that they assumed Taiwanese workplace, cultural, government, and business norms will work here. There's chip manufacturing in the US, so it's not that it can't work. It just won't be the same as Taiwan.

You can re-write your exact same response but replace TSMC with Toyota. When Toyota started to build plants in 90s, people said much the same. In the beginning, yes, there were huge cultural gaps and major issues. Over time, they were fixed. I expect the same for TSMC.

I wonder if he means getting people with PhDs to work grueling hours in an assembly line.

OP never followed up his many claims with responses nor numbers, and had many wrong things that were disproven. I wouldn't bother.

Well EU is busy getting out of having their supply routs and clientele being depended on Russia, can't really exit two failing tyrannic dictatorships at the same time - can you?

Communist China is still in recession and as manufacturing keeps exiting to India and other places, Chinese market won't be that attractive anytime soon.

No wonder real Chinese (Taiwanese) companies want backup plants elsewhere and are not interested into selling-off to failing dystopian dictatorships.

Correct, Europe would be wise to detach from China faster

Exports to the EU fell 11% from a year earlier to $38.3 billion in November compared https://apnews.com/article/china-exports-imports-decline-eco...

China's newly appointed defense chief and Shoigu discussed boosting military cooperation and coordination as the Russia-Ukraine war drags onhttps://www.newsweek.com/china-russia-ukraine-war-dong-jun-c...

They didn't lose it, they volunteered it, supporting not only the embargo but the spirit of the embargo...rightfully.

Opening a factory in the USA means jeopardizing the safety of Taiwan to some extent, they were all for it, and as far as I know as of late last year was reading to begin staffing and production, however the US administration backed off their enthusiasm and support leaving the factory stranded, an absolute fucking catastrophe.

> Opening a factory in the USA means jeopardizing the safety of Taiwan to some extent

It doesn't which is why they are all for it. Diversifying TSMCs production base geographically weakens China's hand.

but also weakens Taiwan's value from US' perspective.

I’m out of the loop, why was building a factory in the US not feasible?

U.S. doesn’t work 996, TSMC tried to import all of the senior staff from Taiwan leading to cultural mismatch and resentment

>The 996 working hour system (Chinese: 996工作制) is a work schedule practiced by some companies in China. It derives its name from its requirement that employees work from 9:00 am to 9:00 pm, 6 days per week; i.e. 72 hours per week, 12 hours per day. A number of Mainland Chinese internet companies have adopted this system as their official work schedule. Critics argue that the 996 working hour system is a violation of Chinese Labour Law and have called it "modern slavery".

Oh good Lord. I wonder if anyone has insight into how this is actually done. Are people really working the 72 hours or is it like here, people goofing off most of the day and hurrying up to get done when they need to?

These posts seem to imply the latter?

https://www.reddit.com/r/China/comments/boq5qe/what_the_996_...

"Contrast that to the US and Europe who keep talking about these things, but don't actually execute(although the US at least tries to throw money at the problem)."

What other things should governments do other than "throw money at the problem"... crazy profitable ventures should have some help but they aren't giving those grants back etc.

I'd say total compensation packages in orgs are the root of why things might work in Japan and not in the US. The pay structure in Japan is very rigid, and by just one metric, top executive pay [1], Japan is 3.4x more efficient.

While a highly-paid CEO may not break the company, the skew in pay upward across an org will.

[1] https://www.jstor.org/stable/26592017#:~:text=After%20contro....

>they immediately acted to bring manufacturing of giants such as Iris Ohyama back to Japan. Contrast that to the US and Europe who keep talking about these things, but don't actually execute

I'm not sure what you mean by this because the US already has Intel 4 while Japan currently has zero advanced fabs. Is this just another Japan good comment or do you have a specific point in mind?

Lots of wrong things with the comments

> Contrast that to the US and Europe who keep talking about these things, but don't actually execute(although the US at least tries to throw money at the problem).

"Our analysis of multiple surveys indicates that as much as 91 percent of U.S. manufacturers have reshored some production in 2022, up from just 7 percent 2012. " https://www.assemblymag.com/articles/98200-a-look-back-at-20...

China's annual exports drop for first time in seven years. Among key trading partners, exports to the U.S. led the decline, down 13% from the previous year https://asia.nikkei.com/Economy/Trade/China-s-annual-exports...

> TSMC lost the Chinese market...South Korean officials on the other hand lobbied heavily to get long term exemptions, which allowed them to turn around their profit situation.

Samsung profit tumbles 35% as chip weakness persists. https://www.scmp.com/tech/big-tech/article/3248969/china-rem...

Over 50% of Korean firms missing earnings target in China this year: survey https://www.koreaherald.com/view.php?ud=20230830000611

>Our analysis of multiple surveys indicates that as much as 91 percent of U.S. manufacturers have reshored some production in 2022, up from just 7 percent 2012.

'Some' is doing a lot of work. What does that really mean? If 99% of manufacturers each onshore 0.01% of their manufacturing, all that has really happened is that everyone can probably now label things "made in america".

here are some numbers

U.S. manufacturing construction spending reached a 20-year high, hitting a $194 billion annual rate in April 2023, nearly double the $107 billion annual rate from a year ago. https://thinkkc.com/news/blog/kc-smartport-blog/2023/08/01/i...

In 2021, Intel announced more than $43.5 billion in new manufacturing investments across Arizona, New Mexico and Ohio to bolster U.S. chipmaking and R&D leadership. https://www.intc.com/news-events/press-releases/detail/1638/...

Walmart previously announced a $350 billion investment to make U.S. manufacturing more “affordable and feasible,”

https://www.sme.org/technologies/articles/2023/october/resho...

> U.S. manufacturing construction spending reached a 20-year high

A nominal 20-year high and the increase is since the pandemic. Also this is for construction. Here are some other time series: https://fred.stlouisfed.org/series/OUTMS

Walmart $350 billion investment is over 10 years and mostly is for agriculture and some small electronics.

Also this source is an American manufacturing lobbying group.

here are some more

companies have announced over $166 billion in manufacturing in semiconductors and electronics, and at least 50 community colleges in 19 states have announced new or expanded programming to help American workers access good-paying jobs in the semiconductor industry. https://www.whitehouse.gov/briefing-room/statements-releases...

Apple commits $430 billion in US investments over five years https://www.apple.com/newsroom/2021/04/apple-commits-430-bil...

Microsoft will buy enough U.S.-made solar panels to power 1.8 million homes https://www.greenbiz.com/article/microsoft-will-buy-enough-u...

Tesla plans to spend $3.6 billion more on battery and truck manufacturing in Nevada https://www.cnbc.com/2023/01/24/tesla-plans-to-spend-3point6...

I feel like this comment is GPT-ed. You are just putting random numbers out there. They are meaningless. The previous poster gave you an indicator "Real Sectoral Output for All Workers". The indicator adjusts for inflation.

US manufacturing never recovered from the 2008 crisis. And in 2005, the output of manufacturing was higher than it is today (in real value terms, not funny dollars).

Of this 194 billion, how much are one-offs? If intel builds one plant (in 2011) for 43 billion and average investmenr in 2011 was 109 billion, then intel is like 45% of all.

So in 2023 maybe there was some other big one off?

Of course a factory is still a factory, so the one-off thing is a bit misleading, but what I try to say that it seems one big factory takes 40% of investment. What would be the spending for 10? 400 bilion? One is still better than zero though and probably has subcontractors.

> Our analysis of multiple surveys indicates that as much as 91 percent of U.S. manufacturers have reshored some production in 2022, up from just 7 percent 2012.

You should always be skeptical of statistics written this way. It is a very unintuitive way to aggregate and suggests that this is the strongest number they could find.

Perhaps, however, the re-shoring is in fact happening. There has been a massive build out of new manufacturing facilities in the US over the last few years.

Near $120 billion spent on new manufacturing facilities in 2022 alone. The surge looks like this:

https://i.imgur.com/Bydq6Hb.png

The US manufacturing sector has added 900,000 jobs since 2014 (according to the US Bureau of Labor Statistics). It was at 12,081,000 manufacturing jobs in January 2014, and January 2024 is at 12,979,000. For the US, which is supposed to be a wilting manufacturer, that's a huge gain over a decade (including the pandemic hit, which slashed 700,000 jobs out temporarily; one guesses the figure would just be even higher minus the pandemic).

Manufacturers don't add a million jobs if they're not expanding presence. Even if they were expanding very slowly, they would do everything possible to avoid adding jobs/labor (most US manufacturing expansion in decades past came from productivity gains).

The US has modest corporate taxes, a good business regulatory environment, amazing capital markets, enormous economic scale, a single giant market, ports that can easily get you to Asia/Europe/Latin America, consumers, and labor. It makes perfect sense that at a time of growing risk of conflict with China, that the US would be reshoring.

Also, wouldn't be surprised if the drive towards minimal-latency shipping is having some evening effect.

Increased labor costs... versus rapid, $$ last-minute transport / port fees to hide latency across the Pacific.

At some point, it's just cheaper to pay the manufacturer more, if they can more rapidly respond to demand.

Paying more just doesn’t work if a chokepoint is clogged up or closed. Chinese ports locking down, congestion at LA/Long Beach, Panama running dry, Suez being blocked, the Red Sea crisis, etc. It is making just-in-time inventory untenable.

Canada and Mexico have multiple entry points without as severe risk.

> https://i.imgur.com/Bydq6Hb.png

Looks like an inflation graph to me.

Without any source information it may as well be written in crayon. (No offense. At least op provided something to back their claim.)

What people arguing against your points don’t understand is that globalism is coming to an end. It started when China initiated decoupling and started their ill advised wolf warrior diplomacy.

in favor of 'friend-shoring' not really 're-shoring'

Is it? From what I see, global trade is free as never been in previous century or any time in history before.

Yes, but that’s not the trend.

The US can no longer afford to be world police with its navy.

Russia is now outside of the US economic system, while relations with China continue to deteriorate. It will deteriorate further once China attacks Taiwan.

US-China tensions are not that huge at the moment.

Russia is not really cut off from the global trade.

Africa becomes more united and collected.

Globalization is still a trend.

US China relations continue to deteriorate because of China’s aggression in the South China Sea and continued decoupling aka de-risking.

Russia is cut off from the developed Western economy. Some NATO commanders are also warning that Russia will not stop at Ukraine.

Even if Russia and China weren’t on a war path, the US simply cannot continue to subsidize global defense of trade via our navy. Currently, only the UK and Japan have a sizable fleet of long range ships, but they still don’t compare to the US navy.

Yes, globalism is still here, but the trend is that it will be on a major decline in the coming years. Destabilizing geopolitics will only hasten it.

Russia is cut off only on paper. In real life they do trade very well.

And yes, they will never stop because they have an empire mentality. Make Russian empire great again! Russian are ready to kill anyone and die for promise to get 2000$ per month from their government.

Being able to get supplies from black markets at a high markup isn’t the same as being connected to the markets. Russia also doesn’t want yuan or rupees, and most of their new customers haven’t even paid them yet.

Common misconception but the US does not pay for its navy by itself. That is done by the world using a ever overprinted currency for reserve and trade. The US can print aka lend that money to set a navy to work. The world pays for its police.

Some hicks in the rust belt contribute and receive very little benefits from a fee trade empires existence and are thus usually left to there own devices by both parties.

> Common misconception but the US does not pay for its navy by itself. That is done by the world using a ever overprinted currency for reserve and trade. The US can print aka lend that money to set a navy to work. The world pays for its police.

Common misconception that the US controls the world dollar. The Eurodollar has been where the world's trade has been pegged for half a century, the US simply benefits from a massive economy and it being their native currency.

This is like saying the US controls the world through ubiquity of "English" as a business language. No, but they definitely benefit from it. You can choose to speak any other language, or use any other reserve currency that you like.

> Russia is not really cut off from the global trade.

Some people confuse the "global" trade to the American & Friends trade.

Russia is not even cut off from trade with USA. There are even messages that USA is buying some amount of Russian oil every month.

From the other hand, Russians can buy anything western they want, just with more friction and higher price. Even dual-purpose western components for army are freely available for them.

They are cut off from the American system. The products will leak from both ways unless you impose a strict naval blockade. For some premium, many countries will dodge the sanctions.

> a strict naval blockade

I think if USA would naval blockade itself then many, many people would give a sigh of relief.

Imagine the anti liberal backlash that will come if the first gi dies from forces armed up by neoliberale trading. They will ride the ex-elites out of town on a rail, if they are lucky..

Nah, the brakes have definitely been thrown on over the last few years. The US has had back-and-forth with China and Canada, e.g. the huge tariff on that Embraer plane, export restrictions on computer chips, and I think the steel tariff dispute is still simmering away. The UK has pulled out of the EFTA and that's affected supply chains a lot. China has cut off the workarounds that allowed offshore ownership of stock in their companies, a whole lot of what was tolerated in Hong Kong is not being tolerated any more. And that's before you get into the Russia situation - cutting off SWIFT is pretty much unprecedented, and the oil price cap stuff is also new. Global trade is very much less free than a decade or two ago, and that trend is accelerating.

Russia is not fully cut from western financial system. Even few EU banks are there.

Price cap is poorly enforced. All economic sanctions applied to Russia are poorly enforced thanks to globalization and lack of political will of western countries.

Sanctions are severe on paper, but in real life just adds some extra friction.

Unfortunately we are living in times when there are only weak populists in the government.

I wish the next is president will be someone like Reagan, but it is impossible.

A lot of the "it's time we moved away from China" corporations are moving their manufacturing from the authoritarian communist country of China to... the authoritarian communist country of Vietnam. Vietnam has loose regulations due to being undeveloped and seeking business just like China was 30 years ago. In a decade MBAs will be amazed when they find out Vietnam exerts total absolute control of their businesses just like China does.

Manufacturing hubs are shifting hands, but globalism isn't ending. Companies can still get away with paying $10 a day or less to people in some countries and they're never giving that up.

True but Vietnam hates China, so that will work. Also, the Vietnamese can pretend to be a democratic country in 10 years; so that they look cute in the eyes of your average American. You know, kinda like what the Korean or Taiwanese did.

Taiwan is a democracy

Unlike China, I don't see how Vietnam would ever be some kind of threat to world peace and order or western powers in general. It's not that large, it doesn't have an ongoing war with an important high-tech trading partner, it doesn't seem to have much interest in being a superpower, etc.

Companies don't care about threats to world peace. They cared that China locked down cities, their factories, and exports. If world peace were their concern, there are a lot of countries they'd refuse to do business with. They usually only stamp their feet and whine when countries do things that affect their profit margins, like push for increased wages, environmental regulations, and locking up a factory to prevent disease spread.

One thing to keep in mind: a few years ago, South Korea with North Korea to allow corporations to do manufacturing in North Korea. [1] The result: mega corporations like Korea's Hyundai and Japan's Family Mart flooded in to take advantage of cheap (probably even slave) labor. It closed not because companies felt morally wrong about it--it closed because the governments forced it to close.

[1] https://en.wikipedia.org/wiki/Kaesong_Industrial_Region

> The US has modest corporate taxes, a good business regulatory environment, amazing capital markets,

Yep!

> ports that can easily get you to Asia/Europe/Latin America, consumers,

Yep!

> and labor

Nope. Labor costs in the US are ridiculous and make it non-economical for most consumer manufacturing. This is why real output remains low and we are exploring 'friend-shoring.'

Well… I believe labor in the US is cheaper than Australia, NZ, UK, and most of the EU. (I’m not sure about Japan). Basically all comparable countries in terms of wealth. So that’s something?

There's 0 chance US labor is cheaper than the UK at this point.

US: >Real median household income was $74,580 in 2022

UK: >In the financial year ending (FYE) 2022, median household income in the UK before taxes and benefits was £35,000, increasing to £38,100 after taxes and benefits.

NZ is even lower per https://en.wikipedia.org/wiki/List_of_countries_by_average_w... so unless there is a big difference in the manufacturing sector specifically, it's probably also cheaper there

That combined with the house prices... ouch.

Wait - the median income in the UK is higher AFTER tax is removed? The median person receives more benefits than pay tax?

Something not quite right there, because at £35,000 you're not eligible for that many benefits. Unless this is somehow miscounting pensioners?

US labor is in no way, shape or form cheaper than EU labor, even with the higher taxes in the EU. Have you checked the average or median salary in the US and Germany, France, Spain, let alone any Eastern European country?

None of which are the UK, Australia or NZ.

> and most of the EU.

...

Tesla pays 20-30$ per hour. Is it ridiculous and non-economical?

There is Canada/Mexico for labour if you just want to get things on the same continent which still protects the US from supply chain shocks somewhat.

> It was at 12,081,000 manufacturing jobs in January 2014, and January 2024 is at 12,979,000.

That is 7.4% growth since 2014.

US population grew by 6.4% during that same period. Hard to see the "huge gain" here.

So like, if you're going to present contrary evidence, quoting an article by a party with a vested interest that claims they "analyzed multiple surveys" and then fails to link to them or even name them is not exactly helpful.

I'm not arguing the point, mostly because I still don't have reliable evidence to argue about.

> "Our analysis of multiple surveys indicates that as much as 91 percent of U.S. manufacturers have reshored some production in 2022, up from just 7 percent 2012. " https://www.assemblymag.com/articles/98200-a-look-back-at-20...

With no numbers or rules for the numerator and the denominator, and no clear methodology, and no clear understanding of the biases and the agendas of the authors, these ratios are meaningless.

Annual exports dropped on a dollar basis but not when using RMB. On the other hand if you were to check Japan's exports, which newspapers are more likely to report on a Yen basis, it states exports are up due to weaker yen by quite a bit, but if you use dollar basis, it's down by more than 10%.

Most Chinese factories in 2023, due to way less orders and needing to clear out their over inventory, had to cut their margins dramatically. thus, more shipped out, but making way less. They won't be able to do that in 2024, with factory shutdowns due to no margins and no more inventory. and that's why:

- Chinese stock market has dropped 11% this year, with 3 year cumulative loss of 6 trillion in 3 years https://www.cnn.com/2024/01/23/investing/china-stock-market-...

- Chinese economy is suffering from deflation

- China’s youth unemployment rate hit consecutive record highs in recent months. From April to June, the jobless rate for 16- to 24-year-olds reached 20.4%, 20.8% and 21.3% respectively

Why focus on youth employment? Not that I disagree China economy is the toilet. But I do think they have a more nimble economy that can turn around quicker.

If your thesis is that world manufacturing is leaving China to return (primarily) to the US that seems unlikely. More likely it will move to places like Africa and South America, no? We would expect that to happen as China economy transitions and they become the dominant world super power.

> they have a more nimble economy that can turn around quicker

That's not evident in the persistent high youth unemployment rate. There is also something called middle income trap https://en.wikipedia.org/wiki/Middle_income_trap

> they become the dominant world super power.

There's no sign that that is guaranteed. No sign in culture power. No sign in economic power. No sign in demographics power. No sign in innovation power. No sign in technology power. No sign in military power.

> There's no sign that that is guaranteed. No sign in culture power. No sign in economic power. No sign in demographics power. No sign in innovation power. No sign in technology power. No sign in military power.

They're on course to become the biggest economy in the world. Popularity of their culture is surging (though from a very low baseline) - you hear a lot more about Chinese movies/novels/games, or even software, than even a couple of years ago. Innovation and technology are hard to measure, but on measures like scientific papers published or patents registered they're also surging. There's plenty to criticise about the Chinese government and Chinese culture, but there are definitely significant signs of success.

> They're on course to become the biggest economy in the world

They are not:

https://www.bloomberg.com/news/articles/2024-01-25/us-extend...

> Popularity of their culture is surging

Not as popular as Korean or Japanese, or arguably Indian.

> but there are definitely significant signs of success.

This is true. The very fact that China is considered a super power shows that they have made tremendous progress.

>More likely it will move to places like Africa and South America, no?

No. Africa and South America largely lack in basic infrastructure, political stability, and workforce education. You can't do advanced modern manufacturing in places like that. Brazil has some promise, and maybe Nigeria and Chile, but that's about it. Brazil still has major stability problems, not sure about the others.

> Our analysis of multiple surveys indicates that as much as 91 percent of U.S. manufacturers have reshored some production in 2022

I'm not touching the rest, but this seems like an obvious fig leaf for largely failing to bring production home.

>I think a while back TSMC finally understood that building a factory in the US is just not feasible

TSMC always understood this. This wasn't a realization as the cost benefit analysis is obvious: US workers don't have the skill involved and they demand higher pay. The combination of the two makes the US plant a net loss.

TSMC did what they did because of political pressure. So the plan was always for the US plant to just satisfy that political pressure as much as possible. There was never a plan for the US plant to do anything profitable, it's more of a forced "technology transfer".

It's nothing like a forced technology transfer at all. The US isn't acquiring TSMC tech in any manner what-so-ever. It could hack TSMC - as China does the US - if that's what it wanted.

The US is adding such domestic manufacturing for strategic national security reasons: namely, a modern tech-heavy economy can't function without the chips that power it. And yes, of course it pressured TSMC to contribute to that.

It's technology transfer. And it won't be obtained all through legal means.

Hacking isn't enough the amount of skill involved can't be gleaned from a computer file.

You need US based employees who can be poached who are doing espionage and all that. This is entirely what it's for. CIA and other defense agencies do this worse stuff. What I'm saying isn't out of the blue. And it's also quite obvious. There's really no real difference between manufacturing here or abroad so that is the only reason why they bring it here.

Most of the IP that TSMC make comes from US-based companies anyway. There's no "technology transfer".

You're describing technology transfer from the US to Taiwan. Which is absolutely true and happened. But the literal fact that the US can't build what Taiwan builds shows it's currently a literal transfer in the Other direction.

What I don't understand, is how thr initial US transfer to Taiwan negates the technology transfer I'm describing here? It doesn't.

I think you lack objectivity. Your American pride is getting in the way. Same with the downvoters I guess.

My comment has nothing to do with whether or not Taiwan or the US deserves or does not deserve to be a participant or victim in this "technology transfer" it's just a statement about reality.

Having a facility that’s not on the ring of fire and avoiding any earthquakes or similar natural disasters are also big benefits. Apple also invested with TSMC by buying out all 3nm production.

> a while back TSMC finally understood that building a factory in the US is just not feasible

Why would that be? There are plenty of fabs in the US.

They probably won’t work for $30-40k per year and think it is a lot of money

Well, that’s certainly a narrative.

But a more likely reason is Japan has excellent infrastructure and the Yen is at a record low (and falling).

> manufacturing giants such as Iris Ohyama

??? Giant? No. They have less than 3,000 employees and make mostly cheap plastic home goods. They have annual revenue (not profits) of less than 1B USD.

> TSMC lost the Chinese market, because their government went along headfirst with US trade war policy

Does TSMC really have a choice given their dependency on IS and Western European suppliers to make chips at all? The plan for Taiwan being taken over is to simply stop supplying Taiwan fabs, they aren’t self sufficient (or Korean or Japanese etc…).

I don't think that's right. TSMC will not change their corporate style and management style to what American engineers are used to. We won't work for less wages than competitors while also working 25-30% more. There is absolutely no reason they can't build fabs in the USA, sure it's not as easy as in Taiwan, but that because they are brittle about how to get it done. There's also the rather positive fact that the USA isn't out to end their civilization and way of life.

>they immediately acted to bring manufacturing of giants such as Iris Ohyama back to Japan.

Finally people on HN are talking about this. Along with a lot of Supply Chain / raw material shifting as well post 2020. So it wasn't just manufacturing.

>Contrast that to the US and Europe who keep talking about these things, but don't actually execute

Contrast to Apple, Which keeps feeding PR and Media about they are moving out of China but dont actually execute ( until very very recently in 2023 ).

So, maybe someone here can explain this to me. I anyways hear about how the entire semiconductor industry is completely dependent on TSMC, and nothing can operate without them, thus their geopolitical importance.

But then what are Intel, Arm, etc in this picture? I don't understand semiconductor manufacturing in enough detail -- I assume TSMC occupies a different part of the supply chain? But chip manufacturing seems like a pretty integrated process top to bottom; what's the division between them? In concrete terms, what is it that TSMC is doing that nobody else is?

ARM does not manufacture own chips, they just design and license them. Outside of Intel, pretty much everyone else uses TSMC to build their computing chips.

Not even amd or TI?

AMD's main manufacturer is TSMC. They also use other companies AFAIK.

Texas Instruments has its own factories in the USA but AFAIK they make other kinds of ICs than the popular ones that TSMC is known for, they do mostly analog and embedded (still important stuff).

The sibling post to yours lists other factories. There are definitely other companies.

> Outside of Intel, pretty much everyone else uses TSMC to build their computing chips.

this isn’t correct.

UMC, SMIC and TI have more than 10 fabs each.

Nanya 3 fabs.

Micron > 10 fabs.

Intel > 10 fabs.

Tower Semiconductor 7 fabs.

Rohm > 10 fabs.

Fuji 4 fabs.

Fujitsu 5 fabs.

Kioxia 10 fabs.

Renesas 9 fabs.

and the list goes on and on.

Most of these are very large node sizes though [1]. In ultra-small node sizes for processors, there is basically only Intel, Samsung, Globalfoundries (ex AMD) and TSMC.

[1] https://en.wikipedia.org/wiki/List_of_semiconductor_fabricat...

except that GF isn't leading edge any more. Their smallest node is (I think) 12nm which is about 8 years behind.

Even Intel builds their GPUs at TSMC.

For <10nm specifically.

And it is TSMC, Samsung and Intel.

> I anyways hear about how the entire semiconductor industry is completely dependent on TSMC

I don't think it's as strong dependance as many commenters assume here. Samsung is 2 years behind TSMC and intel 3-4 years behind in terms of fab capability. While losing 2 years of progress is not great, it's definitely nothing like world can't function without TSMC.

Obviously they need few years to ramp up but I assume it's not like Taiwan geopolitics situation would change in a day.

They focused on the manufacturing part https://youtu.be/r_8XClnnvIk

Currently TSMC has the only leading edge chip fabrications plants (fabs) on the planet and they're all located in Taiwan. They account for all new chips for all new Apple products, all new AMD products, most new Nvidia products, etc. Most companies design the the chips, but then outsource the manufacturing of them to TSMC as building a fab has astronomical upfront costs.

TSMC has acquired a lead in this area through a number of different methods. One of the main things is that they focus deeply on manufacturing. Another is that they work 24 hours a day in R&D, running 3 shifts so they basically have the lights on all the time. And as mentioned above, the upfront costs are incredibly high with a fab costing on the order of 20+ billion dollars to construct.

Intel is attempting to catch up, but it will likely be another 3 to 5 years before they are able to do so. Honestly just having R&D up and going all the time is probably a huge advantage for TSMC and probably a big reason behind their success. Regardless, suffice to say basically all cutting edge product shipments would cease in a matter of months if TSMC fabs were destroyed.

Multi-shift R&D is a weird proposition. Is it for best using machinery?

According to this, it has something to do with the movement of the wafers:

https://news.ycombinator.com/item?id=35667282

It also has to do with competitive boasting: theirs goes to eleven.

I'd translate "move wafers" as run test batches. What I understand from that is that he can run more tests on his machines.

Honestly, I'm not sure anything of that is real. I can't believe other fabs don't run tests 24/7, and I can't believe they have people that rarely meet changing the same machines instead of only running tests without changing anything.

I used to work in R&D for leading edge node development and we had a couple night shift technicians to unblock long running high priority tests. However, we could have iterated much faster if we had dozens of engineers running additional tests at night. Some tests require you to be there at the tool to change temp and so on. And there are ton of possible tests you can do. If you get a result back in the middle of the night and have engineers to review the results and configure a new test then and there, that’s a much faster learning cycle time.

Another advantage of TSMC is how they have enough fab space dedicated to R&D that they can run prototypes through quicker because they aren’t competing with manufacturing to get processed.

So if it’s so successful why doesn’t Intel etc just copy this process?

Cost and culture? I don't know that most 20 year intel employees in Beaverton are going to be willing to do night shifts.

Worse yet, Moore's law was basically an automatic monopoly for anybody that had a much more productive R&D than the others. But we are only noticing the gains now, that the law is gone for years already.

Maybe the culture thing is really pervasive. It wouldn't be a first.

Some things are hard.

To add to what others said here, another salient factor is that in Taiwan it's pretty possible to get late night food, drink, and other things. These things are almost non-existent in the US which would make working night shifts extra rough.

Depends on where the fab is. I agree if we're talking about rural Arizona, but if the demand is there, I guarantee you that someone will open a franchise (or six) down the street and operate overnight to meet that demand. In and around larger cities though, it's not really a problem. There is usually _something_ open, even if it might not be a long list.

apart from NYC (and maybe Las Vegas), I don't know any other US cities that has functioning food places that operate throughout the night abuduntly

do you have examples?

Are we discounting fast food? I can't speak for everywhere, but in Dallas, many chains are open 24/7. Whataburger is a Texas franchise that has been 24/7 for basically it's entire history in all locations. They are closed for Christmas and Thanksgiving only. McDonalds is 24/7 in most major cities. In Dallas proper, there are some smaller (local) chains that are frequently open until 2 or 3am but admittedly they mostly cater to the bar crowd who wants something to eat before heading home after drinking all night. Café Brazil is a local Denny's/IHOP type diner that is open 24/7 including major holidays!

Your addition of "abundantly" makes things more ambiguous, but I think the choices available in most large US cities is _decent_ after midnight. Can you get absolutely anything like NYC? No, probably, not. But burgers, diners, and fast food are all generally available. That's also ignoring 24/7 grocery stores like Walmart that can fill in the gaps if you're craving something hyper specific. Perhaps you can't buy it ready-made, but you can buy the ingredients!

> Currently TSMC has the only leading edge chip fabrications plants (fabs) on the planet and they're all located in Taiwan.

What about Samsung? I thought they also made leading edge.

They're close, but I believe they're not doing any 3nm manufacturing at the moment but I could be wrong about that.

NVIDIA, Apple, AMD, Qualcomm exclusively use TSMC for their best chips. Even Samsung prefers to use TSMC-fabbed Qualcomm SoCs over their in-house design and manufacturing for their flagship phones.

>Intel is attempting to catch up, but it will likely be another 3 to 5 years before they are able to do so

According to them they will do it in next 12 months.

Sounds like hopeful words for investors

What makes you think so? I have their stocks and been following their roadmaps and it seems to not be that out of touch with reality

Actually TSMC has 2 fabs in Mainland China (a 12" fab and an 8" fab), and one in the USA (8" fab) [1].

[1] https://www.tsmc.com/english/aboutTSMC/TSMC_Fabs

They do, but they don't produce any of their leading edge chips in mainland China, just as they won't be in the USA either. Looking at the wording of my post I should have made the leading edge part more clear. My mistake there!

What happens if a war with Taiwan does break out? Who's poised to pick up the slack?

Literally no-one – the majority of high-tech consumer electronics would stop being produced as soon as existing chips run out.

That’s not true - with the exception of the leading edge nodes, there are other fabs.

The question is how easily could they migrate and prioritise the workload?

There's very few of these other fabs outside China. They exist, but aren't able to deliver enough at a global scale. That's what we learned the hard way during the chip shortage a few years ago, where car production for jbstance basically came to a crawl.

If that happens, you won't care. You'll be concerned with your personal safety.

This is part of the reason why you're hearing posts about TSMC expanding out of Taiwan. As it stands today it would be a fairly large economic hit to have advanced processors stop production. Building out redundancy seems to be a top priority.

It's also worth noting that in the event of a war the US is very, very likely to bomb the shit out of the TSMC plants.

https://www.businessinsider.com/us-would-destroy-taiwan-semi...

https://www.theregister.com/2023/03/14/us_china_tsmc_taiwan/

Perhaps Korea or Japan, but in practice nobody, and that's partly by design.

Taiwan being the core producer of the super high end chips is guaranteing them that if a war ever happens, they won't be left as sacrifice to the opponent while the rest of the world goes business as usual.

They critically need to be a strategic and non replaceable producer.

It's very unpredictable. It might spill to the whole East Asia and then who knows.

Where can someone dig into the data on this?

> Currently TSMC has the only leading edge chip fabrications plants (fabs) on the planet

No, Samsung has some in South Korea also.

> I anyways hear about how the entire semiconductor industry is completely dependent on TSMC, and nothing can operate without them, thus their geopolitical importance.

This is quite hyperbolic. While losing TSMC will have a massive global impact, there are alternatives in many different market segments (e.g. Intel for CPUs and GPUs for computing, be it personal or datacenter), STMicro/NXP/Infineon/Bosch for industrial applications, etc. They may not be as advanced, or as good, or as cheap, or in enough quantities, but it's flat out false to say that TSMC is the entire semiconductor industry. Just a major portion of the bleeding edge.

Intel before Patrick P. Gelsinger only made chips for Intel.

ARM doesn't make chip

on the low-end and mid-tier chip, you have plenty of fabs like Global Foundries, UMC...etc.

on the high end, you have Intel, TSMC, and Samsung

Intel used to have the fab lead over everyone else, generally being 1-2 generations ahead. That persisted for at least a decade. Then they had a huge whiff. Their 10nm node was due in 2016 and they didn't start volume production until 2019. That gave TSMC a huge opportunity to catch up and pass them which is exactly what they did... TSMC went volume on 10nm in 2017.

Samsung is behind a generation (or thereabouts). Not sure why exactly.

AMD spun off their fab as GlobalFoundaries. The separate company bowed out of the bleeding-edge fab business. Their 14nm node was licensed from Samsung and they completely cancelled their 7nm node.

As bleeding-edge fabs kept getting more and more expensive lots of companies decided to go fabless. They were able to do that because unlike everyone else TSMC doesn't make chips for itself so it doesn't represent a competitive threat. This is the real key to TSMC's success.

Other companies like GlobalFoundaries and ON Semi are filling volume with older process nodes like 14nm. There's a large number of chips needed by millions of manufacturers large and small. Only a small proportion of those need the power/perf you get from the latest and greatest. Manufacturing on an older node is cheaper thanks to lower capex and much much better yields.

Eventually I think we'll see litho equipment make its way into the niche market similar to 3d printers. We've seen some people hacking together stuff to make their own chips in a garage... they're tiny 5-2000 transistor affairs but it would be quite interesting to be able to churn out a custom chip with a million transistors on it.

A cottage industry does seem to be within the realms of possibility, and 3D printing is an apt comparison. However, it’s likely to be prone to the same challenges - input materials (filament and wafers) will be subject to variable quality and environmental factors (moisture, heat, dust), and the machines themselves will take some time to mature (how many iterations of incompatible 3D printer ‘standards’ have cropped up in the two decades since DIY printing emerged in the scene?).

The nirvana of a household printing capability never really materialised.

Please explain this to me, isn't the lithography equipment from ASML that does all the hard work of imprinting the wafers? If Samsung and Intel buys the same machines why can't they make the chips with the same transistor density?

Michelangelo had the same chisel and tools as every other artist of his time.

Why didn't they just make art with the same quality?

It's not just what tools you have, but how you use them.

You don't just go plug in one of these machines and it starts pumping out working chips.

Increasing density means better controls and processes, better knowledge of running and calibrating the machines.

Tons of processes before and after the ASML machines.

You don't just buy the machines or build a fab.

Each fab is the development of an entire "process". Furthermore how the machines are used varies greatly. Everything from exact doping and cleaning processes to metal deposition parameters, plus the masks you feed into the system which these days are all interference-based - that is you rely on the interference patterns because you can't actually make masks nor use light of a wavelength small enough to create 14/10/7nm feature sizes. The mask you make is nonsense but the interference pattern it creates results in the actual patterns you want.

How many layers, how you can intersperse metal vs silicon layers, allowed dimensions for gates, wires, distances between components, and a whole host of other things are unique to the specific process. Oh and parameters vary based on what you are making on a given process (eg SRAM vs logic). Different layers can have different requirements as well. You might not even be able to take a design to another fab from the same company on the "same" process node, let alone a competitor.

This stuff is all extremely closely guarded by every fab. If you're a serious customer (read: $$$$) the fab will share these parameters with you so your engineers can design to that spec. Then your first run will likely have problems and need to work with the fab engineers to tweak layouts and such to get usable yields - depends on how much you are relying on the fab for design/layout. TSMC will happily license you certain blocks to use in your design along with guarantees on yield since the design is made for their process from the start. You can also license designs (eg from ARM) that are pre-customized for a given fab's process.

I wonder why none of the trillion-dollar club members with "infinite" money, e.g., MSFT or AAPL, are building bleeding-edge fabs in locations that are geopolitically more stable than Taiwan. If TSMC tanks, then these companies will also tank by a massive amount, much more than 20 bln.

> Eventually I think we'll see litho equipment make its way into the niche market similar to 3d printers...

Won't you be better served with a FPGA? Are there any cases, besides hobby projects, where making you own silicon on a process many generations behind the state of the art would make sense?

With 3D printers, there are parts that are simply impossible to make without them, and there is no programmable plastic, so it makes sense to use them. In fact, we are starting to see entire factories churning out plastic parts from FDM printers not so different from what consumers buy. In the case of Prusa, they even make the parts using the same printers they sell to consumers.

Good answers here, but look at the book "Chip Wars" if you want to learn a lot more.

Asianometry has a number of explainer videos on TSMC. I recommend them:

https://www.asianometry.com/p/tsmc-briefly-explained

Generally, they're not (or should not be), and its mostly hype tunnel vision (imo). Here's a link to the List of Semiconductor Fab Plants worldwide. [1] Note: This is a 1000 line table that lists production per plant at 10,000-100,000 / month for most fabs with numbers. There are a lot of other chip sources.

However, TSMC has the largest capacity worldwide. Intel's probably second for capacity. Samsung's probably third. Micron, Texas Instruments, Infineon, Epistar, and several others are lower down yet significant.

If you need small chips, out of plants already constructed, based on the data in the list, Intel and Samsung appear to have the "current" smallest node size at 7 nm. TSMC has three plants under construction with 3-5 nm node size. Intel and Samsung also have a lot of capacity in the 7-20 nm range.

Most companies are in the 20+ nm size ranges, although definite exceptions with Nanya (10 nm), SK Hynix (10 nm), nanoPHAB (10 nm), Win Semiconductor (10 nm), GlobalFoundries (12 nm), SMIC (14 nm), UMC (14 nm), Kioxia/SanDisk (15 nm).

There's also technology and product variety, although most list 3-5 for each plant, so there's a bunch of specializations.

[1] https://en.wikipedia.org/wiki/List_of_semiconductor_fabricat...

>TSMC occupies a different part of the supply chain.....

This is an interesting but not so obvious question, because it is so basic, I would thought it wont be on HN at the same time judging from comments on Hardware topic it really should have been spelled out clearly in the first place.

So first you have IP companies like ARM. They have lots of different IPs. ( Intellectual Properties ) From ISA ( Instruction Set Architecture ), think x86, ARMv8 etc to the actual CPU design, think Intel P-Core, E-Core, Skylake, AMD's Zen core etc. Along with many other IPs from GPU to Network, CPU Interconnect or even DRAM controller. ARM will work with TSMC ( or any other popular Fabs for specific market ) to have those design specific for the Fab and specific to certain Fab technology.

So in an extremely over simplified situation, you buy a few IPs, or there are even reference designs. Stitch them together. Send them over to TSMC and say hey can you print them ( or Fab ) for me.

What TSMC doing that nobody can do ( in the same time frame ) is that they could Fab it using industry leading technology, or the smallest transistor currently available as long as you are willing to pay for it.

There are plenty of chips that doesn't require TSMC to print. From Automotive to low cost semiconductor using very very old semi-conductor Fab's technology. Think about the small chips that goes into toys or electronic appliance. Or simple calculators.

With all this capital investment and the physics approaching an asymptote, my gut says we're going to see fabbed chips become more commoditized in 20 years (with lower pricing and more competition), and I'm excited for that! Imagine if you could order a fab run as frictionless as ordering business cards.

I thought companies were doing this for old but useful chips? There are chip manufactures who buy old fabs and then do runs for chips that go in automobiles/boats/whatever. At least, it was my assumption.

Canon's new litho might make low-volume very frictionless:

https://global.canon/en/technology/nil-2023.html

> with lower pricing and more competition

Every node shrink since 7nm (some say 28nm) has increased the price per transistor.

So if we keep going for smaller and smaller nodes expect prices to keep on rising. Once we reach some physical limit that we can't figure out how to solve is when you can expect prices to stagnate/lower (and progress stops too)

Does this account for macroeconomic effects?

Thing is, fab building, integrating production machinery, fab supply chain, and operations still has a lot of special sauce in it. Some of those factors are not affected by transistor physics converging at a limit.

But, for less than cutting-edge chips, if your budget is a bit more than for printing on paper, it's getting there.

We may be getting close to the limit for 2D density, but have just barely started moving in the 3rd dimension.

Chips are also quite small, limited partly by the ability to cool them once they enter a computer and partly because a single defect often means the whole chip must be discarded (which means large chips generatel lower yields).

I suspect we will see much more development in all of these directions, with individual chips extending deeper into 3d and getting improved tolerance to defects allowing them to get larger, as well as with chiplet, die-2-die, stacking and similar methods of combining chips in a package continuing to move forward at a rapid pace.

I don't think we should expect foundry development to stagnate in the near future. If anything, as AI starts to be used in developing new chips, it may well accelerate.

>> We may be getting close to the limit for 2D density, but have just barely started moving in the 3rd dimension.

Nah, 3D is already well under way. Micron is stacking 232 layers of memory cells to get high density: https://www.electronicdesign.com/technologies/embedded/artic...

Flash memory is also increasing the number of bits per cell by essentially going analog. So those two thing are why you can get 2 TB on your keychain.

Then AMD is doing their 3D V-cache, putting a bunch L3 or L4 cache on a die stacked on top of the CPU. The issue there is that both the CPU and cache dissipate too much heat to keep stacking higher much higher.

GPUs are using HBM memory which is also stacked, but again power dissipation is going to limit how far that can go.

Even in low end devices - Raspberry Pi - we've had DRAM stacked on top of the SoC die for many years now.

20 Years ago 3D was seen as a way to higher levels of integration once scaling came to an end, but it's already been happening in more and more places as the end of scaling is near. Innovations will continue to squeak out slowly for some time.

> layers

Thinking in terms of layers is still essentially 2D.

The manufacturing process uses layers, so I'm not sure how that follows.

Tell that to my 3d printer

I'd say the 3D printer is an example of my point. You can design and think in 3D despite the manufacturing process being layered. In contrast, chips are designed from a 2D standpoint. As an example, does the ALU circuitry span layers or is it just on one?

This is all just memory and not the processor itself

Because logic dissipates a lot more power. You can't stack CPU die because the power density goes up linearly with the number of chips.

Edit: power per area goes up.

A major difficulty with 3D is heat dispersion. It will be interesting to see how this will evolve.

I don't think we'll see much of a change in how we're approaching an asymptote for the foreseeable future. In order to do something like that, you would need a significant innovation that disrupts the whole trillions of dollars industry and it's unusual for a maturing industry to see that kind of change. If any one of these innovations were significant enough, I think it would have been prioritized above the other innovations that have been done instead which only get us smaller incremental gains.

I imagine we'll still see gains of 10-15% a generation with whatever improvements come over the next few decades but I don't see us going back to the gains of the 80s and 90s where performance for the end user was doubling regularly, if that's what you're hoping for.

Edit: Also I think the price for the performance will continue to rise. If we're looking at performance per dollar, I think gains will only be in the 5-10% range

3D makes every part of your heat story much worse.

You don't want to just stack layers... modern chips already have tons of layers and each one is an opportunity to screw something up. Stacking of dies has its own problems but people are shipping stacked dies and have been for some time and as noted the heat is a major problem.

Things will improve but not like they used to. It doesn't take that many more scaling nodes before gates are a handful of atoms and quantum effects dominate. But really gate sizes stopped shrinking a while ago - scaling has been terribly uneven for some time.

I am not in the chip industry at all, but I understand that chips offload so much heat in part due to cramming as much as performance as possible into a small 2D space. If you increase the number of layers by 10-20x you now have much more "surface area" to work with. Could we see chip designs that operate at a much lower voltage (thus minimizing heat) and are "slower" as measured by chip frequency, but have greater overall bandwidth? Maybe a chip for servers that has 256 cores on it, each with their own caches?

From a Moore's law perspective this would continue the transistor count doubling trend too.

It’s a larger surface area only if you increase the area without increasing the heat generation. That is a heat sink. If you want all those layers computing and therefore producing heat, then you might as well keep it flat and have more fluid moving heat away. In a 3d space that fluid is going to be started between different layers and reach its carrying capacity of heat quicker.

No not really.

Dynamic switching power consumption is proportional to frequency which is why things slowed down dramatically once we hit 1Ghz.

Of course you must reduce voltages to reduce current and losses... but at a certain point your transistors no longer switch between clear on/off states because the voltage swing is too small. This is a quantum effect of all semiconductors (related to their valence band energy IIRC). The tiny leakage this represents only makes your heat dissipation story much worse. So you end up trapped between two walls: higher voltage is required to drive the transistor between full on-off states but lower voltage is required to keep the gate from burning itself up. No matter what material you pick (silicon, germanium, etc) there is some minimum voltage required.

Also your insulating layers become a problem as they shrink: quantum tunneling allows electrons to jump the insulating layer. There's a tradeoff here too... alternative oxide layers reduce leakage at the gate but have worse barrier constants and so are prone to more quantum tunneling (related to conduction band energy).

Adding more layers just makes more problems, as each layer you stack risks mangling the layer below it. Plus the risk of introducing a defect. The more defects you have the worse your yields and the more expensive the chip gets.

The short version is: we are at the point where no matter which direction we go some quantum effect kicks in to cheat us out of further optimizations in that direction.

Imagine if you can produce 3D compute substrate as 1 liter modules that can be built as Lego's. With holes to run water-cooling pipes through.

If you have, say, 100l of such substrate for less than $10,000, with a density similar today's tech, just in 3 dimensions, you can probably afford to run most of it at very low frequencies to conserve energy, with only non-parallelizable operations running at GHz-level frequencies.

Such a machine could still probably provide as much compute as a small datacenter can today.

Now I don't know if it will take 20 years or 200 to be able to do something like that, but I'm certain that we have a looong way to go before we reach the physical limit to how much compute we can stuff into a rack.

The funny thing about the 3D chips is that if extrapolated far out, we end up with cubes. Definitely sci-fi territory. And maybe questions from kids like "but if they are shaped like cubes, why are they called chips?".

Some semiconductor segments have been like this forever. You don’t really care who is making your LM317 voltage regulator —- it’s a straight up commodity and vendors compete on price and availability.

This was the cause of a number of boom/bust cycles in semiconductors.

I'm also wondering whether we're in for another bust. 5ish years? Maybe the demand situation is just different this time.

General rule: No, it's not different this time.

Slightly more sophisticated rule: Yes, it's different. But it's not much different.

6nm and 7nm technology nodes and 100k WSPM capacity. Other investors are Sony, Toyota, and Denso, they will also be main customers.

Taiwan already has a fab in japan 40 nm, 28 nm, 22 nm, 16 nm, and 12 nm process technologies working with 55k WSPM capacity.

We talk a lot about software bloat. How good/fast do chips really need to be for defense purposes? Chips are so good now it seems you could be 2-3 generations back and still get the job done.

These chips are probably aimed first and formost at civil production, in particular for car/plane/boats (for domestic and export needs) and consumer electronics, from rice cookers to door bells etc.

For Japan it's a crucial bet, a lot more than other countries in America or the EU as it's the crux of the economy.

> consumer electronics, from rice cookers to door bells

I doubt rice cookers and toasters are using the latest chips for TSMC. E.g. the ESP32 microcontrollers are manufactured by TSMC on 40nm, and there are a bunch of different companies that have similarly sized/powerful microcontrollers being manufactured (e.g. STMicro).

If I get your point, the question would why have TSMC make these low end cheaper chip in Japan when they abund on the market ?

The whole effort of bringing TSMC factories to Japan was triggered by the shortage a few years ago, when I think a Toyota factories had to slow down as they couldn't get enough chips to meet the production goals.

They'll probably rely on TSMC to make the deal work even if external competitors offer better prices or availability in the short term.

Isn't this about 6 and 7mm nodes? Aren't those still pretty good nodes with lower costs than the cutting edge (3nm and lower)?

It's planned to produce from 40nm down to 6~7nm. Now, the 6~7nm might end as an empty promise, if TSMC wants to keep more advantage at home.

It seems like RADAR can continuously be analyzed and improved with more and more compute power.

When RADAR is the eyes and ears on the modern battlefield, you want as many computers on it as possible to find as many targets as possible and to differentiate them.

If one is serious about it, there is a clear distinction between verifiable and trusted computation. There's also a less clear distinction between cheap and expensive to verify ones.

Just because most equipment must be trusted, it doesn't mean you can't use lots and lots of untrustable hardware.

As drones (both airial, ground based and naval) take over for human soldiers, state-of-the-art electronics may increase in strategic importance rather than decrease.

Drones are becoming more important, sure, but will never take over for humans. Humans are immune to cyber-attacks in ways that drones can't be (humans are capable of major autonomy so maintaining connection is less important, for example), but all of that is overshadowed by perhaps the most important strategic feature of human soldiers:

Human soldiers can be killed. If e.g. Russia bombed a US military base in <country>, then the US has an instant cassus belli (and is politically forced to respond drastically and be drawn into the war) and <country> knows this, which makes the US military base a far more effective red line against Russia as a result. In other words, human solders act as a political tripwire that more effectively binds allies together.

Clearly you're unaware of what's going on in Ukraine.

Drones are not there to take over for humans, drone-based weapons are ridiculously effective compared with anything before them and there's plenty of space for more computing power to make them even more effective.

> humans are capable of major autonomy

Actually, this is the main reason why drones will be needing more and more compute; to ensure that they can be as autonomous as possible.

Whether they will be able to reach or surpass humans in this regard is still disputed by some, especially those who think that the human brain is powered by some kind of non-material soul.

> human solders act as a political tripwire

This really only applies before a war starts. Once it has started, you want to have the best equipment you can within your budget.

Now there are other political reasons why autonomous drones may be resisted, primarily that they can be seen as very scary or even evil by onlookers. But the same could be seen by current remotely piloted drones, but that hasn't stopped the US from frequently using them for assassinations at least since Obama's period.

Once drones become able to operate without a human to control them remotely, they will. Especially if jamming limits their ability to communicate.

This makes sense. The counter argument would be that war is a competitive endeavor. So if you do have chips that are 2-3 generations behind what your opponent has access to they will try to design weaponry that takes advantage of this gap.

Sure but nobody is going to build a brand new factory that implements an old process

Disassembly of crashed Russian missiles in Ukraine show that they use multiple consumer-grade microcontrollers and DSPs. Western systems engineers would typically have opted for a single aerospace/defense-grade FPGA instead of having so many different chips and interconnects complicating the system.

Using Russia's approach you can easily stay many semiconductor manufacturing generations back. Using the Western approach you will prefer staying up to date so you can continue using the latest the latest proprietary manufacturer-supported FPGA tooling.

Russia's approach has historically always been zerg rushing - they view both people and military tech as disposable, so it has to be extremely cheap, but it doesn't matter if the accuracy loses out.

The US, NATO in general, and especially Israel, are the other end of the spectrum: they prefer expensive, but powerful and very accurate weapons (and in the case of vehicles, prioritize survivability). For Western countries, wars are unpopular so they want to keep the fatality rate low, and Israel doesn't have that many people in the first place so their Merkava tank designs focus survivability even more.

The Merkava has seen some time losses but that was due to open back hatchets and IEDs big enough to flip the entire vehicle.

I'd argue that Russia's approach is not necessarily just human wave tactics. There were a lot of planning done on for, say, the Battle of Kursk.

Also from last year:

"TSMC to build US$11 billion chip manufacturing plant in Germany" https://www.scmp.com/news/china/article/3230440/tsmc-build-u...

How earthquake resistant is all the lithography equipment? The alignments are so sensitive..

Like Japan, Taiwan is also on the Ring of Fire, and is no stranger to earthquakes.

Is "tectonically stable" a natural resource that relevant regions could use to their economic advantage?

Geo/political stability.

Yes, and it's one of the reasons why companies choose Arizona for their fabs.

https://arstechnica.com/gadgets/2021/06/why-do-chip-makers-k...

I hear it's one of the factors regarding the location of Intel's upcoming German fab.

We know how to make earthquake resistant buildings and foundations for decades now.

Fabs still stop when there is an earthquake and it is expensive. Being earthquake resistant means extremely expensive equipments don't break when there is an earthquake, it doesn't mean there is no interruption.

Not when you're talking about tolerances below nanometres.

Earthquakes last about a minute max, how many semi-conductors have been produced in Japan already? Many.

A lot of this is from memory so...possibly wrong...

Very but also its solveable. You stop manufacturing, realign the tools, and then move on. With the 2011(?) japanese earthquake, there were impacts of the earthquake vibrations on lithography well beyond Japan - and it took a while for the earth to settle back down to the point that lithography was stable again. Aftershocks, tsunamis, reverberation of earthquake energy are all sources of vibration that can affect lithography.

The bigger issue (to me with what you are saying) is actually within the semiconductor supply chain. At least when I was involved, many/most/a huge fraction of the basic wafers come from Japan. Two of the major players were (are?) SUMCO and Shin-etsu, and they had huge fleets of CZ furnaces because power was relatively cheap, and more importantly, generally very stable. The process time for ingot (the precursor to wafer) growth is on the order of like weeks/months and is very energy intensive. So, power stability is a big deal. The power issues after the earth quake had huge impacts on wafer availability over the following year or so.

They won't break, but they may screw-up in-progress wafers and need to be realigned. You're right that they're incredibly sensitive, and they've built this into the design of the fab. They float the masks, objectives, and wafers on pneumatic isolators (separate stages for each, I believe). Extra-sensitive machines will also be individually isolated from the rest of the fab to prevent adjacent machines and footsteps from interfering.

Here's an example of a common pneumatic isolator, the kind of which can be found in almost every lab making optics:

https://www.newport.com/f/pneumatic-vibration-isolators-with...

And you can see what seismic isolation in a datacenter (as an example) looks like in this video https://www.youtube.com/watch?v=GXwQSCStRaw

So what's cool about japan is people stay in the job for a long time and they are affordable, yet the country is very high tech and educated. This seems like a stronger match than America. The TSMC chairman said they retain people for 10 years in taiwan

Interesting why Japan and not some EU country. Risk wise, I think EU is a safer bet compared to Japan if China attacks Taiwan

1. Japan is a vassal state of the US.

2. Japan has a different work and compensation culture than European countries.

3. China is no discernible threat to Japan.

3: Hmmm.

Not according to Japan's 2023 defense white paper.

https://www.mod.go.jp/en/publ/w_paper/wp2023/DOJ2023_Digest_...

I wouldn't put much stock in the strategic implications of these publicized papers outside of the strategic implications being that they want you to think China is a discernible threat.

It's similar to the Israel/Iran thing with the end goal being trying to drum up the warmongers.

Just reading a few sentences of that, it reads exactly like warmonger rhetoric that we all know intimately rather than inside baseball.

Uh, are you saying that Iran wouldn't erase Israel from the Middle East if they had the capability to do so? Where are you going with this? It's indisputable that Iran sponsors anti-Israeli militias across the Middle East, which would qualify them as a threat.

Israel has nukes, I don't think any country within missile range is going to wipe them off the map

I believe that whether or not Iran would erase Israel from the Middle East, which they probably would, along with a certainty than Israel not only would erase Iran from the Middle East, but has been actively trying to get the US to do so for some time -- the reality is that the only erasure thus far that has been actually executed in earnest is by the Israelis on behalf of the Gazans.

Which would, you know, qualify Israel as "a threat". And they certainly are, in many ways.

To what or whom, you haven't really made clear, but certainly a threat.

Anyhow, I'm really not interested in this level of discussion because it feels extremely superficial. It also feels quite gross to either take seriously or promote warmonger rhetoric, but that's just my personal opinion.

Also, please do not predicate your posts with "uh", it's a bit silly.

You're the only user here trying to aggravate everyone.

>having a different opinion is "aggravating"

Why did you post this when it is clearly non-productive at best and an obvious terms violation?

If you think Iran isn’t a threat to Israel I believe it is your opinion we shouldn’t put much stock in

Israel is a larger threat to Iran than Iran is to Israel, I think that's relatively obvious and can't really be argued.

Israel is not an existential threat to the existence of Iran. What world do you live in?

Iran has started no wars in recent memory, killed no civilians, etc.

Iran has, nevertheless, been the constant target of Israel attempting to drag the USA into a cataclysmic war.

What world do you live in?

Iran hasn’t directly started any wars, but has been fighting proxy wars across the Middle East and Africa for over a generation. You are seriously misinformed here and should not be discussing this stuff until that changes. The entire spectrum of conflicts in the Middle East right now are all coordinated by Iran and its revolutionary guard.

> Iran has started no wars in recent memory, killed no civilians

By that measure, America’s never done any mischief in Latin America!

They bombed Pakistan the other day.

>and can't really be argued.

Challenge accepted.

Iran is pretty darn hard to actually invade - unlike Iraq, Iran has tons of mountainous terrain and a huge army, and even the US hasn't tried it. So while it would be very unfortunate for Iran if they got into that war, the regime might survive. The US could probably do a lot more, but if they did then Iran could go metaphorically-nuclear and block the strait of Hormuz, destroying global oil supply and supply of LNG to asia, causing mass blackouts there. So the US would be quite leery about doing so.

Israel could nuke Iran, but then they'd be utterly fucked politically. Not just for breaking the nuclear taboo, but because Hezvollah et al would cream themselves and have basically infinite recruitment.

In contrast, Israel has recently had some of its citizens killed by the Iran-backed Hamas, and are bordering both Lebanon and Syria. This didn't hurt Iran from political backlash, not at all, Iran's been pretty open in their hating Israel. If for any reason the US withdrew its military backing for Israel, Iran could and would support an extended proxy war between Israel and its neighbors. Not Gaza (Israel controls their water and fuel supply, so if they're not concerned about pissing off the US and rest of the world then they can kill them all fairly easily) but Lebanon and Syria via Hezbollah and some 'renegotiation' over Golan heights.

What's more, Iran could just invade Israel outright, with Syria's support. Iran has 9x the population size and their army is reasonably modern. Without US backing, Egypt's current dictator could decide to provide military support of their own. And if they win that war then Israelis will be ethnically cleansed. Which is arguably a worse threat than what Israel could do to Iran.

There you go, it can be argued! I wouldn't normally argue for it, I'd say they're about equal personally. It's kind of weird to compare them; neither is really in a good position to fight the other in the first place.

Good post, I basically agree - hence the current policy Israel and America have cooked up of isolation and embargo.

Regarding ethnic cleansing, while you speculate, the Gazans suffer under it in reality.

I'd give Iran the implicit moral highground because they are pretty peaceful and haven't attempted to ethnically cleanse anyone, unlike Israel in both aspects.

Just on the basis of what has actually happened, Israel is an egregiously immoral state. Rich man, eye of a needle and all that.

Now, as for whether or not they would... I think they probably would. But that's speculation, not reality.

>3. China is no discernible threat to Japan.

China is known to make spurious territorial claims. What happens if they produce an ancient fishing map tomorrow claiming Okinawa as their own?

Unless China plans on a war of mutual total destruction, it can't take Okinawa let alone stand against the rest of Japan.

A vassal state?

Obviously

Using a definition of "vassal" that's so broad as to be meaningless, yes

How is it meaningless? Japan was conquered and has had its constitution written by the US while hosting a large number of its troops. The only difference between this and any other historical vassal state is the US paints an illusion of not being one and its economic system doesn’t require tribute to profit greatly as it favors trade.

It also is free to amend and revise its constitution, to terminate the US-Japan defense treaty (Article 10), to pursue its own foreign policy goals, etc. "Vassal state" is an old term with specific connotations that fits Japan if you sort of squint the right way but really doesn't.

If Japan ended its defense treaty and began pursing goals counter to US interests, you are mistaken if you think it would lead to a conflict.

To military conflict? I doubt it. To political and economic conflict? Probably. The same thing would be true of France or Poland. That's generally what happens when an ally country stops being an ally country, but it doesn't imply that they're a vassal state.

France and Poland are also US vassals lmao. France can make a better case but that’s because it admitted that all of these alliances subserve it to the US in the 60s and has fought hard to retain some semblance of independence.

Are there any US allies you don't consider vassals? Is every NATO member a vassal?

Here's an interesting and often used Chinese government propaganda word they use: vassal. It's meant to insult any U.S allies that work together with U.S against Chinese aggression, and to try to drive a wedge between the countries. Nevermind that it's so obvious it never works. And they never call Russia or North Korea a Chinese vassal officially - although some Chinese netizens do

> It's meant to insult any U.S allies

Lol, there's no such thing as "US allies". All US presidents agree, but only Trump acknowledged it.

Perhaps some ideology makes you ignore the otherwise obvious similarities between the peoples of Taiwan and Japan - including but not limited to intelligence, work culture and education - but to me, Taiwan and "some EU country" seem to be a world apart.

I don't think they were talking about the people, but the geography. The Taiwan and Japan are both physically close to China. Taiwan is 125 miles away. Japan is 500+ miles away. Germany is 5,500 miles away.

There are select EU nations that fit that bill. I'd be more suspect of the fact that the EU's political future (not existence, but future) is more uncertain than is communicated to the public. East-West political instability would be a concern in terms of ip control and espionage, at minimum. For more context, historically Europe is a powder keg.

Not just work culture; in my opinion, Taiwan is culturally closer to a Chinese-speaking Japan than an extension of mainland China.

imo Switzerland/Germany could be a good fit too(tsmc will open a fab but for bigger nm so I was wondering why not open another fab but for smaller nm, but maybe you are right- timezone sync can play a big factor)

It's easier to hire in Japan when you already have Sony and the optics makers (Nikon, Canon etc) with past knowledge in fabbing than in many other countries.

That said, I think TSMC had plans for factories basically anywhere they could, not laying they eggs in any single basket. The difference will probably be in process levels.

makes sense, I thought that the situation is similar in Germany (Zeiss, experience in fabbing too* ) but I guess Japan is more stable now *afaik TSMC will open a fab in Germany, but for bigger nm's for car industry

Place like France or the Netherlands are safer than Japan, but not by _that_ much, and Japan is far less likely to end up in a trade war with the US. Europe may have had a tough few years, but it's pretty much the one place on Earth that could be capable of going toe-to-toe with North America.

>I think EU is a safer bet compared to Japan

EU's members in NATO have the same safety guarantee provided by the USA as Japan. The US and Japan have a direct alliance that is a treaty guaranteed by the US if Japan is attacked compared to Ukraine's Budapest Memorandum.

EU is no safer than Japan. see the current Ukraine-Russia war.

Japan and Taiwan have similar work cultures and are far closer to Taiwan compared to the US. A 45-minute plan ride for TSMC's higher-up to check on their fabs in Japan is far easier

>EU is no safer than Japan. see the current Ukraine-Russia war.

Ok, IDK why I keep seeing this but I have yet to see any basis in objective reality for it.

Two years ago? Sure. But now with Russia having expended a minimum of half of all of its military stores on a proxy war with a previously insignificant former Soviet vassal, there is no rational reason to consider them to be a security threat in the region. Yes they still have enough nukes to level pretty much everything but if that happens, everywhere becomes insecure so it's not a real consideration.

The worst possible case for the EU (aside from general nuclear exchange, but again...) at this point is some sort of migrant tsunami (pun intended I guess) from a failing Russian state and subsequent loss of stability in neighboring regions to the south but I'd absolutely take that over being within rock throwing distance of an exponentially militarizing China, particularly given that the EU and China aren't even really adversaries in any meaningful way.

> there is no rational reason to consider them to be a security threat in the region

Putin is a dictator. Actions irrational for Russia could be rational for him. Would NATO risk broad conflict if he puttered around in the Baltics? I think so. But do I know so? And wouldn’t challenging NATO like that play rather well domestically?

Japan is one of the most armed countries in the world and one of the most defensible. I highly doubt that Japan is less safe than most of the EU in practical terms.

Kumamoto, where TSMC's plant is located, is rich in water resources. I don't know much about it, but it seems that water resources are essential to the semiconductor manufacturing process.

The way I see it, which could be a naive way, is that building chips in the USA is never going to make strict economic sense. The cost of living is just too high here. But what is the value of chip autonomy, supply chain security, IP security, etc and who will pay for it?

Question I've had for a while now: Why doesn't the UK have any serious chipmaking facilities?

They'd be the perfect location wouldn't they? Abundant cheap water, absence of earthquakes, the usual business friendly environment...?

Those 3 are not that important. You need either a significant high tech manufacturing industry (US post-war, Japan, Korea, Taiwan are all significant manufacturing nations), you need significant capital (debatable for UK), you need talent (UK has no history of chip manufacturing), etc.

Relevant: https://www.youtube.com/watch?v=AE0DEOictVg

Is it feasible to a third world country to build a foundry like this? I've been curious about this for a long time. There are very few high end foundries in the world. It seems like a major supply chain risk.

Perhaps In-Q-Tel will mysteriously get $20b in new AUM.

"Abenomics" was a serious failure. Maybe Japan is starting to turn the corner, finally.

Japan not gonna go to war for you when China invades, bros

hahahahahaha

Nice choice, no way they will allow Chinese communists to get their hands on high tech.

Japan still not gonna go to war for you when China invades, bros

Reply on news.ycombinator.com