According to Money Today [1] the chip center is actually to be located in Yongin, Gyeonggi-do---about a hour away from the center of Seoul. Samsung is the primary investor to this center and will invest 300 trillion won (that's probably where $230B came frome) over next 20 years.
For those who don't know much about SK, Yongin is a city of over a million that's right next to/south-east of Seoul -- considered sort of the major suburb region. It's about as large as Seoul area-wise, but is one of the fastest growing urban areas in the country. Thus it has generally better infrastructure planning, roads, newer buildings, slightly more affordable housing, etc. making it very desirable as a location for families to live in with not terrible commutes into Seoul if needed (about an hour).
Seoul, while an amazing and dynamic city, can be a bit of an urban snarl. The layers of urban development, especially the post-war decades can be a bit dodgy, expensive and not all that great to raise a family in.
Living-in, or commuting to Yongin is far more desirable than Seoul at the moment.
Comparing "Seoul" to "Yongin" is not very good in the grand scheme of things. Seoul has a metro area which doesn't include Yongin, but it would be unwise to consider Yongin outside of Seoul since it's basically a satellite city connected to Seoul. The name for this is Gyeongi but that's another thing. Most of Gyeongi is still basically Seoul in natural form. The same way Greater London is part of London but separate from the City of London.
Because Seoul is so big, it's more akin to calling Yongin a "suburb" of Seoul, and Yongin has its own suburbs.
Seoul and Yongin compared area wise would be the worst since you have to first set the definition of Seoul. Seoul covers Yongin in most natural definitions of a city, but in reality the Seoul inner region is its own and is obviously tiny. It's better to call it Seoul metro area (Seoul Capital Area).
Yongin is an old city, it doesn't have better infrastructure planning at all, but it doesn't have a lot of super old infrastructure which is great for future planning. Be it due to budget or just slow crawl of improvement, who knows. It's desirable because it has everything you need, now including Costco, IKEA, etc. and it's got land. That's all.
Your last answer is weird. There's nothing better about Yongin than living in Seoul, except for possibly commutes. It's all a matter of perspective and negatives. I don't think either are objectively better but living in Seoul is definitely more sought after and "higher" status.
North Korea's not a threat to anything except rogue ROK fishermen. They can hardly get fuel into the country. The citizens have to result to eating tree bark. When one of their soldiers fled to the south he had ten different intestinal parasites...Their entire existence is a charade and it's being milked to sell weapons. Shit is inhumane. And I've been to Seoul dozens of times. A lot of the locals are afraid of reunification because it might tank their condo values (or at least that's what I was told).
I was saying that because -- in my humble opinion -- the ongoing farce with North Korea is as much on the ROK as it is with the US...There's a significant segmentation of South Korean's that are just as unlikely to push for re-unification as the US. They're glad to play the game with the Pentagon if that means their bottom line isn't hurt. Which is wild. There's a huge portion of Korea that's been locked into hell for sixty+ years. Enough with the bullshit.
Easily within range of bigger rocket artillery. The security point prolly wasn't considered when selecting this location otherwise they would have build it somewhere on their southern coast. Of-course most of the engineering talent is near Seoul, so that leaves them with no choice.
It may matter tremendously if SHTF, as a difference between few missiles that get trough Patriot defenses hitting some random part of fab vs fab looking like Mariupol steel plant or Bakhmut. russian war in Ukraine gives us some very good lessons is east vs west current warfare, if you care to look for them
I don't think "production during the war" is the point. I believe the point GP is making is that production can continue after hostilities cease if and only if the plant is intact.
Being beyond the range of 155mm artillery makes that much more likely.
Sure, but not their conventional missiles. That your high value targets are outside of range of the opponents very cheapest weapon systems seems like a very mild benefit.
It is crazy to me, that this investment will be all in Seoul.
It is not very hard to make the case that Seoul is overly concentrated for both politics and the economy at the expense of other areas of South Korea. And as far as security balwark goes, Seoul is only 40km from the DMZ.
The location will actually be Gyeonggi-do, south-east of Seoul.
This area is close to the existing semiconductor regions of S.Korea, such as Pyeongtaek, Giheung, etc.
So it's quite an organic decision to select such a location, as also all the suppliers and customers are already located in this region.
Relocating to an entirely different region would just delay the process and actually risk to collapse the economy for quite mundane reasons.
Other Industries have their center of weight located more south, like Automotive and Energy.
One "mundane reason" to imagine the complexity:
People now working in semiconductor sector all live around Seoul. Apartments in Korea are not rented but predominantly purchased on loan.
Just announcing to pull a whole industry further away from Seoul would immediately drive up housing prices of the destination and collapse the housing prices in Seoul. So people selling to relocate could not pay off their loans with that money and could not acquire at the destination due to huge debt.
Seoul real-estate market could collapse, and since most of it is owned on loan, it would at least immobilize the whole Seoul region (as people could not relocate without risking bancrupcy), at worst cause banks to no longer accept the apartments as sufficient security for the loan.
I think you're grossly exaggerating the real estate implications. There are 25 million people in the capital region, and only a small fraction of them work in the semiconductor industry. Relocating 200-300K people to a new industrial center might ease the pressure on housing a little bit, but there are millions more waiting to move in.
Moreover, Seoulites really, seriously, desperately don't want to move out. The government has tried many times to create new cities away from Seoul with stable public-sector jobs; but a lot of people who work in Sejong, for example, just keep living in Seoul and make the 250km round trip every day. The 300K who actually moved to Sejong didn't make a dent in housing prices anywhere else, either.
> Just announcing to pull a whole industry further away from Seoul would immediately drive up housing prices of the destination and collapse the housing prices in Seoul.
That’s a win for the average person in Seoul. Housing prices are destroying this generation of young people.
Are housing prices in Seoul seeing the same inflation we see in many western cities? I'd be mildly surprised. (Anecdotally) It seems like Asian cultures are much less likely to treat housing as an investment and/or allow the few to dictate to the many (specifically in the housing arena).
>It seems like Asian cultures are much less likely to treat housing as an investment
Isn’t China one of the most prominent examples of using real estate as investment? There’s even whole ghost towns of apartment buildings that are constructed for this purpose.
yea all my Chinese (in China) relatives assume you should only invest in housing. And then when we tell them we have zero investment properties, they get really, really concerned for our future welfare and think that we're not doing very well.
No, domestic savers have capital controls and do not have the means to move money easily outside the country. The stock market also is not reliable in China as an avenue of investment. Government pensions are also small. So that mostly leaves real estate.
Evergrande Group is emblematic of the real estate industry’s trouble there. Property is 15-30% of Chinese GDP and much of it goes towards investment property.
You dig into a lot of statements that don't really relate to this. It's not about housing prices. It's been proven that building more stuff outside of Seoul's greater region will not decrease Seoul housing prices...
It's just that leaving Seoul is not worth it for anyone involved.
Leaving aside the DMZ/NK Risk... Cities work. Bigger cities work better than smaller cities. They're cheaper, cleaner, greener, more productive, with better health and social outcomes and more access to education etc etc etc.
That's why basically any large project basically NEED to be within X minutes of a major city. The saving in land cost setting it up in a rural area is nothing compared to having no staff, no services, no infrastructure, local political problems etc.
Over half of South Koreans live in the Seoul Capital Area. That’s a much higher proportion than countries other people often accuse of being overcentralized and neglecting the regions, like the UK with London or France with Paris.
And to some degree what you describe is a self fulfilling prophecy; big cities hollowing out the smaller ones, shrinking their economies and leaving their governments in deficit and the services worse, rinse and repeat.
You are completely right that this doesn't have a good outlook, but a self-fulfilling prophecy works in practice because of the societal friction.
SK government had probably one last chance to reverse its course, when it tried to move the administrative capital to Sejong in 2003, but it was met with hard backlashes including a very creative constitutional decision. Keep in mind that Sejong is not very far from Seoul anyway, and yet it failed; Sejong does host many administrative facilities, but is short from being the capital in any sense.
It's a nice self fulfilling prophecy. Smaller cities shouldn't exist. Imagine the internship opportunities available to high school students that kids can get in a big city vs small ones. My peers had tech companies in their hometowns whereas I didn't. Small cities put children in a disadvantage and should not exist. Big cities also have the scale to have specialized schools that match students with their interests while small cities just have a classroom for the whole grade.
When we are saying smaller cities in South Korea we are still talking about million plus cities. There are five of them not in Seoul’s orbit.
Germany famously lacks a dominant city and has a pretty good economy. So does the US. Decentralization is also diversification; if one region is falling behind it does not take the whole country with them.
decentralization provides certain benefits at the cost of pure efficiency. People with your mindset are why our supply chains are still fucked up years after the Covid lockdowns. Maximizing efficiency kills durability and reliability
Big cities are way more expensive. People who live there have almost no kids. Rural and small cities have kept the US population from crashing completely. Big cities are currently dismantling their specialty high schools. I don't think the facts on the ground support that big cities are better for kids.
“People who live there have almost no kids” - that’s just not true and super North America specific. Some personal data points - my entire family was born and raised in city of >3M population, everyone whom I know in Europe with kids live in fairly big cities.
> Small cities put children in a disadvantage and should not exist
This is the kind of radical bat shit craziness, from posters who are so intelligent they can rationalize anything, craziness that I love to find on HN.
I went to high school in a specialized class in Sweden anyway and I didnt have any internship because if you study all your waking hours to get max grades you don't have time for interning.
Who said internships can't be fun? What I do as a software develop building CRUD apps is basically what I did as a teen, except working in a team is way more fun and there's senior people to help you if you have issues instead of spending 3 days trying out stackoverflow suggestions.
It's not a value judgement, it's just life. Villages should shrink to feed towns which should shrink to feed cities which should shrink to feed the largest regional conurbation. Trying to force the flow backwards like the UK does just makes everyone poorer and leads to worse outcomes in social terms.
Bigger cities are just more efficient and effective.
There isn’t any good reason why the outlying regions of the UK should have lower GDP per capita than Eastern Europe.
Seoul is extremely large, at 26M. Reversing even some of the flow would largely still be aiming at million plus cities in South Korea, of which there are five outside the Seoul Capital Area.
How does moving people from high productivity jobs in high productivity regions to low/low ones help anyone? That's the question here. There is no reason one area should have a higher or lower GDP (GDR) than any other. And if they do, that is not a moral judgement on that region. It's just a fact. People seem to want to reduce everything to the lowest level to make it "fair", but regions are not people, regions should NOT be treated fairly. People should be treated fairly. Regions (and countries) are just made up...
Who says it’s high/high and low/low rather than high/high and high/low?
Making people move, to a place that already has well documented housing and transportation shortages, is not really fair. Unequal GDP per capita is not fair either, particularly since that means less ability to pay for services, which are generally lower quality and much further away from homes.
For fabs it seems like just a bit outside of a city is the optimum location. Chandler (Phoenix) for Intel's new fab and outside of Portland for the other US ones. I expect it's similar in Taiwan and South Korea.
Side note but isn't water a critical input for a fab? Arizona's water problems are going to continue to get worse so this seems like a bad move by Intel.
Yeah, it makes sense to put it not in the centre but within (say) 1h of the centre. Then you can recruit from an 8 figure workforce including multiple universities, offer a huge list of places to live and amenities to workers, meet reliability with everyone from regulators to ambassadors to investors to lawyers, have gold standard infrastructure, etc.
I'm never sure when it comes to water. I think a lot of places in the west chose to have shortages by refusing to build infrastructure and wasting supply on subsidizing things like (export orientated, unnecessarily high water use forms of) agriculture. But I am no expert...
Chip supply chains are some of the most complex things humans have ever built and the amount of skilled work you need is crazy, so it really doesn't make much sense to put it anywhere else.
In general also many East-Asian cities don't really suffer from dysfunction and lack of development so agglomeration really has a lot of upsides.
I wish I had ready-to-go URLs I could reference, but the complexity is just stunning. For example, the purity of the feed chemicals has been a key limitation, and steady improvements have been made by suppliers over decades. There are orgs that specialise in things such as shipping ultra pure chemicals in specially cleaned teflon containers. Just trucking and piping this stuff around must have had millions (billions?) of dollars of research sunk into it. You don't get to 99.9999% or better purity while pumping stuff into a dirty old diesel truck's rusty tank!
The "tools" themselves are an extra level of crazy. For one, the ASML x-ray ("EUV") laser lithography machines are just an unbelievable piece of precision engineering. If twenty years ago you had told me that we'd have high-power x-ray lasers in an industrial setting with nanometer focusing precision I would have laughed in your face, or assumed that it's a function of a particle accelerator the size of a decent town.
I remember seeing an add for a new-generation tool (machine) designed for the permanent vacuum part of the production line. They were proudly showing off their key improvement, which was that its MTBF was something like 7 hours. This was amazing apparently, because both them and their competition couldn't do much better than 40 minutes previously!
Think about how demanding the requirements must be that a 40 minute MTBF was state of the art and totally acceptable in large-scale manufacturing, and raising that to "still less than a day" is worthy of advertising in an industry magazine!
They have tried building hubs further out of Seoul, but it just doesn't work. Why would people leave and tie a billion liabilities to their workplace which is not even easy to maintain in Seoul itself?
They want workers and they want Seoul. Workers want Seoul and they don't want to rely on their employers. This is always going to continue the fueling of Seoul border growth.
But workers want Seoul because already all the eggs are in the Seoul basket. Universities? Seoul. Jobs? Seoul. Services? Seoul.
The country has been overconcentrated in Seoul since at least the beginning of the dictatorship. Breaking the cycle will not happen overnight, and you don’t even need to build new cities, just invest in the other ones. (Sejong was probably a bad idea.)
So that's about 2.3x more money in 20 years than the U.S. plans to spend on the plant in Ohio that will supposedly create 50,000 jobs.
The minimum wage is $10.10 in Ohio vs. $7.28 with the latest increase in South Korea for 2023. That's almost 1.5x more wages for workers in Ohio.
And I wouldn't say we had the edge on chip manufacturing vs. S. Korea, so given the numbers above, how can the U.S. compete? What chips are we making vs. what is intended to be made in S. Korea?
Semiconductor manufacturing is not a zero sum game. We can increase the market by increasing the supply, and there are plenty of niches in the market for each manufacture to habit. More chips being produced is a net good for the world. I am more concerned that the chips are being used for something better than telemetry spyware or ad-tech spam than which stable democracy they are produced in.
A semiconductor fab does not require that many people so the wages should not matter that much.
And the relative amount invested is probably not that important. For the US it's important that there is at least some fairly high volume produced of the latest manufacturing nodes.
Right now nothing is produced in the US of the latest node, as they are manufactured in Taiwan and South Korea.
Whatever the number, relatively speaking, it's basically peanuts compared to the number and value of the wafers and chips they are spitting out.
It's advanced machines doing most of it, and there is an operator handling each machine. The machines are extremely expensive, the building is extremely expensive, the ventilation/filtration/air control system is expensive etc.
These are not smartphone factories.
What is expensive is running these fabs/lines at half the capacity.
I used min. wage as a Fermi solution - back of the envelope - since I am not an industry expert or insider and the product was an HN comment.
The article about the Ohio plant was quoted for the 50,000 jobs figure.
I am interested in manufacturing in general coming back to the U.S., so it intrigued me that S. Korea is making the "world's largest chip center" contrasted with all of the news I've been hearing over the past 5 years about chip manufacturing as an investment strategy in the U.S.
> fairly high volume produced of the latest manufacturing nodes
By the time a fab is completed, it is often no longer in the "latest manufacturing nodes", specifically when it was not in the "latest manufacturing nodes" even at planning stages.
Politicians of every nation increasingly understand that domestic chip production is a matter of national security (if not outright national sovereignty). In the long term, the upper bound on funds for chip production is "the entire defense spending budget".
USA can compete on quality and supply chain assurance. I'm not saying that SK chips are poor quality, but I imagine that a more expensive chip, or more expensive device is a small price to pay for reliable, accountable, and implant free assurance to the military or itar compliant corps.
also in the new and upcoming space age that is budding, quality assurance is a number one concern.
I think it's already mostly immigrants working in TSMC's Taiwanese foundries. I don't see it being very different in the US either as the wages are unlikely to be much higher than where the chips are currently made in Asia.
Not sure if that will be the case. I’d imagine they would pull from the same pool as auto manufacturers and other “anyone can apply” plant/manufacturing jobs.
It looks like a FAB worker makes about $36k. In my state the average UAW yearly income is about the same, in Arizona it’s $46k.[1]
Some google searches says a fab worker in Taiwan makes about $17k.
> I’d imagine they would pull from the same pool as auto manufacturers and other “anyone can apply” plant/manufacturing jobs.
What is this pool?
The cost of living in the US is (according to my very lazy Google search) about 4 times as high as in Taiwan on average. From what I understand from afar (I'm in the EU), $36k/year really isn't enough to buy a good quality of life in the US. I don't see anyone but those who have very little bargaining power working in these fabs...
My point was that whether it's the US, South Korea or Taiwan, the pool of available workers is probably the same and I suspect the birth rate of the country has very little impact on it.
This 230 b$ announcement is private funding over 20 years.
For a fair comparison, South Korea government's budget is around 500 b$ per year, and this chip center will be funded by circa 10 b$ per year from Samsung (with commodities and tax exemptions granted by the government).
There is a difference between creating a wafer with more chips on it than scaling up production of said wafers to be profitable. Think of it like producing cars. Many people can make cars, and on a small production scale. It's an entirely different ballgame for mass production.
It wasn't proposed as a immediate replacement of the current world model though.
This is like saying those new car manufacturers shouldn't bother because they'll never make then as fast and at high scale as the already established ?
Not so much. If we believe there is a 20% chance that Taiwan gets annexed, but the result of that would be catastrophic in terms of access to the chips that are the life blood of the modern world, it would make sense to prepare for that eventuality even if it weren't the most probable outcome, assuming the cost of that preparation is affordable.
I don't have a viewpoint per se. It is just that the size and location of the investment is unusual and comes against the backdrop of geopolitical tensions.
If you already have a foothold in South Korea, moving to anywhere south won't give you much safety (especially given NK's, although crude, launch capability). That's why Seoul is so close to the DMZ and still perfectly functioning.
The amount seems unprecedented and the timing. I understand competing for marketshare with TSMC, but it made me think like they are scrambling in case TSMC suddenly goes poof.
You have a viewpoint from the area that you come from. South Korea has a very different viewpoint to yours. Samsung is competing with TSMC, who is ahead and also investing massively. The only surprising thing about this headline is the amount, but makes sense when it is spread out over literally decades.
“The concentration of so much crucial chip manufacturing in just two places has caused concerns over global supply chain stability especially as South Korea and Taiwan are both militarily threatened by neighbors, North Korea and China respectively.”
Always with the Western bias.
If anything, given that there are U.S. troops in South Korea 10,000km and across an ocean from the U.S. mainland who is militarily threatening who? Long term Korea will be reunited, one hopes without conflict.
“The Korean conflict is an ongoing conflict based on the division of Korea between North Korea (Democratic People's Republic of Korea) and South Korea (Republic of Korea), both of which claim to be the sole legitimate government of all of Korea. During the Cold War, North Korea was backed by the Soviet Union, China, and other allies, while South Korea was backed by the United States, United Kingdom, and other Western allies.
The division of Korea by the United States and the Soviet Union occurred in 1945. […] The U.S. maintains a military presence in the South to assist South Korea in accordance with the ROK–U.S. Mutual Defense Treaty. In 1997, U.S. President Bill Clinton described the division of Korea as the "Cold War's last divide".”
As for Taiwan the cornerstone of U.S.–China relations is the one-China policy, a policy which officially states that the China–Taiwan problem is an internal Chinese problem – whatever you think about the status of the tensions between the mainland and the island is irrelevant, I'm talking about official U.S. foreign policy
“The United States formally acknowledged that "all Chinese on either side of the Taiwan Strait maintain there is but one China".”
As such the U.S. is violating this policy by having troops stationed in Taiwan, by expanding that troop presence, and by the constant shuttling back-and-forth of high level politicians.
My point is. Why does CNN insist on this throwaway paragraph when we all know that there are (at least) two sides to every conflict (if there weren't by definition we wouldn't have a conflict)? It is saddening that only the Western perspective is given. Why has mainstream establishment journalism in the West given up all pretenses to an unbiased neutral point of view?
The same goes for HN. I know this is a U.S.-centric site but what people say on here regarding China and Russia is myopic and jingoistic most of the time, as are the articles linked to. All you have to do is search for the terms "Russia" and "China" and sort by recent date to be confronted with comment after comment and page after page of repulsive Western chauvinism. Like it or not these are nuclear superpowers and we have to share the planet with them. The sooner the United States government and its people acknowledge this reality and step away from the constant brinkmanship and "national security" hypocrisy. The conflict in Ukraine is testament that the rest of the world has had enough of the West's arrogance.
So you think South Korea should just surrender and let themselves be ruled by a dictatorship? Any time someone complains about "western chauvanism", their answer to things is always to dismantle democracy and allow dictators to rule everything. Fuck that.
There is no violation of policy when both sides of the strait have differing views on what "one China" means. China can warn all it wants. It has been threatening Taiwan for decades. Frankly, Taiwan and the rest of Asia have had enough of China's arrogance.
I would add that recognition of the "One China" policy is just lip service for trade. Talk privately with any politician or leader of industry and they'll admit as much. Another way to view it is that it's merely acknowledgement of what China believes, the same strategy you might employ dealing with a petulant toddler.
>Frankly, Taiwan and the rest of Asia have had enough of China's arrogance.
Living in East Asia this is something I've observed when the subject comes up with locals. You're also going to hear similar if you watch the local news. It's certainly not an "anti-China" bias limited to the west.
Taiwan was not near enough to a communist dictatorship, so of course a government will invest to make chips even closer to an even crazier communist dictatorship.
And sorry for being political, I guess there are many big brain reasons for this (unironically), but it's funny
So where would you put it exactly? I'm guessing you think it could be put in some rural place in North America where there's no infrastructure and no workforce...
Or alternatively, in a country where doing any kind of labour organizing is liable to get you jailed and where a 70-hour working week is about to be enshrined in law.
Samsung is a South Korean company so...of course they'll build it in South Korea? Not really any reasonable places to build such a project in SK without being close to their local crazy communist dictatorship.
[1] https://news.mt.co.kr/mtview.php?no=2023031511112875044