IMF projects $43,511 for Korea this year, 49-50K next year so it's almost certain Korea will overtake Italy by 2027.
The nominal GDP also grew 26.2% in Q2 and IMF is projecting 2.25 trillion by 2026. Nominal GDP could also overtake Canada next year if the chip boom continues and the Won continues to strengthen against the Canadian dollar.
The per capita gap with Japan would also widen to almost 20,000 dollars by 2031
It is, specifically massive inflation in the price of chipsets being exported.
Over half of the growth in Real GDP was literally just chipsets. But the nominal figure exploded due to the massive increases in prices there. What's interesting is that doesn't really flow to broad inflation - so the real GDP figures are sorta understating just how much money is flowing in to Korea right now.
Real GDP isn't "nominal - inflation rate (like CPI)", it's the actual increase in real activity at a product/industry level. So the inflation adjustment here happened at the level of chipset prices, not broad inflation. To make an example: let's say you sold 100 iphones last year and sold 102 this year. Last year an iphone was $1,000, this year it's $1,500. Your GDP growth is 2% real. But your income went from 100,000 to 153,000. So a 53% increase. Many mistakenly presume that the calculation would be something like "53% increase in income - 3% increase in CPI = 50% increase" but that's not how GDP works - it deflates at more specific levels.
Basically, what this means is that while GDP had significant inflation, that doesn't mean the korean economy saw significant inflation, so the end result is that real GDP is really underselling how much growth is happening here. That's a super rare condition so most of the time this would be irrelevant (GDP Deflator generally runs slightly cooler than CPI for instance), but in this case it's very important to look at.
So yeah, the GDP deflator can show 22% inflation while Korea's broad measure of consumer inflation is something much more reasonable, and both of those can be perfectly accurate.
That's how the GDP deflator works. It's generally a very good gauge of inflationary impacts because GDP is usually not super isolated to one category that sees insane price appreciation in a short period of time.
Even still, it's not inaccurate, the measure is telling you exactly what it should be - how much of GDP is due to heightened output and how much is due to price appreciation. It's just that there's massive price appreciation in one sector that makes up a large portion of the output, but that has a small bearing on actual day to day costs for citizens.
So I can be tremendously richer but “real gdp” increase is 0 is this what you mean? Like say i am a qatar citize oil price double but barrels per year stay the same, everyone in qatar make significantly more this year but real GDP almost unchanged?
I mean, in this specific circumstance, yes that's possible. Oil could be another example. When GDP is more broad it's not an issue since the GDP deflators tend to track broad inflation somewhat close - for instance in the US the GDP deflator is generally a bit under CPI.
But yeah, ya just gotta understand what the data is telling you and why.
Most countries do this afaik. What’s hard to believe about it? Economists would want to know how much real activity is going on in the economy seperately from the nominal prices and CPI will not be as simple as nominal - real.
Yeah but is iphone 18 and 17 comparable? Say i sold the same amount of iphone this year, but ip18 comes with new technology like Apple intelligent or faceid(just an example) with a higher price i will assume gdp to increase
Korea TFR as of August= 0.914 (growing 3 straight years)
Spain TFR as of August = 1.12
Its not even that different considering Spain is being propped up by refugees and immigrants
Turns out your government being rich does help pull up birth rates
Immigration to continue replacing low wage dead end jobs are just kicking the can down the line. Europe's economic industries are looking pretty grim, just like S Korea's demographics. I wouldn't say one is particularly looking into a brighter future than the other.
Insane turnaround. In 1960 Korea's GDP per capita was lower than Ghana. Now overtaking Spain ? That's 60 years of industrial policy that actually worked.
Meanwhile Spain's productivity has been flat for 15 years.
Spain and most of southern Europe has had some real slow burn economic problems for a while now. Their median incomes continue to dwindle or remain stagnant, overall GDP growth is slow, productivity increases are nearly nonexistent. This trend started in the early 00s with the tech boom in the US not being felt in southern Europe, but really accelerated post GFC when many euro countries embraced austerity and dove head first in to a double dip recession. Many have never really recovered from those hits.
Exactly, and that's the key difference. Korea doubled down on capex and R&D during every crisis, southern Europe doubled down on austerity.
Look at 2008. Breaking point. Korea's corporate operating surplus was up 48% in Q2 this year alone, while Spain/Italy never rebuilt their industrial base. One bet on semiconductors and AI, the other on tourism and real estate.
The 2026 chip boom just made that gap visible in GDP per capita.
tbf, a major major portion of Korea's growth as of late has been NVDA specific orders and chipsets in general. While this is a good thing as it's generating significant wealth and growth there, it is also very narrow and tied to the AI craze in the US. Something like over 50% of their total GDP growth as of late has been the expansion in chipset exports.
So while this is a fantastic expansion in activity and wealth for them, it also may not be sustainable.
Great point, concentration risk is real. Over 50% of recent GDP growth from chip exports is textbook Dutch disease risk.Spain/Italy got tourism windfall and built more hotels. Korea gets AI windfall and builds more capacity. One is cyclical, one is structural. That's why I think this overtake sticks even if NVDA orders cool.
Yeah I would suggest that becoming a leader in chipset manufacture, and specifically one that's more isolated from the geopolitical risks that exist in Taiwan, is going to be very beneficial for them in the long run. But the context is important.
True, this year was brutal. If it goes back to 1330 range like Bank of Korea expects, Korea already clears $40k this year. Exchange rate was masking the real productivity gap.
If the average falls below 1457 Won, Korea clears 42K. It's currently 1456 Won and falling. GDP per capita is likely going to hit 44~45K by the end of 2026 given that Q3 and Q4 exports are coming in even stronger
That's huge. 1456 Won right now means we are already looking at $42K+ for 2026, not just $40K. If Q4 exports stay strong with HBM demand, $44-45K is totally in play.
That's the fastest nominal jump in almost 50 years right? 26.4% YoY nominal GDP in Q2. Wild.
Thanks for posting the original source btw, Sedaily had the best breakdown.
There's a lot of uneducated comments under this post that either pretend Spain was dramatically more developed than SK just a few decades ago or that assume Spain didn't have any productivity gain in the last 15 years. Both of those assumptions are wrong.
Yes, SK grew faster, but in the above graph I added Germany and Mexico as a "rich" and "poor" economy for scale. Clearly Spain and SK are both in the "up and coming" category.
Yeah but Korea's per capita is about to leapfrog all the way to 49.5K next year, passing Italy, and this might even be underestimating it considering the Won is strengthening against the USD. KRW this year was extremely weak from Jan~July, which inflated the average YTD currency rate used for the per capita calculations.
The average is currently 1456 Won for 1 USD whereas the currency today is 1340 Won
Spain isn't going to change much from this years' 41K
Korea is investing 1.3 trillion dollars to advance semiconductors and build new fabs so yes, more jobs are coming, but it takes time: https://fortune.com/2026/06/28/samsung-sk-hynix-chip-fabs-investment-fabs-1-3-trillion-10-years-nand/
Korea also announced 745 billion towards renewables, SMRs, new grids, power chips, green steel, fusion, and fission: https://www.koreajoongangdaily.com/korea/korea-commits-745-billion-for-carbon-neutrality-sustainable-growth/12909522
If the increase is largely due to semiconductors, how sustainable is it? When the cycle goes bust, we might see a drop in per capita GDP.
Apparently the above sentence is considered too short and not engaging with the sub or something like that, which is interesting. I mean, you can say a lot of things, but what you type is mostly fluff, or you can type something that's short and sweet, but quite meaningful. So going with number of words to gauge engagement is not quite what I consider very meaingful. I repeat, my initial post is considered too short and not engaging with the sub or something like that, which is interesting. I mean, you can say a lot of things, but what you type is mostly fluff, or you can type something that's short and sweet, but quite meaningful. So going with number of words to gauge engagement is not quite what I consider very meaingful.
I was wondering the same. But that applies to a number of different countries - for example gulf states with commodity prices. Or Canada for that matter.
The answer to that question will come down to how diversified their economy is.
Ok, but How much of this is related to AI overspending? Samsung, hynix, ... If all additional GDP generated from AI boom would have stopped, how much they will fall?
Pretty dependent, which is why today's chip selloff matters for the GDP story. IMF puts Korea near 43500 dollars per capita this year and almost 49000 next on the back of semiconductors. They upgraded growth for the same reason. If AI spend slows, that jump in living standards is mostly an export cycle, not a broad domestic boom.
It's really not, Samsung and SK hynix got paid huge prices for their RAM, that's past tense now. OpenAI or Anthropic may or may not crash but in either case Samsung got paid. Of course the KOSPI would decline but hardly GG
They have contracted revenue and order commitments, they haven’t gotten paid yet. Deposits probably 10-20% of purchase order so if they don’t ever get the money then it’s going to be a bit of an issue
Even if S Korea's economy regresses back to 2024, when memory chips were going through its worst cycle, it isn't all that bad. Just means the workers get back to being paid $60K a year, and even that's unlikely to happen. Even if AI development stops now, it already increased the baseline for memory demand by a lot.
The "bubble" mostly exists in companies like Antropic and OpenAI being valued at 2T dollars with no revenue, not Samsung and SK being valued at 1T profiting $280 billion dollars this year. Both Samsung and SK shares trade far below premium.
South Korea is having a very good decade not only in their semiconductors, but also military, shipbuilding, petroleum refining, cosmetics doing very well. Surprisingly well diversified for a country of their size. Let's see if they can defend well from China's pursuit.
So what’s your point? Rich getting richer and houses getting more expensive are universal in developed countries. That has nothing to do with the title.
It’s on par with Western Europe. So many of my friends in Korea get a lot of free handouts. They will work for a few months and quit and get unemployment checks while they go make art and music. The Korean government is also very generous to people who have kids.
People really blow shit out of proportion because they watched Squid Game and Parasite. Koreans also love to openly criticize their society and government.
Korea's wealth inequality and income inequality is very mild if you compare it to the U.S. Don't tell this to Koreans tho, they don't believe this at all
It is not as simple as that. You’re just seeing the tip of the iceberg. There are about 70 countries in the semiconductor supply chain. It’s one of the most delocalized end products that exists, only possible through vast local specialization across the world. Everything from raw materials production, to processing, to manufacturing equipment manufacturing, to production of specialty items like photoresists, to education, to chip design is distributed across the world. It would be impossible to bring all that under one flag. But that said, in the US we are bringing production in-house.
Because of a lot of asymmetries along the supply chain. The smile curve, IP rights, margins of various steps in the supply chain, stability of local markets, etc.
There are 3 main fabs (chip mfg facilities) in the world : Samsung, Intel and tsmc. Tsmc has about 80% of the market. Samsung is based out of Korea and Intel is based out of the USA.
The memory makers also have their own fabs, albeit they’re not as relevant. The two largest memory makers are in South Korea: Samsung and Sk Hynix.
But when it’s all said and done, it’s actually not the manufacturing of chips that makes the most money. It’s the design of them, which the US is still a clear #1.
But the US is also aware of the supply
Chain vulnerability by not having your chips made in the US as exposed by Covid, hence the chips act and serious attempts to bring chip manufacturing to the US. The thing is it takes time, but it’s happening. Tsmc has been building massive fab in Arizona and Samsung just built their biggest fab in Texas. The government has also essentially subsidized Intel with a blank check to get their mfg capability up.
Because the Western world literally gave up chip manufacturing in the late 20th century. It’s ironic that most people (except Taiwanese and Korean) have no clue and interests about chips manufacturing business 5 years ago and now everyone think it’s the spice from Dune.
Taiwan and Korea produce way more per capita PhD’s in hard comp sci and electrical engineering than anywhere else in North America or Europe. It’s also dramatically cheaper to employ them there.
They are being built in North America. Oregon (Intel), Arizona (Intel, TSMC), Texas (Samsung, Texas Instruments), New York (Micron, Global Foundries), and Idaho (Micron). all have massive chip making operations.
The issue is that all the revenue generated through exports is usually in foreign currency, so you cannot easily use this to invest into your own country. They'll have to invest the excess profits in other countries which doesn't neccessarily benefit the general population.
[OP] rdh2dmd | 7 hours ago
IMF projects $43,511 for Korea this year, 49-50K next year so it's almost certain Korea will overtake Italy by 2027. The nominal GDP also grew 26.2% in Q2 and IMF is projecting 2.25 trillion by 2026. Nominal GDP could also overtake Canada next year if the chip boom continues and the Won continues to strengthen against the Canadian dollar.
The per capita gap with Japan would also widen to almost 20,000 dollars by 2031
BrokenManOfSamarkand | 6 hours ago
Japan really fumbled the bag. It's insane to think they were on track to be an economic superpower.
Green_Space729 | 5 hours ago
Wasn’t it the US that rat fucked their chips industry and currency?
BrokenManOfSamarkand | 5 hours ago
They've been declining for a long time
Caster0 | 5 hours ago
Japan still has some extremely solid companies though and their talent is still incredible.
It truly is weird how their economy is still flat
ProbablySatan420 | 3 hours ago
It’s been flat because the overall economy has been stuck in the year 2000 since 1980.
Caster0 | 2 hours ago
Yeah, it seems like a glaring issue is their conservative economic structure, which inhibits rewarding risk takers.
It's a shame too as they have the talent and resources to succeed and was easily the most dominant Asian country for a while even after ww2
Thadlust | 5 hours ago
No, Japan had structural issues beyond the Plaza Accords. As a counterfactual, West Germany did just fine
The_Admiral_ | 5 hours ago
Letting yourself get rat fucked isn't a good strategy? Someone needs to inform Europe.
Iron-Fist | 6 hours ago
Not sure what numbers this is pulling but they are forecasted to 3.3% for the year. Not bad for a developed economy (Spain is at 2.5%) but not 26% lol
https://finance.biggo.com/news/4d42a987-2052-4527-a219-6631c05178ad
Suspicious_Row_6136 | 6 hours ago
Nominal vs Real. Nominal growth is 26%, real growth is 3.3%. Different things
Zinch85 | 6 hours ago
That would mean a 22% inflation. I don't think that's the case
RIP_Soulja_Slim | 6 hours ago
It is, specifically massive inflation in the price of chipsets being exported.
Over half of the growth in Real GDP was literally just chipsets. But the nominal figure exploded due to the massive increases in prices there. What's interesting is that doesn't really flow to broad inflation - so the real GDP figures are sorta understating just how much money is flowing in to Korea right now.
Real GDP isn't "nominal - inflation rate (like CPI)", it's the actual increase in real activity at a product/industry level. So the inflation adjustment here happened at the level of chipset prices, not broad inflation. To make an example: let's say you sold 100 iphones last year and sold 102 this year. Last year an iphone was $1,000, this year it's $1,500. Your GDP growth is 2% real. But your income went from 100,000 to 153,000. So a 53% increase. Many mistakenly presume that the calculation would be something like "53% increase in income - 3% increase in CPI = 50% increase" but that's not how GDP works - it deflates at more specific levels.
Basically, what this means is that while GDP had significant inflation, that doesn't mean the korean economy saw significant inflation, so the end result is that real GDP is really underselling how much growth is happening here. That's a super rare condition so most of the time this would be irrelevant (GDP Deflator generally runs slightly cooler than CPI for instance), but in this case it's very important to look at.
So yeah, the GDP deflator can show 22% inflation while Korea's broad measure of consumer inflation is something much more reasonable, and both of those can be perfectly accurate.
JoJo_Embiid | 3 hours ago
Is this the way SK calculated it or every country do that? Hard to believe. I never know this in my life
RIP_Soulja_Slim | 3 hours ago
That's how the GDP deflator works. It's generally a very good gauge of inflationary impacts because GDP is usually not super isolated to one category that sees insane price appreciation in a short period of time.
Even still, it's not inaccurate, the measure is telling you exactly what it should be - how much of GDP is due to heightened output and how much is due to price appreciation. It's just that there's massive price appreciation in one sector that makes up a large portion of the output, but that has a small bearing on actual day to day costs for citizens.
JoJo_Embiid | 3 hours ago
So I can be tremendously richer but “real gdp” increase is 0 is this what you mean? Like say i am a qatar citize oil price double but barrels per year stay the same, everyone in qatar make significantly more this year but real GDP almost unchanged?
RIP_Soulja_Slim | 2 hours ago
I mean, in this specific circumstance, yes that's possible. Oil could be another example. When GDP is more broad it's not an issue since the GDP deflators tend to track broad inflation somewhat close - for instance in the US the GDP deflator is generally a bit under CPI.
But yeah, ya just gotta understand what the data is telling you and why.
JoJo_Embiid | 2 hours ago
thannks, learnt something new today
ProbablySatan420 | 3 hours ago
Most countries do this afaik. What’s hard to believe about it? Economists would want to know how much real activity is going on in the economy seperately from the nominal prices and CPI will not be as simple as nominal - real.
JoJo_Embiid | 3 hours ago
Yeah but is iphone 18 and 17 comparable? Say i sold the same amount of iphone this year, but ip18 comes with new technology like Apple intelligent or faceid(just an example) with a higher price i will assume gdp to increase
RIP_Soulja_Slim | 2 hours ago
In the real world there's quality adjustments to account for new features and what not that also impact these price measures.
Suspicious_Row_6136 | 6 hours ago
https://www.koreaherald.com/article/10865710
[OP] rdh2dmd | 6 hours ago
That's real gdp. Nominal is 26% https://m.koreaherald.com/article/10866539
probably_4_human | 45 minutes ago
Spain works fewer hours and have more births than Korea though
They think more of the future
[OP] rdh2dmd | 31 minutes ago
Korea TFR as of August= 0.914 (growing 3 straight years)
Spain TFR as of August = 1.12
Its not even that different considering Spain is being propped up by refugees and immigrants Turns out your government being rich does help pull up birth rates
probably_4_human | 27 minutes ago
South Korea has spent 200 billions in family support and the fertility rate has crashed below 1.
Spain isn't doing great. But immigration can alleviate the crisis.
Ok-Explanation5605 | 30 minutes ago
isn't Spain's birth rate at like 1.1? Spain works fewer hours because there's basically no job outside of tourism and banking.
probably_4_human | 23 minutes ago
They've been adding up immigrant population though. Their population trajectory is much more healthy than Korea
Ok-Explanation5605 | 20 minutes ago
Immigration to continue replacing low wage dead end jobs are just kicking the can down the line. Europe's economic industries are looking pretty grim, just like S Korea's demographics. I wouldn't say one is particularly looking into a brighter future than the other.
YieldBug | 6 hours ago
Insane turnaround. In 1960 Korea's GDP per capita was lower than Ghana. Now overtaking Spain ? That's 60 years of industrial policy that actually worked. Meanwhile Spain's productivity has been flat for 15 years.
RIP_Soulja_Slim | 6 hours ago
Spain and most of southern Europe has had some real slow burn economic problems for a while now. Their median incomes continue to dwindle or remain stagnant, overall GDP growth is slow, productivity increases are nearly nonexistent. This trend started in the early 00s with the tech boom in the US not being felt in southern Europe, but really accelerated post GFC when many euro countries embraced austerity and dove head first in to a double dip recession. Many have never really recovered from those hits.
YieldBug | 6 hours ago
Exactly, and that's the key difference. Korea doubled down on capex and R&D during every crisis, southern Europe doubled down on austerity. Look at 2008. Breaking point. Korea's corporate operating surplus was up 48% in Q2 this year alone, while Spain/Italy never rebuilt their industrial base. One bet on semiconductors and AI, the other on tourism and real estate. The 2026 chip boom just made that gap visible in GDP per capita.
RIP_Soulja_Slim | 6 hours ago
tbf, a major major portion of Korea's growth as of late has been NVDA specific orders and chipsets in general. While this is a good thing as it's generating significant wealth and growth there, it is also very narrow and tied to the AI craze in the US. Something like over 50% of their total GDP growth as of late has been the expansion in chipset exports.
So while this is a fantastic expansion in activity and wealth for them, it also may not be sustainable.
[OP] rdh2dmd | 2 hours ago
This will change as custom HBM orders increase from AMD, Google, Broadcom, OpenAI, and others
YieldBug | 6 hours ago
Great point, concentration risk is real. Over 50% of recent GDP growth from chip exports is textbook Dutch disease risk.Spain/Italy got tourism windfall and built more hotels. Korea gets AI windfall and builds more capacity. One is cyclical, one is structural. That's why I think this overtake sticks even if NVDA orders cool.
RIP_Soulja_Slim | 5 hours ago
Yeah I would suggest that becoming a leader in chipset manufacture, and specifically one that's more isolated from the geopolitical risks that exist in Taiwan, is going to be very beneficial for them in the long run. But the context is important.
Suspicious_Row_6136 | 6 hours ago
SK was actually slightly higher than Spain for a few short years before the KRW tanked
YieldBug | 6 hours ago
True, this year was brutal. If it goes back to 1330 range like Bank of Korea expects, Korea already clears $40k this year. Exchange rate was masking the real productivity gap.
[OP] rdh2dmd | 2 hours ago
If the average falls below 1457 Won, Korea clears 42K. It's currently 1456 Won and falling. GDP per capita is likely going to hit 44~45K by the end of 2026 given that Q3 and Q4 exports are coming in even stronger
YieldBug | an hour ago
That's huge. 1456 Won right now means we are already looking at $42K+ for 2026, not just $40K. If Q4 exports stay strong with HBM demand, $44-45K is totally in play. That's the fastest nominal jump in almost 50 years right? 26.4% YoY nominal GDP in Q2. Wild. Thanks for posting the original source btw, Sedaily had the best breakdown.
shatureg | 4 hours ago
This is not that big of a story? Spain's and SK's GDP/capita curves followed each other closely for decades and Spain just overtook SK a couple years ago: https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?locations=ES-KR
There's a lot of uneducated comments under this post that either pretend Spain was dramatically more developed than SK just a few decades ago or that assume Spain didn't have any productivity gain in the last 15 years. Both of those assumptions are wrong.
Both Spain and SK have undergone significant growth in the last few decades and have had similar levels of development and GDP/capita (both nominally and PPP adjusted) since at least the 90s: https://data.worldbank.org/indicator/NY.GDP.PCAP.PP.CD?locations=ES-KR-DE-MX
Yes, SK grew faster, but in the above graph I added Germany and Mexico as a "rich" and "poor" economy for scale. Clearly Spain and SK are both in the "up and coming" category.
[OP] rdh2dmd | 3 hours ago
Yeah but Korea's per capita is about to leapfrog all the way to 49.5K next year, passing Italy, and this might even be underestimating it considering the Won is strengthening against the USD. KRW this year was extremely weak from Jan~July, which inflated the average YTD currency rate used for the per capita calculations. The average is currently 1456 Won for 1 USD whereas the currency today is 1340 Won
Spain isn't going to change much from this years' 41K
Tkins | 2 hours ago
Will this be similar to Ireland though where the GDP per Capita skyrockets but your average joe doesn't really see much of that wealth?
[OP] rdh2dmd | 2 hours ago
Ireland was propped up by foreign big tech setting up shop. Most of that boom wasn't even really "theirs"
Tkins | 2 hours ago
So in this case, are we seeing more and higher paying jobs or is most of the wealth going to higher chip prices?
[OP] rdh2dmd | 2 hours ago
Korea is investing 1.3 trillion dollars to advance semiconductors and build new fabs so yes, more jobs are coming, but it takes time: https://fortune.com/2026/06/28/samsung-sk-hynix-chip-fabs-investment-fabs-1-3-trillion-10-years-nand/
Korea also announced 745 billion towards renewables, SMRs, new grids, power chips, green steel, fusion, and fission: https://www.koreajoongangdaily.com/korea/korea-commits-745-billion-for-carbon-neutrality-sustainable-growth/12909522
Ok-Disaster-551 | 6 hours ago
If the increase is largely due to semiconductors, how sustainable is it? When the cycle goes bust, we might see a drop in per capita GDP.
Apparently the above sentence is considered too short and not engaging with the sub or something like that, which is interesting. I mean, you can say a lot of things, but what you type is mostly fluff, or you can type something that's short and sweet, but quite meaningful. So going with number of words to gauge engagement is not quite what I consider very meaingful. I repeat, my initial post is considered too short and not engaging with the sub or something like that, which is interesting. I mean, you can say a lot of things, but what you type is mostly fluff, or you can type something that's short and sweet, but quite meaningful. So going with number of words to gauge engagement is not quite what I consider very meaingful.
OK_x86 | 6 hours ago
I was wondering the same. But that applies to a number of different countries - for example gulf states with commodity prices. Or Canada for that matter.
The answer to that question will come down to how diversified their economy is.
Junkererer | 3 hours ago
Semiconductors, datacenters etc are the future, there's no going bust. Some stocks may be overhyped but chips are a quite safe bet
Michal_F | 6 hours ago
Ok, but How much of this is related to AI overspending? Samsung, hynix, ... If all additional GDP generated from AI boom would have stopped, how much they will fall?
Just check this year's Kospi crash ... ;)
Rumblestillskin | 6 hours ago
I expect large but I also think that chip spending will continue into the future even if and when AI stocks crash.
Rabitai_Trades | 5 hours ago
Pretty dependent, which is why today's chip selloff matters for the GDP story. IMF puts Korea near 43500 dollars per capita this year and almost 49000 next on the back of semiconductors. They upgraded growth for the same reason. If AI spend slows, that jump in living standards is mostly an export cycle, not a broad domestic boom.
[OP] rdh2dmd | 5 hours ago
KOSPI went from 2300 in 2025 to 6000 today.
It's a pullback not a crash
MelancholyTurtle95 | 5 hours ago
Yeah but broader AI bubble hasn’t popped yet. When it does it’s GG for South Korea
Titan-Titillator | 4 hours ago
It's really not, Samsung and SK hynix got paid huge prices for their RAM, that's past tense now. OpenAI or Anthropic may or may not crash but in either case Samsung got paid. Of course the KOSPI would decline but hardly GG
MelancholyTurtle95 | 4 hours ago
They have contracted revenue and order commitments, they haven’t gotten paid yet. Deposits probably 10-20% of purchase order so if they don’t ever get the money then it’s going to be a bit of an issue
Ok-Explanation5605 | 35 minutes ago
Even if S Korea's economy regresses back to 2024, when memory chips were going through its worst cycle, it isn't all that bad. Just means the workers get back to being paid $60K a year, and even that's unlikely to happen. Even if AI development stops now, it already increased the baseline for memory demand by a lot.
The "bubble" mostly exists in companies like Antropic and OpenAI being valued at 2T dollars with no revenue, not Samsung and SK being valued at 1T profiting $280 billion dollars this year. Both Samsung and SK shares trade far below premium.
TopBalance438 | 4 hours ago
It's gg for the world at that point
WeedForReddit | 3 hours ago
South Korea is having a very good decade not only in their semiconductors, but also military, shipbuilding, petroleum refining, cosmetics doing very well. Surprisingly well diversified for a country of their size. Let's see if they can defend well from China's pursuit.
[OP] rdh2dmd | 2 hours ago
Also EVs and entertainment IP
KJongsDongUnYourFace | 7 hours ago
Hasn't Koreans wealth disparity steadily risen over the last 20 years?
Houses become unaffordable and the top 10 percent hold more wealth than ever before
[OP] rdh2dmd | 7 hours ago
Korea's Gini coefficient is actually pretty good relative to other Western nations
lqIpI | 7 hours ago
Either you are in the RAM/NAND loop or not
The closer in you are, the larger the gap to the outside is
SK Hynix workers are on track for a nearly $1M bonus next year
charliehu1226 | 7 hours ago
How affordable are houses in Europe and America?
KJongsDongUnYourFace | 7 hours ago
Not affordable in the slightest to the lower working class. The same can be said about Korea?
charliehu1226 | 7 hours ago
So what’s your point? Rich getting richer and houses getting more expensive are universal in developed countries. That has nothing to do with the title.
KJongsDongUnYourFace | 7 hours ago
The rich getting richer increases GDP per capita. It doesnt appropriately measure how people live and function
charliehu1226 | 6 hours ago
There’s no causation between wealth disparity and GDP per capita. I don’t know where did you get that idea.
KJongsDongUnYourFace | 6 hours ago
I dont have that idea. Im saying that GDP per capita isnt a great measure of success.
ProbablySatan420 | 3 hours ago
It generally is. You can look at median per capita income which is also growing significantly.
Ferreman | 7 hours ago
When it comes to median wealth per citizen, Korea is only #18. When it comes to home ownership they only have like 50% of its citizens that own homes.
KJongsDongUnYourFace | 7 hours ago
They also have the highest OECD poverty rate for old age citizens (pushing 45 percent).
Firm-Sprinkles-7702 | 4 hours ago
Tbh it makes sense when you realize South Korea was the 2nd poorest country not that long ago
ProbablySatan420 | 3 hours ago
Yes and that is due to South Korea’s background. It’s good to see the development from a 3rd world to a 1st world countries 3 generations
Ferreman | 7 hours ago
Wow thats insane
mnugget1 | 6 hours ago
In Korea there's something that exists between ownership and renting called jeonse that a lot of people use in lieu of full monthly renting.
CarpenterAlive5082 | 5 hours ago
It’s on par with Western Europe. So many of my friends in Korea get a lot of free handouts. They will work for a few months and quit and get unemployment checks while they go make art and music. The Korean government is also very generous to people who have kids.
People really blow shit out of proportion because they watched Squid Game and Parasite. Koreans also love to openly criticize their society and government.
fistoic | 6 hours ago
Korea's wealth inequality and income inequality is very mild if you compare it to the U.S. Don't tell this to Koreans tho, they don't believe this at all
Clone95 | 2 hours ago
Housing affordability declines with increasing wealth globally. It’s the transition away from a laborer economy to an info one.
GuyWithLag | 6 hours ago
And this is why GDP per capital is a shit indicator and used to intentionally misrepresent the per person income.
Independent-Size-258 | 7 hours ago
Why are these chip building so concentrated in Korea and Taiwan? Why can't these be built in North America or Europe?
If AI is truly the future, the Western world needs to bring high tech manufacturing in house.
GurProfessional9534 | 7 hours ago
It is not as simple as that. You’re just seeing the tip of the iceberg. There are about 70 countries in the semiconductor supply chain. It’s one of the most delocalized end products that exists, only possible through vast local specialization across the world. Everything from raw materials production, to processing, to manufacturing equipment manufacturing, to production of specialty items like photoresists, to education, to chip design is distributed across the world. It would be impossible to bring all that under one flag. But that said, in the US we are bringing production in-house.
VCeral | 5 hours ago
But then why is the value concentrated in a few countries?
GurProfessional9534 | 5 hours ago
Because of a lot of asymmetries along the supply chain. The smile curve, IP rights, margins of various steps in the supply chain, stability of local markets, etc.
UnderstandingThin40 | 6 hours ago
There are 3 main fabs (chip mfg facilities) in the world : Samsung, Intel and tsmc. Tsmc has about 80% of the market. Samsung is based out of Korea and Intel is based out of the USA.
The memory makers also have their own fabs, albeit they’re not as relevant. The two largest memory makers are in South Korea: Samsung and Sk Hynix.
But when it’s all said and done, it’s actually not the manufacturing of chips that makes the most money. It’s the design of them, which the US is still a clear #1.
But the US is also aware of the supply Chain vulnerability by not having your chips made in the US as exposed by Covid, hence the chips act and serious attempts to bring chip manufacturing to the US. The thing is it takes time, but it’s happening. Tsmc has been building massive fab in Arizona and Samsung just built their biggest fab in Texas. The government has also essentially subsidized Intel with a blank check to get their mfg capability up.
charliehu1226 | 6 hours ago
Because the Western world literally gave up chip manufacturing in the late 20th century. It’s ironic that most people (except Taiwanese and Korean) have no clue and interests about chips manufacturing business 5 years ago and now everyone think it’s the spice from Dune.
theerrantpanda99 | 5 hours ago
Taiwan and Korea produce way more per capita PhD’s in hard comp sci and electrical engineering than anywhere else in North America or Europe. It’s also dramatically cheaper to employ them there.
Ok_Title5534 | 3 hours ago
In the past, when semiconductor prices were low, price competition ensued among companies, and those unable to absorb the losses went bankrupt.
SpicyRice99 | 2 hours ago
Cheap yet educated labor
Every_Mobile3968 | 2 hours ago
proximity of resources cheap manufacturing and convenient export logistics
bihari_baller | 6 hours ago
>Why can't these be built in North America
They are being built in North America. Oregon (Intel), Arizona (Intel, TSMC), Texas (Samsung, Texas Instruments), New York (Micron, Global Foundries), and Idaho (Micron). all have massive chip making operations.
Glupscher | 4 hours ago
The issue is that all the revenue generated through exports is usually in foreign currency, so you cannot easily use this to invest into your own country. They'll have to invest the excess profits in other countries which doesn't neccessarily benefit the general population.
[OP] rdh2dmd | 2 hours ago
Yes but corporations pay taxes.
The EXCESS tax revenue from corporations alone is expected to come in at 40 billion USD. This is expected to more than double next year
https://www.taipeitimes.com/News/biz/archives/2026/09/28/2003864988