Something I wish was discussed in the article was the fact that shipbuilding salaries in Japan, while low compared to developed countries at the time, was around 160% the average salary a Japanese employee could demand in the post-war era [0]
A similar relative difference didn't seem to exist in the Canada (which I'm using to extrapolate for the US) around that time [1].
It appears that this incentivized higher skilled workers to work in the shipbuilding and steel industry in post-war Japan. Essentially, shipbuilding in Japan in the 50s and 60s would have been the equivalent of being a Software Engineer in the US today, and it was treated as an engineering/STEM disciple [2] instead of as a blue collar semi-skilled discipline in the US.
I can safely say a similar trend happened in South Korea in the 1970s-80s as well according to one of my professors back in the day who specifically specialized in Korea and Japan policy and advised Hyundai and Samsung back then, which lead to a similar decrease in shipbuilding capacity in Japan.
Unsurprisingly, this trend can be seen to this day in any industry - be it chip design, VFX, battery manufacturing, etc.
I've noticed a similar trend in distributed systems/infra/os/networking/cybersecurity as well, where American schools skip teaching systems or architecture fundamentals in order to overindex on Theory/Applied Math whereas NAND-to-Tetris is the default philosophy in programs in Israel, India, and the CEE.
I'm reading this blog, and the debate regarding the role of industry vs deindustrialization came to mind...was decline of shipbuilding in the US due to better paying jobs doing less dangerous work elsewhere?
The US was never a commercial shipbuilding hub once the transition was made to metal ships. During the timber ship era, the huge quantities of quality lumber the US had meant that they were world class at shipbuilding, but metal ships? Not so much outside of direct production for the World Wars(1). I mean, the fact that the US passed the Jones Act in 1920 is a pretty strong signal that even then the US was not viable in commercial shipbuilding.
The US Navy, which has been tied for the largest or exclusively the largest in the world for over a century, meant that military shipbuilding was in a better state(2) but still not super competitive- few countries have ever bought the direct products of American yards- yards in the US mostly serve the USN and then sometimes the USN will transfer ships they don't need to other nations.
1: And even then, a whole lot of it was unskilled people learning by doing, and making mistakes.
2: There are many caveats, but it appears that British shipyards were probably more manpower efficient building warships than the US was in WW2. D.K. Brown discusses the evidence in _Nelson to Vanguard_. It's not perfectly clear- because hours were charged to contracts differently in the two countries, the fact that British ships charged fewer hours to a ship doesn't necessarily mean that they used fewer hours in actuality (while everyone counted laborers on a dock for hours, how do you handle hours spent doing accounts payable? Or material deliveries, or training? There are differences in how they were counted between countries, and Brown was unable to correct for them).
It is also true that the US produced warships faster chronologically, so it is possible that the US threw so many bodies at ship construction that they went past the point of manpower efficiency to get less time spent in the yards. But, overall, British shipbuilders seem to have been more manpower efficient than US yards during WW2, as far as warship production.
One thing that really hurt US commercial shipbuilding after WW2 was the glut of surplus Liberty and Victory ships in circulation. Nobody was buying new ships for a long enough time that the shipyards couldn’t stay in business even if they tried.
That's based on the premise that we were a major blue water shipbuilder.
We weren't, as data from the mid-20th century showed [0]
Japan had a larger blue water shipbuilding capacity than the US by 1910 in both sail and steam. We were on par with newly independent and much poorer Greece (Greece's developmental indicators didn't catch up with Western Europe until the last 3-5 years).
To be fair, we heavily invested in railroads for transcontinental transportation during that era so we didn't have much of a need for long distance seafaring until the World Wars.
The US was never a major commercial shipbuilding hub aside from during WW2 when manufacturing capacity was "reallocated" to shipbuilding. We were a very insular regional power that concentrated on our continent and some archipelagos with limited strategic value until WW2.
It was the UK that was the center of commercial shipbuilding before the Japanese, Koreans, and Chinese took that mantle.
The US had a decent military shipbuilding industry, but they atrophed after the Cold War ended, and the incentive for naval warfare capacity receded until recently.
> Something I wish was discussed in the article was the fact that shipbuilding salaries in Japan, while low compared to developed countries at the time, was around 160% the median salary a Japanese employee could demand in the post-war era [0] ... It appears that this incentivized higher skilled workers to work in the shipbuilding and steel industry in post-war Japan.
It would be interesting to use wage and salary controls to drive talented workers out of certain industries and into more productive ones. For instance, cap total compensation and benefits for people working in advertising (including adtech) at say, $130k/year. People working on cryptocurrency could be capped at 2x minimum wage. It's sometimes hard to identify industries that should be supported, but it seems like it would be much easier to identify the handful of well-compensated but problematic industries.
Japan, Korea, and China kind of did this, but not via direct wage and salary controls. Window guidance is the informal practice where government/finance officials tell banks which industries to prefer for loans. https://en.wikipedia.org/wiki/Window_guidance
It only really works where capital markets are undeveloped or heavily restricted; and it does starve out the rest of the economy, which is bad for the not-chosen-few (e.g. South Korea's chaebols) and can also backfire if you end up being bad at picking winners.
Window guidance doesn't have a great correlation with wage power, as was seen with the loss of shipbuilding capacity in Japan in the 1980s to South Korea as wages rose leading to automation (remember the whole Japanese "robots" trend).
>It's sometimes hard to identify industries that should be supported,
It's sometimes controversial to claim this industry or that industry should be pursued at a national level, but this isn't the same thing as difficult. It's just that politics gets in the way.
And yes, shipbuilding should be one of those industries we pursue.
For that matter, I think you may have nailed the one industry we should punish.
> seems like it would be much easier to identify the handful of well-compensated but problematic industries
That would only incentivize a brain drain. A good example of this is software engineering in UK, Germany, Canada, South Korea, and Japan, as SWE salaries in those countries are not significantly different compared to other vocations.
Also, I'm not a fan of cryptocurrencies, but they are one of the few industries left in the US that incentivizes NAND-to-Tetris level knowledge, and it did help subsidize the GPU buildout that made foundational models easier to train cost effectively.
You can't "command economy" innovation - it can only be nudged.
The solution I've seen most industrial planners use is provide tax holidays and subsidizes for targeted industries, as this helps reduce the upfront cost of hiring, and does give wiggle room to raise compensation. Linking that with production, timeline locks, or even tariffs tends to help force an ecosystem to develop - which is what Japan used to build their shipbuilding and automotive industries in the post-war era.
> Also, I'm not a fan of cryptocurrencies, but they are one of the few industries left in the US that incentivizes NAND-to-Tetris level knowledge,
And what does it do with that knowledge?
> and it did help subsidize the GPU buildout that made foundational models easier to train cost effectively.
Crypto and AI both use GPUs, but I'm not under the impression that AI people repurposed old crypto mines for anything (e.g. crypto mines used janky racks of consumer GPUs, AI typically uses specialized high-end equipment).
Notice how I talked about HPC, distributed systems, and cybersecurity in my comment?
> I'm not under the impression that AI people repurposed old crypto mines for anything
The crypto boom helped incentivize the scaling out of GPU design and fabrication, just like how video games helped with the first iteration in the 2000s.
There's a reason the concept of "dual use technology" has gained currency
That's an interesting take. I read the article and it sounds like the US invented modern shipbuilding during WW2, and the Japanese just copied it and ran with it. But ok.
The us invented modern management in wwii - but it was perfected in japan after the war. Ships were built all over to old management processes but modern management made a big difference in cost, time, and quaity.
Shipbuilding is strategic, but it's unglamorous and projects go wildly over budget. This is why the UK has so few ships in rotation now. In the US shipbuilding is anemic, particularly submarines. The "AUKUS" project was really an attempt at life support.
The work is really hot (or cold), dust and noise everywhere. Most people don't tolerate it for long. Recently a major contractor was caught for skipping required procedures for welds.
The USS Columbia was authorized in 2016, I haven't seen it or the new steam turbine that it is supposed to have. Grumman was awarded the turbine contract in 2014, it was supposed to be delivered in 2021, and it still isn't ready. They had to prepare a huge report to Congress. Each Columbia boat is estimated to cost $15.2 billion.
Trump keeps complaining about getting manufacturing back and we can't even build ships that are fully funded + incentive bonuses.
I think the biggest issue is that shipbuilding sucks as a business. Unless you are a large company with many shipyards, you are producing a small number of ships every year, and the ships take a long time to build. Any particular shipyard will likely have gaps between orders due to a number of essentially random factors. If you can't transfer the workers to other shipyards, you have to lay them off. And then you may have hard time getting new orders or hiring the people back once you have work for them.
If you are from a country with some shipbuilding, you've probably seen the same news cycle repeat from decade to decade. A shipyard gets a big lucrative order. It delivers a fancy new ship. This repeats for a while. Then the shipyard is in financial trouble, as there are no new orders. People get laid off, but the shipyard goes bankrupt anyway. And then some foreign company buys it, and the cycle starts again.
Until you've visited an actual shipyard, you won't realize how much it sucks as a workplace
There's a reason the moment Japan became a developed country, shipbuilding left for South Korea, and then when SK became developed, it left for China. Now it's leaving for Philippines (Japan), Indonesia (SK), India (SK, UAE), and Vietnam (SK, UAE) after the Korean and Japanese players got pushed out of China.
> There's a reason the moment Japan became a developed country, shipbuilding left for South Korea, and then when SK became developed, it left for China. Now it's leaving for Philippines (Japan), Indonesia (SK), India (SK, UAE), and Vietnam (SK, UAE) after the Korean and Japanese players got pushed out of China.
The list of top shipbuilding countries is mostly wealthy, including SK and Japan at #2 and #3.
1 China
2 South Korea
3 Japan
4 Philippines
5 Italy
6 France
7 Germany
8 Finland
9 Taiwan
10 Russia
The share that Japan has today is a fraction of what it was at 50 years ago, and South Korea's is a fraction of where it was 20 years ago. Both Japan and SK's shipbuilding capacity shifted to China by the 1990s and 2000s when their keiretsus and chaebols like Mitsui and Hyundai began investing in CN (and Singapore's Temasek and GIC co-invested in port management and logistics), but both began shifting rapidly away from China in the mid-2010s due to the Senkaku Diaoyu standoff and the THAAD standoff respectively.
For example, Japanese shipbuilding conglomerates like Mitsui and Kawasaki began a large partnership with the Philippines to ToT much of their technology and invest in building an alternative capacity in Subic 15ish years ago [0] which helped increase PH's share. HD Hyundai and Hanwa began a similar expansion in Subic over the past 3-4 years as well.
HD Hyundai and Hanwha began doing something similar in Vietnam in the past 5 years due to the FTA and VN's industrial incentives for shipbuilding [1] and India's recent PLI over the past year [2][3]. The UAE has been playing a major role on this front as well because DP World has been co-investing in shipbuilding and port capacity in Vietnam (Saigon), India (just about every container port), and SK (BJFEZ) - the largest ports in all 3 countries are either owned or managed by DP World. One of the best strategies I've used in my career is to follow where and what Abu Dhabi and Dubai affiliated SWFs and SOEs invest or partner in - they have become similar to what Temasek and GIC were 20-30 years ago.
I am not sure whether it sucks as a workplace. My memories from Greece were that shipyard employees were complaining they don’t get to build ships anymore. What sucks is probably the cost management that comes with the unique combination of capital intensive alongside specialist skills.
I'm not sure about the labor conditions in European shipbuilding hubs like Greece, but in SK labor and safety laws remain fairly laggard. When I visited the Ulsan port a decade ago it was fairly common to see migrant workers from China, Thailand, and Vietnam working in not great conditions (low pay, long hours, weak unionization, and a surprising lack of safety equipment like harnesses)
It's something I noticed a lot in Korean manufacting to be honest - once you scrape behind the Hallyu Wave you can see standards remain laggard, as was seen with the battery factory explosion a couple months back.
I've never seen a convincing explanation from the laissez-faire neoliberal crowd how their view of the world squares with the fact that all the Asian miracles-- Japan, Taiwan, China, South Korea, Singapore-- involved heavy and widespread state intervention. They generally handwave it away and try to change the subject as quickly as possible.
All industry everywhere is government subsidized. Germany has always had auto tariffs, and all of German industry heavily benefited from the long term, very cheap piped natural gas deals with Russia.
America's remaining industries are subsidized or protected with significant non trade barriers- medicine, rockets, agriculture, light truck manufacturering, financial services.
Software services are the one industry that mostly seems to stand on its own two feet, but there was a ton of nurturing of that in the past, as well as onerous export controls on cryptography.
When you have the blueprint of what to build with a large export market ready to buy, it's easier to plan an economy.
Modern economies, unfortunately, don't have that. The capital markets are already mature. What ventures left are uncertain. When governments pick winners there, the pie really doesn't grow.
Well regulated fair markets remain the best system in history to find a way forward to grow the pie.
Unfortunately, regulators have a habit of becoming captured by winners. The markets become no longer fair. Then real growth stagnates.
> all the Asian miracles-- Japan, Taiwan, China, South Korea, Singapore-- involved heavy and widespread state intervention.
Hong Kong didn’t. Counter example disproves rule.
It’s really not surprising that people followed a successful model, given that people are fashion followers. See how there was so little variation in COVID response across countries.
Hong Kong operated as a highly independent special economic zone for doing business in China (at it was very close to the Shenzhen SEZ). Its success was due to the way China partially opened up, and the role it played as a crossroads into China.
HK wasn’t an “Asian Miracle”. It was a trading hub with the benefit of being a British protectorate. That’s very different to the other named countries which built industrial bases.
I mean, the simple answer is that people in Japan, Taiwan, South Korea, and China are all poorer than the US and EU. Japan had decades to catch up and failed to do so. The laissez-faire approach gave people happy, low-stress, and materially rich lives.
Now, we're in a new era where national industrial capacity will at times take precedence over quality of life, and we have to make tradeoffs. But frankly it wasn't obvious that would happen again — China's policies over the past decade made it a reality.
I genuinely think an average person in Japan has "richer" life than a person in NA. No real data points, just vibes. I lived in NA, EU and Japan. It is so relative that, I haven't been able to take GDP per capita statistics seriously. The most awful thing for an average, is the work ethics. But that has been rapidly changing in the last 5 years, I think.
> The laissez-faire approach gave people happy, low-stress, and materially rich lives.
The usual schtick conveniently ignoring the absolute marauding of resources from non-western countries and the effective slave labor extracted from Asia.
You can't wax poetic about the merits about turning your own workforce into "slave labor" (the entire OP point), and then complain about countries buying things produced by the slave labor.
Taiwan and South Korea are both richer than the EU on aggregate measured by GDP (PPP) per capita. So are Singapore and Hong Kong, but it’s not fair to compare city-states. Even in nominal terms, Taiwan and SK are right in the middle of the EU member states between Slovenia and Czechia.
A similar relative difference didn't seem to exist in the Canada (which I'm using to extrapolate for the US) around that time [1].
It appears that this incentivized higher skilled workers to work in the shipbuilding and steel industry in post-war Japan. Essentially, shipbuilding in Japan in the 50s and 60s would have been the equivalent of being a Software Engineer in the US today, and it was treated as an engineering/STEM disciple [2] instead of as a blue collar semi-skilled discipline in the US.
I can safely say a similar trend happened in South Korea in the 1970s-80s as well according to one of my professors back in the day who specifically specialized in Korea and Japan policy and advised Hyundai and Samsung back then, which lead to a similar decrease in shipbuilding capacity in Japan.
Unsurprisingly, this trend can be seen to this day in any industry - be it chip design, VFX, battery manufacturing, etc.
I've noticed a similar trend in distributed systems/infra/os/networking/cybersecurity as well, where American schools skip teaching systems or architecture fundamentals in order to overindex on Theory/Applied Math whereas NAND-to-Tetris is the default philosophy in programs in Israel, India, and the CEE.
[0] - https://www.jstor.org/stable/2519602
[1] - https://publications.gc.ca/collections/collection_2017/statc...
[2] - https://www.usni.org/magazines/proceedings/1966/august/japan...
The US Navy, which has been tied for the largest or exclusively the largest in the world for over a century, meant that military shipbuilding was in a better state(2) but still not super competitive- few countries have ever bought the direct products of American yards- yards in the US mostly serve the USN and then sometimes the USN will transfer ships they don't need to other nations.
1: And even then, a whole lot of it was unskilled people learning by doing, and making mistakes.
2: There are many caveats, but it appears that British shipyards were probably more manpower efficient building warships than the US was in WW2. D.K. Brown discusses the evidence in _Nelson to Vanguard_. It's not perfectly clear- because hours were charged to contracts differently in the two countries, the fact that British ships charged fewer hours to a ship doesn't necessarily mean that they used fewer hours in actuality (while everyone counted laborers on a dock for hours, how do you handle hours spent doing accounts payable? Or material deliveries, or training? There are differences in how they were counted between countries, and Brown was unable to correct for them).
It is also true that the US produced warships faster chronologically, so it is possible that the US threw so many bodies at ship construction that they went past the point of manpower efficiency to get less time spent in the yards. But, overall, British shipbuilders seem to have been more manpower efficient than US yards during WW2, as far as warship production.
One thing that really hurt US commercial shipbuilding after WW2 was the glut of surplus Liberty and Victory ships in circulation. Nobody was buying new ships for a long enough time that the shipyards couldn’t stay in business even if they tried.
We weren't, as data from the mid-20th century showed [0]
Japan had a larger blue water shipbuilding capacity than the US by 1910 in both sail and steam. We were on par with newly independent and much poorer Greece (Greece's developmental indicators didn't catch up with Western Europe until the last 3-5 years).
[0] - https://www.vliz.be/imisdocs/publications/288698.pdf
It was the UK that was the center of commercial shipbuilding before the Japanese, Koreans, and Chinese took that mantle.
The US had a decent military shipbuilding industry, but they atrophed after the Cold War ended, and the incentive for naval warfare capacity receded until recently.
It would be interesting to use wage and salary controls to drive talented workers out of certain industries and into more productive ones. For instance, cap total compensation and benefits for people working in advertising (including adtech) at say, $130k/year. People working on cryptocurrency could be capped at 2x minimum wage. It's sometimes hard to identify industries that should be supported, but it seems like it would be much easier to identify the handful of well-compensated but problematic industries.
It only really works where capital markets are undeveloped or heavily restricted; and it does starve out the rest of the economy, which is bad for the not-chosen-few (e.g. South Korea's chaebols) and can also backfire if you end up being bad at picking winners.
It's sometimes controversial to claim this industry or that industry should be pursued at a national level, but this isn't the same thing as difficult. It's just that politics gets in the way.
And yes, shipbuilding should be one of those industries we pursue.
For that matter, I think you may have nailed the one industry we should punish.
That would only incentivize a brain drain. A good example of this is software engineering in UK, Germany, Canada, South Korea, and Japan, as SWE salaries in those countries are not significantly different compared to other vocations.
Also, I'm not a fan of cryptocurrencies, but they are one of the few industries left in the US that incentivizes NAND-to-Tetris level knowledge, and it did help subsidize the GPU buildout that made foundational models easier to train cost effectively.
You can't "command economy" innovation - it can only be nudged.
The solution I've seen most industrial planners use is provide tax holidays and subsidizes for targeted industries, as this helps reduce the upfront cost of hiring, and does give wiggle room to raise compensation. Linking that with production, timeline locks, or even tariffs tends to help force an ecosystem to develop - which is what Japan used to build their shipbuilding and automotive industries in the post-war era.
And what does it do with that knowledge?
> and it did help subsidize the GPU buildout that made foundational models easier to train cost effectively.
Crypto and AI both use GPUs, but I'm not under the impression that AI people repurposed old crypto mines for anything (e.g. crypto mines used janky racks of consumer GPUs, AI typically uses specialized high-end equipment).
Notice how I talked about HPC, distributed systems, and cybersecurity in my comment?
> I'm not under the impression that AI people repurposed old crypto mines for anything
The crypto boom helped incentivize the scaling out of GPU design and fabrication, just like how video games helped with the first iteration in the 2000s.
There's a reason the concept of "dual use technology" has gained currency
The work is really hot (or cold), dust and noise everywhere. Most people don't tolerate it for long. Recently a major contractor was caught for skipping required procedures for welds.
The USS Columbia was authorized in 2016, I haven't seen it or the new steam turbine that it is supposed to have. Grumman was awarded the turbine contract in 2014, it was supposed to be delivered in 2021, and it still isn't ready. They had to prepare a huge report to Congress. Each Columbia boat is estimated to cost $15.2 billion.
Trump keeps complaining about getting manufacturing back and we can't even build ships that are fully funded + incentive bonuses.
https://news.usni.org/2024/10/31/hii-fewer-than-2-dozen-ship...
https://news.usni.org/2024/04/10/late-turbines-have-major-im...
https://www.congress.gov/crs-product/R41129
If you are from a country with some shipbuilding, you've probably seen the same news cycle repeat from decade to decade. A shipyard gets a big lucrative order. It delivers a fancy new ship. This repeats for a while. Then the shipyard is in financial trouble, as there are no new orders. People get laid off, but the shipyard goes bankrupt anyway. And then some foreign company buys it, and the cycle starts again.
Until you've visited an actual shipyard, you won't realize how much it sucks as a workplace
There's a reason the moment Japan became a developed country, shipbuilding left for South Korea, and then when SK became developed, it left for China. Now it's leaving for Philippines (Japan), Indonesia (SK), India (SK, UAE), and Vietnam (SK, UAE) after the Korean and Japanese players got pushed out of China.
The list of top shipbuilding countries is mostly wealthy, including SK and Japan at #2 and #3.
For example, Japanese shipbuilding conglomerates like Mitsui and Kawasaki began a large partnership with the Philippines to ToT much of their technology and invest in building an alternative capacity in Subic 15ish years ago [0] which helped increase PH's share. HD Hyundai and Hanwa began a similar expansion in Subic over the past 3-4 years as well.
HD Hyundai and Hanwha began doing something similar in Vietnam in the past 5 years due to the FTA and VN's industrial incentives for shipbuilding [1] and India's recent PLI over the past year [2][3]. The UAE has been playing a major role on this front as well because DP World has been co-investing in shipbuilding and port capacity in Vietnam (Saigon), India (just about every container port), and SK (BJFEZ) - the largest ports in all 3 countries are either owned or managed by DP World. One of the best strategies I've used in my career is to follow where and what Abu Dhabi and Dubai affiliated SWFs and SOEs invest or partner in - they have become similar to what Temasek and GIC were 20-30 years ago.
[0] - https://www.offshore-energy.biz/japanese-shipbuilders-encour...
[1] - https://www.mof.gov.vn/webcenter/portal/btcen/pages_r/l/news...
[2] - https://oceanpress.co.kr/news/article.html?no=26769
[3] - https://magazine.hankyung.com/business/article/202412269312b
It's something I noticed a lot in Korean manufacting to be honest - once you scrape behind the Hallyu Wave you can see standards remain laggard, as was seen with the battery factory explosion a couple months back.
America's remaining industries are subsidized or protected with significant non trade barriers- medicine, rockets, agriculture, light truck manufacturering, financial services.
Software services are the one industry that mostly seems to stand on its own two feet, but there was a ton of nurturing of that in the past, as well as onerous export controls on cryptography.
Modern economies, unfortunately, don't have that. The capital markets are already mature. What ventures left are uncertain. When governments pick winners there, the pie really doesn't grow.
Well regulated fair markets remain the best system in history to find a way forward to grow the pie.
Unfortunately, regulators have a habit of becoming captured by winners. The markets become no longer fair. Then real growth stagnates.
Nothing works forever.
Hong Kong didn’t. Counter example disproves rule.
It’s really not surprising that people followed a successful model, given that people are fashion followers. See how there was so little variation in COVID response across countries.
Now, we're in a new era where national industrial capacity will at times take precedence over quality of life, and we have to make tradeoffs. But frankly it wasn't obvious that would happen again — China's policies over the past decade made it a reality.
The usual schtick conveniently ignoring the absolute marauding of resources from non-western countries and the effective slave labor extracted from Asia.