Twenty years ago, my daughter had a social studies teacher who taught about the evils of sweatshops. It gave me an opportunity to have a conversation with my daughter about the alternative scenario where the workers didn't have that employment opportunity. That sometimes what we might consider to be unacceptable through our POV is a better alternative. This is a fair look at this issue and I appreciate the nuanced approach to nudging these countries toward better working conditions.
Even now many poor, indebted farmers in Vietnam still fled the countryside when bad harvests happen, or agricultural prices stay low.
Many ex-farmers ended up return the land to the collectives or selling land for factories, so quite a lot of them now complained that they had to buy grains for their chickens!
I basically made Noah’s same argument, but with a lot vaguer sourcing, in a Student Congress round in 2023. “The number one killer in Bangladesh is poverty!”
Great post. It’s disappointing that many subscribers are calling this post “ghoulish.” That reaction reflects a flawed mental model of how high-margin, high-wage industries actually emerge in a nation-state.
In their minds, these industries appear fully formed—without any understanding of how low-wage, low-margin sectors typically lay the groundwork. Over time, as physical and human capital accumulate and skilled labor pools deepen, these sectors move up the value chain into higher-value, higher-wage activities.
India’s IT and BPO industries began with low-margin call centers but gradually evolved into hubs for R&D and full product divisions (GCCs). Similarly, much of East Asia’s electronics manufacturing started with basic assembly before advancing into complex OEM production.
When developing countries legislate with this progressive mental model, the results are often disastrous. Labor protections that are sustainable in high-margin industries can render low-margin sectors unviable—effectively preventing poor countries from ever climbing the value chain in the first place.
The difference between a Bangladesh and a Mali is that Bangladesh at least attracts foreign investment, Mali has never netted fdi over a billion dollars since independence. The people there are either cotton & peanut subsistence farmers or they are artisinal miners for gold. There wont be any scaling up the value chain in Mali unless a multinational takes advantage of their cheap labor to make t-shirts like Bangaldesh did in the 2000s. Just like what China did in the 1980s or Taiwan did in the 1950s.
It builds a straw man "progressives don't want poor countries to have factories"
Then it finally and reluctantly engages with the actual position which is "poor nations should have safer and better paid factories". Then it concedes that these campaigns work.
Also, Noah doesn't say at what point lack of safety and exploitation becomes too much.
Nike and other sports brands got in trouble because they were buying soccer balls from Pakistan. These soccer balls were being sewn by children who were chained to the looms and beaten regularly.
So these companies cleaned up their act and stopped buying the child labor soccer balls.
Does Noah think this cleanup is a tragedy?
Are sweatshops good if they employ teenagers? Probably. What about if they employ 12 year olds? What about 10 year old children?
If we relaxed all labor and safety regulations and didn't bat an eye when companies hired indentured children abroad to make cheap clothing, man their economies would grow fast!
Then poverty would drop even faster, so in the long run it was OK that the kids were enslaved and their arms ripped off.
This piece purposefully doesn't engage with the actual practices of these sweatshops, even as it concedes that pressure to clean up from customers and western countries improved worker lives WITHOUT killing the golden goose.
Acknowledging tradeoffs isn’t “ghoulish”, it’s just part of living on planet Earth. Obviously chaining children to looms lies at one extremely steep portion of the tradeoff curve, but that doesn’t mean that there’s no margin at which improving working conditions will lead to worse overall human welfare. I would love for the world to settle into a different equilibrium where people who can afford it just pay more for their clothing and the poor people who make the clothing get better working conditions, but the path from Point A to Point B is littered with unintended consequences.
Does banning SROs lead to their residents getting nicer apartments, or do they end up on the street? Does raising the minimum wage result in a redistribution from owners of capital to workers, or does it result in more automation and less employment? These are empirical questions with real tradeoffs and can’t be answered by emotional appeals to what is more or less ghoulish.
Not wanting children to be chained and beaten until they sew enough soccer balls and not wanting women to be sexually abused and mistreated in the workplace unto suicide is an exclusively "Lefty assertion"? Good to know. That does make the case that the alternative "Righty assertion" that all that abuse is perfectly fine...well, ghoulish.
Those same women are equally likely to be sexually abused and beaten when married off as teens in their home village, where an not insignificant number also commit suicide. Sweatshops can be awful but a) they are range in conditions, not all are the same. b) they also provide a way for young women to avoid early marriage, earn some money and thus a higher status in their family, and get a bit of exposure to a the wider world. You do have to take into account what the alternative to work in a sweatshop is and for village women it is often early marriage and being a virtual slave to a mother-in-law, long hours of hard labor, and inadequate food.
If someone is going to engage in dual logical fallacies of Strawmen and Excluded Middle, it's unlikely to be of great utility - but yes as Noah own writing evokes
A. range of conditions is large (and in my experience one needs to look at both Formal and Informal employment, as informal is often the real abusive and equally the least or not subject to regs)
[The Lefty discourse on this rather ressembles that of the 19th century Marxists and Marx - collapsing a wide range of cases (truly awful to not great) into one, using Worst Case as Base Case and extrapolating from there
B. the other actual real world choices, which are not the idealised ones the Lefty professional activitsts see - as in indeed (a) barely surviving rural labor on yes inadequate food and family overall barely getting by (although nicely photogenic poverty for OxFam walls) with early marriage unlikely own choice for either young men or young women, (b) urban informality like rag picking etc in nastier conditions than typical formal factory, (c) unemployment and non-legal (as like prostitution)
Collapsing this down to "beaten to sew soccer balls" is emotional sloganeering.
This isn’t a very charitable interpretation. Noah literally uses increased health of children as an example of the benefits. Also, a few terrible factories shouldn’t doom the whole project. Close the terrible ones and keep the good ones, right?
This goes to an idea of how much the ends justify the means. Perhaps a country could attract 3 billion in FDI if they agreed to allow companies to come in, build a factory, and enslave the workers for 5 years.
For just 5 years of a few thousand literal slaves, a poor country could have a capital endowment of 400 shiny new factories that could then start hiring workers at good wages for good jobs. This means health and wealth and money for MILLIONS of the people of that country.
Most people would say that this isn't worth it. Even if you could unequivocally show that the country post slavery would be X amount richer, the ends in this case don't justify the means.
Noah really doesn't want to put a line on "At what point are factories too unsafe, are the workers too exploited?" Because then he would have to be on record as saying that "X number of new factories is worth the death of Y number of Bangladeshis."
Should we feel good if we improve labour standards and the exploitive garment countries moves out of Bangladesh to another country and people fall back into extreme poverty.
Then we ban the exploitive garment companies and people buy their Tshirts directly from exploitive Chinese companies.
Then we ban any tshirt sales from China and people buy directly from other exploitive countries.
I’d like to believe that we can successfully play whack-a-mole with human nature but nothing in the real world seems to suggest it works that neatly.
I think the current process where the US government and people push for incremental reforms in countries where we buy goods is probably the best we can do as much as it sucks to admit that we’re powerless as a group to do more (individuals of course can and should choose to do what they think is right).
I remember one time I was taking an uber in Singapore. The driver, in his day job, was a sand inspector for a company that imported sand into Singapore.
Singapore needs sand because it has become 22% larger since 1965 and all of that is done by dumping sand into the ocean and building new land.
But not all sand is created equal, depending on what you need the sand for, it needs to have the right grain size, the right mineral composition, the right weight... it was all super technical.
Now, he was hired by his company, because the Singapore government has a policy. If the government import sand, and find out that the sand is not up to spec, than the Singapore government will refuse to take it or pay for it. This leaves the import companies with a boat with 1000 tons of sand they can't sell.
To insure against that possibility, it's worth it for the company to pay someone who's only job is to inspect sand quality.
Does it make the sand slightly more expensive? Probably. Does it mean that Singapore doesn't import bad sand? Definitely.
If you make companies that sell in the US to some degree liable for labor abuses/safety standards in their supply chains, than they will assert pressure as the final buyers to clean up said supply chains.
Will their be leakage? Sure.
Is it an easy question to decide how much liability they should have? No.
How much pressure should countries require the companies who sell in their markets to put on cleaning up their supply chains?
In this very piece, Noah admits that this kind of pressure works to raise labor and safety standards. He also admits that some amount of this pressure is compatible with industrializing out of poverty. Even Bangladesh in their successful modern garment story never did the 18th century British tradition of having 8 year olds pulling bits of cotton out very fast spinning looms. Bangladesh said, "This is a horrifically bad idea, so, even though we could be profitably employing our elementary schoolers to the benefit of foreign capital, let's not." Maybe they would have made a lot more factories in the 90's, been richer today, with less NET child poverty today had they done so...
But this is a real question that is worth a real response and discussion.
This peace avoids this much more difficult and technical discussion.
Totally a fair critique. I do think Noah’s post serves a really important purpose. We can’t take our eye off the ball of how effective industrialization is at reducing extreme poverty. That should be the north star of anyone who care about reducing global suffering.
The question we’re asking is how much pressure can America and Europe put on supply chains before the economic incentives break.
I think we’d both agree its a bad outcome if governments exert too much pressure and you end up with a situation with an industry dominated by Chinese companies who fake certifications or who ship directly to uncaring consumers.
The unintended consequences of overly strict regulations will likely be an even worse global standards for sweatshops.
I’d love to see a progressive movement that did a better job grappling with these trade-offs. So rarely do I see a reckoning with our own political limitations. “We can push this hard on the Bangladeshi garment industry to improve standards, if we push any harder people are going to seal out cheaper and more unethical options.”
One challenge here is that we're often paternalistically substituting our own values and ideas about these tradeoffs for those of the people actually working in the factories and their governments.
How the ethics of doing that shakes out can maybe be debated, but we should be aware that there are actual humans involved who may not share our values about these tradeoffs.
Noah was being ghoulish. Posts like his is what gives Neoliberals (like me) a bad name. It's ghoulish to minimize the barbarity inflicted on the workers of worlds developing nations and pretend it's a necessary trade off to their advancement. It's not.
I was actually going to write a sarcastic reply about how it would be even better for a nation to employ slave labor.
A nation like Bangladesh has a huge comparative advantage over advanced nations. The margin between them is large enough that it does not require forcing women to work 16 hour days to make their garments competitive to dominate our markets. Bangladesh could have had some very limited workers rights by western standards and they still could have undercut and bankrupted any western manufacturers.
What Bangladesh does have to face is another poor nation willing to gain a comparative advantage by abusing workers. Poor nations face a race to the bottom to be competitive. For example, Bangladesh's garment industry could be put out of business if a nation employed slave labor... IF the west would buy from them.
But we won't. (well we do sometimes, but only if we can be in denial about it. ) It's illegal. The point is, We, the advanced nations, set the floor on the race to the bottom.
Advanced nations do set the floor higher than slave labor. We could set it higher We could demand very minimal standards of workers rights. We could prohibit imports of products that force women to work 16 hour days, or work in unsanitary conditions, or be exposed to toxins. We could demand a path unionization.
We could demand all kinds of baby steps that would remove much of the barbarity workers endure while still letting poor nations maintain a huge comparative advantage and their path out of poverty.
But our clothes would cost a bit more.
I have long considered myself a neo liberal, but I have always lamented how brutal it was in practice. It didn't have to be.
I've also met real life communists and anarcho capitalists. It doesn't make putting them as the "typical" mainstream movement member not a strawman move.
The ghoulish thing to say is that some work is so bad no one should be "allowed" to do it. If someone is free, then whatever they are doing, the alternative is worse. Taking away the thing you find inhumane makes their lives worse, not better.
Obviously chaining someone to a desk and beating them to work, so in other words slavery, is wrong.
Yeah, the thing so many outsiders don't want to grapple with is that the alternative to the horrible factory job can sometimes be that person, or their child, actually dying.
I don’t think it’s a strawman. In a student congress round in 2023, the resolution we were debating was a tax on all imported Bangladeshi goods until the fire safety situation was improved. Real professionals in this field are probably doing good work to get companies to guarantee their supply chains are free from child labor, but amateur Progressive commentators regularly get this very wrong. Their first impulse is to just make it harder to for American companies to import garments from global south countries. And, as seen on the right, often if the rank and file get an issue wrong, the elite can’t stop them. So posts like Noah’s are useful.
Noah's laying his mainstream economist cards on the table here--this is, and has been for some time, the conventional wisdom among mainstream economists for several decades now: the best (or only) way for poor countries to get rich is by allowing foreign direct investment from rich countries to pour in, and to leverage whatever comparative advantages they can in the global marketplace. Much has been written about whether, either empirically or analytically, this is even true, but I'll just insert my own personal objection to the premise: just because you can measure one thing--GDP per capita--or a few things, like longevity or infant mortality, and show improvement over time doesn't mean you can say definitively that sweatshops are a necessary ingredient for progress.
If because your land or whatever becomes polluted or unusable or sold out from under you by your government to some multinational corporation, and you're forced into laboring a 996 (9am to 9pm, 6 days a week) schedule instead of working at your own pace with no overseer on the family or community farm, I'm pretty sure your productivity is going to go up a whole bunch, but is that a real improvement in your quality of life? No doubt your country's GDP is going to shoot up too as subsistence agricultural workers are driven into factory jobs, but is all that worth it if people lose autonomy over their lives? We in the West seem to think that prosperity as measured by national GDP is THE measurement by which all else should be measured, but for more traditional communities and peoples, that is just not the way they think and operate. What is the price of a traditional way of life, one going back perhaps centuries, ripped away? You can't measure that in dollars and cents.
This is not to say that I'm not a believer in economic or social progress, but the people that are the supposed beneficiaries of the wonderful bounty of foreign investment and integration into our global market system should at least have a say, and as far as I can tell they do not. They don't even have a say on whether or not they can go to the bathroom.
Obviously conditions and situations in different places can be wildly different and my own experiences, though real, are just anecdotes really. But I'll note two things I learned from doing volunteer work in the developing world:
1) Many of those family or community farms are so marginal that the difference between living and starving to death is determined by tiny differences in rainfall or in how the randomness of annual floods shakes out from year to year. Meeting so many people literally on the knife edge between sustainable life and death was a shock that I'm not really over, some 20 years later.
2) Outsiders need to be self-aware about their paternalism and their often casual willingness to cancel out the wishes and values of local people and replace them with their own values. This is especially true for managing heartbreakingly tough tradeoffs and around how to manage risk-taking in general.
Yeah for pre-industrial societies, having a stable wage is already a pretty good deal though, compared to not knowing that you would starve or not next year!
My company has a large office in Delhi and sponsors a lot of charitable organizations, among which is a sewing machine training center. Extremely poor women, including those from scheduled castes, are trained on industrial machines. They can either get certified as sewing machine operators, which allows them to apply to work in the sweatshops, or can receive a high-quality domestic machine so that they can do home piecework or alterations.
I visited the center, met some of the women, and even learned a little about running the machines. It's hot, power is iffy, the machines can be dangerous... but the alternative for many of these women is literally begging on the street, which is also hot and dangerous. It's very clear that the Indians who sponsor and work at this center view it as essential for lifting these people out of poverty (which is a strong cultural obligation).
Much of what I've learned about India from my travels and interactions is that I, as a Westerner, know very little about the situation and have a lot of preconceptions which don't really apply.
The point about economic growth is fair, but I think the points about labor conditions are far from bullet-proof. The US government, corporations, and other developed nations have immense buying power. They can use that power to set standards across-the-board for what is acceptable level of negative externality.
I think reframing this as “trade policy” opens a lot wider range of tools to manage worker conditions as well as climate impact (“carbon tariff”).
In the extreme, wouldn’t the US have a trade policy about trading with a hypothetical country who used literal slave labor? Similar for egregious polluters?
The politics of raising consumer prices is challenging, but it’s happening today to protect other interests/values.
I’m curious what you think about reframing this as a trade policy issues about curbing major abuses, and how worker conditions parallels pollution!
I understand and completely agree that sweatshops are better than starvation, and definitely an improvement for the extremely poor. I also embrace that it has been a very successful rung on the ladder to middle income countries.
I agree with you -- Developed countries shouldn't stop the very poor countries from getting a foothold.
My question is: Suppose a trading bloc of countries with a lot of buying power wanted BOTH to buy cheap clothes from very poor countries, AND to require some minimum standards on their labor practices and environmental impact. Is there a trade policy that could achieve both goals? If so, at what cost? Maybe these would all have side effects that are politically nonstarters (higher consumer prices), but curious if you had the political will, what options would be on the table? Maybe:
- Yes, we could require some minimum wage (much lower than $7.25/h), some environmental factors etc. This would raise consumer prices some, because those poor people are paid slightly more than they
- No, then buyers will nearly always pick a richer country. The poor country will never get the foothold
"My question is: Suppose a trading bloc of countries with a lot of buying power wanted BOTH to buy cheap clothes from very poor countries, AND to require some minimum standards on their labor practices and environmental impact"
I guess it's theoretically possible. I guess the question is what happens when another very poor country comes along and says hey we want to do this too.
I do agree there has to be some minimal standards. People dying in fires, or being chained to work stations clearly cross the line. Dying in a fire isn't better than dying by starvation.
But I would just be careful where drawing the lines. Good intentions can still create bad results.
I think it’s fascinating that MAGA has adopted the left’s “sweatshops are bad” narrative to excuse the inflationary and other negative effects of tariffs and essentially argue for the repatriation of low paying, labor-intensive employment to the US.
As with many things MAGA, they've identified a real problem (eg working class communities got destroyed by unskilled manufacturing job losses), but associated the problem with the wrong causes (globalization hollowed out manufacturing communities).
The real problem is that US industry moved up the value chain with automation that's raised productivity, but also significantly raised the skill requirements for the marginal factory worker.
We have like 800,000 unfilled manufacturing jobs in the US and a whole bunch of unhappy MAGA folks without the skills and/or work ethic to get hired in those jobs (electricians, pipefitters, machinists, welders, mechatronics tech).
The problem is a mismatch between the skills needed and the skills available in the labor pool and nothing MAGA or the Trump Administration is doing seems likely to address this.
To me, the politically smart answer to this is 1) recognizing the issue is that we messed up by telling everyone to go to college, and 2) building pathways to allow people to get help/achieve success learning marketable trades (loans for training, but w/institutions held accountable for success, maybe public/private training partnerships, etc)
Well no problem now as you will have a load of white collar ex tech wanting to retrain as pipe fitters and plumbers etc . At least they have maybe 10 years at that until robots replace them. Oh and of course robots can be chained to their work station.
That's fair, but most of those $20 -> $1 an hour jobs got offshored in the 80s and 90s. Recent job losses from globalization (say in the past decade) are lower, where automation is taking a bigger and bigger share of the responsibility for job losses. I think in the last decade, automation accounts for something like 85-90% of manufacturing job losses.
This is why I think the idea of tariffs to bring back jobs is so fraught. As I said in another comment below, Chinese EVs are made w/10-15 hours of labor, while US ICE cars use 25-35 hours. some of that is EVs vs ICE, but a lot is just more automation.
To make things worse, not only does a Chinese EV factory use 1/2 the number of labor hours per car, their labor is 60-80% skilled compared to US factories that are more like 20-30% skilled. Right now the US has around 800,000 skilled manufacturing job openings that industry can't fill.
So we're using tariffs to bring back jobs to the US, but they're going to be fewer than some people think and they're going to require skills our workers don't have, and we're going to make the skilled labor shortage even more acute. This will be great for those skilled workers, but also will raise costs for consumers even more on top of tariffs.
But I agree tariffs aren't going to bring back a large number of jobs.
I think we should be pushing for free trade agreements with other friendly nations (Canada, the EU, Japan etc) as long as they open up their markets to ours.
I'm ok with tariffs on China. Say an additional 10% each year until they
I agree with most of that, except I'd probably limit China tariffs mostly just to strategic industries, rather than using tariffs to try to change them into something they're not going to become and making us poorer for no reason.
Of course, in the grand scheme of things I'd be thrilled to get your plan as is instead of what we've got now.
What I'm curious about though is how many of those jobs were lost to China (or other low-wage countries) and how many of them were lost to automation?
I also wonder if the US and UK suffered needless deindustrialization because their unions were organized on the basis of occupation (which made them hostile to change) rather than by firm (as is common in Japan) or on an industry-wide basis (as is the case in Germany)?
First, I agree that globalization played a role in hollowing out manufacturing, but I'd say it played a much smaller role than automation (like 70-80% automation/20-30% offshoring, with even more automation more recently).
But here's the thing, a Chinese EV requires 10-15 hours of labor per vehicle while a US ICE car requires 25-35 hours. So even if the 20-30% of jobs lost to globalization had stayed, they'd be on their way out by now as we move to more automated manufacturing. Again, the solution to this problem is getting our workforce better trained and helping them deal with substance abuse. Nothing the Trump Administration is doing seems geared toward helping with that.
On your other point, I am very much in favor of controlling the border. I think if Biden didn't fumble that so badly Trump wouldn't be president.
China created an economy for 1.5 billion people based on manufacturing...obviously manufacturing creates jobs.
You need people to make the machines that automate the manufacturing, people to make the steel, people to mine the ore, people to transport everything all around.
You need an entire economy to manufacture stuff. It's not just 100 guys in a warehouse cranking out cars.
First, close to half the labor-intensive jobs in industries like textiles have already been moved from China to poorer countries. EVs were just a crisp example. The trend is across industries.
My assertion is that automation is primarily what's been eating unskilled jobs recently, both here and all around the world. It's only going to get worse as robots and machines get even more capable.
New modern factories across industries need fewer workers overall plus a relatively higher ratio of skilled to unskilled workers. That ratio will continue moving higher as automation gains capability. That doesn't mean less jobs overall, but likely less demand for unskilled labor versus supply.
Everyone likes a scapegoat and globalization seems to be what the MAGA populists have latched onto. That's a mistake, not just because their remedies are going to make us all poorer. It's also a mistake because they're chasing the wrong villain and not doing anything to address the real problem, which is the skills mismatch (and prob. also substance abuse in hard hit communities).
Noah, don’t economies grow as laborers become better trained and more productive? Surely in Bangladesh, they have people who can fix sewing machines by now. It should mean that more trained people who can work with machinery and computers can work with more sophisticated machinery higher up in the value chain. They will earn more income. The more computers in the country, the higher value jobs will be offered due to having a trained workforce.
Your charts on the lowering of poverty, along with the chart showing higher GDP, aren’t they not the story of trade and other programs that have lifted the entire world out of poverty that was seen only 40 years ago?
As for the stupidity of many on the left, if Zoe Saldana had been the model the clothing seller had chosen, would anyone say they are saying black genetics are superior? It would be tough to argue that neither woman is beautiful.
He did argue in his post about Bangladesh that they should move up to higher value chain though.
The point is that between extreme poverty and a lower-middle income, sweatshops are unfortunately necessary for many countries. (India tried to leapfrog to service economy, and this caused another bunch of issues, such as high youth unemployment and low labor participation of women, led to high gender inequality).
I would say much of the difficulties of poorer countries around the world are their attitude towards women working and holding jobs. Afghanistan will never be a richer nation. Oh, they may sell some rocks to China, but that will go into the corrupt pockets of the Taliban. Asia, Sub Asia has the same problem. When women work and the pill is available, they prefer to work and make their own money rather than pop kids out one after another.
I wonder how many people in the West contrast Afghanistan with the ex-Soviet "-stans" to the north and think "maybe in the 1980s we should have just let the Soviets win"?
Whatever its failings, communism was at least a modernising force. (Pashtun nationalism, not so much). Unfortunately, communism's failings helped bring modernisation into disrepute. And now, all Afghan's have to live with Jurassic conservatism.
The problem was that they really didn't. The Mujahideen were another level of brutal. A friend of mine started reading a history of the Soviet invasion and had to stop a third of the way through because it was making him physically sick. Barbarism without constraint. It will be like N. Korea from now on: throw a cordon sanitaire around it and retreat to a safe distance.
Oh com’on. If the Taliban are still in charge, I think we can expect exactly what they currently have. North Korea and Russia are exactly the same because their rulers are the same. Something very different would have to happen in Afghanistan for it to change.
That’s incredible I was thinking the same thing myself about both White Lotus hotties! Give Paula her own jean commercial! And I think black jeans need a comeback fashion wise 😎
I’d have a lot more sympathy for the whole “sweatshops are a necessary evil” narrative if they’d at least be open to unionizing the sweatshops and letting the workers collectively-bargain their way to a decent local equilibrium. Otherwise the whole situation looks like classic economic duress.
Also, what happens when there’s no longer a bottomless pool of low-skilled labor to tap into? I know we still have Sub-Saharan Africa to count on, at least until their populations age out like the rest of us, but what’s the industrialization path for poor countries in the aging, low-fertility future?
For your last questions, I think the answer is that if the pool decreases, then the price of labor goes up, at which point automation becomes a more competitive option. That is to say, we'll invest more in robots that can sew our clothes, requiring less labor.
There is a game (Victoria 3) about economic development of countries in Victorian era, and your arguments about opening for foreign investments are similar to the debate in-game about allowing foreign investments though!
Side note: One of the "meta" in that game to boost GDP growth is to only allowing foreign investment to help countries escaping from poverty, then not allowing foreign countries to invest before nationalizing their industries and selling it to local private owners. Yet I can't recall any countries in modern day who actually succeeded when doing this; are there any particular reasons behind this?
(I think that would be due to the institutional instability/sovereign risks that these actions created for future investors + potential for corruption and cronyism, but not sure about other reasons).
Presumably any country looking to develop will need to buy capital equipment from foreigners that it doesn't (yet) know how to produce itself, and it will be extremely difficult for them to get the foreign currency to do this if they have a track record of expropriating foreign investors?
(This makes me think of how the Holodomor happened because Stalin wanted to export the Ukrainian grain to pay for American industrial equipment, even if the Ukrainian peasants starved.)
That was covered in my section about "sovereign risks" already though!
(A better example is Argentina, being rich before the Great Depression due to agricultural exports and British investments, but after institutional crises, Peron came to power, and got factories/railroads sold by the Brits. His mismanagement of these assets, plus his protectionist model and export duties for agricultural goods for industrialisation, led to the start of Argentine decline).
Cuba post revolution is also a good example for two reasons I stated earlier, but I wonder whether there are any other reasons:
Wouldn't Argentina's failure to industrialize be more down to its small population and geographically isolated location? Same reason why Australia is now primarily a commodity exporter, even if an unusually wealthy one.
And the decline in Argentine agricultural exports (which were mostly to the UK) was likely more because the British decided to modernize their own agriculture, after WWII's Battle of the Atlantic showed how dangerous an extreme dependency on imported food could be, even for a country that notionally "ruled the waves".
Argentina tried to industrialize before the Great Depression, but at that time its share of GDP for manufacturing was half of Australia!
And no, the British didn't just decide to modernize agriculture; if anything the UK in WW2 was self-sufficient in food, the only things they lacked were luxury foods and meat (especially beef and mutton - that they bought from Argentina a lot).
The decline of UK imports with Argentina can be traced back to Ottawa Declaration, that raised tariffs for entire British Empire; as well as the Roca-Runciman treaty, in which Argentina got to keep beef export quota but had to open the market for British manufactured goods.
Fantastic post. I am surprised that you would hae these opinions. It gives me hope.
The so-called "Progressives" want to pull the rug out from under developing countries...ending hydocarbon fuels, ending jobs, ending their ascent out of poverty that is the MOST dramatic story of the last 30 years.
Sweatshops is a neologism, a word made up, for people who are poor working at jobs that fortunately, you and I and our children will hopefully never work at. But that doesn't mean these jobs are bad. All work has dignity so long as it is a voluntary exchange between employer and employee.
In 20 years, these jobs will be gone, and hopefully the people working in these shops will have their own businesses, and will employ robots and be living an amazing life.
Also, this makes some arguments bashing OpenAI (or big AI firms) about their low pay for data labellers in Africa (2 USD/hour) wrong; of course these labellers had to see brutal contents and as such, having better compensation would be great, but having a consistent wage and working in house is a dream for many people there, plus it is still a significant wage compared to the alternative!
Being a sweatshop worker sucks, but being a subsistence farmer is usually even worse - otherwise people wouldn't choose to leave the farm in the first place. And as the other opportunies for people get better, the sweatshops either stop being so terrible or start paying a lot more money.
"Bangladesh’s conquest of extreme poverty hasn’t come from any dramatic program of government redistribution, but from rapid economic growth. In the years since 1990, the country has more than quadrupled its living standards, blowing past Pakistan"
The fact is that race-to-the-bottom conditions that foster sweatshop abuses can be alleviated (or worsened) through government programs, such as land reform (or enclosure), education, worker rights to unionize, etc. All the "evidence" you have listed has occurred under race-to-the-bottom conditions.
Not to mention that injury, illness, and loss of educational opportunity are not counted by many of these development studies simply because they ignore losses to human capital.
Industrialization through factory production doesn't need to occur through a sweatshop phase. It can occur with modestly paid employees with unions, health and safety protections. Claims that the "alternative" is worse and that employment is "voluntary" generally overlook the fact that government policies or cultural traditions (such as permitting landlordism) are responsible for the horrible alternatives. Indeed, it is OBVIOUS that lots of industries develop without a sweatshop phase.
"Industrialization through factory production doesn't need to occur through a sweatshop phase. It can occur with modestly paid employees with unions, health and safety protections. "
Really? Say you are a dirt poor country, with no industry. You want to build one. The only thing you can really compete on it price. How do you get that lower price?
Establishment of a social-democratic welfare economy with land reform seems to have worked very well for Kerala, India. https://en.wikipedia.org/wiki/Kerala. Their high human development index allows them to compete outside of poverty level labor.
Really, ideologues really should look at the real world a bit more for counterexamples. If you paid more attention to the works of Amartya Sen, you might learn something.
You vastly overestimated the sincerity of someone like Derek Guy on this sort of topic. You touched on it when you said, "“sweatshops are bad” is just one of those things that American progressives are supposed to believe".
Derek Guy is a progressive fashion influencer. At best he can be sincere about his fashion opinions. There is no possible way he can be sincere in is economic opinions. Meaning even if he believed that sweatshops had some role to play in a country's development, he couldn't actually say that because then he would lose influence with progressives...which is his whole job.
As a practical observation from personal experience, it is further to be observed that the application of rich world standards blindly - in labor codes and other written factors often is a Potemkin village effort where formal employment and labor flows into the informal sector and hidden / off-the-books empoyment and even mini hidden shops doing jobbing w zero protection.
State capacity lacking, and corruption allowing, this becomes a whole hidden industry.
(as an additional factor besides pure unemployment and poverty or alternative alterantive, bare subsistance farming although the latter with the fetishizd 'small holder farmer' provides the OxFam types with nice photogenic poverty porn).
One doesn't get besides the economic carrying capacity the state capacity in one go, it has to be built, and Potemkin village facade law rather often just becomes corruptoin rents for officialdom.
Twenty years ago, my daughter had a social studies teacher who taught about the evils of sweatshops. It gave me an opportunity to have a conversation with my daughter about the alternative scenario where the workers didn't have that employment opportunity. That sometimes what we might consider to be unacceptable through our POV is a better alternative. This is a fair look at this issue and I appreciate the nuanced approach to nudging these countries toward better working conditions.
Even now many poor, indebted farmers in Vietnam still fled the countryside when bad harvests happen, or agricultural prices stay low.
Many ex-farmers ended up return the land to the collectives or selling land for factories, so quite a lot of them now complained that they had to buy grains for their chickens!
I basically made Noah’s same argument, but with a lot vaguer sourcing, in a Student Congress round in 2023. “The number one killer in Bangladesh is poverty!”
Great post. It’s disappointing that many subscribers are calling this post “ghoulish.” That reaction reflects a flawed mental model of how high-margin, high-wage industries actually emerge in a nation-state.
In their minds, these industries appear fully formed—without any understanding of how low-wage, low-margin sectors typically lay the groundwork. Over time, as physical and human capital accumulate and skilled labor pools deepen, these sectors move up the value chain into higher-value, higher-wage activities.
India’s IT and BPO industries began with low-margin call centers but gradually evolved into hubs for R&D and full product divisions (GCCs). Similarly, much of East Asia’s electronics manufacturing started with basic assembly before advancing into complex OEM production.
When developing countries legislate with this progressive mental model, the results are often disastrous. Labor protections that are sustainable in high-margin industries can render low-margin sectors unviable—effectively preventing poor countries from ever climbing the value chain in the first place.
The difference between a Bangladesh and a Mali is that Bangladesh at least attracts foreign investment, Mali has never netted fdi over a billion dollars since independence. The people there are either cotton & peanut subsistence farmers or they are artisinal miners for gold. There wont be any scaling up the value chain in Mali unless a multinational takes advantage of their cheap labor to make t-shirts like Bangaldesh did in the 2000s. Just like what China did in the 1980s or Taiwan did in the 1950s.
It's because too many people can't accept that they FEEL the world should work and the way the world actually works are different.
Which is why they don't like economics. Because economics is about how the world actually works, not how they want it to.
This post is ghoulish.
It builds a straw man "progressives don't want poor countries to have factories"
Then it finally and reluctantly engages with the actual position which is "poor nations should have safer and better paid factories". Then it concedes that these campaigns work.
Also, Noah doesn't say at what point lack of safety and exploitation becomes too much.
Nike and other sports brands got in trouble because they were buying soccer balls from Pakistan. These soccer balls were being sewn by children who were chained to the looms and beaten regularly.
So these companies cleaned up their act and stopped buying the child labor soccer balls.
Does Noah think this cleanup is a tragedy?
Are sweatshops good if they employ teenagers? Probably. What about if they employ 12 year olds? What about 10 year old children?
If we relaxed all labor and safety regulations and didn't bat an eye when companies hired indentured children abroad to make cheap clothing, man their economies would grow fast!
Then poverty would drop even faster, so in the long run it was OK that the kids were enslaved and their arms ripped off.
This piece purposefully doesn't engage with the actual practices of these sweatshops, even as it concedes that pressure to clean up from customers and western countries improved worker lives WITHOUT killing the golden goose.
Acknowledging tradeoffs isn’t “ghoulish”, it’s just part of living on planet Earth. Obviously chaining children to looms lies at one extremely steep portion of the tradeoff curve, but that doesn’t mean that there’s no margin at which improving working conditions will lead to worse overall human welfare. I would love for the world to settle into a different equilibrium where people who can afford it just pay more for their clothing and the poor people who make the clothing get better working conditions, but the path from Point A to Point B is littered with unintended consequences.
Does banning SROs lead to their residents getting nicer apartments, or do they end up on the street? Does raising the minimum wage result in a redistribution from owners of capital to workers, or does it result in more automation and less employment? These are empirical questions with real tradeoffs and can’t be answered by emotional appeals to what is more or less ghoulish.
Sorry but your common sense amazes me sir 😎
The post is clear minded rational data based analysis.
Of course for those more interested in virtue signaling it is "ghoulish" in not following a priori Lefty assertions.
Not wanting children to be chained and beaten until they sew enough soccer balls and not wanting women to be sexually abused and mistreated in the workplace unto suicide is an exclusively "Lefty assertion"? Good to know. That does make the case that the alternative "Righty assertion" that all that abuse is perfectly fine...well, ghoulish.
Those same women are equally likely to be sexually abused and beaten when married off as teens in their home village, where an not insignificant number also commit suicide. Sweatshops can be awful but a) they are range in conditions, not all are the same. b) they also provide a way for young women to avoid early marriage, earn some money and thus a higher status in their family, and get a bit of exposure to a the wider world. You do have to take into account what the alternative to work in a sweatshop is and for village women it is often early marriage and being a virtual slave to a mother-in-law, long hours of hard labor, and inadequate food.
If someone is going to engage in dual logical fallacies of Strawmen and Excluded Middle, it's unlikely to be of great utility - but yes as Noah own writing evokes
A. range of conditions is large (and in my experience one needs to look at both Formal and Informal employment, as informal is often the real abusive and equally the least or not subject to regs)
[The Lefty discourse on this rather ressembles that of the 19th century Marxists and Marx - collapsing a wide range of cases (truly awful to not great) into one, using Worst Case as Base Case and extrapolating from there
B. the other actual real world choices, which are not the idealised ones the Lefty professional activitsts see - as in indeed (a) barely surviving rural labor on yes inadequate food and family overall barely getting by (although nicely photogenic poverty for OxFam walls) with early marriage unlikely own choice for either young men or young women, (b) urban informality like rag picking etc in nastier conditions than typical formal factory, (c) unemployment and non-legal (as like prostitution)
Collapsing this down to "beaten to sew soccer balls" is emotional sloganeering.
This isn’t a very charitable interpretation. Noah literally uses increased health of children as an example of the benefits. Also, a few terrible factories shouldn’t doom the whole project. Close the terrible ones and keep the good ones, right?
This goes to an idea of how much the ends justify the means. Perhaps a country could attract 3 billion in FDI if they agreed to allow companies to come in, build a factory, and enslave the workers for 5 years.
For just 5 years of a few thousand literal slaves, a poor country could have a capital endowment of 400 shiny new factories that could then start hiring workers at good wages for good jobs. This means health and wealth and money for MILLIONS of the people of that country.
Most people would say that this isn't worth it. Even if you could unequivocally show that the country post slavery would be X amount richer, the ends in this case don't justify the means.
Noah really doesn't want to put a line on "At what point are factories too unsafe, are the workers too exploited?" Because then he would have to be on record as saying that "X number of new factories is worth the death of Y number of Bangladeshis."
Should we feel good if we improve labour standards and the exploitive garment countries moves out of Bangladesh to another country and people fall back into extreme poverty.
Then we ban the exploitive garment companies and people buy their Tshirts directly from exploitive Chinese companies.
Then we ban any tshirt sales from China and people buy directly from other exploitive countries.
I’d like to believe that we can successfully play whack-a-mole with human nature but nothing in the real world seems to suggest it works that neatly.
I think the current process where the US government and people push for incremental reforms in countries where we buy goods is probably the best we can do as much as it sucks to admit that we’re powerless as a group to do more (individuals of course can and should choose to do what they think is right).
It's not about banning specific countries.
I remember one time I was taking an uber in Singapore. The driver, in his day job, was a sand inspector for a company that imported sand into Singapore.
Singapore needs sand because it has become 22% larger since 1965 and all of that is done by dumping sand into the ocean and building new land.
But not all sand is created equal, depending on what you need the sand for, it needs to have the right grain size, the right mineral composition, the right weight... it was all super technical.
Now, he was hired by his company, because the Singapore government has a policy. If the government import sand, and find out that the sand is not up to spec, than the Singapore government will refuse to take it or pay for it. This leaves the import companies with a boat with 1000 tons of sand they can't sell.
To insure against that possibility, it's worth it for the company to pay someone who's only job is to inspect sand quality.
Does it make the sand slightly more expensive? Probably. Does it mean that Singapore doesn't import bad sand? Definitely.
If you make companies that sell in the US to some degree liable for labor abuses/safety standards in their supply chains, than they will assert pressure as the final buyers to clean up said supply chains.
Will their be leakage? Sure.
Is it an easy question to decide how much liability they should have? No.
How much pressure should countries require the companies who sell in their markets to put on cleaning up their supply chains?
In this very piece, Noah admits that this kind of pressure works to raise labor and safety standards. He also admits that some amount of this pressure is compatible with industrializing out of poverty. Even Bangladesh in their successful modern garment story never did the 18th century British tradition of having 8 year olds pulling bits of cotton out very fast spinning looms. Bangladesh said, "This is a horrifically bad idea, so, even though we could be profitably employing our elementary schoolers to the benefit of foreign capital, let's not." Maybe they would have made a lot more factories in the 90's, been richer today, with less NET child poverty today had they done so...
But this is a real question that is worth a real response and discussion.
This peace avoids this much more difficult and technical discussion.
Totally a fair critique. I do think Noah’s post serves a really important purpose. We can’t take our eye off the ball of how effective industrialization is at reducing extreme poverty. That should be the north star of anyone who care about reducing global suffering.
The question we’re asking is how much pressure can America and Europe put on supply chains before the economic incentives break.
I think we’d both agree its a bad outcome if governments exert too much pressure and you end up with a situation with an industry dominated by Chinese companies who fake certifications or who ship directly to uncaring consumers.
The unintended consequences of overly strict regulations will likely be an even worse global standards for sweatshops.
I’d love to see a progressive movement that did a better job grappling with these trade-offs. So rarely do I see a reckoning with our own political limitations. “We can push this hard on the Bangladeshi garment industry to improve standards, if we push any harder people are going to seal out cheaper and more unethical options.”
One challenge here is that we're often paternalistically substituting our own values and ideas about these tradeoffs for those of the people actually working in the factories and their governments.
How the ethics of doing that shakes out can maybe be debated, but we should be aware that there are actual humans involved who may not share our values about these tradeoffs.
Because he thinks new factories result in fewer deaths, not more?
Noah was being ghoulish. Posts like his is what gives Neoliberals (like me) a bad name. It's ghoulish to minimize the barbarity inflicted on the workers of worlds developing nations and pretend it's a necessary trade off to their advancement. It's not.
I was actually going to write a sarcastic reply about how it would be even better for a nation to employ slave labor.
A nation like Bangladesh has a huge comparative advantage over advanced nations. The margin between them is large enough that it does not require forcing women to work 16 hour days to make their garments competitive to dominate our markets. Bangladesh could have had some very limited workers rights by western standards and they still could have undercut and bankrupted any western manufacturers.
What Bangladesh does have to face is another poor nation willing to gain a comparative advantage by abusing workers. Poor nations face a race to the bottom to be competitive. For example, Bangladesh's garment industry could be put out of business if a nation employed slave labor... IF the west would buy from them.
But we won't. (well we do sometimes, but only if we can be in denial about it. ) It's illegal. The point is, We, the advanced nations, set the floor on the race to the bottom.
Advanced nations do set the floor higher than slave labor. We could set it higher We could demand very minimal standards of workers rights. We could prohibit imports of products that force women to work 16 hour days, or work in unsanitary conditions, or be exposed to toxins. We could demand a path unionization.
We could demand all kinds of baby steps that would remove much of the barbarity workers endure while still letting poor nations maintain a huge comparative advantage and their path out of poverty.
But our clothes would cost a bit more.
I have long considered myself a neo liberal, but I have always lamented how brutal it was in practice. It didn't have to be.
It’s not a strawman argument if you’re heard it in real life. And I sure have as a progressive.
I've also met real life communists and anarcho capitalists. It doesn't make putting them as the "typical" mainstream movement member not a strawman move.
The ghoulish thing to say is that some work is so bad no one should be "allowed" to do it. If someone is free, then whatever they are doing, the alternative is worse. Taking away the thing you find inhumane makes their lives worse, not better.
Obviously chaining someone to a desk and beating them to work, so in other words slavery, is wrong.
Yeah, the thing so many outsiders don't want to grapple with is that the alternative to the horrible factory job can sometimes be that person, or their child, actually dying.
exactly
I don’t think it’s a strawman. In a student congress round in 2023, the resolution we were debating was a tax on all imported Bangladeshi goods until the fire safety situation was improved. Real professionals in this field are probably doing good work to get companies to guarantee their supply chains are free from child labor, but amateur Progressive commentators regularly get this very wrong. Their first impulse is to just make it harder to for American companies to import garments from global south countries. And, as seen on the right, often if the rank and file get an issue wrong, the elite can’t stop them. So posts like Noah’s are useful.
You know what is provably worse than factory employment? Rural poverty, which is one of the worst existences on earth.
Assuming those kids weren't kidnapped of the streets, then the question is why were they working at the factory.
The answer is usually that it was better than the alternative. Which is usually either starvation for their family, or prostitution.
Noah's laying his mainstream economist cards on the table here--this is, and has been for some time, the conventional wisdom among mainstream economists for several decades now: the best (or only) way for poor countries to get rich is by allowing foreign direct investment from rich countries to pour in, and to leverage whatever comparative advantages they can in the global marketplace. Much has been written about whether, either empirically or analytically, this is even true, but I'll just insert my own personal objection to the premise: just because you can measure one thing--GDP per capita--or a few things, like longevity or infant mortality, and show improvement over time doesn't mean you can say definitively that sweatshops are a necessary ingredient for progress.
If because your land or whatever becomes polluted or unusable or sold out from under you by your government to some multinational corporation, and you're forced into laboring a 996 (9am to 9pm, 6 days a week) schedule instead of working at your own pace with no overseer on the family or community farm, I'm pretty sure your productivity is going to go up a whole bunch, but is that a real improvement in your quality of life? No doubt your country's GDP is going to shoot up too as subsistence agricultural workers are driven into factory jobs, but is all that worth it if people lose autonomy over their lives? We in the West seem to think that prosperity as measured by national GDP is THE measurement by which all else should be measured, but for more traditional communities and peoples, that is just not the way they think and operate. What is the price of a traditional way of life, one going back perhaps centuries, ripped away? You can't measure that in dollars and cents.
This is not to say that I'm not a believer in economic or social progress, but the people that are the supposed beneficiaries of the wonderful bounty of foreign investment and integration into our global market system should at least have a say, and as far as I can tell they do not. They don't even have a say on whether or not they can go to the bathroom.
Obviously conditions and situations in different places can be wildly different and my own experiences, though real, are just anecdotes really. But I'll note two things I learned from doing volunteer work in the developing world:
1) Many of those family or community farms are so marginal that the difference between living and starving to death is determined by tiny differences in rainfall or in how the randomness of annual floods shakes out from year to year. Meeting so many people literally on the knife edge between sustainable life and death was a shock that I'm not really over, some 20 years later.
2) Outsiders need to be self-aware about their paternalism and their often casual willingness to cancel out the wishes and values of local people and replace them with their own values. This is especially true for managing heartbreakingly tough tradeoffs and around how to manage risk-taking in general.
Yeah for pre-industrial societies, having a stable wage is already a pretty good deal though, compared to not knowing that you would starve or not next year!
https://acoup.blog/2025/01/03/collections-coinage-and-the-tyranny-of-fantasy-gold/
My company has a large office in Delhi and sponsors a lot of charitable organizations, among which is a sewing machine training center. Extremely poor women, including those from scheduled castes, are trained on industrial machines. They can either get certified as sewing machine operators, which allows them to apply to work in the sweatshops, or can receive a high-quality domestic machine so that they can do home piecework or alterations.
I visited the center, met some of the women, and even learned a little about running the machines. It's hot, power is iffy, the machines can be dangerous... but the alternative for many of these women is literally begging on the street, which is also hot and dangerous. It's very clear that the Indians who sponsor and work at this center view it as essential for lifting these people out of poverty (which is a strong cultural obligation).
Much of what I've learned about India from my travels and interactions is that I, as a Westerner, know very little about the situation and have a lot of preconceptions which don't really apply.
The point about economic growth is fair, but I think the points about labor conditions are far from bullet-proof. The US government, corporations, and other developed nations have immense buying power. They can use that power to set standards across-the-board for what is acceptable level of negative externality.
I think reframing this as “trade policy” opens a lot wider range of tools to manage worker conditions as well as climate impact (“carbon tariff”).
In the extreme, wouldn’t the US have a trade policy about trading with a hypothetical country who used literal slave labor? Similar for egregious polluters?
The politics of raising consumer prices is challenging, but it’s happening today to protect other interests/values.
I’m curious what you think about reframing this as a trade policy issues about curbing major abuses, and how worker conditions parallels pollution!
I think the point is that countries that have those minimal standards are usually better developed and have a cost of X
Now a really poor country comes along and says we will ignore all those standards and do X - 20%
The really poor country is fine doing that because the alternative is often starvation.
So should Western countries stop the really poor country from getting a foothold in the industry?
Thanks for the thoughtful response.
I understand and completely agree that sweatshops are better than starvation, and definitely an improvement for the extremely poor. I also embrace that it has been a very successful rung on the ladder to middle income countries.
I agree with you -- Developed countries shouldn't stop the very poor countries from getting a foothold.
My question is: Suppose a trading bloc of countries with a lot of buying power wanted BOTH to buy cheap clothes from very poor countries, AND to require some minimum standards on their labor practices and environmental impact. Is there a trade policy that could achieve both goals? If so, at what cost? Maybe these would all have side effects that are politically nonstarters (higher consumer prices), but curious if you had the political will, what options would be on the table? Maybe:
- Yes, we could require some minimum wage (much lower than $7.25/h), some environmental factors etc. This would raise consumer prices some, because those poor people are paid slightly more than they
- No, then buyers will nearly always pick a richer country. The poor country will never get the foothold
- Something else?
"My question is: Suppose a trading bloc of countries with a lot of buying power wanted BOTH to buy cheap clothes from very poor countries, AND to require some minimum standards on their labor practices and environmental impact"
I guess it's theoretically possible. I guess the question is what happens when another very poor country comes along and says hey we want to do this too.
I do agree there has to be some minimal standards. People dying in fires, or being chained to work stations clearly cross the line. Dying in a fire isn't better than dying by starvation.
But I would just be careful where drawing the lines. Good intentions can still create bad results.
I think it’s fascinating that MAGA has adopted the left’s “sweatshops are bad” narrative to excuse the inflationary and other negative effects of tariffs and essentially argue for the repatriation of low paying, labor-intensive employment to the US.
The lower class is saying they would rather have low paying, labor-intensive jobs.
What do you think of the progressives response of, "no you wouldn't." Then being surprised when they actually vote for it.
As with many things MAGA, they've identified a real problem (eg working class communities got destroyed by unskilled manufacturing job losses), but associated the problem with the wrong causes (globalization hollowed out manufacturing communities).
The real problem is that US industry moved up the value chain with automation that's raised productivity, but also significantly raised the skill requirements for the marginal factory worker.
We have like 800,000 unfilled manufacturing jobs in the US and a whole bunch of unhappy MAGA folks without the skills and/or work ethic to get hired in those jobs (electricians, pipefitters, machinists, welders, mechatronics tech).
The problem is a mismatch between the skills needed and the skills available in the labor pool and nothing MAGA or the Trump Administration is doing seems likely to address this.
To me, the politically smart answer to this is 1) recognizing the issue is that we messed up by telling everyone to go to college, and 2) building pathways to allow people to get help/achieve success learning marketable trades (loans for training, but w/institutions held accountable for success, maybe public/private training partnerships, etc)
Well no problem now as you will have a load of white collar ex tech wanting to retrain as pipe fitters and plumbers etc . At least they have maybe 10 years at that until robots replace them. Oh and of course robots can be chained to their work station.
Probably going to need some "learn to weld" programs for out of work coders :)
It's quite clear that globalization has had an effect on manufacturing jobs.
Paying someone in a poor country $1 an hour is cheaper than paying an American $20 an hour (even after taking into account shipping).
But American consumers have also benefited by getting much cheaper stuff.
Also automation has had at least as big of an effect.
That's fair, but most of those $20 -> $1 an hour jobs got offshored in the 80s and 90s. Recent job losses from globalization (say in the past decade) are lower, where automation is taking a bigger and bigger share of the responsibility for job losses. I think in the last decade, automation accounts for something like 85-90% of manufacturing job losses.
This is why I think the idea of tariffs to bring back jobs is so fraught. As I said in another comment below, Chinese EVs are made w/10-15 hours of labor, while US ICE cars use 25-35 hours. some of that is EVs vs ICE, but a lot is just more automation.
To make things worse, not only does a Chinese EV factory use 1/2 the number of labor hours per car, their labor is 60-80% skilled compared to US factories that are more like 20-30% skilled. Right now the US has around 800,000 skilled manufacturing job openings that industry can't fill.
So we're using tariffs to bring back jobs to the US, but they're going to be fewer than some people think and they're going to require skills our workers don't have, and we're going to make the skilled labor shortage even more acute. This will be great for those skilled workers, but also will raise costs for consumers even more on top of tariffs.
I've seen some research that says a lot of the job losses really accelerated around 2000 when China entered the WTO (page 2)
https://www.jec.senate.gov/public/_cache/files/94bf8985-1e87-438b-9a3a-e3334489dd30/background-on-issues-in-us-manufacturing-and-supply-chains-final.pdf
But I agree tariffs aren't going to bring back a large number of jobs.
I think we should be pushing for free trade agreements with other friendly nations (Canada, the EU, Japan etc) as long as they open up their markets to ours.
I'm ok with tariffs on China. Say an additional 10% each year until they
open up their economy
stop stealing tech
stop threatening Taiwan
basically join the liberal world order.
I agree with most of that, except I'd probably limit China tariffs mostly just to strategic industries, rather than using tariffs to try to change them into something they're not going to become and making us poorer for no reason.
Of course, in the grand scheme of things I'd be thrilled to get your plan as is instead of what we've got now.
What I'm curious about though is how many of those jobs were lost to China (or other low-wage countries) and how many of them were lost to automation?
I also wonder if the US and UK suffered needless deindustrialization because their unions were organized on the basis of occupation (which made them hostile to change) rather than by firm (as is common in Japan) or on an industry-wide basis (as is the case in Germany)?
Globalization did significantly contribute to the hollowing out of our manufacturing communities. We should learn from that mistake as well.
If as you say we have a significant number of unskilled workers, shouldn't we stop importing foreign unskilled workers?
First, I agree that globalization played a role in hollowing out manufacturing, but I'd say it played a much smaller role than automation (like 70-80% automation/20-30% offshoring, with even more automation more recently).
But here's the thing, a Chinese EV requires 10-15 hours of labor per vehicle while a US ICE car requires 25-35 hours. So even if the 20-30% of jobs lost to globalization had stayed, they'd be on their way out by now as we move to more automated manufacturing. Again, the solution to this problem is getting our workforce better trained and helping them deal with substance abuse. Nothing the Trump Administration is doing seems geared toward helping with that.
On your other point, I am very much in favor of controlling the border. I think if Biden didn't fumble that so badly Trump wouldn't be president.
China created an economy for 1.5 billion people based on manufacturing...obviously manufacturing creates jobs.
You need people to make the machines that automate the manufacturing, people to make the steel, people to mine the ore, people to transport everything all around.
You need an entire economy to manufacture stuff. It's not just 100 guys in a warehouse cranking out cars.
First, close to half the labor-intensive jobs in industries like textiles have already been moved from China to poorer countries. EVs were just a crisp example. The trend is across industries.
My assertion is that automation is primarily what's been eating unskilled jobs recently, both here and all around the world. It's only going to get worse as robots and machines get even more capable.
New modern factories across industries need fewer workers overall plus a relatively higher ratio of skilled to unskilled workers. That ratio will continue moving higher as automation gains capability. That doesn't mean less jobs overall, but likely less demand for unskilled labor versus supply.
Everyone likes a scapegoat and globalization seems to be what the MAGA populists have latched onto. That's a mistake, not just because their remedies are going to make us all poorer. It's also a mistake because they're chasing the wrong villain and not doing anything to address the real problem, which is the skills mismatch (and prob. also substance abuse in hard hit communities).
Noah, don’t economies grow as laborers become better trained and more productive? Surely in Bangladesh, they have people who can fix sewing machines by now. It should mean that more trained people who can work with machinery and computers can work with more sophisticated machinery higher up in the value chain. They will earn more income. The more computers in the country, the higher value jobs will be offered due to having a trained workforce.
Your charts on the lowering of poverty, along with the chart showing higher GDP, aren’t they not the story of trade and other programs that have lifted the entire world out of poverty that was seen only 40 years ago?
As for the stupidity of many on the left, if Zoe Saldana had been the model the clothing seller had chosen, would anyone say they are saying black genetics are superior? It would be tough to argue that neither woman is beautiful.
He did argue in his post about Bangladesh that they should move up to higher value chain though.
The point is that between extreme poverty and a lower-middle income, sweatshops are unfortunately necessary for many countries. (India tried to leapfrog to service economy, and this caused another bunch of issues, such as high youth unemployment and low labor participation of women, led to high gender inequality).
https://www.youtube.com/watch?v=Lvzn41Lv-O4
I would say much of the difficulties of poorer countries around the world are their attitude towards women working and holding jobs. Afghanistan will never be a richer nation. Oh, they may sell some rocks to China, but that will go into the corrupt pockets of the Taliban. Asia, Sub Asia has the same problem. When women work and the pill is available, they prefer to work and make their own money rather than pop kids out one after another.
I wonder how many people in the West contrast Afghanistan with the ex-Soviet "-stans" to the north and think "maybe in the 1980s we should have just let the Soviets win"?
Whatever its failings, communism was at least a modernising force. (Pashtun nationalism, not so much). Unfortunately, communism's failings helped bring modernisation into disrepute. And now, all Afghan's have to live with Jurassic conservatism.
My point is more that Soviets (unlike the modern West) had the requisite level of ruthlessness to modernise a tribal Muslim society like Afghanistan.
The problem was that they really didn't. The Mujahideen were another level of brutal. A friend of mine started reading a history of the Soviet invasion and had to stop a third of the way through because it was making him physically sick. Barbarism without constraint. It will be like N. Korea from now on: throw a cordon sanitaire around it and retreat to a safe distance.
LOL, I am positive they are not thriving economies. Putin wouldn’t allow the competition.
Never is a very long time. In the 1950s, South Korea was the poorest in the world. Who knows where Afghanistan will be in 70 years.
Oh com’on. If the Taliban are still in charge, I think we can expect exactly what they currently have. North Korea and Russia are exactly the same because their rulers are the same. Something very different would have to happen in Afghanistan for it to change.
That’s incredible I was thinking the same thing myself about both White Lotus hotties! Give Paula her own jean commercial! And I think black jeans need a comeback fashion wise 😎
I’d have a lot more sympathy for the whole “sweatshops are a necessary evil” narrative if they’d at least be open to unionizing the sweatshops and letting the workers collectively-bargain their way to a decent local equilibrium. Otherwise the whole situation looks like classic economic duress.
Also, what happens when there’s no longer a bottomless pool of low-skilled labor to tap into? I know we still have Sub-Saharan Africa to count on, at least until their populations age out like the rest of us, but what’s the industrialization path for poor countries in the aging, low-fertility future?
For your last questions, I think the answer is that if the pool decreases, then the price of labor goes up, at which point automation becomes a more competitive option. That is to say, we'll invest more in robots that can sew our clothes, requiring less labor.
This is a bit out of date, but the US militarily is or was investing in sewing robots: https://www.npr.org/transcripts/488611449?storyId=488611449
"Also, what happens when there’s no longer a bottomless pool of low-skilled labor to tap into"
wages rise. and consumer goods get more expensive.
There is a game (Victoria 3) about economic development of countries in Victorian era, and your arguments about opening for foreign investments are similar to the debate in-game about allowing foreign investments though!
https://www.reddit.com/r/victoria3/comments/1lsc09d/open_foreign_investment_in_vic3_imperial_russia/
Side note: One of the "meta" in that game to boost GDP growth is to only allowing foreign investment to help countries escaping from poverty, then not allowing foreign countries to invest before nationalizing their industries and selling it to local private owners. Yet I can't recall any countries in modern day who actually succeeded when doing this; are there any particular reasons behind this?
(I think that would be due to the institutional instability/sovereign risks that these actions created for future investors + potential for corruption and cronyism, but not sure about other reasons).
Presumably any country looking to develop will need to buy capital equipment from foreigners that it doesn't (yet) know how to produce itself, and it will be extremely difficult for them to get the foreign currency to do this if they have a track record of expropriating foreign investors?
(This makes me think of how the Holodomor happened because Stalin wanted to export the Ukrainian grain to pay for American industrial equipment, even if the Ukrainian peasants starved.)
That was covered in my section about "sovereign risks" already though!
(A better example is Argentina, being rich before the Great Depression due to agricultural exports and British investments, but after institutional crises, Peron came to power, and got factories/railroads sold by the Brits. His mismanagement of these assets, plus his protectionist model and export duties for agricultural goods for industrialisation, led to the start of Argentine decline).
Cuba post revolution is also a good example for two reasons I stated earlier, but I wonder whether there are any other reasons:
https://snowdentodd.substack.com/p/cuba-and-venezuela-at-the-end
Wouldn't Argentina's failure to industrialize be more down to its small population and geographically isolated location? Same reason why Australia is now primarily a commodity exporter, even if an unusually wealthy one.
And the decline in Argentine agricultural exports (which were mostly to the UK) was likely more because the British decided to modernize their own agriculture, after WWII's Battle of the Atlantic showed how dangerous an extreme dependency on imported food could be, even for a country that notionally "ruled the waves".
Argentina tried to industrialize before the Great Depression, but at that time its share of GDP for manufacturing was half of Australia!
And no, the British didn't just decide to modernize agriculture; if anything the UK in WW2 was self-sufficient in food, the only things they lacked were luxury foods and meat (especially beef and mutton - that they bought from Argentina a lot).
The decline of UK imports with Argentina can be traced back to Ottawa Declaration, that raised tariffs for entire British Empire; as well as the Roca-Runciman treaty, in which Argentina got to keep beef export quota but had to open the market for British manufactured goods.
https://en.wikipedia.org/wiki/British_Empire_Economic_Conference
https://en.wikipedia.org/wiki/Roca%E2%80%93Runciman_Treaty
Short version, it’s complicated. It’s also deeply studied because of Argentina’s unusual economic path.
https://en.wikipedia.org/wiki/Economic_history_of_Argentina
Fantastic post. I am surprised that you would hae these opinions. It gives me hope.
The so-called "Progressives" want to pull the rug out from under developing countries...ending hydocarbon fuels, ending jobs, ending their ascent out of poverty that is the MOST dramatic story of the last 30 years.
Sweatshops is a neologism, a word made up, for people who are poor working at jobs that fortunately, you and I and our children will hopefully never work at. But that doesn't mean these jobs are bad. All work has dignity so long as it is a voluntary exchange between employer and employee.
In 20 years, these jobs will be gone, and hopefully the people working in these shops will have their own businesses, and will employ robots and be living an amazing life.
Everyone has to start somewhere.
Also, this makes some arguments bashing OpenAI (or big AI firms) about their low pay for data labellers in Africa (2 USD/hour) wrong; of course these labellers had to see brutal contents and as such, having better compensation would be great, but having a consistent wage and working in house is a dream for many people there, plus it is still a significant wage compared to the alternative!
Being a sweatshop worker sucks, but being a subsistence farmer is usually even worse - otherwise people wouldn't choose to leave the farm in the first place. And as the other opportunies for people get better, the sweatshops either stop being so terrible or start paying a lot more money.
It's all about the alternatives.
"Bangladesh’s conquest of extreme poverty hasn’t come from any dramatic program of government redistribution, but from rapid economic growth. In the years since 1990, the country has more than quadrupled its living standards, blowing past Pakistan"
You are conveniently ignoring Bangladeshi land reform: https://en.banglapedia.org/index.php/Land_Reforms We might observe that most developed nations went through a
Land reform give peasants better income alternatives to sweatshop employment than they have without land reform.
Not to mention government mandated education: https://factsanddetails.com/south-asia/Bangladesh/Education_Health_Transportation_Infrastructure_Bangladesh/entry-8199.html
The fact is that race-to-the-bottom conditions that foster sweatshop abuses can be alleviated (or worsened) through government programs, such as land reform (or enclosure), education, worker rights to unionize, etc. All the "evidence" you have listed has occurred under race-to-the-bottom conditions.
Not to mention that injury, illness, and loss of educational opportunity are not counted by many of these development studies simply because they ignore losses to human capital.
Industrialization through factory production doesn't need to occur through a sweatshop phase. It can occur with modestly paid employees with unions, health and safety protections. Claims that the "alternative" is worse and that employment is "voluntary" generally overlook the fact that government policies or cultural traditions (such as permitting landlordism) are responsible for the horrible alternatives. Indeed, it is OBVIOUS that lots of industries develop without a sweatshop phase.
"Industrialization through factory production doesn't need to occur through a sweatshop phase. It can occur with modestly paid employees with unions, health and safety protections. "
Really? Say you are a dirt poor country, with no industry. You want to build one. The only thing you can really compete on it price. How do you get that lower price?
It's not with unions I can tell you that.
Establishment of a social-democratic welfare economy with land reform seems to have worked very well for Kerala, India. https://en.wikipedia.org/wiki/Kerala. Their high human development index allows them to compete outside of poverty level labor.
Really, ideologues really should look at the real world a bit more for counterexamples. If you paid more attention to the works of Amartya Sen, you might learn something.
The book “The Travels of a T-shirt in the Global Economy” goes into this issue in some depth. Very much recommended.
You vastly overestimated the sincerity of someone like Derek Guy on this sort of topic. You touched on it when you said, "“sweatshops are bad” is just one of those things that American progressives are supposed to believe".
Derek Guy is a progressive fashion influencer. At best he can be sincere about his fashion opinions. There is no possible way he can be sincere in is economic opinions. Meaning even if he believed that sweatshops had some role to play in a country's development, he couldn't actually say that because then he would lose influence with progressives...which is his whole job.
It's basically impossible to convince someone of something if their livelihoods depend on them not believing it.
As a practical observation from personal experience, it is further to be observed that the application of rich world standards blindly - in labor codes and other written factors often is a Potemkin village effort where formal employment and labor flows into the informal sector and hidden / off-the-books empoyment and even mini hidden shops doing jobbing w zero protection.
State capacity lacking, and corruption allowing, this becomes a whole hidden industry.
(as an additional factor besides pure unemployment and poverty or alternative alterantive, bare subsistance farming although the latter with the fetishizd 'small holder farmer' provides the OxFam types with nice photogenic poverty porn).
One doesn't get besides the economic carrying capacity the state capacity in one go, it has to be built, and Potemkin village facade law rather often just becomes corruptoin rents for officialdom.