'Michael Pettis also argues that cheap Chinese stuff actually makes Americans poorer, by reducing their domestic production so much that Americans actually end up consuming less. I’m highly skeptical of this argument, since a basic principle of economics is that people don’t voluntarily do things that make them poorer.'
I am not sure how this principle holds when you take large and heterogeneous groups such as 'Americans' or the 'Chinese' (and even for smaller groups it fails if you'd ask me). There are plenty of Americans who would do something that would make others poorer, as long as it makes them richer, even if total wealth decreases. Pushing addictive painkillers (or illegal drugs) on the market was a voluntary choice of some, even if drug (or painkiller) addictions reduce productivity of the addicted and thus total productivity and wealth creation. Moving production to the other side of the world has certainly made CEO's and shareholders wealthier, but not factory workers, nor plenty of local communities (and who says the shareholders and CEO's are domesticated in the US, meaning that even more corporate profits move elsewhere). The increasing call for onshoring and the potential economic benefits thereof shows that these past actions might have been a voluntary mistake, from a national point of view.
Buying a cheaper Chinese car does seem like an individually rational choice, but when a sufficient number of people makes this choice this could push some other manufacturer over the brink, meaning that an even larger number of cars won't be made in the US. And considering the size of automotive plants and companies these days, the bankruptcy of a company like Ford, however unlikely it seems, would certainly make Americans poorer.
Even on an individual level: How many people have voluntarily bought NFT's?
This doesn't say anything about whether or not we need tariffs though, it's just that your 'basic principle of economics' can work if, and only if, people would be fully 'rational' (which we're not) and have full information of the future effects of their actions (which we don't), and you see a nation as a homogeneous entity (which it is not). So maybe that basic principle needs to be revised if you're trying to make sense of the real world?
This is one of my favourite topics but one I don't fully understand. Here is my understanding and I would love it if you or anyone else could answer some questions that I have:
The Chinese government (especially the CCP) has amassed a great war chest in terms of funds/assets through taxation of individuals and companies which, in turn, received payments from around the world in the first two decades of the 21st century. In essence, 1 out of every 100 dollars or their equivalent created around the world by various central banks probably went into the CCP's pockets. China became the factory of the world as companies increasingly used its massive labour force and pro-capital policies working with thousands of contract manufacturers, especially as the products of Japanese and Korean companies became more expensive and American and European companies outsourced jobs and factories to China to produce goods at scale for an increasingly globalised world.
Given the state-owned banking system and a snakehold over financial executives, the CCP used some of this tax revenue as well as international dollar bonds issued by local governments through local government financing vehicles (LGFVs) to subsidise the manufacturing sector when it could have spent it instead on improving healthcare. (The high taxation levels also probably stopped Chinese workers from building wealth through their incomes so they went on a house buying binge.) Chinese companies, on the other hand, began to become global exporters in their own right developing their own 'brands' targeting the less affluent markets initially and then entering European and American markets. The emergence of the internet and improved shipping networks allowed them to export directly to the end consumer (e.g. Shein) or else export directly to wholesalers/retailers.
The People's Bank of China also controls the exchange rate preventing it from appreciating as renmimbi increasingly became an important currency in international trade.
So my questions are:
1) Why don't other large developing countries (India, Brazil, Indonesia) copy this?
2) If you think of the CCP or the Chinese government as basically a big bank that makes sub-optimal lending decisions then won't the money run out sometime? When?
3) China needs to increase domestic demand or consumption but what does that mean in practice? What are they unable to do that people take for granted in developed countries? Is this just a roundabout way of saying the CCP or the Chinese government needs to tax less and banks need to reduce interest rates for personal loans or credit card loans or give out more loans than they currently do etc.
4) Much of China's trade with the world actually involves intermediate goods, not final goods. Given companies rarely like to change suppliers and it's hard to find ones that can produce at scale, speed and at cost then can there ever truly be any alternatives?
5) Wouldn't the world suffer if tariffs on Chinese intermediate goods leads to higher prices for consumers worldwide? Doesn't everyone deserve an Alexa or an iPhone or their knock-offs?
6) If the PBOC controls the exchange rate like this and it has largely worked out for them then why do we advocate free float of exchange rates for developing countries? Provided they're good at managing their reserves, shouldn't they pick an optimal exchange rate and stick to it? How is it that some countries like Japan and China can control exchange rates but others like Sri Lanka or Argentina or Indonesia or India can't?
Brazil, most of East Asia and the Gulf countries did try this in the lead up to the GFC. Low US interest rates and fiscal deficits and over investment increased these countries trade surpluses and raw material prices (oil). Rather than letting their currencies appreciate they intervened, buying dollars and treasury bonds, which depressed yields, boosted liquidity and kept the investment boom going.
It is much easier to pursue this strategy when you have a trade surplus , not a lot of external debt, interest rates are falling and your export clientele are booming. Most EM countries these days aren’t in very strong financial positions and can’t replicate China’s strategy. They also don’t have the domestic savings.
We will can quibble about the details, but the real story here is that more and more of the (supposedly) smart people on both sides of the political aisle are willing to admit something that has been verboten for decades: "mercantilism works".
The best (non-specialist) book on this subject was published almost 25 years ago (before the rise of China!): Global Trade and Conflicting National Interests by Gomory and Baumol. They explain how classical trade models developed for English wool and Portuguese wine break down into multiple stable equilibria when you're talking about dozens of countries and thousands of products. The "national interest" part of the title refers to strategies (mercantilism) to shift the global equilibria in your nation's favor.
As an Australian importer that has been dealing with China and going there 2-3 times a year for 15 odd years, i read a lot of the economic theories and comments posted ( Eg Michael Pettis etc) but as is usually the case, those are typically academic and observational, but not directly transactionally knowledgeable.
What is missed in a lot of this ( and i am certainly not a booster for the Chinese CCP particularly) is just how brutally competitive the domestic Chinese market is. For any product.
1.5 billion consumers and many tiers of business customers. Cut your teeth there and survive and you are world class. The Chinese middle and upper class ( numbering far more than the US )
are now very discerning.
The " government" definitely promotes, helps fund, and opens trade doors for certain sectors that it wants to promote. But it is "hunger games" there let me tell you.
The efficiency that is embedded in their processes weeds out pretty quickly the weaker and lesser competitive and innovative manufacturers. ( Note almost no designer/power brands in concept industries )
The product innovation cycle there is 1/2 to 1/3 of the time involved in pretty much any western country and Australia is way down the line as we have basically no domestic manufacturing.
What has me baffled with the idea that Trump promoted, and apparently convinced the MAGA crowd and others generally about tariffs, was that the Chinese / other countries paid the tariffs. Well as we know, the US importers and consumers do. Hhmmm go figure.
So if i was China, Mexico, Canada, Europe etc, and Trump applied 10%, 20%, 60 % tariffs, i would ring the Donald and say ok, understood, we will put up our prices to the USA 10%, 20%, 60 % Before ... we invoice the importers, and take the pain here, and you can tell the US citizens we are paying. But in fact capture the profit margin that is far more than they work on now.
The BS and smoke and mirrors that is all this incredibly bright academic posting would come down to the same thing. The US consumers pay more and the exporters make more margin.
The multivariate challenge that trade surpluses represent for macro-economics reminds me of the old joke about the three academics stranded on a desert island with only tinned food to eat and nothing to open the cans with: the engineering professor sets about making a stone tool to puncture the cans; the physicist starts calculating whether they can heat the cans to a high enough temperature to blow the lids off; and the economist assumes a tin opener.
I’ve been reading Pettis for a long time. 15 years for sure, has he been around for 20? Been awhile. He’s been saying the same thing (or close to it) for a long time and has not yet reached the stopped clock stage of being right twice.
His observations and diagnoses are spot on, but from observations one must be able to make useful real world predictions (the world that bond and currency traders operate in) or to build a mathematical model which might make useful predictions under certain very limited circumstances (the imaginary world of economists).
What is clear is that China for a long time used fixed investment/real estate as “stimulus” to keep things ticking over even as debt piled up faster than GDP did. Pettis and others observed these diminishing returns were unsustainable and begged a Japan-style crisis, but this cycle sustained itself for a lot longer than most thought possible. Then Xi decided (late 2010s) that manufacturing exports and renewables were going to be what was pushed (and all of the European and American “stimulus” worked like a charm for Xi’s goods exports during Covid, reaching record levels during the alleged “supply shortages” of 2021-22). This does tie in with Xi’s strategic shift years ago to stop buying treasury bonds and instead use his surplus to invest and lend in emerging markets. These infrastructure projects abroad create demand for Chinese goods (or rather, were run by cadres for the purpose of importing Chinese goods, increasing Chinese influence and securing raw materials - much like old time colonialists) in addition to being a way to recycle surpluses (and perhaps to squander them, over the long run, though maybe they’d be squandered in treasury bonds over the long run if placed there instead…..or seized, like Russia’s).
Many observers consider Xi’s export strategy and EM lending/white elephants to be wasteful and unsustainable, and most pols consider it unfair. I doubt that tariffs will stop it because, as Churchill was claimed to have quipped: now we’re just negotiating the price. Neither price nor profit has been a major obstacle in China’s policy making. China was willing to go through wage and price deflation during Covid and who knows when all the pigeons in the banking system come home to roost. You can always make more pigeons.
Some people spend trillions to pay off donors or forgive student loans or hand out subsidies (free health insurance for those making 300k) to drive consumption. Others use it to build (too many) things (whether apartments or solar panels).
There are obviously fractures in Chinese society. A middle and professional class has been created but there aren’t enough jobs (and the best ones are reserved for cadres). When young people can’t find good jobs and can’t afford flats there will be some tension. However, the savings of 4 grandparents and 2 parents will eventually flow to one young person. Also, the working class is still a lot larger than the middle class and Xi’s manufacturing subsidies are keeping them busy.
The move toward tariffs and trade bans (which I support in certain cases) may end up being well-timed with Xi’s manufacturing extravaganza failing. If jobs are lost, who will the working class blame? Xi or the Gweilo?
Noah writes: "In order for cheap Chinese imports to actually impoverish Americans, there would have to be some kind of externality or coordination problem involved. That might be the case, but Pettis or the MAGA people need to explain what they think that externality is. It’s not readily apparent to me what it might be."
Isn't forced de-industrialization exactly such an "externality"? It acts indirectly to impoverish a country by stripping it of its international influence and ultimately even the ability to defend its own interests and territory.
I don't think that's what Michael Pettis means, probably, and I don't think it's really what the MAGA people mean either. Trump seems to focus on "lost business", which is perhaps why he is still sore at the Japanese, and on American jobs generally. But it's definitely part of what drives concern about that supply chains that supply the US DoD, and why there is so much concern about the origin of semiconductors these days.
I'm no economist but I always find myself wondering if there isn't a sort of critical mass of diversity that an economy needs to be healthy. You see economies around the world that apparently thrive but depend too much on tourism for example. It invariably reveals itself as hollowness and fragility when times get tough. Similarly an economy in which everyone is a clerk at Walmart or flipping burgers at McDonalds seems hopelessly fragile and imbalanced, not in a trade balance sense, but in a robustness-in-the-face-of-difficulties sense.
America used to build ocean-going vessels. We don’t do that. It went to Japan, then Korea and I’m sure that manufacturing will devolve and go somewhere else as those economies about make more valuable products. Japan used to produce most electronics, then it also moved to South Korea and then on to China. What would Taiwan export if not for semiconductors?
America couldn’t affordably build those ships anymore. We can barely build aircraft carriers or submarines anymore. This problem and history is best illustrated by the American car manufacturers. The Big Three cannot create a $25,000 EV and make money. Their labor costs and legacy costs choke them. China can, but we want to protect our car-making ability for jobs.
Those cars get subsidies that studies show largely went to the wealthy and car manufacturers. It was a job program. Our military-industrial base is mainly gone, a shadow of its former Cold War self. Even if we were double our military budget, we do not yet know what to build. Drones or manned weapons? Which will be more effective. The Pacific, the body of water China wants to control, would need ships. Either way, you need big ships. An EV drone ship cannot traverse the Pacific. America doesn’t know what to build, the government is an awful prognosticator, and industrial policy produces favored industries and companies, not effciencies.
GM manufactures ICE cars well below $25K. We just don't have the supply chains to build EVs at those prices -- yet. We can and should be building them, but we're probably about to stop those plans and hand the (entire automobile) industry to China. It doesn't really matter what we "want to build" in the future if we don't invest in building the supply chains necessary for it.
I'm going to say this then in response to my post. The Big Three cannot build ICE vehicles that Americans want. Most of the Big Three no longer build sedans Americans want. I think you’ll find the actual purchase price of well-equipped vehicles after adding taxes, the vast amount of those cheap vehicles come closer to $30,000. The world has yet to flock to the Chevy Volt. GM doest’ not sell a crapload of Chevy Malbu’s. They make money selling trucks.
I have been in the automotive sector for 35 years. You learn a few things. Trucks make profits for the Big Three.
"GM no longer manufactures..." is a funny thing to hear in 2025. My liberal father has been saying the same things since the 1970s. Ironically he's now on his second Volt and absolutely loves it. I recently leased a cheap Honda Prologue, which is actually manufactured by GM, and it's a shockingly fantastic car.
Other than that I'm not going to defend US car manufacturers. Where we seem to disagree is that I think this is due to bad management, and you think it's due to labor. (And also: I think that losing that industry would be a crushing blow to the US from which we would have difficulty recovering.)
The emergent war technologies leave governments in a dilemma. They don't know if they are building targets or weapons or both. They will have to build both until a major peer to peer war occurs and sorts out the issue. It is obvious they will have to build a sh*t ton of equipment no matter what they build.
The failure of Trump I tariffs to enhance domestic manufacturing is of a piece with the failure of the corporate tax cuts to do the same. After complaining for years that corporate taxes are holding back business investment and claiming that a reduction would bring back the jobs “stolen” by China, American companies did a lot of stock buybacks and consolidation. We make many welfare recipients work to justify their temporary assistance checks. Why not do the same with tax expenditures, conditioning cuts on domestic investment?
China is the geopolitical rival, but the US's biggest trading partners are Mexico and Canada. I'd like to see an article with this depth of analysis on effects of tariffs on Mexico and Canada, including the effects on all 3 countries, impacts on household income, etc.
Because the northern Mexican states and virtually all Canadian provinces trade more with the US than within their own country, they are all a completely integrated part of the American economy, and Trump's plan seems similar to (and as sensible as) erecting tariffs between US states.
Would a shift to more domestic consumption in China create a dynamic of rising expectations and more "liberal-thinking" among the Chinese people that Xi Jinping might view as harder to control and a threat to his goal of militarizing China to threaten and eventually overcome the dominance of the West?
"Xi is making the Chinese economy look a little bit more like the old Soviet one, where production was determined by plans instead of by the market." Well it's good to see Noah saying this explicitly. Two articles ago Marx and Lenin's ideas were dead. I guess the distinction is that we believe this central planning is driven by expediency and not ideology?
Back to the subject of this article: what concerns me about Pettis's ideas is that they aren't really new. We've been hearing about conservative ideas to support the US government via a consumption tax (Google "national retail sales tax") for at least fifteen years now. Economists never had great things to say about this proposal. Is the "tariff everything we import in order to reduce US consumption and pay for income tax cuts" idea just the same pig dressed up in different lipstick?
Everybody could tariff China together, if the United States used the threat of its own tariffs to achieve that coordination. But that's not going to happen. Trump believes in tariffs but not international coordination, while most Democrats have the reverse beliefs.
Thanks for the post. I've been waiting for somebody to highlight how much Pettis has gotten right. Too bad that naive American tariffs won't fix things!
The impression I get is that Democrats have been pretty happy using targeted tariffs -- but they didn't have the political courage to actually do much about it until Trump shifted the Overton window.
I think the inflation angle is really important in explaining the last few years' of trade data. China has been having much lower inflation than the rest of the world, depreciating its real exchange rate, making trade surplus worse, while also depressing domestic demand.
Which is why I kinda both agree and disagree with Pettis. Lack of domestic consumption can definitely be a syndrome of a bigger problem that also manifests in trade imbalance, but, as in this case, you have to specify the underlying mechanism moving both, otherwise, it's not really a theory. and I don't see Pettis doing so.
"Today, [unlike in the 1930s], Americans consume far too large a share of what they produce, and so they must import the difference from abroad." What? Note to Pettis from 5th grade teacher, to take a difference you have to have two entities, A-B is a difference. In my field, physical chemistry, someone writing this would be forever considered a quack. There is apparently an implicit factor "consumption of goods (CG)" such that CG = CG_d + CG_i. The premiss of the argument is that Americans consume too many US produced goods. The logical result of his argument would be to tax (e.g. VAT) US made goods to reduce the share consumed. Calling for tariffs on imports when the premise of your argument is that US consumption of US production is too high is so logically flawed that no person who understands logic would accept it. My 7th grade mathematical logic teacher (back when CA schools put money into gifted programs) would have given me an F if I was that stupid.
Then he fully goes into irrational land: “By taxing consumption to subsidize production, modern-day tariffs would redirect a portion of U.S. demand toward increasing the total amount of goods and services produced at home. That would lead U.S. GDP to rise, resulting in higher employment, higher wages, and less debt. American households would be able to consume more, even as consumption as a share of GDP declined…”
Tariffs are taxes paid in the US. They are not production subsidies. Are we to assume in his mind all tariff revenue will be used to subsidize US production. How would that work? The US is roughly at full employment. How are we going to get higher employment? Production has inputs of physical capital, intellectual capital, and physical labor. If labor is near its maximum (and potentially decreasing with deportations) then to produce more goods in the US can only occur through increasing physical and intellectual capital inputs, both of which are slow processes even with the assumption of new funding (the “subsidies”).
So it is almost impossible for employment to increase in the US for native born workers because of demographics. We can, and have, increased employment through immigration. Higher wages? Mathematically real higher wages can only occur with increases in real GDP except through large scale economic redistribution from non-wage earners to wage earners. The trend in the US has been in the opposite direction. Less debt? He offers zero support for this claim. Again we see arguments that no intelligent person could accept. Claims need to be substantiated by facts, logic and mathematics for any intelligent critical thinker to accept. And finally: “American households would be able to consume more, even as consumption as a share of GDP declined…”. Once again a claim with zero facts, logic and mathematics. And again mathematically and logically impossible. 8% of the workforce is in manufacturing and we consume ca. 30% of expenditures on goods (and part of this is basic food stuffs). So ceteris paribus taxes, spending, …, try some simple math. He makes no connection between consumption and GDP but implicit is that consumption is a near linear function of GDP. Product consumption is ca. 30% of spending (apparently linear in GDP in his non-model). A 1% decrease in spending on goods (again “even as consumption as a share of GDP declined” [see implicit linear GDP spending requirement] ) requires a 3% increase in GDP all coming from the new subsidized manufacturing production. 8% of the non-farm workforce (FedFred) is in manufacturing. A 3% increase in GDP from 8% of the workforce would require a manufacturing productivity increase of 1180%.
Apparently many people can’t do 7th grade logic and high school level math. Anyone who has these skills understands this guy is a complete quack. I could do this math and logic in 7th grade. Apparently much of the economics world didn’t pass 7th grade. Yet Noah claims “Whatever you think of Pettis’ theories, I think he’s probably the single most important and influential international economics theorist in the world today. “. Noah thinks someone who couldn’t pass my 7th grade and do very simple algebra is an important economics theorist. No, he is a logically and mathematically incompetent quack.
'Michael Pettis also argues that cheap Chinese stuff actually makes Americans poorer, by reducing their domestic production so much that Americans actually end up consuming less. I’m highly skeptical of this argument, since a basic principle of economics is that people don’t voluntarily do things that make them poorer.'
I am not sure how this principle holds when you take large and heterogeneous groups such as 'Americans' or the 'Chinese' (and even for smaller groups it fails if you'd ask me). There are plenty of Americans who would do something that would make others poorer, as long as it makes them richer, even if total wealth decreases. Pushing addictive painkillers (or illegal drugs) on the market was a voluntary choice of some, even if drug (or painkiller) addictions reduce productivity of the addicted and thus total productivity and wealth creation. Moving production to the other side of the world has certainly made CEO's and shareholders wealthier, but not factory workers, nor plenty of local communities (and who says the shareholders and CEO's are domesticated in the US, meaning that even more corporate profits move elsewhere). The increasing call for onshoring and the potential economic benefits thereof shows that these past actions might have been a voluntary mistake, from a national point of view.
Buying a cheaper Chinese car does seem like an individually rational choice, but when a sufficient number of people makes this choice this could push some other manufacturer over the brink, meaning that an even larger number of cars won't be made in the US. And considering the size of automotive plants and companies these days, the bankruptcy of a company like Ford, however unlikely it seems, would certainly make Americans poorer.
Even on an individual level: How many people have voluntarily bought NFT's?
This doesn't say anything about whether or not we need tariffs though, it's just that your 'basic principle of economics' can work if, and only if, people would be fully 'rational' (which we're not) and have full information of the future effects of their actions (which we don't), and you see a nation as a homogeneous entity (which it is not). So maybe that basic principle needs to be revised if you're trying to make sense of the real world?
This is one of my favourite topics but one I don't fully understand. Here is my understanding and I would love it if you or anyone else could answer some questions that I have:
The Chinese government (especially the CCP) has amassed a great war chest in terms of funds/assets through taxation of individuals and companies which, in turn, received payments from around the world in the first two decades of the 21st century. In essence, 1 out of every 100 dollars or their equivalent created around the world by various central banks probably went into the CCP's pockets. China became the factory of the world as companies increasingly used its massive labour force and pro-capital policies working with thousands of contract manufacturers, especially as the products of Japanese and Korean companies became more expensive and American and European companies outsourced jobs and factories to China to produce goods at scale for an increasingly globalised world.
Given the state-owned banking system and a snakehold over financial executives, the CCP used some of this tax revenue as well as international dollar bonds issued by local governments through local government financing vehicles (LGFVs) to subsidise the manufacturing sector when it could have spent it instead on improving healthcare. (The high taxation levels also probably stopped Chinese workers from building wealth through their incomes so they went on a house buying binge.) Chinese companies, on the other hand, began to become global exporters in their own right developing their own 'brands' targeting the less affluent markets initially and then entering European and American markets. The emergence of the internet and improved shipping networks allowed them to export directly to the end consumer (e.g. Shein) or else export directly to wholesalers/retailers.
The People's Bank of China also controls the exchange rate preventing it from appreciating as renmimbi increasingly became an important currency in international trade.
So my questions are:
1) Why don't other large developing countries (India, Brazil, Indonesia) copy this?
2) If you think of the CCP or the Chinese government as basically a big bank that makes sub-optimal lending decisions then won't the money run out sometime? When?
3) China needs to increase domestic demand or consumption but what does that mean in practice? What are they unable to do that people take for granted in developed countries? Is this just a roundabout way of saying the CCP or the Chinese government needs to tax less and banks need to reduce interest rates for personal loans or credit card loans or give out more loans than they currently do etc.
4) Much of China's trade with the world actually involves intermediate goods, not final goods. Given companies rarely like to change suppliers and it's hard to find ones that can produce at scale, speed and at cost then can there ever truly be any alternatives?
5) Wouldn't the world suffer if tariffs on Chinese intermediate goods leads to higher prices for consumers worldwide? Doesn't everyone deserve an Alexa or an iPhone or their knock-offs?
6) If the PBOC controls the exchange rate like this and it has largely worked out for them then why do we advocate free float of exchange rates for developing countries? Provided they're good at managing their reserves, shouldn't they pick an optimal exchange rate and stick to it? How is it that some countries like Japan and China can control exchange rates but others like Sri Lanka or Argentina or Indonesia or India can't?
Brazil, most of East Asia and the Gulf countries did try this in the lead up to the GFC. Low US interest rates and fiscal deficits and over investment increased these countries trade surpluses and raw material prices (oil). Rather than letting their currencies appreciate they intervened, buying dollars and treasury bonds, which depressed yields, boosted liquidity and kept the investment boom going.
It is much easier to pursue this strategy when you have a trade surplus , not a lot of external debt, interest rates are falling and your export clientele are booming. Most EM countries these days aren’t in very strong financial positions and can’t replicate China’s strategy. They also don’t have the domestic savings.
We will can quibble about the details, but the real story here is that more and more of the (supposedly) smart people on both sides of the political aisle are willing to admit something that has been verboten for decades: "mercantilism works".
The best (non-specialist) book on this subject was published almost 25 years ago (before the rise of China!): Global Trade and Conflicting National Interests by Gomory and Baumol. They explain how classical trade models developed for English wool and Portuguese wine break down into multiple stable equilibria when you're talking about dozens of countries and thousands of products. The "national interest" part of the title refers to strategies (mercantilism) to shift the global equilibria in your nation's favor.
As an Australian importer that has been dealing with China and going there 2-3 times a year for 15 odd years, i read a lot of the economic theories and comments posted ( Eg Michael Pettis etc) but as is usually the case, those are typically academic and observational, but not directly transactionally knowledgeable.
What is missed in a lot of this ( and i am certainly not a booster for the Chinese CCP particularly) is just how brutally competitive the domestic Chinese market is. For any product.
1.5 billion consumers and many tiers of business customers. Cut your teeth there and survive and you are world class. The Chinese middle and upper class ( numbering far more than the US )
are now very discerning.
The " government" definitely promotes, helps fund, and opens trade doors for certain sectors that it wants to promote. But it is "hunger games" there let me tell you.
The efficiency that is embedded in their processes weeds out pretty quickly the weaker and lesser competitive and innovative manufacturers. ( Note almost no designer/power brands in concept industries )
The product innovation cycle there is 1/2 to 1/3 of the time involved in pretty much any western country and Australia is way down the line as we have basically no domestic manufacturing.
What has me baffled with the idea that Trump promoted, and apparently convinced the MAGA crowd and others generally about tariffs, was that the Chinese / other countries paid the tariffs. Well as we know, the US importers and consumers do. Hhmmm go figure.
So if i was China, Mexico, Canada, Europe etc, and Trump applied 10%, 20%, 60 % tariffs, i would ring the Donald and say ok, understood, we will put up our prices to the USA 10%, 20%, 60 % Before ... we invoice the importers, and take the pain here, and you can tell the US citizens we are paying. But in fact capture the profit margin that is far more than they work on now.
The BS and smoke and mirrors that is all this incredibly bright academic posting would come down to the same thing. The US consumers pay more and the exporters make more margin.
The multivariate challenge that trade surpluses represent for macro-economics reminds me of the old joke about the three academics stranded on a desert island with only tinned food to eat and nothing to open the cans with: the engineering professor sets about making a stone tool to puncture the cans; the physicist starts calculating whether they can heat the cans to a high enough temperature to blow the lids off; and the economist assumes a tin opener.
The engineer looks for stones to make a computer.
The physicist assumes the equation is reality.
I’ve been reading Pettis for a long time. 15 years for sure, has he been around for 20? Been awhile. He’s been saying the same thing (or close to it) for a long time and has not yet reached the stopped clock stage of being right twice.
His observations and diagnoses are spot on, but from observations one must be able to make useful real world predictions (the world that bond and currency traders operate in) or to build a mathematical model which might make useful predictions under certain very limited circumstances (the imaginary world of economists).
What is clear is that China for a long time used fixed investment/real estate as “stimulus” to keep things ticking over even as debt piled up faster than GDP did. Pettis and others observed these diminishing returns were unsustainable and begged a Japan-style crisis, but this cycle sustained itself for a lot longer than most thought possible. Then Xi decided (late 2010s) that manufacturing exports and renewables were going to be what was pushed (and all of the European and American “stimulus” worked like a charm for Xi’s goods exports during Covid, reaching record levels during the alleged “supply shortages” of 2021-22). This does tie in with Xi’s strategic shift years ago to stop buying treasury bonds and instead use his surplus to invest and lend in emerging markets. These infrastructure projects abroad create demand for Chinese goods (or rather, were run by cadres for the purpose of importing Chinese goods, increasing Chinese influence and securing raw materials - much like old time colonialists) in addition to being a way to recycle surpluses (and perhaps to squander them, over the long run, though maybe they’d be squandered in treasury bonds over the long run if placed there instead…..or seized, like Russia’s).
Many observers consider Xi’s export strategy and EM lending/white elephants to be wasteful and unsustainable, and most pols consider it unfair. I doubt that tariffs will stop it because, as Churchill was claimed to have quipped: now we’re just negotiating the price. Neither price nor profit has been a major obstacle in China’s policy making. China was willing to go through wage and price deflation during Covid and who knows when all the pigeons in the banking system come home to roost. You can always make more pigeons.
Some people spend trillions to pay off donors or forgive student loans or hand out subsidies (free health insurance for those making 300k) to drive consumption. Others use it to build (too many) things (whether apartments or solar panels).
There are obviously fractures in Chinese society. A middle and professional class has been created but there aren’t enough jobs (and the best ones are reserved for cadres). When young people can’t find good jobs and can’t afford flats there will be some tension. However, the savings of 4 grandparents and 2 parents will eventually flow to one young person. Also, the working class is still a lot larger than the middle class and Xi’s manufacturing subsidies are keeping them busy.
The move toward tariffs and trade bans (which I support in certain cases) may end up being well-timed with Xi’s manufacturing extravaganza failing. If jobs are lost, who will the working class blame? Xi or the Gweilo?
Noah writes: "In order for cheap Chinese imports to actually impoverish Americans, there would have to be some kind of externality or coordination problem involved. That might be the case, but Pettis or the MAGA people need to explain what they think that externality is. It’s not readily apparent to me what it might be."
Isn't forced de-industrialization exactly such an "externality"? It acts indirectly to impoverish a country by stripping it of its international influence and ultimately even the ability to defend its own interests and territory.
I don't think that's what Michael Pettis means, probably, and I don't think it's really what the MAGA people mean either. Trump seems to focus on "lost business", which is perhaps why he is still sore at the Japanese, and on American jobs generally. But it's definitely part of what drives concern about that supply chains that supply the US DoD, and why there is so much concern about the origin of semiconductors these days.
I'm no economist but I always find myself wondering if there isn't a sort of critical mass of diversity that an economy needs to be healthy. You see economies around the world that apparently thrive but depend too much on tourism for example. It invariably reveals itself as hollowness and fragility when times get tough. Similarly an economy in which everyone is a clerk at Walmart or flipping burgers at McDonalds seems hopelessly fragile and imbalanced, not in a trade balance sense, but in a robustness-in-the-face-of-difficulties sense.
America used to build ocean-going vessels. We don’t do that. It went to Japan, then Korea and I’m sure that manufacturing will devolve and go somewhere else as those economies about make more valuable products. Japan used to produce most electronics, then it also moved to South Korea and then on to China. What would Taiwan export if not for semiconductors?
America couldn’t affordably build those ships anymore. We can barely build aircraft carriers or submarines anymore. This problem and history is best illustrated by the American car manufacturers. The Big Three cannot create a $25,000 EV and make money. Their labor costs and legacy costs choke them. China can, but we want to protect our car-making ability for jobs.
Those cars get subsidies that studies show largely went to the wealthy and car manufacturers. It was a job program. Our military-industrial base is mainly gone, a shadow of its former Cold War self. Even if we were double our military budget, we do not yet know what to build. Drones or manned weapons? Which will be more effective. The Pacific, the body of water China wants to control, would need ships. Either way, you need big ships. An EV drone ship cannot traverse the Pacific. America doesn’t know what to build, the government is an awful prognosticator, and industrial policy produces favored industries and companies, not effciencies.
GM manufactures ICE cars well below $25K. We just don't have the supply chains to build EVs at those prices -- yet. We can and should be building them, but we're probably about to stop those plans and hand the (entire automobile) industry to China. It doesn't really matter what we "want to build" in the future if we don't invest in building the supply chains necessary for it.
I'm going to say this then in response to my post. The Big Three cannot build ICE vehicles that Americans want. Most of the Big Three no longer build sedans Americans want. I think you’ll find the actual purchase price of well-equipped vehicles after adding taxes, the vast amount of those cheap vehicles come closer to $30,000. The world has yet to flock to the Chevy Volt. GM doest’ not sell a crapload of Chevy Malbu’s. They make money selling trucks.
I have been in the automotive sector for 35 years. You learn a few things. Trucks make profits for the Big Three.
"GM no longer manufactures..." is a funny thing to hear in 2025. My liberal father has been saying the same things since the 1970s. Ironically he's now on his second Volt and absolutely loves it. I recently leased a cheap Honda Prologue, which is actually manufactured by GM, and it's a shockingly fantastic car.
Other than that I'm not going to defend US car manufacturers. Where we seem to disagree is that I think this is due to bad management, and you think it's due to labor. (And also: I think that losing that industry would be a crushing blow to the US from which we would have difficulty recovering.)
The emergent war technologies leave governments in a dilemma. They don't know if they are building targets or weapons or both. They will have to build both until a major peer to peer war occurs and sorts out the issue. It is obvious they will have to build a sh*t ton of equipment no matter what they build.
Indeed, all that and with a $38 Trillion dollar debt whose interest is more than the defense budget....Ummm
The failure of Trump I tariffs to enhance domestic manufacturing is of a piece with the failure of the corporate tax cuts to do the same. After complaining for years that corporate taxes are holding back business investment and claiming that a reduction would bring back the jobs “stolen” by China, American companies did a lot of stock buybacks and consolidation. We make many welfare recipients work to justify their temporary assistance checks. Why not do the same with tax expenditures, conditioning cuts on domestic investment?
Really good one, Noah. I like hearing the questions experts ask themselves.
China is the geopolitical rival, but the US's biggest trading partners are Mexico and Canada. I'd like to see an article with this depth of analysis on effects of tariffs on Mexico and Canada, including the effects on all 3 countries, impacts on household income, etc.
Because the northern Mexican states and virtually all Canadian provinces trade more with the US than within their own country, they are all a completely integrated part of the American economy, and Trump's plan seems similar to (and as sensible as) erecting tariffs between US states.
So what does that look like?
Would a shift to more domestic consumption in China create a dynamic of rising expectations and more "liberal-thinking" among the Chinese people that Xi Jinping might view as harder to control and a threat to his goal of militarizing China to threaten and eventually overcome the dominance of the West?
"Xi is making the Chinese economy look a little bit more like the old Soviet one, where production was determined by plans instead of by the market." Well it's good to see Noah saying this explicitly. Two articles ago Marx and Lenin's ideas were dead. I guess the distinction is that we believe this central planning is driven by expediency and not ideology?
Back to the subject of this article: what concerns me about Pettis's ideas is that they aren't really new. We've been hearing about conservative ideas to support the US government via a consumption tax (Google "national retail sales tax") for at least fifteen years now. Economists never had great things to say about this proposal. Is the "tariff everything we import in order to reduce US consumption and pay for income tax cuts" idea just the same pig dressed up in different lipstick?
Everybody could tariff China together, if the United States used the threat of its own tariffs to achieve that coordination. But that's not going to happen. Trump believes in tariffs but not international coordination, while most Democrats have the reverse beliefs.
Thanks for the post. I've been waiting for somebody to highlight how much Pettis has gotten right. Too bad that naive American tariffs won't fix things!
The impression I get is that Democrats have been pretty happy using targeted tariffs -- but they didn't have the political courage to actually do much about it until Trump shifted the Overton window.
Great post!
I think the inflation angle is really important in explaining the last few years' of trade data. China has been having much lower inflation than the rest of the world, depreciating its real exchange rate, making trade surplus worse, while also depressing domestic demand.
Which is why I kinda both agree and disagree with Pettis. Lack of domestic consumption can definitely be a syndrome of a bigger problem that also manifests in trade imbalance, but, as in this case, you have to specify the underlying mechanism moving both, otherwise, it's not really a theory. and I don't see Pettis doing so.
"Today, [unlike in the 1930s], Americans consume far too large a share of what they produce, and so they must import the difference from abroad." What? Note to Pettis from 5th grade teacher, to take a difference you have to have two entities, A-B is a difference. In my field, physical chemistry, someone writing this would be forever considered a quack. There is apparently an implicit factor "consumption of goods (CG)" such that CG = CG_d + CG_i. The premiss of the argument is that Americans consume too many US produced goods. The logical result of his argument would be to tax (e.g. VAT) US made goods to reduce the share consumed. Calling for tariffs on imports when the premise of your argument is that US consumption of US production is too high is so logically flawed that no person who understands logic would accept it. My 7th grade mathematical logic teacher (back when CA schools put money into gifted programs) would have given me an F if I was that stupid.
Then he fully goes into irrational land: “By taxing consumption to subsidize production, modern-day tariffs would redirect a portion of U.S. demand toward increasing the total amount of goods and services produced at home. That would lead U.S. GDP to rise, resulting in higher employment, higher wages, and less debt. American households would be able to consume more, even as consumption as a share of GDP declined…”
Tariffs are taxes paid in the US. They are not production subsidies. Are we to assume in his mind all tariff revenue will be used to subsidize US production. How would that work? The US is roughly at full employment. How are we going to get higher employment? Production has inputs of physical capital, intellectual capital, and physical labor. If labor is near its maximum (and potentially decreasing with deportations) then to produce more goods in the US can only occur through increasing physical and intellectual capital inputs, both of which are slow processes even with the assumption of new funding (the “subsidies”).
So it is almost impossible for employment to increase in the US for native born workers because of demographics. We can, and have, increased employment through immigration. Higher wages? Mathematically real higher wages can only occur with increases in real GDP except through large scale economic redistribution from non-wage earners to wage earners. The trend in the US has been in the opposite direction. Less debt? He offers zero support for this claim. Again we see arguments that no intelligent person could accept. Claims need to be substantiated by facts, logic and mathematics for any intelligent critical thinker to accept. And finally: “American households would be able to consume more, even as consumption as a share of GDP declined…”. Once again a claim with zero facts, logic and mathematics. And again mathematically and logically impossible. 8% of the workforce is in manufacturing and we consume ca. 30% of expenditures on goods (and part of this is basic food stuffs). So ceteris paribus taxes, spending, …, try some simple math. He makes no connection between consumption and GDP but implicit is that consumption is a near linear function of GDP. Product consumption is ca. 30% of spending (apparently linear in GDP in his non-model). A 1% decrease in spending on goods (again “even as consumption as a share of GDP declined” [see implicit linear GDP spending requirement] ) requires a 3% increase in GDP all coming from the new subsidized manufacturing production. 8% of the non-farm workforce (FedFred) is in manufacturing. A 3% increase in GDP from 8% of the workforce would require a manufacturing productivity increase of 1180%.
Apparently many people can’t do 7th grade logic and high school level math. Anyone who has these skills understands this guy is a complete quack. I could do this math and logic in 7th grade. Apparently much of the economics world didn’t pass 7th grade. Yet Noah claims “Whatever you think of Pettis’ theories, I think he’s probably the single most important and influential international economics theorist in the world today. “. Noah thinks someone who couldn’t pass my 7th grade and do very simple algebra is an important economics theorist. No, he is a logically and mathematically incompetent quack.