63 Comments

I didn't follow the explanation, but it did not seem to touch on two seemingly reasonable reasons for wanting to build automobiles instead of grow them.

1. Military implications - The ability to build vehicles & infrastructure seems important for a military campaign. It seems likely to be a problem for the US that China dominates drones.

2. Innovation implications - Reportedly, people build electronics in China because the density of expertise makes it feasible, not just cheaper. In principle, it seems plausible that there would be important benefits to those who fabricate something to improve it further. Wanting to have innovation happen in the US makes sense.

And really, a lot of the high value jobs performed in the US are in the world of finance and bits. As in the case of Trantor, that can leave the seat of an empire vulnerable when it depends on other worlds for the sustenance of its population and elevated industry.

Expand full comment

Note that these benefits disappear if you use a naive model of knowledge which says, knowledge moves for free because it’s all in books. If knowledge has to be embodied and is expensive to transfer, then this benefits (such as localized knowhow) become much more salient. I think lots of academics model knowledge as being much more transferable than it really is.

Expand full comment

This, in reality, is everything. Consumer-oriented heavy industry is directly transformable into military-oriented heavy industry in the event of a war. This includes concentrations of both industrial plant and manufacturing process knowledge, and subsidies to distort a country's economy in favour of relevant industries are far, far cheaper than constantly buying total-war-quantities of armaments to keep the factories running.

This is then relevant to whether China can repossess the Erie Canal. Europeans probably couldn't have done that to America in the 1840s. But they, the Americans and the Japanese did do it to China on a massive scale in the 19th and early 20th centuries.

Expand full comment

Two hundred fifty years of international trade and monetary theory in a single lesson! Wonderful tour de force, badly needed.

My experience of half a life time teaching this to people for whom it was required, not voluntary, was like teaching all of economics: They don't want to know! They actively reject. Just like lay people actively reject. [It's different for those who study such stuff voluntarily.]

Why might this be? The question keeps me up some nights. Evolutionary biology? If they have it, we don't? Who is this we? If you have it, I don't have it? Anyway, it's not fair, another way of saying "I deserve more".

Many years ago, I talked to my daughter's third grade class, at her behest, trying to explain comparative advantage. I constructed an example of a mother and father growing tomatoes and potatoes, each in their own patch, and then they specialized, with Papa less efficient in both than Mama. The little kids got the arithmetic one helluva lot more quickly than grownups!

Expand full comment
author

Part of the problem with teaching economics is that people think they already know it, aside from the details. The subject is something they have interacted with all their lives and think they understand. The language uses terms like "competition" and "efficiency" that they think they know the meaning of. One way I try to deal with that problem is by offering examples where the economic conclusion is the opposite of their intuition, such as the effect of legalizing polygeny or polyandry on men or women. They may think my conclusion is wrong but they can't think they understand my argument, since their version has the opposite conclusion from mine.

A second problem, in this case, is that absolute advantage is simpler, easier to understand, than comparative advantage. You have to think carefully about it to realize that absolute advantage is not an internally consistent concept. If they have heard the term "comparative advantage" they think they can guess the meaning from the words and guess wrong. Also, the trade they are familiar with is within one country, where there is no exchange rate and absolute advantage makes sense for competition although not for understanding gains from trade.

Expand full comment

Words don't mean what you think they mean in economics and in mathematics.

I used to have this problem when I was reading math texts. They would use a word that was in common or almost common usage -- familiar to me anyway -- and then the rest of the explanation didn't make any sense.

I finally learned that whenever I ran across a word like that in math to just mentally substitute farblegastic, or some other nonsensical term and wait for it to take on a mathematical meaning from the context. My math studies went much better after that.

If there's blame to be laid, I would put it with the mathematicians and economists who hijack perfectly good words and load them up with obscure meanings, although some of the confusion may be historical (word had somewhat different meanings or connotations in the past).

Expand full comment
author

Would it be better if we invented new words instead of using an existing word with a related meaning?

Expand full comment

I often think it would be, since that's essentially what I'm doing when I substitute my own. But I'm just one person and that's how learning has gone for me. Maybe others find it different. Or maybe so many people bounce off higher maths because they can't get past this particular challenge. Who knows?

Expand full comment

"One argument sometimes offered for restricting foreign investment is that if the Chinese own a lot of US assets that gives them power over us." " ...their assets were sitting on our territory under our control. It wasn’t like they could repossess the Erie canal."

In today's world, money is political power. As China or other foreign (or corporate) interests accumulate assets in the USA, they can use this to leverage political power, until instead of us repossessing the canal, they're using the canal and the laws they've bought around it to possess our political process.

The bribery and power in the political process may be a confounding factor in the "pure" economics of supply and demand.

Expand full comment
author

They don't have to own US assets to bribe US politicians. They can buy dollars with yuan for that purpose just as to buy US goods.

Expand full comment

A great article, bookmarked it, thank you for this. I read the same arguments from you before but it is worth going through it again as I find that I haven't really internalized it well enough to be able to explain it well to other people (which shows that the argument is nontrivial and that is most likely the reason why most discourse about this hasn't moved anywhere since mercantilism).

Still, I like playing the devil's advocate (it's good to test limits of a good principle), so I will mention some caveats that others have hinted on as well:

Let's say that the Chinese subsidize all their high tech exports so much that there is a comparative advantage in going all in on wheat in the US. The US is better off in the short run, effectively subsidized by Chinese tax payers as you say.

But it also has two additional effects:

1. The US stops accumulating technical know-how and human capital over time. If the Chinese stop with these subsidies, it will take decades to re-establish that. This gives China a political weapon against the US and since China is an autocracy its government doesn't have to worry as much about things which are clearly making (most of) its population worse off. Even worse if there are some strategic resources in which China is close to being the only producer (or the only producer in quantity). I guess the recent investments in microchip production in the US go into this category.

2. If the subsidies are really so extreme that they shift the US economy towards a few products, it becomes fragile even if there is no bad intent on the Chinese end. A new disruptive technology can make the advantage Iowa has in soil irrelevant and while the US economy can reorganize itself, it won't happen overnight or even in a matter of a few years if most of the non-agricultural know-how has gone overseas.

I suppose that this is an extreme scenario which is unlikely to happen if simply because China doesn't have the resources to really do this for long enough to really degrade certain industries in the US (or in the EU where new tariffs on Chinese electric cars are being considered). And it would probably not be enough to do that in the US alone, but in the whole world, to really make sure that if it changes its policy, nowhere in the world will be able to produce chips/electric cars/whatever else for at least a couple of years. If China cannot afford to do this, it is unlikely that any other country will (other countries are either more accountable to their citizens and so not likely to engage in such policies or too poor/small to matter).

Expand full comment
author

I agree that a sufficiently wise dictator could tweak trade to take account of such long-run indirect effects, although if we assume the same wisdom in investors, it probably wouldn't be necessary — if you know that there is a 2% probability of needing to make tanks each year you can subsidize your car factory to keep it around as a gamble on making a fortune when tanks suddenly become very profitable — providing you can trust your government not to expropriate the factory or the income if you win the bet.

But tariffs are not imposed by an all wise benevolent ruler, they are an outcome of the political market, and there is no reason to expect them to push you closer to the all-wise long term optimum instead of away from it. The predictable effect is to make us poorer.

Expand full comment

> if you know that there is a 2% probability of needing to make tanks each year you can subsidize your car factory to keep it around as a gamble on making a fortune when tanks suddenly become very profitable

At that point you run into a version of the "market remaining irrational longer than you can stay solvent" problem.

Expand full comment

The our costs dollars , theirs yuan seems a wee bit slippery. Until and still lately though it's looking like such may change, any international trade, both sides have their calculators handy figuring what's that in petro-dollars. I see what you're saying but still...

Personally I think it's simply good husbandry to grow wheat and at least build some shoes and autos. Say anything happens; Bat flu, the chairman has a stroke, the Kamel's elected, shipping's affected.

Sure would be great to have wheat growers and shoe and auto builders very close at hand.

Expand full comment

The main consideration that's moved my sympathies towards (some degree of) mercantilism is careers: expertise/networking/branding takes time to build up in every system tried so far, and preventing fluctuations in trade from forcing generalism on your population (if a government) or your associates (if an individual) is an investment that plausibly has its place.

Put another way: *some* degree of stability is necessary for specialization (proportional to how risk-averse the specializers are), which in turn is necessary for the best gains from trade. Too much stability, i.e., the specialists stifling most-to-all innovation, is empirically a pretty easy trap to fall into, though.

Expand full comment

I think this looks entirely right, I found it extremely informative and opinion-changing, and I'd thoroughly recommend it to anyone who isn't American. For Americans, I suspect it's missing the far weirder dynamics of the US dollar.

Everything about US fiscal and monetary policy should be far more inflationary than it is. A large part of why it's not is the Chinese and Japanese governments hoarding dollars in order to distort their economies in favour of industrial development so they become manufacturing economies instead of agricultural economies, which isn't done for purely rational economic reasons but for strategic and social policy reasons, and doesn't impact this point.

However the value of the dollar must be impacted by the fact that it's used in trade outside the US between non-Americans in transactions that don't involve them.

My proof of this would be Ecuador, which does (did?) use the US dollar as its currency. Clearly, Ecuador's balance of trade can't be the sole determinant of the value of the dollar, and presumably can't be a major part of it. I'd suggest that it can't seriously have much impact on the value of the dollar at all, so can end up running permanent, real trade deficits until their economy becomes drained of capital altogether.

Half of US cash is circulating outside the US (source: just googled it). The eurodollar market is... big (source: everyone on google disagrees by order of magnitude), and creates at least quasi-exogenous demand for dollars. My suggestion would be that the effect of the international dollar on the US should be similar to the effect of the US on Ecuador as a determinant of exchange rates.

I'm not going to pretend I really understand the effects of this for a moment, but my tentative hypothesis is that the US dollar doesn't lose value fast enough to balance trade deficits. This lets the US create dollars without compensating inflation (or rather, much of the inflation happens outside the US, so within the US there's less inflation than there otherwise would be; a bit like if Ecuador printed its whole GDP's worth of US dollars - historically the Pope did this with debased in the LMU).

I think the long term effect is that it becomes cheaper for a while for the US to simply print Hondas (or let banks create them out of thin air) instead of growing them. But this presumably can't last forever, and the end state is a US that no longer grows or makes anything. At that stage (realistically, long before that stage), the international dollar is probably toast.

Hence, the "competitiveness" argument isn't necessarily farmers vs factory workers. It could well be farmers and factory workers vs banks and the bureau of engraving and printing. It's still distortionary, but probably a good idea in the long term.

Caveat: I'm very far out on a limb her and all of the details will probably be wrong. I think something of this sort should be a real problem though.

Expand full comment

I recently read Woody Holton's 'Unruly Americans and the Origins of the US Constitution'. He explains why the US went with tariffs for federal taxes.

The federal government was broke and in debt to people they could not welsh on. (That is, the government was in debt to America's governing class). They could not issue fiat currency, because everyone knew fiat currency from the Continental Congress was not worth a continental dollar. So they tried taxing everyone for specie and having the states collect it. This meant taxing farmers. Most of the country was farmers, who had very little money and no specie. So they tried having tax collectors seize farms to get blood from these stones.

Whiskey Rebellion. Dead tax collectors. Tax collectors going to debtor's prison because they could not collect any money and when they tried to seize farms, armed farmers objected. The farmers voted in a bloc against anyone collecting taxes from them.

So, after a secret Constitutional Convention, our founding fathers went with tariffs collected by federal customs agents at ports where merchants had some specie and could be bullied one at a time. Also a Bill of Rights so the annoyed farmers would accept the Constitution.

Yes, the farmers were paying higher prices for goods. But they kept their farms. And they expected to be poor anyway. Yes, tariffs were welfare for local manufacturing. Most US manufacturing was in New England, which was on the edge of secession until sometime after the War of 1812. Welfare for manufacturing owned by New England's governing class may have kept Boston in the US through 1815. Cheap at the price.

Expand full comment

> US costs are in dollars, Chinese costs in Yuan, and what determines the exchange rate between them is the cost of producing things.

It hasn't "occurred to" them because it is not true. China aggressively modifies yuan exchange rate, making it disconnected with what yuan would "honestly" be worth by comparison of costs of production.

Expand full comment
author

In order to modify the exchange rate the Chinese government has to buy dollars with yuan, which means buying US assets, as I described. Just announcing "a dollar is now worth four yuan instead of three so as to make our goods cheaper to Americans" doesn't work unless they are actually willing to exchange yuan for dollars at that rate, and since it is above the supply equals demand rate that means they accumulate dollars.

Expand full comment

The cost of producing new dollars is essentially zero-ish, though (I know about second-order consequences like inflation, but that doesn't seem relevant here). That's not what determines the exchange rate. China can afford to hoard dollars.

More generally, this predicts that purchasing power parity comparisons shouldn't exist because they would be baked in in exchange rates. That's just... not true?

Expand full comment
author
Jul 23·edited Jul 23Author

If China hoards dollars it is loaning money to the US at zero nominal interest, which is a pretty bad deal for them, good for us. And if it literally hoards dollars the extra money printing doesn't cause inflation because those dollars are not circulating.

More likely, the Chinese government uses its dollars to buy US bonds, which pay interest. That's the capital flow situation already discussed.

The reason purchasing power parity comparisons exist is that the exchange rate is driven by prices of traded goods. Purchasing power parity is determined by prices of consumed goods. Having cheap servants available, or cheap housing, lowers the cost of living but doesn't affect trade, aside from tourist or expatriate expenditures.

Expand full comment

1. You are then postulating that buying US bonds is a transaction equivalent to buying US pieces of land. Weird and probably wrong.

2. It isn't a bad deal if their explicit purpose is to keep yuan at the "wrong" exchange rate for various second-order effects.

Expand full comment

The actual prediction is surely that differences between purchasing power and exchange rate should wholly reflect transaction costs in trade flows, which sounds like it must be right. If I can buy laundry detergent in Guatemala for the equivalent of 10c a gallon (at the nominal Guatemalan exchange rate; for the sake argument that can be 1c=1 peso) but it costs me 60c a gallon in the US, it's worth me doing so iff the cost of bringing it in from Guatemala (tariffs, shipping costs and the cost of currency conversion) is less than 50c a gallon. If the transaction costs are only 40c, that should force down the detergent price in the US 50c a gallon.

If this was uniform for everything, the market value of the Guatemalan peso in dollars would start as 60c=50 pesos, falling back to parity once US prices fall.

Expand full comment

A bit more sophisticated, but still clearly at odds with what actual federal banks (or whatever institution controls the currency in each country) can and regularly do, essentially with transactions restricted. It's not just China: for instance, do you _really_ believe that ruble's fair price in 2024 after making the adjustment you describe becomes 87,78/dollar, i.e. virtually the same as in 2022?

Expand full comment

I've no idea if it's 'fair,' for testing the theory it depends on what's happened to the purchasing power of the ruble. I'm not sure a central bank can keep manipulating a currency forever though, particularly to keep its value up, as they'd end up draining all their foreign exchange reserves and having to start buying dollars at the market price to sell them at the manipulated price. Russia may be doing this, but then again the ruble may be worth that anyway given people are still buying Russian oil/gas; that's surely part of its purchasing power (it could be an expensive currency to buy a Big Mac in, but if you can get a discount on oil it's still worth having).

Expand full comment

Price of oil/gas hasn't gone up that much in the last two years to compensate for all the rest of what's effectively inflation.

As for whether they can keep it up forever... maybe not, but forever ain't on their timescale. As the meme goes, "I don't need to outrun the bear, I just need to outrun you".

Expand full comment

I think more economics should be taught in schools. Maybe as an unit in math classes. It's not intuitive stuff to most people.

Expand full comment
author
Jul 23·edited Jul 23Author

I think there are more things worth knowing than can be included in a high school, or college, syllabus. Economics should be available for those that want to learn it. I don't believe in having a body of things that all kids are supposed to learn. The extreme version of my approach is unschooling, which we used for our kids.

https://daviddfriedman.substack.com/p/unschooling-1

https://daviddfriedman.substack.com/p/unschooling-2

Expand full comment

I was very fortunate in that my high school had what amounted to a college catalogue and while there were requirements, they were of the nature (we were on a quarter system, 3 courses per normal year), 9 English credits, 9 math credits, 6 PE credits, etc.

It was left up to the student to select which courses would fill those credit requirements.

There was also a prerequisite tree, so some things you couldn't avoid. You were pretty much going to have to take Paragraph Writing in English and whatever the pre-algebra was in math, unless you tested out. And there were placement tests which could give you credit for the bottom of the tree.

But after that, you could focus on statistics or business math or literature or composition. I particularly liked the structure of Paragraph Writing, Beginning Essay Writing, Intermediate Essay Writing, Advanced EW: Research Paper, Advanced EW: Persuasive Writing, Advanced EW: Article Writing, etc.

We also had a second year of biology and chemistry for those inclined.

After I finished they switched to a semester system and it was all replaced with almost entirely requirements. English 1, English 2, English 3.... Do the teachers even know what topics are meant to be covered when courses are titled so generically?

The argument was htat the earlier system was good for kids who knew what they wanted but that everyone else needed to be told what to learn.

Bunch of hogwash. I'm more and more convinced that the education system (teacher education) is driven by latest fads and whatever Cool Aid has caught the literature's eye than actual science about how people learn.

Expand full comment

I agree that there are many things worth knowing. I do think schools could do a better job doubling up for efficiency purposes though. Instead of teaching math with arbitrary equations, combine it with physics and chemistry and economics. Instead of English class covering arbitrary fiction to write essays on, combine it with history class. It'd be a lot more efficient and free up time for other subjects.

I do agree that too many subjects are currently mandated, but I think mandating a few very important ones are okay. I think many kids and/or their parents make very poor decisions, and it's sometimes good to have a little bit of a nanny state make decisions for the same reasons you'd want a nanny to make decisions for a child.

Expand full comment
author

That might make sense if the nanny state had the knowledge and incentives to make the right decisions for each child, as a nanny might. It doesn't, as demonstrated by both your example of what it should do and doesn't and the uniformity of actual public school syllabi — a little opportunity to choose electives, but mostly the same subjects for all kids.

Expand full comment

If it'd be possible to completely dismantle the nanny state and replace it with anarcho-capitalism, I think it'd be easier to first replace it with a smaller nanny state that makes better decisions.

Expand full comment

> I think it'd be easier to first replace it with a smaller nanny state that makes better decisions.

I don't think so. That would require somehow filling the bureaucracy with competent nannies.

Expand full comment
author

In this case you don't have to dismantle the state, just its control over schooling, which you can do with a voucher program, something that actually exists in some states already.

Expand full comment

> The only reason Americans want to buy yuan with dollars is to buy Chinese goods, the only reason Chinese want to sell yuan for dollars is to buy American goods.

I wouldn't have any way to know if it was true, but I keep hearing that the price of the USD in other currencies is supported by the number of dollars tied up in exchange, related to oil contracts, between countries that have their own currencies. (The motive to write contracts in dollars is that USD are not vulnerable to deliberate inflation by the buyer, nor deflation by the seller.)

Expand full comment

At an individual level selling off all your assets to consume today seems like a reckless long term plan. Even if you believe that your creditors will have no leverage to seize those assets, it’s a risky gamble. I knew someone that ran up a bunch of credit card debt and managed to get away with not paying most of it, but I wouldn’t recommend it.

I think people feel the same about selling all the nations assets to the Chinese to fund consumption. Yes, in theory the treasury could just say one day “we default on any treasuries currently owned by China” to avoid paying those debts and what are they going to do about it, but I think you can agree that has a lot of potential impacts and it’s not something people are in a mood to try out.

So yes we could think about trade deficits as the equiviliant of “tribute” because we are selling them “assets” we will just seize if we ever feel like it. But the actual process seems a lot messier.

This isn’t to say I think we should be in a trade war with China. I would like to buy a cheap EV. But I’d rather pay for it the same way I pay for my car, with cash or at least a loan I can reasonably afford.

Expand full comment

There is one part of the explanation that I found confusing (could be because I am reading this at 1am).

You write that (in referenco to imposing a tariff on imports):

"Assets that an investor will use himself, a house in the Bay Area that a Chinese millionaire buys for his son at Stanford, is now more attractive, since the changed exchange rate means that it costs less in yuan than before."

But if a tariff on Chinese imports is imposed / increased, this will make Chinese goods more expensive in US dollars, hence the demand for yuan decreases, hence yuan becomes cheaper in dollars (and so dollars become more expensive in yuans). But this means that the Bay Area house, priced in dollars, is going to be more expensive in yuan than before ... or am I missing something?

Expand full comment
author

I think I am the one missing something, that I got that backwards. I will think about it again and if so remove the sentence.

Expand full comment

If a tariff is introduced, purchases of Chinese goods by Americans decrease (which causes the $ to rise) and therefore sales of American assets to the Chinese decrease (which causes the $ to fall). The $ must remain stable.

Expand full comment

I think it might be slightly tangential to your point (or the point of people on the thread) but it does seem to me that a country can regulate itself into economic ruin by making it too expensive to produce there, just like bandits left unchecked can kill economic production. The Soviet Union was not able to keep pace with the rest of Europe, for instance. That does seem to suggest that one can do damage to ones economy through regulation that prevents people from producing.

Expand full comment
author

Of course you can. As I said, you can make yourself poorer, but not less competitive, if "less competitive" means less able to export as much as you import.

Expand full comment

Ok, I see what you mean. I was thinking of competitive in terms of "being a place people want to produce things, or build a factory etc." I think the notion of "competitive" is very different across people, probably across time for the same person as well. If one is thinking of competitive in terms of "American businesses are better off subcontracting to Chinese plants because production is too expensive in America", regardless of import/export ratio, that is a little different.

Expand full comment

I'd really like to see some numbers. I know that roughly wheat can fairly easily produce 75 bu/A at a sale price of $6.30. And on large farms only 2-3 workers can produce that on 1500 acres. So, about $700,000 worth.

How many auto workers does it take to produce $700,000 worth of vehicles? I have no idea.

I guess I'm thinking about how many jobs are involved in producing equivalent amounts of product. Is that something to consider?

I suppose I cold try to find, say, the value of light vehicles (autos and light trucks) produced by GM and divide by the unmber of total employees.

It just seems to me like farming needs fewer jobs to produce the same amount of value. But it's just a top of my head guess.

Expand full comment
author

If it does, that means we can grow cars more cheaply than we build them. Following out your argument, we are better off with less efficient technologies — should dig ditches with spoons instead of shovels.

Expand full comment

I consider a different issue, in that in cases where goods were manufactured in the United States, all stages of procurement, labor, and sale are taxed by the government. Those costs funded by the taxes still need to be paid even if the money is expatriated.

The issue isn't that 'trade is or isn't balanced' it's that when enough of it moves overseas in a commercial-tax-funded country without capturing that tax value by some other means puts us in the hole faster.

I understand the point that the cost of goods doesn't effect our productive competitiveness (except in an indirect way) but it definitely effects our pricing competitiveness.

Expand full comment
author

The government taxes production of the goods we export, the wheat, just as it would tax production of the cars that the imported cars the wheat buys replace.

Expand full comment