Curing international trade confusion

Can you translate that into supply and demand?

In one of my favorite childhood movies, Remember the Titans, Coach Boone (Denzel Washington) describes his offense. “I run six plays, split veer” he says. “It’s like Novacaine. Just give it time, it always works.” That’s how I think about price theory. There are a few plays, and if you stick to them, you can keep marching down the field of knowledge, while others come up short.

In price theory, my go-to play is simple: can I say this in supply and demand? What would it even mean? You can apply this to almost any economic argument, and it cuts through a lot of confusion.

There are two parts that are helpful about “saying something in supply and demand.” First, you start thinking about how multiple sides of the market determine what happens. The outcome arises from those forces. There’s feedback.

Second, you are forced to think about causation. The magic isn’t that supply and demand curves tell you what the price will be—they don’t. I could draw the curves anyway I want; the equilibrium could be anywhere. Instead, what drawing the curves does is that they force you to think in terms of comparative statics: if something changes, which curve moves and why? That turns vague claims into empirical claims with testable predictions. Instead of hand-waving, you have to specify what shifted, what caused the shift, and how prices will respond.

This approach works because most economic confusion comes from the same few mistakes: ignoring how markets adjust, confusing accounting identities with causal relationships, and treating prices as irrelevant details rather than the signals that coordinate decisions. You see these patterns everywhere in economic commentary, not just from one person or perspective.

Trade doesn’t negate supply and demand

These mistakes show up constantly, especially in discussions about trade, which often get caught up in terms like currency depreciation, capital flows, and global imbalances.

Michael Pettis’s recent Financial Times piece on capital flow restrictions provides a perfect example of what happens when you don’t run the supply-and-demand play. Pettis, a widely published commentator in elite outlets, makes a fundamental—I’m not going to say error—different step than I would that runs through much of his work: he mixes accounting identities and economic arguments and treats prices as irrelevant.

In other words, he’s not thinking in terms of supply and demand.

Let’s start with his claim that “a country’s investment is constrained not by scarce saving but rather by inadequate domestic demand.” This seems amenable to translating into supply and demand. He uses the word demand. So he’s arguing that quantity is constrained by demand, not supply (savings). But that makes no sense—it’s constrained by both! That’s what determines equilibrium.

Pettis consistently misses how sides of the market adjust and interact. When he writes that “increasing the supply of foreign capital may not spur investment,” he’s describing a world where supply increases have no effect on quantities. But that’s an extreme assumption. Normally, if you increase supply while demand remains a constraint, you still increase the equilibrium quantity.

Because Pettis almost surely means the quantity supplied of foreign capital. I doubt he’s thinking in these abstract curves, but, instead, the observed quantities. So more foreign capital entered the market. But to know the effect, we should also ask why the supply of foreign capital increased? This is where it gets murky with Pettis. Here, he doesn’t explain the underlying cause. Maybe Chinese savers have excess savings due to policy distortions, maybe U.S. assets became more attractive, maybe foreign central banks are intervening. Each would imply different adjustments.

How will prices respond? It depends. Let’s suppose Chinese savers have increased savings due to policy distortions (a frequent claim). Then those savings bid up the price of investments (lowering the interest rate). Lower interest rates should increase the quantity of investment demanded. That’s basic supply and demand.

In all of these cases, investment goes up generally. But Pettis claims increasing capital supply “may not spur investment.” That’s impossible in a normal market. The only way this makes sense is if he’s assuming the demand curve for investment is perfectly vertical (completely inelastic) - meaning firms won’t invest more even when capital gets cheaper. Does anyone think that is true? It could be, and he did say “may”. But we can see how translating into supply and demand reveals the assumptions and the core of the argument.

Pettis tries to escape this by invoking exchange rate effects, but that just pushes the problem up a level. When capital flows in, it doesn’t just sit there doing nothing, as Pettis suggests. It bids up asset prices, including the exchange rate. Those higher asset prices make further inflows less attractive—that’s still just the equilibrating mechanism that balances supply and demand. Pettis treats this as some kind of force that throws everything out the window. But that’s just prices adjusting.

Think about literally any other market. When more people want to buy coffee, prices rise. Those higher prices make other previous buyers drop out; they switch to tea or just drink less coffee. But the net effect is still higher prices and more coffee sales than before the demand increase. The price rise pushes back against the original demand shock, but it doesn’t eliminate it. Same thing happens with capital flows: the exchange rate adjusts to make further inflows less attractive, but that adjustment doesn’t magically make the original shift disappear or reverse the net effect.

He argues that “taxing capital inflows will indeed reduce trade deficits for countries like the US, it will not do so while raising domestic interest rates.” Okay. Lots going on. Interest rates can be confusing. Let’s think about the price of assets, instead. Can we translate this into supply and demand? What shifts when you restrict foreign capital supply? The demand curve of domestic assets decreases. There are fewer bidders. But how do prices respond? The price of assets must drop, and the interest rate must rise. Why do we know that? The domestic market already showed it wasn’t willing to invest at the old interest rate. If they were, they would have before! Restricting foreign supply means domestic savers will need to earn higher returns to change their behavior.

Always reasoning from an accounting identity

Pettis consistently reasons backwards from accounting identities. That’s a recipe for lots of confusion in my experience. I think we can avoid this if we think in terms of supply and demand. Instead, he sees that the current account must equal the capital account and concludes that capital inflows “force” trade deficits. But accounting identities don’t tell us about causation—they balance through price adjustments that coordinate decisions.

Take this example from a different piece: “In today’s world, foreigners do not fund U.S. trade deficits. They direct their excess savings into the U.S. financial market mainly to take advantage of its depth, liquidity, and governance... If foreigners pour capital into the United States, the United States must run an equivalent trade deficit.” Foreigners “pour” capital into the U.S. because they find American assets attractive at current prices. If those investments bid up asset prices (including the exchange rate), that changes the relative attractiveness of American vs. foreign goods. People respond to those new prices by buying more foreign goods. That’s not mechanical—it’s choice responding to price signals. Policy affects that and maybe US policy could improve things. I’m not arguing with that. But if we want to improve things, we better think through the equilibrium adjustments.

This pattern of reasoning from accounting identities appears throughout Pettis’s work. Consider: “it is not the case that the United States runs a current account deficit because Americans save too little. It is the reverse: Americans save too little because the United States runs a current account deficit.”

Both the current account deficit and American savings rates are equilibrium outcomes—they’re jointly determined by the interactions in markets, not one causing the other. It’s like arguing that a high apple price “causes” low apple consumption, when really both price and quantity are determined simultaneously by supply and demand. Never reason from a price change! We can ask what policies would shift savings or investment, but the equilibrium outcomes themselves don’t cause each other any more than price causes quantity.

This approach eliminates human choice entirely. Pettis treats markets as foreigners imposing their will: “the United States has no choice but to run a corresponding trade deficit.” Capital flows are just forced upon you like the weather if the government doesn’t do something about it. In his telling, Americans are passive victims who must automatically adjust their saving and spending when foreigners decide to invest here.

The starkest example: “If a country organizes its economy in such a way that its savings vastly exceed its investment, the rest of the world must automatically adjust either its savings or its investment.” I mean that must be true, but how does that framing help us? If I sell goods, does it make sense to say the rest of the world “must” buy them? Only under weird definitions of “must.” In both cases, we are looking at an outcome (savings > investment, or my sales >0), not some abstract goal. These are the traded quantities. And, again, it removes any choice. Why am I selling the goods? Can policy change my sales? Sure. I don’t need to think everything is some “free market” for my argument to go through.

Now, maybe some of these foreign capital flows stem from policy distortions—Chinese state banks parking reserves, or mercantilist export subsidies creating artificial savings. Those are legitimate concerns. But recognizing that some foreign actions might be distorted doesn’t mean we should throw basic economics out the window. In any market, you’re subject to other people’s actions, including their policy choices. That doesn’t mean accounting identities turn into causal relationships

Sometimes there are just gains from trade

As I said above, there is an implicit assumption that investment is basically fixed, so foreign capital doesn’t really help. In another piece, Pettis acknowledges that isn’t true for developing countries: “In a world where domestic saving is insufficient to finance high-return investment opportunities, foreign capital is a welcome addition. In such a model—applicable, for example, to nineteenth-century America and many developing countries today—external inflows support growth by funding much-needed investment in infrastructure, industry, and productivity-enhancing projects that would otherwise go unfunded.”

But then he dismisses this entire possibility for modern developed economies! Why? What’s so different about 2025 America that foreign investors can’t possibly see opportunities that domestic investors miss? What if foreign investors have different risk preferences, longer time horizons, or complementary skills? What if a Chinese company sees potential in American farmland that American farmers don’t, or a German firm has manufacturing expertise that makes a U.S. factory more productive?

Pettis treats all capital as homogeneous—just money sloshing around with no accompanying ideas. But capital comes bundled with knowledge, networks, and capabilities. Foreign investment often brings new technology, management practices, or market access. That’s not captured in his mechanical accounting framework, but it’s exactly the kind of thing supply and demand can help us think through. Different types of capital, serving different purposes, at different prices, with different productivity implications.

Also, it’s not some written rule that “domestic saving is insufficient to finance high-return investment opportunities.” That can’t be true in a closed economy where domestic savings must finance all investment. So what possibly could be the claim?

High-return is a relative term. The price on the y-axis is always relative. What we mean is that there is some outside investment willing to take lower returns. So when all is settled, at the market price/interest rate, some of that investment comes from the outside. At the market rate for daycare, I buy some and don’t provide all the childcare. At a much higher price, that would not be true. Luckily, someone outside my household is willing to supply childcare services at a lower price.

Is it all really just supply and demand?

Okay. I’ll confess! I’ll confess! No. Supply and demand doesn’t tell us everything.

Pettis has already dismissed the “Econ 101” framework as irrelevant to current conditions. But as Tyler Cowen points out, the critique runs much deeper—PhD-level international macroeconomics models also undermine Pettis’s claims.

The big thing is that words can cover up a lot of unclear thinking. Without a model—even a simple one—it’s easy to make claims that seem plausible but you don’t actually know what they claim is. The biggest thing is just thinking through a model.

But I’m sticking with simple supply and demand anyway, not because the advanced models aren’t useful, but because they, too, can let people hide unclear thinking behind lots of moving parts. If Pettis’s core logic doesn’t work when stated in basic terms, if his claims about automatic causation and mechanical adjustments fall apart when you ask “which curve shifts and why?” I don’t see how adding more financial frictions, sticky prices, market power, whatever, would save the argument. More sophisticated models will slow down adjustment, add complications, and capture institutional details. But they rarely reverse the fundamental direction of causation or make demand curves slope upward. Better to get the basic story straight first.

We still need to know elasticities—how responsive people are to price changes. We need to understand timing—how quickly markets adjust. We need institutional details about how exactly these adjustments happen. If foreign capital really does depress domestic savings, is that because Americans are highly sensitive to interest rate changes? Does it happen over months or decades? Through what specific mechanisms?

The beauty of supply and demand isn’t that it has all the answers. It’s that it forces you to be explicit about your assumptions about big, looming topics like “capital flows” and “global imbalances.” When Pettis says capital inflows “automatically depress U.S. savings,” he’s smuggling in huge assumptions about elasticities, timing, and causation. Those may be true, but I need to see them laid out to make any sense of it. Better to get the basic story straight first.