A Balance of Payments Primer, Part II: The Dollar and All That

And what is the “Mar-a-Lago Accord”?

Source: Federal Reserve

Last week I published a primer on the balance of payments, which I have now made free. It laid the groundwork for today’s post, which is about the (sort of) fun stuff: the international role of the dollar and how it does or doesn’t affect the U.S. balance of payments. At the end, I’ll talk about the fallacies underlying how some Trump officials are thinking about trade.

The international monetary system inspires a lot of mysticism, because it sounds both mysterious and important. As a result, it’s easy to get hung up about the dollar’s role in the world economy. For example, back in the day, some people on the left claimed that the real reason we invaded Iraq was to prevent Saddam Hussein from moving away from the dollar. More recently, Elon Musk has issued dire warnings that the dollar may lose its reserve status, causing runaway inflation. And now there’s talk of a “Mar-a-Lago Accord”, based on the belief that US trade deficits reflect the special international role of the dollar, and that we can magically revive US manufacturing through financial engineering.

Well, as my old teacher Charles Kindleberger used to tell students, “Anyone who thinks too much about international money goes mad.” So the goal of this primer is to contain the madness and explain the following:

1. Why the dollar plays a special role in the world

2. How the dollar’s role affects the balance of payments

3. What’s with the “Mar-a-Lago Accord”?

The money of moneys

It’s true, as you often hear, that the U.S. dollar is the world’s preeminent reserve currency. But saying this alone actually defines the dollar’s role too narrowly, and implies that its role mainly depends on government policies. In fact, the dollar is special among national currencies in multiple ways, mostly involving private-sector rather than government decisions. It’s sometimes called a “dominant currency” or a “key currency”: the dollar is to other currencies the way money is to other goods and assets. The dollar is basically the money of moneys.

Why is the dollar the money of moneys? Partly this reflects the size of the US economy. But while we have a big economy, it’s not dominant: both the euro area and China are of comparable economic size.

The dollar, however, is dominant in a way the U.S. economy isn’t, largely as a result of circular causation: people around the world use dollars to make transactions and set prices, as well as holding a lot of dollars, and because other people do the same. In a 1967 essay Kindleberger compared the role of the dollar as an international currency to the role of English as an international language, an analogy that still works.

If you want to know more, see Krugman 1984 and Gopinath and Stein 2018.

So the dollar plays a special role in global trade and investment, mainly as a result of private-sector decisions. However, this specialness is reinforced by government actions.

For various reasons, many governments around the world, accounting for around half of world GDP, try to manage the rate at which their currency exchanges for other currencies. Indeed, some governments try to “anchor” their currencies by keeping their values constant in terms of other currencies, usually the dollar. Most notably, China tries to keep the yuan stable, in principle against a basket of currencies but in practice against the dollar. And in order to be able to support their currencies, governments hold reserves of foreign currency. These reserves aren’t entirely held in dollars, but as the chart at the top of this post shows, the dollar predominates: the share of global reserves held in dollars is much larger than the U.S. share of world GDP.

How the dollar’s role affects the balance of payments

Since the 1970s most advanced countries have allowed free movement of capital in or out and also allowed the values of their currencies to fluctuate. They haven’t had much choice. When capital is free to move, even massive intervention in the currency markets won’t stabilize a currency if speculators don’t find the stabilization credible. For example, Britain’s attempt to stabilize the pound in 1992 famously failed, despite around $50 billion spent trying to prop the currency up, because speculators, including but not limited to George Soros, didn’t believe the British government would be willing to endure the high interest rates and unemployment that would result if it tried to keep the pound pegged to the Deutsche mark.

The corollary to this story is that countries that give up the attempt to fix their exchange rates and let their currencies float freely don’t need large foreign currency reserves, because they have no exchange rate target to defend.

However, many developing countries, above all China, operate under different rules, more similar to those that applied in advanced countries in the 1950s. China maintains controls that limit capital flows, while the yuan is effectively pegged to the U.S. dollar. To maintain this peg, China has generally sold yuan and bought dollars, in the process accumulating large dollar reserves (which basically means a lot of Treasuries — short-term U.S. government debt.) So have other developing countries plus Japan, (which is a special case I don’t want to get into today.)

Why should we care if some countries choose to acquire lots of dollar reserves? Well, some economic commentators, notably Michael Pettis of the Guanhua School of Management at Peking University in Beijing (!) have argued that the accumulation of dollar reserves has distorted U.S. trade. Stephen Miran, now the chairman of Donald Trump’s Council of Economic Advisers, wrote a white paper last fall based on this argument. According to Miran, the foreign accumulation of dollar reserves keeps the dollar persistently overvalued, leading to large trade deficits.

I don’t believe that economic theories are actually driving Trump’s policies; instead, they’re being used the way a drunkard uses a lamppost, for support not illumination. But it’s still worth asking whether these arguments are right.

Miran argues that we are living in a world in which

the reserve asset producer must run persistent current account deficits as the flip side of exporting reserve assets … America runs large current account deficits not because it imports too much, but it imports too much because it must export [U.S. Treasuries] to provide reserve assets and facilitate global growth.

There’s no reason to believe that such views are actually driving Trump administration policy. Still, is Miran right? No. Remember, the dollar has been the dominant reserve currency since the 1940s, yet until the late 1970s the United States actually ran persistent trade surpluses:

How was this possible, when the demand for reserves required that America have ever-increasing liabilities?

The answer was that while U.S. liabilities held overseas were growing, so were overseas U.S. assets. In the 1950s and the 1960s U.S. companies invested substantial sums expanding their foreign subsidiaries. So the dollar’s role as a reserve currency didn’t force us to run trade deficits, because even in the 1960s private capital could and did head the other way. That’s even more possible now, in a world in which capital can and does flow in all directions.

So how much does foreign demand for reserves drive U.S. trade deficits today, when capital movement into and out of the United States and many other countries is unrestricted?

Here’s how I like to think about it: It’s true that foreigners buy a lot of U.S. debt securities every year, which to some extent reflects the role of the dollar as a reserve currency — although most of those purchases are of long-term securities, not the short-term securities usually used as reserve assets. But there’s also a lot of capital inflow to the U.S. that doesn’t involve buying bonds, while at the same time there’s a lot of capital flowing out of the country as American individuals and companies invest abroad. Here’s the picture for 2024:

Suppose that we were able somehow to force foreign governments to stop buying U.S. bonds. The balance of payments always balances, so something else would have to change in response. But why should all or even most of the adjustment involve a reduction in the trade deficit? Why couldn’t it involve an increase in other kinds of capital inflow, or a reduction in capital outflows?

In fact, I’d argue that the best bet is that a reduced demand for dollar reserves would have little effect on net capital flows to the United States, which mostly reflect the perception that the U.S. economy offers good investment opportunities. To be fair, some reasonable economists might disagree. What we can say for sure is that assertions that there is necessarily a one-to-one link between foreign demand for reserves and U.S. trade deficits are wrong.

In fact, as I pointed out last week, America’s move into persistent trade deficits in the 1980s and 1990s reflected domestic events — Reagan’s deficits, then the Clinton-era investment boom — rather than foreign reserve demand.

Yet the idea that foreign demand for dollar reserves is driving U.S. trade deficits is out there, as is the idea that we need to fix this alleged problem. For now, I don’t expect anything to happen. But if the trade war, which is happening, fails to deliver the promised revival of U.S. manufacturing, which it will, Trump might decide to back some kind of grand currency scheme. What might that entail, and what would it do?

About that “Mar-a-Lago Accord”

During the first half of the 1980s a soaring dollar caused the U.S. trade deficit to explode to a level that seemed shocking at the time, leading to demands that something be done about it. So in September 1985 finance ministers from the world’s leading economies met at the Plaza Hotel in New York and agreed to take steps to weaken the dollar. The dollar did, in fact, decline after that meeting. It has never been clear how much of this decline was caused by policy and how much was simply the natural deflation of a market bubble, but the Plaza Accord became famous as an example of apparently successful international cooperation to resolve a balance of payments issue.

This is why Miran’s white paper talks about a “Mar-a-Lago accord” that would similarly weaken the dollar in a way that would supposedly benefit the United States. But what might such an accord involve?

Well, if you think that the trade deficit is a problem caused by the dollar’s role as a reserve currency, the straightforward solution would be to try to end the dollar’s special role. That would be a coherent policy, although not a wise one.

But that isn’t what Miran calls for. Why not? Because Trump is adamantly against it. He has, in fact, threatened to impose tariffs on countries that try to move away from the dollar.

What, then, is the idea? Miran’s paper doesn’t offer any specifics, and given the lack of any evidence that he has actual policy influence trying to figure out what he would propose if he were making proposals doesn’t seem like a good use of our time.

It does seem possible, however, that Trump may at some point try to use the threat of tariffs to force other countries to allow the dollar to depreciate versus their currencies. There is precedent for such action — not the Plaza Accord but the “Nixon shock” of 1971. Nixon’s actions did weaken the dollar, but as you can see from the net exports chart above, they didn’t lead to a long-term improvement in the U.S. trade balance. In fact, the era of big deficits was still to come.

It's hard to see why a MAGA replay of the Nixon shock would yield better results.

Maybe the important thing to understand about Trump’s policy framework is that he doesn’t have one. He wants tariffs and thinks they’ll reindustrialize America, and might turn to currency policies when they don’t. But his views about currency are contradictory: he wants a weaker currency but threatens anyone considering moving away from the dollar. Indeed, I’m pretty sure that he wants both a smaller trade deficit and increased capital inflows, because arithmetic has a well-known globalist bias.

The idea that Trump’s international economic policy reflects sophisticated thinking about the international role of the dollar is, as Adam Tooze says, sanewashing. But this is par for the course. Remember Kindleberger’s dictum: Anyone who spends too much time thinking about international money, which includes thinking about how other people might be thinking about international money, goes mad.