2026 Financial Markets Conference – Evening Keynote Transcript – May 18, 2026

May 18, 2026

Evening Keynote – The Third Wave: Consequences of Global Imbalances

Transcript

Paula Tkac: Good evening, everyone. Welcome back; welcome to dinner. Welcome to our online audience. We're really thrilled that you joined us here tonight for what proves to be, I think, a very interesting conversation. I do want to remind everyone before we start the session, please use your Cvent app to submit questions for Q&A. I'll have access to those up here, and so Gita can opine on what is most on your mind.

I don't think Gita needs any introduction, really—but just in case you've been living under a rock, she is the Gregory and Ania Coffey Professor of Economics at Harvard, specialties in international economics and macroeconomics, and of course she was the first deputy managing director and chief economist of the IMF. So, without further ado; Gita?

Gita Gopinath: Thank you, Paula. And thank you so much for having me give this keynote. I was told to talk about something about the global economy, and I decided to talk about global imbalances—it's called the third wave, and you'll see why I'm calling it The Third Wave: Consequences of Global Imbalances. The reason I picked this is because global imbalances have been driving, in my view, the global agenda of these last 12 months.

The April 2 "Liberation Day" tariffs were attributed to global imbalances. After the Supreme Court struck it down, we had the Section 122 tariffs that got imposed—that was also with reference to global imbalances. What China is doing in terms of sending goods around the world, and in response to tariffs dumping in Europe and Asia, that is also being looked at as a strain on the global economy, and attributed to these global imbalances.

Okay. So, let me walk you through the three waves. The first wave is what ended in the Plaza Accord of 1985. What preceded it—and you can see that on the graph; the red is the United States, the kind of light blue is Japan, and the yellow is China. So, the first wave of global imbalances was the 1980s, when you had, again, the US running large current account deficits, and on the surpluses side it was Japan that was running current account surpluses.

And it got to a point where there was now a big pushback and a call for protectionism in the US, and they said, "Well, we've got to do something about this, or we're all going to end up putting tariffs on each other." And so, you had the Plaza Accord, which was, basically, the G7 coming together and deciding they were going to try and weaken the dollar.

So that was that. That was the Plaza Accord. And then you had the second wave, which is what preceded the Great Financial Crisis. Again, you have the US on the deficit side, running large current account deficits. On the surplus side, you have now some new entrants; you have China, you have East Asian economies, you have Japan still in there. And how did that end? Well, that ended with the Great Financial Crisis.

And then we have now what is the third wave; we have the IMF, whose responsibility it is to tell you whether the imbalance is excessive or not, and they have made the estimation that these are excessive large imbalances. Again, you have the US in the deficit position, China in the surplus position, and Europe a little bit in the surplus position along with Japan.

And so, the question—for the rest of my talk, and what comes next—is: How will this one end, compared to the previous two?

So, to answer that question we all first need to know what global imbalances are; and I think, for me, a good analogy of this question is the story of the blind men and the elephant. I don't know how many of you know of this story. It's actually a pretty popular one. It originates in India—I think the first mention is in 500 BC—but it's about an elephant that comes into a village, and there are these blind men who are supposed to describe what the elephant is; and each of them picks a part of the elephant to touch, and to figure out what it is. And one of them touches the trunk and says it's a snake, one touches the leg and says it's a tree, one touches the body and says it's a wall, and so on.

I think the discussion on global imbalances in the world is a bit like that, which everybody has their favorite view on what is responsible for the imbalance. There are some who believe that it is because of unfair trade practices. There are others who believe it's because of industrial policies. There are some who believe it's because of the dollar being the reserve currency of the world, and therefore by design the US has to run deficits. There are some who think it's fiscal deficits, and so on.

So, part of the challenge, frankly—and which is what France's G7 presidency is focused on—is trying to get everybody to say, maybe can we all agree on what actually we're trying to fix over here? What is the problem that's generating global imbalances?

Okay. Since usually most people don't know what a current account balance is, I thought I would start by doing a bit of a primer on global imbalance. This is just to remind all of us what a current account balance reflects. So, this—you should have learned this in some undergraduate class.

You need to know two things when you're thinking of global imbalances. One is the current account; the other is the net international investment position of the country. So, the current account is defined by—it's an accounting identity. The current account is equal to national savings minus national investment—which, again, through an accounting identity, is equal to the trade balance of the country plus the net factor income from abroad.

So, if a country is running a current account deficit, it means that its national investment exceeds its national savings. It means that it is, in most cases, importing more than it's exporting, but the trade balance is really the predominant piece of the current account. And then you have the net factor income from abroad, which captures the fact that you also hold assets abroad, and foreigners hold assets in your country. Some American citizens go and work internationally; we want to attribute that income to the right country.

Then, the net international investment position is the difference between the external assets that a country has, and the difference with the external liabilities. Now, this is important, because—I'm going to come back to this, which is, again, through pure accounting—the net international investment position is equal to the cumulative current account of the country (just ignore the next one; there's like a tiny detail there), plus the valuation effects.

What is that? So basically, the reason a country ends up, say, for example, having a large negative net international investment position, is because it is investing more than it's saving, and it needs to borrow from the rest of the world. And so, you're borrowing more and your liabilities are going up relative to your assets, so your net international investment position is getting worse.

The second piece of this is important, and has become very important now: the valuation effects, which is what is happening to the value of your assets held internationally versus foreigners' holdings of US assets (so, even if nobody's selling or buying those assets, just the valuation effects). If US stock markets are booming and foreigners hold US equities, then that's just automatically leading to a worsening of the US net international investment position, because of the valuation effect.

So again, past cycles of global imbalances led to protectionism and/or financial crises, and the question to ask ourselves is: How will this cycle play out? The question is, how is the cycle any different from previous cycles? Is it going to be protectionism? Is it going to be a financial crisis? Is it going to be both?

Let's start with financial crisis. What could happen, in terms of implications, in terms of financial vulnerabilities? Before we go, let's just look at this slide in terms of the US. What we have, this third wave is actually a lot closer to the first wave, which is the twin deficit story—which is what's generating the current account deficit, which is US fiscal deficits (combined, and of course, then the current account deficit—so that's the twin deficit).

So, it's a fiscal deficit-driven current account deficit in the US, unlike what preceded the Great Financial Crisis, which was more of household savings coming down very rapidly as opposed to a deterioration in the fiscal balance. And then, the net international investment position of the US, indeed, looks really tough, because it's -90 percent of GDP in 2025. In fact, that's the number that the US administration quotes when it says that the US has a fundamental payments problem, which is why Section 122 needs to be applied—because of this very negative net international investment position, which is a record. It's true; it's never been this negative.

But here's where the little formula I showed you previously is important, because if you just took the additional borrowing that has happened, or the additional increase in liabilities that have come about because the US has been running trade balance deficits and current account deficits, that number is the lighter blue line, which is the cumulative current account—it's fairly flat. It's 50 percent of GDP. So, over the last decade, yes, it has added to and deteriorated the US net international investment position, but it's really tiny.

So, how do you get from 50 percent to 90 percent? That's entirely a valuation effect. That is the fact that the US stock market, over the last ten years, cumulatively, has done much better than stock markets anywhere else in the world, and the foreign holdings of US equities have grown substantially. So, this tells you that if—we don't want this, but—if there was a stock market correction in the US, we would get a really big improvement in the US net international investment position.

So, what should we worry about? I think it's useful to think of what preceded the Great Financial Crisis, and if you remember, Bernanke talked about the global savings glut—where all the East Asian economies, and China, accumulated large amounts of foreign exchange reserves, huge demand for US safe assets.

That, of course, combined with a lot of new financial innovation in the US, all kinds of securitization, to generate the picture that we saw here, which is you saw US household debt rise from 70 percent of GDP in 2000, to like 97 percent of GDP just before the GFC in 2007. Similarly, financial sector leverage shot up, right before the GFC, while federal debt right before the GFC was about 40 percent of GDP.

What does the picture look like now? The picture looks like we're actually in good shape with households. Households have come down; we had the deleveraging, and they're at 67 percent of GDP. The financial sector overall doesn't have a huge amount of leverage. It's the federal balance sheet that looks particularly severe, where we've gone from 40 percent to federal public debt-to-GDP at 108 percent.

A big chunk of foreign gross inflows into the US do go into US Treasuries, but it's really hard to make the case that that is the reason why debt-to-GDP in the US is 108 percent. That's just decisions, fiscal choices, that were made because of all the crises that we've had.

The other thing that's of course changed between then and now, is the greater role of the non-bank financial institutions in this world, and we have less transparency around what happens with them.

So, where are the sources of fragility? So, we don't have a global savings glut, but if there was a glut, it is a fact that there is a huge appetite for US equities, which is different from what it was pre-GFC.

You have about $41 trillion in gross foreign holdings of US equities; that's about 44 percent of world GDP, excluding the US. Before the Great Financial Crisis, the number was about 20 percent of world GDP, excluding the US—so this has doubled. The world's exposure to US equities is at an all-time high, as a share of world GDP. It's really very high.

So, that could help definitely with the valuations that you see in US equity (markets are certainly helping there). You also see, in terms of Treasury purchases, there are a lot of purchases happening; foreign money is coming in to buy. But what's different, which is true even with domestic purchases of US Treasuries, is that it's rotated out of foreign central bank purchases of Treasuries to more private purchases of Treasuries. So, this is becoming a much more volatile source of demand for US Treasuries than compared to the past.

Okay. So, very quickly—in terms of the financial outcomes, the fragility has rotated. Pre-GFC leverage was being built up in households and banks; it's now about governments, potentially NBFIs, especially hedge funds (we know that). The asset bubble, pre-GFC, what was building up was housing; now it's tech/AI equities (you could make a case for it), some private credit. In terms of triggers, pre-GFC it was subprime MBS and inflation (the Fed did raise interest rates by 175 basis points before we had the dot com bust).

Today, it's not necessarily anything housing related. It could be a DeepSeek-like event; AI. And I think something that we need to increasingly worry about is inflation, and the fact that the Fed may actually have to raise interest rates again, as opposed to going on hold or even cutting interest rates.

In terms of outcomes, at one level it looks better, because at least we know that the banks are in much better shape than before the GFC, so we worry less about a banking crisis. So maybe what we will have is more like a stock market wealth correction, like the dot com.

What I have there is a calculation based on, if there was a stock market correction that was similar to the dot com bust, what would be the hit to US GDP (taking into account what you think the marginal propensity to consume is out of wealth, and so on)? It's basically a 2.5 percent point hit to US GDP, so we're talking about a recession, just purely coming from a wealth shock. Again, it's a combination of the fact that there's a lot more exposure, a lot more households that hold equities.

The path that we have a lot less we don't know about, of course, is—one is the NBFIs and what might happen there. And what we don't know anything about, and we haven't had in any of the previous waves, is what happens if the safe asset is no longer the safe asset, and how does that play out? Because as we saw, the biggest concern, in terms of leverage, is in the government and the public sector.

So, what are the implications going forward? I think it's super valuable for all of us to think about how might a stock market correction impact the economy, what are the linkages between banks and non-banks and NBFIs, and so on. We do need to maintain adequate financial regulation and supervision. The previous wave before the GFC was not just about a global savings glut; it was a global savings glut, combined with a lot of financial innovation and not enough regulation and supervision.

The third is, how do we make sure that the Treasury market functions? And I think that the panel earlier in the day really made some great points on that. And how do we preserve central bank independence?

And of course, the absolute best-case scenario would be if we were all working to reduce the US fiscal deficit. So, I was looking at this purely—by the way, all this, everything I've been talking about so far, is looking at it from the perspective of the US, just given the time constraints.

So that's on the financial crisis side. So, how will this end? Again, the good news is, I think we don't have the problem where there are balance sheet issues with households and with banks, but it could be just a stock market correction. The "unknown unknown" is what happens with Treasuries and government debt.

We don't have to see whether this will lead to protectionism. I think we're already obviously in full blown protectionism. This is what's unique about this time relative to the past.

This is from a paper that I have with a few co-authors where we look at trade between different blocs, the Western bloc (which is the US and Europe), and China on the other side. The dark blue line tells you what happened during the Cold War, and how much trade between the blocs dropped off, in addition to all the other factors that typically drive trade. The red line is what we are tracking right now.

And so, I guess the question is, where will this end—and where will we end up in terms of decoupling or fragmentation? Also, I think that now it's a bit different, too, because it doesn't look like there are just two blocs. I think there are more complicated relations between the US and Europe, to put it mildly; and so, whether there will be this kind of a bloc breakdown, who knows?

There are also major shifts underway, in terms of how trade is reoriented. If you look at China's direct trade with the US, I think it's pretty fascinating. Basically, the share of China, in terms of its direct trade into the US, is basically back down to the level it was before China joined the WTO. Basically, back down to 8 percent. So direct imports from China is at 8 percent, which is what it was before China joined the WTO.

Now, what's happening is that there is an important fraction that is getting rerouted—not just simply rerouting, but also China setting up factories in other parts of the world, especially Vietnam. And you see a big increase in that, is what's happening.

Taiwan, by the way, is all AI cycle; the huge increase in its trade—which, by the way, I also think the reason trade is holding up much better than expected is because of the AI cycle that we're in right now. The right-hand graph basically tells you that if you look at the change in China's export share to certain countries, and the change in their exports to the US, it's a positively sloped line, which means that countries that are selling more now to the US are also importing more from China.

But what's interesting is that in the last 12 months, Mexico—I'm not showing you the whole thing, but Mexico used to play a more prominent role there, but Mexico is actually sliding down. So, while Vietnam continues to play that role, in terms of being a connector country, Mexico is no longer doing that as much—at least, as far as we see.

And then I feel like I have to have a slide on the dollar, because everybody keeps asking me: Is the dollar still dominant? What's happening with the dollar? And I think the broad strokes answer is, basically, that the dollar remains dominant. There are, however, changes on the margin.

So, the left graph is, again, another paper I have with co-authors that looks at the share of renminbi invoicing in global trade. Even if you can't see the numbers, the basic takeaway is that Russia has certainly moved a lot more to renminbi invoicing. And then there are other countries that are more distant from the US geopolitically, that are also moving, taking on the renminbi a little bit; but the numbers are tiny. We're talking about like 2 percent, 2.5 percent of their trade. So, this is tiny, tiny levels.

Similarly with FX reserves, there's been this gradual slide down in the US dollar, once you correct for exchange rate movements and all that. You see a gradual slide down. You're certainly seeing an increase in gold holdings that are certainly there, but nothing dramatic in terms of the currency.

The G7 France presidency, we were asked ("we" as in Hélène Rey, Axel Weber, Chong-En Bai—who's from China—and I) were asked to put together a report on, assuming somebody's interested, how do you solve global imbalances? How do you do that? And so, we just put out that report a few weeks ago, and the basic answer is, like the elephant, it is macro. What you need to do to solve it is, is it macro—which is, you need the US to be able to run smaller public deficits that will also reinforce financial stability, you need China to stop this chronic underconsumption and move towards higher consumption spending, and for the European Union to have more productive investment.

We have a few other suggestions there, and I can talk about that. There's also the issue of sectoral imbalances, which I think are important even though they don't feed into global imbalances. The fact that there are some countries that have a big advantage in certain sectors—like, say, rare earths or chips, and so on. Even if they're not a big part of the global imbalance story, they are consequential. They are going to generate even more protectionism in the world, and so we go into the report about how we could have certain reforms to the WTO to make it work better.

Okay, so since I'm out of time, I'm going to end with coming back to the elephant. The reason I like to end with this is because the story actually of the blind men and the elephant does not have one ending. So, from 500 BC until now, the story has been adapted and has ended in many different ways.

So, the original, there are stories where all the blind men just fight with each other, and that's that. Nothing comes out of it. There is a version where a very wise man (or a king) walks by and says, "Okay, I can see the elephant from a distance; I can see it's not what you think it is. You guys are all right, but you need to listen to each other." And of course, there's a version where all the blind men sit together and actually say, "Okay, maybe you've got a point, maybe I have a point; let's figure it out."

I have no idea where the world is going to go, how this is going to end. But this is very important, not just from the near-term financial stability perspective, but from what it will do to living standards, I believe, all over the world if we let this keep unraveling the way it's doing right now. Well, thank you.

Tkac: All right. Thank you. So, we have questions coming in. We're going to get to those in a second, but I want to start with your last slide, and maybe go straight to the elephant in the room, as it were, that we can all see.

In the last slide, you talked about simultaneous policy actions that are needed across the US, EU, and China in order to address these global imbalances, so I'm really curious about your take on how geopolitical alliances in the geopolitical environment right now are affecting the ability to get to some of these policy actions. So, first there's the coordinated part; do you think there's a chance that these things could go on at the same time, including dealing with the current situations that we're faced with?

But also, what if there's progress in one of them, in one of those countries or areas, and not in another? Will that really help the global imbalances problems that you see?

Gopinath: Yes. So, we were very careful to use the words "simultaneous" and "coordinated." There was a reason for that, because we don't think there's much coordination happening, or cooperation happening.

Each of these policies are beneficial for the country; they should be doing this, even if they didn't care about global imbalances. And so, in a sense, it is in each country's self-interest. This is not about, "Please do this because it helps the world." It's about, "Please do this because it helps you."

Tkac: It's incentive compatible.

Gopinath: It is incentive compatible; exactly. So, that's a good thing. The problem is, let's suppose this happens in a "one country does something and another doesn't" scenario. There, we could actually have some serious problems.

So, let me give you an example. If China were to say, "Okay, fine; we're going to now take action to have more consumption spending, and reduce our savings in the economy." That's going to cause global interest rates to go up, if nothing else happens. If global interest rates go up, we have a problem at home, and many countries in the world. That could be very destabilizing.

So, it's meant to be something that everybody takes some action on. It doesn't have to happen overnight. It's mostly the direction of travel, frankly. The level of the current account deficit and the imbalance is nowhere near what it was before the Great Financial Crisis; it's smaller. As long as everybody is moving in the right direction, that would work.

I actually think that the imbalances are going to grow, for the simple reason that investment in China is doing really badly—the property market crash, overcapacity in many sectors—so I suspect that they're going to have larger surpluses. And in the US, investment is doing really well. So again, if you remember our undergrad class—

Tkac: I remember the math that you showed us.

Gopinath: Yes, exactly. The profound math will tell you that you should have bigger deficits.

Tkac: Yes; awesome. Okay, one more question from me, and then we'll go to the audience questions.

So, I'm thinking specifically about your days at the IMF, and thinking about this G7 work you're doing. A lot of people in this room are financial market participants. Some of us are in policy, and we're thinking about how to look forward and anticipate where this may go. How do you think, or what can you provide as advice? I'm thinking about scenario analysis and risk assessment, when we're talking about just such a massive problem, that would require, as you said, coordinated action to really make it much better.

Gopinath: Yes. I think in the near term it's a good idea for doing all the things we were discussing earlier today, which is looking at US Treasury markets, and seeing what happens if there is suddenly a shift in appetite for US Treasuries. How much dysfunction could you possibly have?

You should also think about how you would respond if any of those triggers were to kick in. You had an inflation problem, you had AI bust; I think the picture that we have to keep in mind is that it's not the households' balance sheets, it's not the corporate balance sheets, it's not the financial institutions' balance sheet (as a whole, of course)—but what's unique about this time around is the government balance sheet, and how will Treasuries respond this time around.

That's the big question, because after the Great Financial Crisis, even though the US was running big deficits, you have the GFC, everybody rushes to the US safe asset. The dollar appreciates. Now the US, the current account deficit did shrink, but that was because US investment dropped, by a lot; and household consumption was affected.

But the drop in investment was a big—housing investment was a big driver of why the current account improved. But that was a time when the dollar appreciated, and the Treasury yield curve flattened.

The question is, what will happen this time around? And I think it's a good idea to stress test your models to allow for the opposite to happen, and what that could imply.

Tkac: All right. Okay, first question: Why don't flexible exchange rates help to correct these large imbalances? That is, why don't surpluses cause more appreciation? Is that channel simply weak, or is it prevented by other factors?

Gopinath: Yes. So, there was absolutely no reason flexible exchange rates will close imbalances. I think that's like a mistake that people have in their head. There is no reason why, and the exchange rate is a price. It's determined by all the fundamentals of the economy, and the policy choices of the economy.

But there is nothing that says that because a country is running a current account deficit, its currency should depreciate and therefore you should have smaller imbalances. I don't know if that's like an idea that floats around, but there's nothing.

The only mechanism that truly says that this cannot go on forever, is the statement that if a country is running deficits and continuing to accumulate liabilities, then at some point—unless you're running a Ponzi scheme—there is a problem, and people will have to... either you default on it, or something happens. There is nothing, there is no natural force that makes a surplus country need to adjust in any form or shape.

So, there's an asymmetry here, that the surplus country—really, you could run surpluses for as long as you want to. The only risk is that you may hit the zero lower bound and so on, but that's separate. But the sense that somehow the exchange rate is an automatic correcting mechanism—it's not right.

Tkac: It's not right; okay. All right. We're myth busting up here, is what we're doing. Okay, I guess we got a question from David Wessel, and I'll add a little of my own to it because I had a similar question about, I've termed it industrial policy; but he says: How significant are the increased use of sanctions, export restrictions, and how lasting are these? What are the effects of those?

Gopinath: Of the export restrictions, or the general tariffs, or—

Tkac: Yes, just putting all of it together again. I was thinking of it as just industrial policy in general that you might use; you might use sanctions or tariffs or export restrictions. I'll let you tell us which ones are important, if any.

Gopinath: I think we've certainly moved away, and I think some of it for good reason, from that world where it was about efficiency-driven free trade. I think there was a rude shock with Russia's invasion of Ukraine, with the realization that Germany should not have depended 100 percent on Russia for its energy.

So, I think there are good national security grounds for which, given this is the world we're in, you have to have some more production at home. You have to build in some more redundancy. It can't be just all efficiency. And therefore, industrial policies can play a role, too.

So, that's a sense in which I think we've moved away, permanently. Then of course, there is, given the geopolitical environment that we're in, every country is going to do more of everything they think is going to be important for their national security.

So, Europe is going to do more defense production; they're going to try and build their own payment rails. Any country that has any fiscal space to bring production home is going to do that, so we are going to be in this environment where I don't think you can... anybody will revert back to a world that says that no, no, we'll just go back to how it was in, say, 2016; maybe that's as much as we can go back—where we just bought from the cheapest place, and we didn't really care about these details.

So that's a permanent change. What the impact on the world economy will be... there are many simulations and values, that I think—what I believe is some of the work that the IMF has done, which looks at what the current crop of tariffs and counter tariffs and everything has done, which is shave about 0.7 percentage points from global growth over 2025 and 2026; I think that's the number that they have. That sounds about right, in terms of—and then, AI has offset that, which is why it's... it has helped.

This is way more than just commerce, I think, right now. I think we just have a huge loss of trust in the world. The last time, when we had the Great Financial Crisis, the G20 came together and worked together to try to save the world economy. I'm not sure there will be any of that happening this time around. There was a lot, in terms of financial regulation, that it helps if there's cooperation. John Schindler earlier today talked about the importance of that cooperation. Are we going to end up with all these fragmented rules and interoperability problems?

So, those are the kinds of problems that are going to crop up. So, we're just moving to that kind of a world, which will have many more transaction costs, many more resources devoted to things that you could have otherwise just traded with each other, but that's—

Tkac: Things that would pull us away from the balanced growth, and the...

Gopinath: That's going to pull us away from any kind of reciprocal trade and balanced growth, yes.

Tkac: Okay, we're going to continue on the "myth buster" theme. International trade and finance are really complicated, and counterintuitive. What single fact do you think most media accounts get persistently wrong, and how would you fix that perception?

Gopinath: I think it's still very popular to say that because the dollar is the reserve currency of the world, the US has to run trade deficits. So that's, again, I think hugely problematic because the argument goes along the following lines, which is that because the dollar is reserve currency, everybody in the world wants to hold dollars. And so, the US has to put out dollar assets, and what does it do with all that money that it gets? It has to go and spend it on goods and buy stuff from other countries. That's the logic of the argument, that that's how we end up being such big importers, is because you get all this money and you have to put it somewhere.

Of course, there are many other things you can do with money. You can take all that money, and buy other assets; but you don't have to buy imports. So that just basic idea that somehow because there's a huge safe asset demand, that that will necessarily mean you take that money and you buy other goods, is just wrong.

It's true that it keeps interest rates lower in the US, and that through that channel you can end up with some more borrowing; but that's a small player compared to the other factors, like fiscal deficits or US household savings decisions, investment, and so on. Much, much more than interest rates.

Actually, Maury (Obstfeld) wrote a fantastic piece for the PIIE recently where he basically said, don't blame the current account deficit on the dollar—but I suspect we will keep hearing still, the very important role that, because it's the dollar, we have to run deficits.

Tkac: And once again, that misperception is going to get in the way of all the policy that you talked about.

Gopinath: That misperception is going to generate also some very confused policy; Stephen Miran was of the view that because of that, we should somehow depreciate the dollar through taxing holders of Treasuries. Yes. That's the problem.

Tkac: Okay, let's turn to financial stability. In an AI-driven market correction—you talked about that possibility—how would you begin to distinguish a healthy repricing from a systemic event?

Sometimes it's, we know it when we see it, and you can't forecast that; but, given what you showed, which is increased international exposure to US equities, what we heard from Jeremy Stein this morning, typically maybe we don't think of equity market value as being necessarily systemic. I'm really curious about how you think about, one, what would it take for it to be systemic, and two, what are the international spillovers, perhaps, that could come from a US equity market devaluation that we might not have seen before?

Gopinath: I think we have to appreciate how big a role AI is playing in holding up growth, not just in the US but in other parts of the world. And it's holding it up through the investment that's happening, data centers, chips, and so on. But also, in terms of the stock market valuations and through that wealth effect—also trade. Korea, which should be hammered by what's happening in Iran, at this point is having a booming economy because of its exports, because it's exporting chips that are, or at least a part of, the global supply chain for some semiconductors.

And so, if that were to go bust, we're looking at many economies, that otherwise look healthy at this time, slowing quite considerably. And unlike the previous dot com bust, public debt levels are extremely high, so the ability to come in and do any big stimulus will hit market constraints. And we're in this world where we have made choices of more fragmentation, more inefficiency built up. And so, it's hard. It's like, everything else looks difficult compared to the AI, and the stock market looks great; everything else is a problem, in a sense.

And so, if the good thing—

Tkac: Goes away.

Gopinath: —You would worry about how much worse it could be this time as compared to dot com. And also, I think the consequence for the rest of the world is much larger, because of their much higher exposure to US equities.

Tkac: The trade imbalances with Japan in the '80s and the '90s were an enormous concern that proved to be overblown. What are your thoughts on the imbalances with China? Do you think China will also prove to be a paper tiger, or is this the real deal?

Gopinath: Will China have a... I'm sorry, what was that?

Tkac: Will it prove to be a paper tiger, in that the imbalances between the US and China—that those concerns are overblown and are not really going to become a problem, or is this the real deal?

Gopinath: So, China is on the surplus side, right? And the question is—their bigger problem is what they have with the property markets, the fact that despite the last five years of trying to fix it, they haven't been able to do that, and that's weakened even further, which is a problem, which is why investment has come down.

So actually, compared to, say, 2019, two things happened to improve China's current account surplus. One was COVID, where it was manufacturing everything that we wanted during that time, and selling it to the world. But also, the fact that the property market slowed down, and investment in the property market was huge. So, compared to any other investment that's been done, in EVs or solar panels or anything—those are a blip in comparison to what's happening in the property market.

So, you saw that investment drop; and last year, 2025, was the first year when total investment in China actually was negative—it dropped. So, that's from our saving investment equation. Savings rates haven't changed that much. Investment is collapsing. That's what's generating these large surpluses.

So, the problem for China is the unbalanced growth that it has, which is a growth that's been very reliant on investment—which is kind of interesting, because if you talk to Chinese authorities, they love to talk about investment. They don't like talking about consumption, because they say consumption generates... when the government does anything to increase consumption, it's moral hazard. Investment is good; it's all supply side.

So, the first wave was about investment in properties; then, that's gone, busted out. Then there was investment in all the frontier technologies. That's where they're putting the money, and now the most recent 15th Five-Year Plan is about investment in people, which is about health care and education—which we might call consumption, but they're also like, "this is investment," which is true.

So, at least that direction of travel is helpful. I don't know how fast they will go there, though.

Tkac: So, let's go back to AI. We do have a question here that asks what early signals you might look to, to gauge the probability of an AI bust.

Happy for you to answer that, share your secrets with the rest of the world; but I'm also interested in the aspect of this of drawing on your knowledge about what the AI landscape looks like, across the globe. Because it feels like, it's new emerging technology, it could certainly increase productivity. There's rapid change, but there's also national security concerns. The investment that you showed, much of Taiwan's growth is coming off of AI and chips.

So, what does the race for AI look like across the globe, and how does it play into what you see as global growth prospects?

Gopinath: So, firstly, I don't spend my day staring at the Bloomberg terminal, so I don't know what will make me worry about it. I think because of AI we will certainly have an increase in imbalances in the near term, just because, again, the US is the go-to place for both investment, and it's attracting a lot of capital inflows because it is the market to bet on if you want to be a part of the AI prospects. This is the most dynamic place to do so. There are the most amount of assets that you can actually put your money in.

China has a lot of AI, but nobody's rushing to put their money in China. So, it will increase imbalances, which means I think we will be having the same conversation again next year—unless there's a crisis, which of course nobody wants that. The risk, of course, is that we end up with some more of this divergence, this feature of everybody having a lot more interest in US markets.

It's not something that's just happened in the last few years. It's actually now a feature of the last ten years, and it is because the US economy has been doing much better. If you look at measures of productivity, this is where the dynamism is. It's a good market, in terms of growth prospects.

So, unless there is more of what we call "balanced growth" around the world, which is that there are other places that offer attractive investments, including in Europe, if they can get the single market going, much more and performing even better; they're working towards it. If that can improve, then that would be more opportunities for investors to diversify where they park their money.

If emerging markets also were to become a much more attractive place for foreign direct investment, that would help. So, in 2025 we saw this rotation for the first time after a long time, where capital flows were going to the US, but now also going into Europe and into emerging and developing countries. But most of the flows that were going to emerging and developing countries were basically carry trade-like flows; so, it was mostly portfolio flows. It was not FDI; it was not buying of equity.

So, we do have this situation where there's not enough, in a sense, good growth opportunities in many parts of the world. So, it will remain concentrated, I would say, in the US.

Tkac: In the US; okay. All right. Well, we're at time, so please join me in thanking Gita Gopinath—and thank you, online attendees.

So, a reminder for everyone: we start tomorrow at 9 a.m. We're going to start with the evolution of central banking in the digital era, to continue on the theme of stablecoins and other things, and then move on to the economic value of AI and its effect on financial markets.

So, we'll see you in the morning. Thank you.