From the ECB research conference on geoeconomics, Day 2. My session starts at 10:00 and runs for about 1 1/2 hours. Transcript follows:
Economic Size and Economic Power
ECB Annual Research Conference 2026
Paul Krugman: Thanks to the ECB for having me here. Thank you all for listening. Just a quick story. I decided to do international economics, alarmingly, 50 years ago in graduate school. And at that time, some people in particular advised me not to do international trade because it’s boring; nothing ever happens and there are no exciting things occurring in the field of international trade. Luckily I didn’t take that advice and things are certainly not boring. The topic of this conference and of a lot of the recent literature, aside from obviously being incredibly important, it’s also intellectually exciting. We’re seeing some new tools, and I’m particularly delighted to see people making use of new data sources, new ways of evaluating. But there’s also new questions.
One of my old teachers, Evsey Domar, said that bad times make for good economics, and I guess that’s kind of what’s happening now. We’re getting a lot of interesting work. Unfortunately, the side effect of that is that we have bad times to worry about, But anyway, here we are.
In a way, this talk is going to be two talks, because the first part is my understanding of where we are in terms of how to think about geoeconomics, which I have been frantically trying to get up to speed on. The second part is going to be about size and power issues, which I believe are becoming really critical. These are subjects we heard a lot about yesterday here at the conference and I will offer, I think, a somewhat different take, or at least a different way to look at it. And I think there will be some surprises in all of that.
So when I decided to become—I don’t know how to describe my current role—but I think I’ve become an influencer? That’s what I do these days. Write these daily, or almost daily posts. And one of the things in that kind of domain is that you always have to have a picture, an eye-catching picture at the beginning. You always start with something that will catch people’s eye because, you know, no one is forced to read it. And I tried to think, how do I come up with a picture of weaponized interdependence and geoeconomics?
And the answer is, well, I don’t know if I’m ashamed or not, but I do read science fiction. And I don’t know how many people have read the classic science fiction novel, Dune, which was the subject of several terrible movies before, finally, Denis Villeneuve got it right.
But the premise of the novel is that there’s this planet, Arrakis, which is the sole source of something called spice, which is essential to the galactic economy. And so everything relies upon it. And so basically Arrakis is the Strait of Hormuz with giant sandworms. It is really very much relevant to current issues. By the way, the struggle for control of the spice leads initially to weaponized interdependence and eventually to outright war, which is kind of where we want to probably end today’s talk.
Let me give you two quotations. The first is from the novel, from the protagonist of the novel. “He who can destroy a thing, controls the thing.” —Paul Atreides
And the second is a rather more verbose statement from Albert Hirschman, who is the patron saint of geoeconomics, basically about how gains from trade are also potentially a tool of coercion: “The total gain from trade for any country is indeed nothing but another expression for the total impoverishment which would be inflicted upon it by a stoppage of trade.”
Those are really the same statements. They’re really saying the same thing. Now, what is interesting about both of those claims is that, whether it’s the character in the novel or Albert Hirschman, they’re basically implying that you yourself have to be willing to pay the cost of the destruction. And that is, I think, one of the really big gaps that we will have in trying to understand where we’re going with geoeconomics. (I guess Dune is cosmoeconomics rather than geo-economics, since it’s interplanetary.) But anyway…
So, how do we think about this? It’s a dirty little secret among people who do international trade theory that in some ways it is the general equilibrium field. Everything needs to add up. We all know that we have to consider interlocking markets, competition for factors of production, all of that. And yet, most of the time, when you want to think about policy issues, you end up with supply and demand. You end up going with partial equilibrium. And so that’s what I ended up doing, trying to come up with my own version. And I was both hopeful and afraid that some of the work presented at this conference would basically make this look silly by coming up with a true general equilibrium approach to geoeconomics.
There is a known familiar kludge; a way to do partial equilibrium while saying that it’s general equilibrium, which is to presume that there is an outside good that is somehow not the subject of policy and is perfectly competitive and produced at constant returns. And we all know that that’s a dodge. It’s a way to deal with it and better to do that than to give up on any kind of comprehensible framework. And I thought it was a terrific paper by David Yang and Earnest Liu yesterday on power. But I was both reassured and disappointed to see that they did the same thing. They used the same kluge to deal with this. And so we can do partial equilibrium.
We imagine one country exports a good to another, and of course then we can imagine that there are multiple goods. And the weaponized interdependence is that a country can threaten to cut off this trade. If we do a standard consumer producer surplus model, essentially, ordinarily we would expect that there are gains to both sides. The exporter of the good gains a producer surplus, which is the bottom triangle. The importer gains the consumer surplus. The threat of weaponized interdependence is that, well, we can cut off those gains and that will cut off the gains for both sides. Which do we think is more important?
As economists, although it would take some, it takes a lot of quantification to be sure, but almost surely, most of the time the pale red triangle at the top is going to be bigger. It is much more common for countries to be really dependent upon imports of something and to have no good alternatives than it is for exporters to have nothing else they can do. Not just because there may be many exporters—or other things you can export or other markets—but also because resources can be reallocated. So although there may be a significant number of people employed producing for a foreign market, they could be doing something else. And so strictly speaking, we generally would tend to think it’s the slope, not the elasticity. We typically think that the elasticity of demand for imports is substantially lower than the elasticity of supply of exports and that the welfare losses from losing access to an export market will typically be low.
Two things here: One is that people in actual positions of power, if anything, think the opposite. I try to not be too political here, but obviously we have an unusual, rather baffling trade dispute between my country and its northern neighbor. And the position of my president is that we don’t need Canada for anything, that there are no benefits to importing oil and aluminum and electricity and so on from Canada and that we are subsidizing Canada and that all of the benefits from our trade together are the bottom triangle here. Okay, that is not standard economics.
I do worry that we may be a little too glib about dismissing the producer surplus side or the export side. Partly that’s because there are situations in which the resources that are being used to produce and export do not have readily available alternative uses. And if you want an actual example that is relevant to this current thing: Athabasca tar sands in Alberta, Canada. At the moment, there is no real other place. Your oil is being extracted in Alberta and is shipped to the American Midwest through pipelines. There is limited pipeline capacity that can take it off to the Pacific Ocean, but not much else. In this case there is actually a producer surplus that’s being generated by the ability to export the oil and the cutoff of that trade would hurt Canadian producers. So there would be some welfare loss there.
Time is always a factor. And one of the things in all of these is that given sufficient time, both of these lines get much flatter. So the Canadians are in fact working on expanding their pipeline capacity to the Pacific. So that will be less of an issue. Another thing worth saying, is that this is also true on the demand side. With all of this craziness, one of the things that we’re witnessing happen in the current oil crisis is that on day one there was really no alternative to the Strait of Hormuz. As time goes by, however, you find ways around. You find ways to substitute, alternative routes. In some sense, the elasticity of demand also rises over time so that the power that comes from the ability to disrupt trade is a wasting asset. You can use it for a time, but not indefinitely.
Another point to make—since I am going to be talking about sizes of countries and power that comes from size—the classic, small open economy that produces a homogeneous good that is traded on world markets and has no ability to influence world prices, presumably does not have power in world trade. But another one of the dirty little secrets of international economics is that to a first approximation, there are no small economies. Even economies that are in fact very small, produce differentiated products. We know this, in part, just by looking at the world and what happens when you try and model world trade; try and model international patterns of trade.
CGE models have been Armingtonized from the beginning, so you always have to just assume that countries produce differentiated products when trying to actually model patterns of trade in a more fundamental way. You end up doing something like Eaton-Kortum, where you end up with randomized technological advantages, where almost everybody to some extent has significant market power in trade in some limited respects.
It’s probably fair to say that every economist my age that you’ve heard of was a Rudi Dornbusch student. And one of the many instructions that Rudi taught me that has been very useful in life is, when you’re not writing for graduate students, never begin by saying, “Consider a small, open economy.” You always begin, “In Belgium.” Although Belgium is not in fact a small economy in that sense. It is in fact producing differentiated products.
And even the example I’ve been using of Canadian oil, oil is a globally traded commodity. We do talk about a world price of oil, but in fact, I did say that the Canadians really have at the moment no alternative buyer. Their oil pretty much must be shipped to the US Midwest because that’s where the pipelines run. It’s also true that the U.S. Midwest has basically no alternative to the Canadian oil. A lot of oil being produced is from the Permian Basin, but the pipelines don’t run that way. And also it turns out that oil is not oil. That the Canadian oil is heavier and the Texas oil is light, sweet crude and for certain things, they need the heavier, sticky stuff that comes from Canada.
So, pretty much everybody has some market power. So,I don’t know. Maybe grain exporters? Though, I suspect that if I got into the wheat market I would learn that wheat is not not wheat. That there are different varieties and all of that.
So, generally, everybody has market power, although it’s a wasting asset if you try to use it. So everyone has some coercive power that comes out of global trade. Nonetheless, size clearly matters. Big countries are more likely to produce varieties where there are few if any competitors, they’re more likely to just have more stuff, so there’s more things that the world depends upon them for, and to the extent that we worry about the producer surplus, they have bigger markets. So size is an important part of this discussion. And it showed up in Steve Redding’s paper yesterday where they basically had to have a size factor for the matrices and we’re able to get the weight on it, but the factor was GDP. And it turns out that the elasticity of power with respect to GDP is 0.78, which is kind of interesting. But my immediate thought was “which GDP?” And I’ll get to that in a second.
In another discussion we were talking about hegemonic systems of world trade and what happens when we have a flip and the United States is no longer the dominant economy and how does that affect the system? But that’s not a prospect. That has, by many measures, already happened and a lot of what we’re trying to understand now is what we do in a world where that has already happened. I’ve been writing about this because it’s both important and also I think intellectually very interesting.
So one of the ways we can compare economies is purchasing power of their GDP. Now you might say why not just use dollar values? And the answer mostly is just instability. If you are going to compare the Euro area with the United States, it looks as if something terrible happened to Europe after 2008, but mostly it’s just that the Euro declined against the dollar. It’s not mostly a real movement in the two economies. So you would basically use something like purchasing power parity to smooth it out.
And if that’s the criterion we use, well, there’s a clear number one economy in the world, and it’s not the United States. And if we do the US-European comparison, actually, there are three economic superpowers in the world by this measure. There’s the United States, there’s China, which is clearly number one, and then there’s the EU, which are roughly co-equal. It’s not the world as the U.S. Treasury Secretary would portray it, let’s say, but it is the world that the standard numbers do.
Sometimes people ask, “Okay, how reliable are these purchasing power parity comparisons?” And the answer is, look, they’re hard. It’s a lot of work, but a lot of work is put in. This is tough as coming out of the World Bank’s International Comparisons Program. And so this is not a casual enterprise. This is something where people have put a lot of effort into trying to make these calculations. And they mostly are in accord with other measures that you would think would get the same thing.
So, just how big are these economies? How much stuff do they produce? We are now in a tripartite world, at some level almost co-equals except that by the numbers China is clearly the biggest. And then there are questions about political cohesion—the ability to act as a unit which is arguably the really big difference between the United States and Europe. That’s the world we’re in now and so any discussion about geo-economics, any discussion about hegemonic stability should bear in mind that, look, there is no hegemon, not by the numbers. China is not yet big enough to be the hegemon, but the United States truly is not big enough to be the hegemon. So we are just not in that world anymore.
There are two puzzles involving these comparisons. And some people who’ve been reading my Substack know I’ve been spending a lot of time on them, partly because I think it’s important, but also because it’s interesting. Starting with the one I’ve been spending a lot of time on, which is the US versus Europe. I’ll go back a slide. By purchasing power parity, go back to the year 2000 and the US and the EU 27— the countries that are currently in the EU 27—look at them now and they’re about the same size as total economies. And yet, if we use national income accounts, the United States has vastly outgrown Europe. A much bigger difference. And it turns out demography is much less of a factor than I thought it would be. Although the US has higher fertility and until yesterday had higher immigration, it turns out that it’s mostly productivity. Measured productivity has risen faster in the United States than in Europe. How can that be?
These are both numbers compiled with great care by very careful competent people. National statistical authorities, the International Comparison Program. How can they be telling such different stories? Well, one thing I have learned, having been in this business for a long time, is that when you see a real anomaly, when something is happening that does not seem to fit your models, one strong possibility is bad data. And that has happened to me. They were hopefully forgotten, but there was a huge literature on why the decline in the dollar following the Plaza Accord didn’t translate into rising import prices. And the answer turned out to be bad data. It turned out to actually be bad data in the same way that might be part of the story here.
So, how do you compute economic growth? We often talk as if economies produced a “thing.” We talk about “one-good economies.” And, you know, that’s a very helpful thing, but it is, of course, a metaphor. And sometimes the metaphor can fall apart in a couple of ways. And one of them is that, obviously, many of the goods that we use and live by in the year 2026 didn’t exist in the year 2000 or have been so transformed as to be fundamentally incomparable. So how do you take account of that? And the answer is hedonic measures. You try to measure, in effect, what is the value of the extra stuff that this new good delivers? And again, the statistical agencies put a lot of effort into this, but there is no international standard for hedonic adjustments. We don’t actually all do it the same way. And although it’s hard to document—though it appears to be the case—it does look as if the US statistical agencies are a little more aggressive in their hedonic adjustments than the European agencies are, so that the United States may be assigning greater value to smartphones in some sense than Europe is.
I don’t think that’s the main story, although I think it is part of it. Although it’s based partly on evidence and partly just on what makes a nicer story, I think a lot of it is simply that in a trading world, the “one-good economy” is a really problematic metaphor. We do in fact produce different mixes of goods. I’ve had an amazing amount of trouble explaining this story to some people, including smart people. So let me give you another version of this.
Imagine a world in which there are different rates of technological progress in different goods. A Riccardian one factor world is all you really need to imagine here. For reasons that are outside the model, there are just different rates of technological progress and their persistence. So you have much faster productivity growth in information technology than you do in haircuts, or whatever. And imagine two economies—I’ll call them “America” and “Europe.” Imagine that Europe lags technologically. It doesn’t have to be a large lag, could just be a year or two. But it just lags a little bit behind. What would the pattern of comparative advantage be? Well, it would be that America would have a comparative advantage in the goods that have the most rapid technological progress because of the technology gap. There would be some technology gap in everything, but it would be much larger for goods where productivity rises 30% a year.
In that case, what would the pattern of trade be? Well, America would specialize in the goods with the most rapid technological advance, Europe would be into the goods with less rapid technological advance.
What would happen to relative living standards in that world? Nothing. This is a steady state world. The relative prices of the rapidly advancing goods would fall so that relative consumption levels would be the same.
What would it look like in the national income accounts? The answer is, do it chain linked. Economic growth, productivity growth would be a weighted average of productivity growth in every sector in which you produce. But because the United States is doing the high productivity growth stuff, it will have a higher measured rate of productivity growth. And this is true, it’s happening. And nobody’s making a mistake here. But if you drew the implication that this meant that Europe was falling ever further behind, you would be wrong. This would actually be a perfectly sustainable state of affairs. You might say, “How can Europe close that technology gap?” This would be a good question, but it is not a sign of ever falling further behind. It’s not a sign that Europe is turning into a museum of its past glories.
So, I like that story partly because, well, I like Europe, but also because I think that it happens to fit rather neatly into what we actually see in the data. This should not discount the possibility that we really just have a data problem, but I think there is this real technology issue and the technological lag is pretty obviously real. It’s not huge. It’s almost entirely on the producer side by the way. The application looks the same on either side of the Atlantic. So that’s a story. I’ll try to get back to what that might mean for economic power in a few minutes, but let me move to the other case.
The other place where we have something funny in the numbers is China. I’ll compare China and the US. It’s a little easier. So, on purchasing power parity, China is by far the bigger economy. In dollars, at current exchange rates, China is still substantially smaller than the United States. Although it might sound paradoxical, this one is actually familiar. The stuff I’ve been saying about technology and Europe versus US is stuff that for some reason was neglected. This we understand extremely well. This is Balassa-Samuelson. As countries develop, productivity grows more rapidly on average in traded sectors than in non-traded sectors. There’s a systematic relationship between the level of per capita GDP as measured either way: as measured by purchasing power parity and the apparent price level.
So, electronics are similar in price in the United States and China, but haircuts are a lot cheaper in China. Those are both valid concepts. The dollar value of GDP is still substantially higher in the United States, although the gap is narrowing. But the PPP is much higher in China. I did a scatterplot of Balassa-Samuelson using the most recent data. And I won’t show it to you because it’s embarrassing, because it works too well. It’s so good a fit that I almost wonder whether somebody was cooking the books to make it look right. But anyway, that’s very clear. So there’s no real mystery.
Ok, let me give you another chart that I think is helpful that may just fit in with the rest. But first, when we’re talking about power, despite the slight defense I gave of Trumponomics, it is primarily about dependence upon imports. Which means that it’s about production. And it’s about production of traded goods. And we want to know who commands dominating positions in enough industries that the threat of disruption of trade is really serious, really something to be worried about.
If we look at manufacturing value added—and this is not purchasing power adjusted, this is just current exchange rates, but for some reason the big fluctuations associated with exchange rate movements are not nearly as visible here. (Probably because it’s traded.) No question that China is the workshop of the world, that China has by far the biggest manufacturing sector. It is not quite like the United States in 1950, where there was more manufacturing than the rest of the world put together. The US plus the EU combined are still larger. I didn’t actually put it in here, but it turns out that EU and US manufacturing value added are about the same. So actually the manufacturing numbers look like the purchasing power parity numbers. They basically say that on that basis, the US and the EU are roughly co-equal powers, both individually inferior to China, but bigger combined.
And so, that’s kind of our power ranking in the world. Or it’s our economic weight. How does that translate into power? There are several big questions that we need to ask here. The first is just about these raw numbers. They’re not quite so raw, but these partially massaged numbers don’t necessarily fully measure the extent to which a cutoff of trade will damage one side of the exchange because they need to be adjusted for elasticity of demand, substitutability. And again, the Yang and Liu paper went through that, found some major disjunctions just to the naked eye, obvious points. Taiwan is a quite small economy, and yet, if you really want nightmares about possible disruptions to world trade, the idea that something would happen that would disrupt Taiwan’s supply of a large part of the world’s semiconductors is a really big deal.
There are also, of course, geographical positions. I don’t think we would be talking nearly as much about Iran if it were not for the fact that 20% of the world’s oil was passing through the Strait of Hormuz. So there will be specifics associated with industries. I really applaud the Liu and Yang paper for making a real effort to get at this. I am a little bit nervous about how well we really estimate those demand elasticities, however, which are crucial. Not a criticism. That’s just saying this is really hard. Again, if you spent a lot of time on international trade modeling over the years—the long, mind-numbing history of computable general equilibrium models—you know that elasticities in trade are one of the hardest things to pin down. And do we fully appreciate that?
This, by the way, has a really strong role if we’re trying to think about US versus EU comparisons. I believe that if we’re asking about living standards, the picture is reasonably clear. Northwest Europe, at least, has something like 10% lower productivity than the US (maybe), and 20 to 30% lower GDP per capita. This is overwhelmingly because Europeans take vacations and Americans do not. And that’s nothing to get really worried about. And the difference in productivity growth rates, again, I think that’s largely just because of differences in industrial mix that don’t necessarily matter for living standards.
But I’ve done an exercise. I’m not putting it up here, but some of you may have seen it. I have been comparing two countries, California and not-California. Just divide the United States into California and the rest. Not-California looks a lot like Europe in terms of productivity growth! Basically when we talk about rapid US technological progress, it’s California. Yes, there’s also Seattle and a few other places. But to a remarkable extent, we really are talking about just part of California. We’re really talking about Silicon Valley.
I’m a huge admirer of the Draghi Report. I’m not a little worried about some of the productivity stuff, but I think the recommendations are almost all things that should happen. He does talk about capital and one thing Europe definitely does lack is a venture capital sector comparable to the United States. But where is the venture capital sector in the United States? It’s not just Silicon Valley, it’s Sand Hill Road. It’s probably just a square kilometer or so of the United States, which is where all of that stuff is.
Why is that relevant? Texas is actually a little bit above the US national average in productivity growth. But it’s a lot lower than California. Does Texas worry that California might weaponize its dominance of advanced technology by cutting Texas off from access to the latest AI model? (That might be doing them a favor given where we are right now.) But anyway, does Texas worry that California might cut them off from access to critical technology? The answer is of course not, because we’re a country. We have a specific clause in the Constitution that prohibits doing anything like that.
Should Europe worry that a future or even the current US administration might try to cut off European access to critical technology? Well, it’s not something you can discount. And so there is a power element there. And I’m not sure at all that trying to calculate elasticities, much as we want to do that, is going to get you the answer to all of that. So, there are issues.
One thing that we also don’t know comes up when we talk about how much the European lag in the most advanced technology creates a vulnerability to some kind of weaponized interdependence. It certainly does, but is that unique? How many other things are there? Are there European products that the United States depends upon that we would be in real trouble if that access were cut off? And by the way, it’s just insane to be even talking this way but that’s the world we’re living in.
Silly examples are part of what just makes all of this tolerable. I actually have several shelves in my pantry full of boxes of Italian pasta. And the reason is, there was a point last year when President Trump threatened to impose 100% tariffs on Italian pasta, which I would have been willing to pay, but I knew that what would happen was not that I would have to pay double, but that it would disappear from the shelves. And if there’s one thing which is really not a substitute, it is American pasta. So, you know, that’s the reality. But are there more significant things? You know, everybody talks about ASML. The moat around that technology may not be as deep as people have been assuming, but are there other examples? The answer is I don’t know. Call for further research is what we always do here, right? But I think these things are really important.
So that’s US versus Europe. What about China’s ability to weaponize its position in the world economy? What about weaponized interdependence coming from China? I unfortunately did not have access to the Yang and Liu paper when I was putting this together. But just looking at world trade. This is netting out intra-EU exports. So if we ask, who is the superpower in world trade? It’s actually here. It’s actually the EU, not the Chinese. And we know about the flood of Chinese goods, we know about the surplus. But the reality is that the United States is a quite distant third and that the EU is actually bigger.
This is from the Mercator Institute Center for China Studies. I don’t know them, but the question I want to ask, regarding U.S./E.U. interdependence, they don’t ask. They only look at dependence on China and they have a rather arbitrary measure of dependency on imports from China. Both the two solid lines are respectively US and EU dependence on Chinese exports. And they net it out. I guess they’ve decided that access to Chinese pajamas is not critical. Although one thing I happen to know just from the economic geography stuff is that most of the world’s buttons are produced in Qiaotou, China. I’m not sure, but it’s possible that if the world loses its supply of buttons, that might actually be a bigger deal than people realize.
But anyway, it’s worth looking at Figure B in the Yang-Liu paper from yesterday. They don’t do a European aggregate, but their figure does in fact appear to show the same result: the Western economies are increasingly quite dependent upon China, and if anything, the United States is more dependent than the EU. I don’t know quite how seriously to take this. I was reassured to see that people whose methodology I really understood were coming to roughly the same conclusion. And it would be interesting to delve into the details of all of that.
But let me not pursue that further, except to say that it’s not as obvious as you might think. Who is vulnerable here? And part of the answer may be that we’re all vulnerable, although apparently the Chinese less than the rest of us. The three superpowers are all quite vulnerable to weaponized interdependence.
Okay, let me get back to my starting point. Here we are in a world where the threat of economic damage from somebody cutting off your trade is clearly very real. There are two questions that you might want to ask about this. One of them is: why? Why would you do such a thing? And every paper I’ve seen so far makes a really excellent choice, which is to not try and answer that question. I mean, I’m obsessed with Canada—U.S. right now. And if you try to ask me, what is this about? That’s the end of your next month or so, because we have no idea what that’s about.
The reality, which Albert Hirschman could have told us 80 years ago, is that it does happen. Countries, governments do have objectives. There’s some old economics, international trade literature about non-economic objectives and how best to serve them, which doesn’t really quite get at this exactly. But at least people made the right choice, which is to say, look, that’s not a question we need to answer right away. We need to ask, what can you do, assuming that your government has those objectives? But that is a big question, obviously. If we want to ask why the world changed so much two years ago, it was not because there was a real change in objective conditions that meant that countries should have different objectives. It’s just that countries did have different objectives, and there they are.
So that’s the strategic choice, but at some level, eventually, we do need to try to understand what’s driving this. And that’s where, if I can make a plea, I do think we need to bring the international relations people into these discussions. Yeah, they don’t do quantitative models. They have an annoying addiction to 2 x 2 matrices, but they are smart. They’ve thought about it a lot and have actually picked up some aspects of international interdependence that I think economists have neglected.
The other question, which is really critical in a lot of these issues now arising, is whether we can measure how much pain is created by disruption of trade. We can measure what happens to the economies when a country weaponizes interdependence. But how much do countries care? What is the relative willingness to bear pain? With U.S. vs. Canada, I don’t really know whether we can measure the dollar value of the costs of trade war. It might be less asymmetric than people think. It might actually be almost comparable, at least in terms of dollar losses, but of course as a share of GDP, it will be enormously higher for Canada. On the other hand, the Canadians kind of know why they’re doing what they’re doing. And certainly the great bulk of the U.S. public has no idea why we’re doing what we’re doing.
Obviously, if you’re also trying to figure out how, if ever, this Persian Gulf crisis ends, there’s no question that Iran is suffering an enormous amount of pain. How much are they willing to do that? For better comprehension, we actually need political scientists, probably country expert political scientists to try and assess that. So world power is not simply a matter of the amount of damage you can do, but the amount of damage that you’re willing to accept. And we don’t have a real answer to that.
Last points.
So we do know now that international interdependence is a potential tool of coercion. It can be used by countries to pursue objectives, whatever they are. They have geopolitical objectives. They can use the threat or the actuality of cutting off trade to create pain for other countries in pursuit of those objectives. First of all, they have an incentive to try to ex ante, alter the calculus, to invest in sectors that make them less dependent so that they’re less vulnerable to this kind of coercion. But conversely, to try to shape their trade in ways that make other countries more dependent.
And that’s where Hirschman came in in 1945. By the way, the book [National Power and the Structure of Foreign Trade] was actually written in ‘42. It’s just amazing to think of him being able to do that thinking just after escaping the Nazis and arriving in the U.S. But at least his claim was that the Nazis deliberately created as much economic dependence in Southeastern Europe as they could. So that is something that goes beyond this kind of picture.
The other is, well, what if the coercion isn’t enough? Or if it’s too much? Now, if weaponized interdependence isn’t getting you where you want to go, well, the next step is weapons. And if I go back to my first slide—that’s showing the war that happens when the players are not prepared to accept the results of weaponized interdependence.
I hope there’s somebody who actually knows what’s going on here, but what’s happening right now is that there is some oil flowing through the Strait of Hormuz. The counter blockade is inflicting a lot of pain. So what do you do if you are the IRGC, the Revolutionary Guard? Well, one answer is you encourage your allies to expand the war and you start doing drone attacks on the pipelines that bypass the Strait.
And so, ultimately, this is all really interesting and frightening. But while I worry a lot about world trade being distorted, impeded, all the duplication of effort, the loss of gains from trade due to people trying to avoid weaponized interdependence, the biggest thing, of course, is that once you start turning trade relations into a source of geopolitical power, you might find yourself starting to do other things in pursuit of geopolitical power.
On that happy note, thank you all.
Philip R. Lane (Member of the Executive Board of European Central Bank): All right. Thank you, Paul, for that wide ranging and upbeat presentation. I’m sure there’s going to be plenty of questions and comments from the floor. And rather than having them one at a time, I’m going to call on a few questioners to get us started and see where we go from there.
Conference Attendee 1 (Steve): Thanks very much. Fantastic lecture. One of the points you made is that exercising due economic power involves self-harm because you’re foregoing mutual gains from trade. I wondered about the sort of income distributional effects of trade underlying that. And so one reason why, what might be driving some of these changes is a sort of perception that GDP per capita has grown very robustly since the 1970s, but the real wage of the median worker has not because of rising inequality. That may be mainly driven by technology rather than trade. Yet that could provide one reason why not just in the US where we see populist pressure, but we also see it in Britain with Nigel Farage, we also see it in France with Marine Le Pen. Obviously there are many things happening there, but I wonder to what extent this rise of geo-economics is also related to the income distributional effects of trade and technology.
Conference Attendee 2 (Arnaud): I was actually curious about the discussion of different measures of economic size, purchasing power parity and real GDP. So one way to think about it is: you may be interested in output or you may be interested in consumption and the difference between the two is trade. And so I was curious about whether, for the mechanism you were describing, we should see something in the national statistics in terms of trade for Europe that should be improving. That’s the other part of the story.
Conference Attendee 3 (Kevin): I guess a question is should we be worried about dependencies? And it strikes me that if we don’t get the buttons anymore, we probably will figure out how to make the buttons, you know? And maybe people will figure out how to make the Dutch machines if the Dutch machines are cut off. So I’m wondering whether the criterion for really worrying about it is, is the stuff that’s being cut off stuff that you would need in the event of military conflict? Because by the time you figured out a substitute, it may be too late. You know, and that’s maybe one distinction that is useful.
Paul Krugman: Yeah, okay. First, obviously, the income distribution issues are, I mean, they’re huge in general when talking about international trade. However, it’s not at all clear to me that they’re playing a role in all of this. When it looked as if the trade war was largely directed at China, you could make a case that this was about protecting wages. But now it’s not, and of course that’s partly because the Chinese effectively weaponized rare earths. But I don’t see that there’s any plausible income distribution story in the currently hot trade conflicts between the United States and countries at similar levels of GDP per capita. It’s really hard to tell a Hechscher-Ohlin story about that. And I think that’s enormously important. And I do think that international trade economists were sort of alerted to and hip to distributional issues long before most of the rest of the economics profession. But I don’t think that’s the story here.
That was a very good question about the terms of trade. And I have been trying to figure out if there’s any way to measure it. It is really, really hard. The conventional terms of trade measures just won’t capture what we’re trying to get here. And you don’t see it (being modeled). Mostly what we have for terms of trade either come from the statistical agencies doing some calculations, but the same hedonic stuff applies. In fact, the first time I realized how crippling uncertainty about hedonic adjustments could be was precisely on import prices in the 1980s. So, it should be there, but I’m not sure that we know how to measure it. But I mean, that’s definitely a clearly implied counterpart of the story and I didn’t present any numbers on it because I don’t have any.
With military issues, if Hirschman was writing his book in 1942, for him to be even talking about trade as a source of power in 1942 required a heroic effort to not think about the Battle of Stalingrad. And, if you’d asked me that question a year or two ago, maybe three years ago, I might have actually said that it is probably purchasing power parity GDP, which is kind of like general productive capacity. But given where war is now, you know, with the war that’s raging not too far away from here right now, a large part of the drones being used by both sides have Chinese components. You might want to think about manufacturing capacity in general and maybe specifically certain kinds of capacity.
Back at the beginning of World War II, people were talking a lot about steel production. And GDP barely existed as a concept, I think. But you could compare steel production. There were some people on the Axis side who said, “Look at how big America’s steel production is. We can’t win this thing.” Now I don’t know what the measure is, but it’s not gonna be the obvious measure from the past.
Conference Attendee 4 (Christophe): I wonder what you think Europe should do in this geoeconomic world. You said using geoeconomic tools can increase the risk of military conflicts. Others view it differently and say it’s maybe a substitute to actual war. So I wonder in this world when the US and China use geoeconomic tools, what should Europe do? What’s the right balance of preventive efforts or should it actively use these tools? And what’s the balance between the military investments or investing dollars, and kind of sanctions and losses connected to that.
Conference Attendee 5 (Ilha): Okay, thanks. Fascinating lecture. I was wondering about the different countries, countries with different political systems. How does that affect their willingness to use economic coercion and also their sort of willingness to endure pain? I’m thinking about China, but also some other countries as well.
Conference Attendee 6 (Luc): So thank you for this terrific and terrifying talk. You spoke mostly about trading of goods. We would expand that to financial trade, FDI. Do you mostly see parallels with what you said today, or are there important differences?
Krugman: OK. Let me actually respond to Luc. I had my Road to Damascus moment on all of this stuff from the field of international relations, from Farrell and Newman on weaponized interdependence. Although it’s funny. Their book is brilliant and exposed all kinds of things. But in 2019 when they were writing about it, they were mostly worried about the United States abusing its power. And it’s turned out that, well, yeah, that can happen, but it’s by no means unique. They were very much focused on the financial role. It turns out that, in ways that I had not realized, and I should know these things, but the centrality of the dollar also means that there’s a centrality of the U.S. banking system, because almost everything ends up having to run through U.S. banks at some point. In fact, if you were trying to tell a story of how we got to where we are right now, a lot of it begins with the U.S. attempting to weaponize the role of the dollar against Iran. Then one thing leads to another.
So that’s very much part of it. And again, by the way, these things are a wasting asset. The alternative payment systems, I don’t think we’re anywhere close to seeing the role of the dollar displaced as the world’s key currency, but the availability of a variety of payment systems as workarounds, plus crypto, and Chinese systems and so on, develops over time. It’s fundamentally the same thing. I think the International Relations people are thinking too much about networks and not enough about just plain trade disruption, but they certainly have a point there. That’s definitely part of it. By the way, the odd thing is that it also creates some vulnerabilities. If you want to imagine that we go completely mad and have a US-European economic conflict, if you want to think about the amount of direct investment by US corporations in Europe, you know, we have given the EU hostages here.
So now, what should Europe be doing? I would certainly not be advising the EU to try to enhance its ability to weaponize its position. I don’t think that this is that kind of continental society, but reduce its vulnerability, yes. I mean, it’s odd: Given where we are now, I hate to talk about the United States as a role model, but we have or were doing some things right. The U.S. was pursuing a fairly significant industrial policy until two years ago with the CHIPS Act and all of that; trying to create not full self-sufficiency, but reduce its strategic dependence on electronics, semiconductors from China and other potentially hostile powers. And there really should be a European equivalent to the CHIPS Act. Now, of course, we have killed that act ourselves. So it’s one of those funny things.
And this is what I think is really interesting and ironic. In terms of trade policy—I don’t know how much people know, though probably most people in this room do—the United States has a system of basically quick response trade policy tools that instead of alphabet soup, it’s number soup. It’s section 201s and section 232s and section 301s and so on. You ask what Act are those sections of, the answer is they’re all different Acts, which makes it even worse. But we have these tools which enable quick responses to market disruption by floods of imports. That’s a 201. Unfair foreign practices, that’s a 301. National security, very open-ended. That’s a section 232, which is a good system. It’s designed so you don’t spend years debating about how to respond to a potentially weaponized use of interdependence. You can act quickly.
Unfortunately, the way the system was set up was that it pretty much vested, allegedly temporary but still more or less unilateral power in the executive branch, all of which based upon the assumption that the president would have a wider view than Congress and take a responsible position and be concerned with maintaining good relations with our allies. So it turns out that everything has a flaw and a system which was extremely effective in the past was almost custom designed to empower the current administration to do a lot of really destructive stuff.
Anyone who’s interested in this stuff I hope has or is about to read Chad Bown and Soumaya Keynes on How to Win a Trade War. And it is fascinating to me that Chad and Soumaya, who are very much free trade pro-globalization types, are practically screaming that Europe needs its equivalent of 201s, 232s, and 301s. That Europe does not have a process for responding to these threats. You don’t need to put current US administration officials in charge, but you need to have some kind of system. The sluggishness of the response to China shock 2.0 and all of that really shows you need more.
I guess maybe it comes down to this: Europe should be more like what the United States is supposed to be. There should be effective precautionary industrial policy. There should be effective precautionary trade policies. I don’t know how that will work.
But then the political systems. That’s a really kind of interesting question, right? Because you would normally think authoritarian regimes don’t care and they can get away with anything. I don’t think that’s what history tells us. I think what history actually says is that democratic systems, for good reason, are if anything often more robust, more willing to accept the sacrifices involved in holding their place in geopolitics. There’s a lot of back and forth, but if we ask what happened during World War II, the mobilization was more thorough earlier in the democratic nations than it was in the Axis powers. If we look at the obvious extreme, there’s the hesitance of the Putin regime to go for full mobilization. It’s actually suggesting that, in contrast with democratic consensus, authoritarian regimes have no dissent but are more fragile or perceive themselves to be more fragile than you think. I think on balance, it turns out that democracies are frustrating and they debate and they sometimes are slow, but they are actually much more effective in conflict than people imagine.
Philip Lane: Thank you, Paul. I suppose I have a question which maybe connects the trade and the macro. So, is it clear who are more vulnerable, people running chronic current-count surpluses or chronic current-count deficits? Or does that margin make a difference?
Krugman: That’s a good question. And I would say actually, if we think there’s mostly imports, then it would be the deficit countries that are more vulnerable. But it’s going to depend upon the goods involved, not at all for the reasons that the mercantilists in my country think. Actually, I may be turning the argument around, but look, on the macro side, the stimulus effects of eliminating current account imbalances are basically not relevant now. You know, in 2010 with the world depressed and all kinds of constraints preventing us from doing enough to boost aggregate demand, that mattered. But it’s not at all relevant now. And there’s plenty of room to cut interest rates if necessary. So those are not relevant.
The United States or its current leadership thinks that we’re a winner in a trade war because we can eliminate our trade deficits and that’s good. But the reality is because we are so dependent upon a lot of imports in a number of sectors, we’re actually the losers in the trade war. I think that was a completely nonlinear response, but I hope that someone can ask Claude to disentangle it.
Okay, very good. So we’re at the end of the session. I’m just gonna pause in case someone indeed has a devastating counter question. Ok, Luke.
Conference Attendee 6 (Luc): So you showed us what we will soon as part of Dune 3, I guess in December. This is where you started.
Krugman: No, that was actually a still from Dune 2.
Luc: Yeah, but I understand that it’s coming. So how likely is it that this is coming in the world today? What probability do you put on it?
Krugman: Wow, I mean... Right, no, it’s an interesting question. Look, there’s the catastrophe that has happened, which has turned out to be, in the micro sense, worse than almost anyone imagined, which is the US military collapse in the Middle East. But in terms of the economic consequences, things have been actually much more muted than people feared. We’re actually having an oil shock that by the numbers is worse, at least relative to world oil supply, than in ‘73 or ‘79. But no global recession. So that one, I’m maybe excessively complacent, although check back in again in a couple of weeks.
The one that scares me is Taiwan. And now you would hope, clearly, it would be completely stupid and counter to Chinese interests to start a war over Taiwan. You know, there’s no economic gain. It would only hurt them, definitely. On the other hand, if there’s one thing we’ve learned, if we didn’t know this before the last few years, it’s that governments sometimes do stupid things. And we do know in particular that irredentism, like Italy wanting Trieste back was a completely crazy reason to get involved in World War I and yet... There it was.
So the answer is that there is a nightmare scenario and it isn’t even exactly weaponized interdependence, but because of all the irrational kinds of stuff that Norman Angell told us was outmoded in 1909, because it’s still out there and it will lead to major disruptions. And given time, we can reconstruct a semiconductor supply chain without Taiwan. But my God, the few years it would take to do that would be very, very ugly.
—END—
I am a tree person, a know nothing about economics-let alone international politics.
But I am wondering when climate change enters into the equation because I see Mother Nature holding the trump card (no pun intended).
And is China is having rogue AI issues?
Thank you for letting us into your world Dr. Krugman
The Last Days (of Democracy)
The following is not intended to be a Bible verse ear thumping. It’s intended to apply ancient wisdom to our current situation.
The Last Days (of Democracy)
Second Timothy, chapter 3, 1-5
1But mark this: There will be terrible times in the last days.
2People will be lovers of themselves, lovers of money, boastful, proud, abusive, disobedient to their parents, ungrateful, unholy,
3without love, unforgiving, slanderous, without self-control, brutal, not lovers of the good,
4treacherous, rash, conceited, lovers of pleasure rather than lovers of God—
5having a form of godliness but denying its power. Have nothing to do with such people.
From Peaceful Solution:
We are in the middle of a five alarm fire, and while most of the country remains oblivious, too many in the resistance are wasting time discussing how to organize and implement a fire department. What’s needed from the general public is total non cooperation and a general Spending Strike. NOW!
Spending Strike. Stop spending except on essentials. We, the People, are funding the fascist takeover of America.
For more information on how to accomplish this, visit my profile page. Or:
https://rpmcmurphy1.substack.com/p/how-to-stop-his-royal-heinous-7d6
Please: Resist. Stop cooperating. Stop spending. Stop the fascist takeover.
Thank you for restacking, and sharing, to help Spread the Word!