C. Peter McColough Series on International Economics With Lawrence H. Summers

Monday, June 24, 2024
Jonathan Ernst/REUTERS

Charles W. Eliot University Professor and President Emeritus, Harvard University; Former U.S. Secretary of the Treasury (1999-2001); CFR Member 


President, Council on Foreign Relations

The C. Peter McColough Series on International Economics brings the world's foremost economic policymakers and scholars to address members on current topics in international economics and U.S. monetary policy. This meeting series is presented by RealEcon: Reimagining American Economic Leadership, a CFR initiative of the Maurice R. Greenberg Center for Geoeconomic Studies.

FROMAN: Well, welcome, everybody. My name is Mike Froman, president of the Council, and it’s great to see you all here. In addition to the 150 or so we have here, we have actually 500 members online, which is—which is terrific, a real tribute to our speaker today who you all know, who needs no introduction. You have his full biography. I won’t go into it. But welcome, Larry. It’s great to have you, and we really appreciate you doing this. 

As we were discussing in the other—in the other room, we launched an initiative here two months ago called RealEcon: Reimagining American Economic Leadership. It’s focused on trade and investment, development, economic security—the convergence of national security and economics. And I’d love to dig into some of those pillars with you as we go through this conversation. 

But why don’t we start first with just a basic economic outlook. A couple of years ago you were highly critical of the Fed, feeling they had started too late; talked about how hard it is to have a soft landing when you already have inflation so high and unemployment so high. We’re still growing. Inflation’s coming down—not down to 2 percent yet, but down. We haven’t had a hard landing. The Fed has resisted calls for cuts. What’s happened over the last couple years? What’s gone better than you expected? And how do you explain what appears to be sort of a historic anomaly? 

SUMMERS: We haven’t had a hard landing yet, and we have the American people enraged over the phenomenon of inflation and what it’s meant for living standards. So I don’t take back my sense that they erred badly in 2021 and early 2022. I would stand with that judgment. 

I think that it’s not clear where we are going. My best guess is that they are badly wrong to think that the neutral interest rate is 2 ½ percent—that, given the magnitude of budget deficits, climate investments, resilience investments; given that the economy seems to be going more or less in line with potential at 5 ½ percent fed funds rate; my guess is that the neutral rate is 4 ½ percent, and so monetary policy is not all that epically restrictive. I thought it was kind of crazy when the market and the Fed thought a few months ago that there were going to be six or seven cuts in 2024, and the market and the Fed have adjusted to thinking there are going to be one or two, and it’s harder to object strongly to that. My guess still would be that there will be somewhat less, and that the future level of interest rates will be higher than the market generally supposes. 

The neutral interest rate is 4 ½ percent, and you have a normal term premium to the ten-year of 120 basis points. That’s a ten-year of 5.7 (percent). People think of that as a quite unlikely event. I don’t think it’s nearly as unlikely as people do, given the magnitude of our fiscal challenges. I think there’s—in an environment like this, an elevated risk of accident, and I think there’s a bit of excessive optimism about inflation. 

What people forget is that if the theory is—and this was always my theory—that we were having excess demand-caused inflation, and there were some sectors where there were bottlenecks, and in those sectors prices were flying up—and so there was transitory inflation, and that’s why people saw numbers like 7 (percent)—the prices that fly up ultimately mean revert and come back down. And while that’s happening, the underlying inflation—the inflation figures flatter the underlying inflation rate, and I think there’s been some of that going on. 

So I think we probably never were an underlying inflation economy much above 4 ½ to 5 (percent), and I think we’re probably still a 3-ish percent underlying inflation economy. So we’ve made progress, but the sense that sort of we had a big problem and we’ve 70—we’ve 80 percent solved it I think is way off because we never were in the sevens and eights, and we’re not really in the twos right now. 

So I think it’s a complicated path for the Fed. I think so far the Fed has made a variety of mistakes. But to their credit, they’ve kind of just-in-time corrected them and they’ve been less bad in their cumulative effect than I might have expected. That was true with respect to staying easy for too long. That is true with respect to what they were saying last year. It remains to be seen—you know, the world ought to be more observant of the fact that on a mark-to-market basis or on a future—present value of future losses of transfers that otherwise would have been to the government, they’ve lost between half-a-trillion and a trillion dollars on unnecessary QE operations, and that’s a pretty consequential bit of unsuccessful speculation. 

FROMAN: Let’s maybe talk a little bit about sort of Bidenomics for a minute. National Security Advisor Jake Sullivan gave a speech a year ago April at another think tank where he laid out a policy of industrial policy, protectionism, restrictions on exports, restrictions on foreign investment. He repudiated really, in some ways, eighty years of theory and practice of neoliberalism, and in some camps people blame neoliberalism for the rise of populism and, here in the U.S., the rise of President Trump. What’s your perspective—first of all, do you agree a hundred percent with everything that Jake Sullivan said? What was your reaction to it? And what’s your view on the relationship between neoliberalism and populism? 

SUMMERS: A lot of parts to that question. 

I think most of the—most sentences that include the word “neo” are—(laughter)—tendentious attacks on something that the speaker doesn’t understand very well. (Laughter.) And that’s true whether it’s neoliberalism being attacked or neoconservatism being attacked. It’s just “neo” is a kind of intellectual bit of hostile language that is almost completely content-free. (Laughter.) It is content—it is the first thing—is the first thing that I would say. 

Second thing I would say is, look— 

FROMAN: Going to strike “neo” from the—from the style guide at the CFR. (Laughter.) 

SUMMERS: Second thing I’d say is, look, I don’t—I’m not a political scientist, and I don’t think the political scientists have it figured out, but first approximation, France has the most welfarist welfare state of any major country and in some ways has more nutty right-wing populism going than we do in the United States. It’s present in the Netherlands. It’s present in Germany. It’s present in Sweden. This idea that somehow if we just had a bigger, more welfare state with more social insurance then we wouldn’t have populism feels more like a Democratic Party fantasy than a reasonable reading of any kind of international evidence. 

Look, I don’t think that any sensible national government is indifferent to the composition of the goods and services produced in its economy. That’s true because a defense base is crucial to national security. It’s how we won World War II. And it’s a terrible, ominous, concerning thing that industrial capacity is a constraint on how many artillery shells we’re giving Ukraine. It’s a terrible thing that China produces every four months more drones than we and Europe have produced together in our entire history. So a sensible society is concerned about its defense industrial base. 

A sensible society is concerned about being leveraged by others and what the resilience of its economy is. A sensible society is aware that while economists go on at infinite length about externalities, there are also real issues about chicken-and-egg problems: Who’s going to make the first fax machine? If you don’t have any chargers you’re not going to have any electric cars, and if you don’t have any electric cars you’re not going to have any chargers. And so somebody needs to kind of catalyze that kind of thing getting started. So the kind of religious you’re trying to say something about industries therefore you’re being dumb, I don’t think that was ever anybody thoughtful’s position, and that was not my position. 

That said, I think roughly speaking I tune out any industrial policy advocate who uses the word “jobs” in the first third of their advocacy; that in this world, and certainly in the world of the U.S. economy today or the next five to ten years, we’re going to be fully employed, or if we’re not going to be fully employed it’s going to be a byproduct of other mistakes, and so the idea that if we create demand for something there will be more total jobs rather than people doing activity A rather than activity B is just wrong. And so the case for industrial policy that is premised on job creation or that is premised on higher wages is basically not an intellectually very legitimate case. And that, unfortunately, was the main thrust of the Jake Sullivan speech that you asked about. 

And the basic phrase “middle-class foreign policy” is mostly a prescription for error. It mostly pushes you to goofy mercantilism and export promotion. It pushes you to protection of particular kinds of jobs. It systematically steers you away from recognizing indirect kinds of effects. 

So, yes, we should have an—we should have lots of strategy that is about industries. Yes, we should. But the best generals are the ones who hate war the most but recognize that wars sometimes have to be fought, and if you don’t prepare for war you’ll end up fighting many more of them. And the best industrial policy advocates are the ones who are most reluctant but who see what it is that is necessary, and who choose to recognize what economists call the theory of targets and instruments, which is basically you figure out what the problem you’re trying to solve is and you choose the instrument that is best targeted at solving that problem rather than having a maximum set of collateral consequences. 

That’s something I’m proud of having done when we served together in the Obama administration, Mike. It was one of the few times I was bureaucratically shrewd. I had responsibility for overseeing the team that worked on the automobile industry bailout, which was a kind of industrial policy. And before the rest of the government had gotten itself organized, Steve Rattner and his team produced at my behest a set of principles for how we were going to do this, the essential idea of which was our goal was to put the automobile industry through bankruptcy and back to the private sector as rapidly as possible, period. And that meant no sandbox for the EPA’s experiment with electric cars, no sandbox for the Department of Labor’s experiments with worker participation in running plants, no sandbox for the Department of Commerce’s regional development initiatives; just a narrow focus on what was the necessary mission that was responding to the failure of the nonavailability of credit during the financial crisis. And I think that was the right thing to have done, and I think it contributed to the fact that that was a very successful venture that is looked back on as having been a successful venture. 

And when I see the CHIPS Act press release about major announcements highlighting the daycare center initiative, I think to myself: I wish the CHIPS Act had had a similar set of principles. And when the people in the administration explain, no, no, you don’t understand, there’s really a huge chips shortage, huge labor shortage, and you really have to have childcare to get workers, I say: Yeah, that might be right, and if you give money to Intel and that is right, Intel will figure out that a good way to recruit labor is to have childcare, and they’re probably better at making that judgment about their recruitment of labor than a bunch of bureaucrats in the Commerce Department who have just been lobbied by a set of Democratic interest groups. 

So the answer is, yeah, I’m for some of this. I’m worried about the defense industrial base and I’m worried about resilience. I resonate to arguments about chicken-and-egg problems. But I think you got to have some limiting principles or you get to some very suspect places. 

And I think people have a huge tendency—you know, just because somebody’s a hyper-conservative economist doesn’t mean that everything they say is wrong. And Gordon Tullock, who was way to the right as an economist, wrote what I thought was a brilliant paper many years ago on what he called the transitional gains trap, a point of which was there’s certain things that when you start them become very hard to stop. You know, a simple example from the old days was the tax deduction for three-martini lunches. What a dumb provision; like, we really shouldn’t have it. But once you put it in place, think about what happens. The people who lobbied for it were not the people who drank the three-martini lunches; the people who lobbied for it were all the people who build restaurants to serve those people—who weren’t earning excess profits; lots of people entered the restaurant industry, and you were going to be imposing a big loss on them and destroying their business if you changed the rules. And so when you started it got very hard to stop. Similarly, pay-as-you-go Social Security. Once you’ve started it, there’s a generation at the beginning that got benefits and never paid taxes. But if you want to end it, there has to be a generation that pays taxes and never gets benefits. And that’s kind of impossible. And so you have to think about the fact that lots of these things are much easier to enter than they are to end. 

The classic example? The Jones Act. The Jones Act was the resilience industrial policy of the 1920s. You read the debates about it, it was all about we’d be dependent, we’d be vulnerable, we’d be this, we’d be that if we didn’t have our own merchant marine, and the only way to have our merchant marine—our own merchant marine was to have a captive market. And there may have been some logic to that in the 1920s. I think it could be debated, but there is some logic. But surely the cumulative hundred-year consequence of the Jones Act has been higher fuel prices for New England, less availability of supplies to the Caribbean, and lots of craziness. And I don’t think we’re proceeding with a full awareness of those kinds of costs. 

So, yes, industrial policy, but not supported by jobs-creating mercantilist arguments; done with reluctance; and done with careful thought to the public-choice aspects about what will still be around after the original motivating problem has passed. 

FROMAN: Well, let me—let me push you a little bit on this, Larry, because so much of our policy the last eight years and so much of the economics profession has been rooted in efficiency and the goals of efficiency, zip in consumer welfare. That’s one important value, but there are other important values like resilience, redundancy, diversification, national security. And how do we think about—maybe it’s a question of what are the limiting principles. What are we—what is the economic theory that helps us put guardrails around the tradeoffs that need to be made between efficiency on one hand—because we’ve seen efficiency just—if that’s the only value, then you may sacrifice secure supply chains, or you may sacrifice national security. How do we—and that’s not the right outcome for society either. How do we figure out that balance? And what are the guardrails that one can put around it? 

SUMMERS: I am—look, I think it’s very hard to formulate a rigorous cost-benefit analysis that shadow-prices—which is the economic term—national security or resilience. But I think that it’s entirely reasonable for policymakers with judgment to weigh those kinds of considerations. 

I was an enthusiastic supporter of the automobile bailout. There was a moment during the automobile bailout when the question was presented, what about the rest of the industrial base in all of those places, in the places that were most affected, and their relationship to the automobile companies. And maybe we needed to make this available to everybody. And I was—I drove a decision that we would do that insofar as it was necessary to maintain a viable automobile industry, but we would not do it for its own sake, despite the fact that there were jobs arguments in the short run for doing so. 

So I think it’s basically a matter of what advocacy arguments you consider legitimate, and what advocacy arguments you consider illegitimate. And almost any set of arguments that are about national security, or about national security broadly defined to include the insecurity of non-resilience—you know, people will make judgments, and you have to make judgments. I may not agree with every judgment that’s reached, but I think almost any judgment that’s reached is likely to be a reasonable one.  

I think when the policies are advocated by arguing that they are going to create millions of jobs, and that because we have them there will be millions more jobs, or the policies are advocated without consideration of the consequences for prices, and without an awareness that real wages are wages divided by prices, and that when you push prices up by stopping people from getting access to lower suppliers, then you are likely to get bad decisions. So I want to be heard as not opposing industrial policy, but as suggesting an alternative grammar of argument for pondering the various ideas that are developed, and I think a rather different grammar of argument than the Biden administration has often used. 

Now, look, I’m sort of aware that sometimes the best arguments in the political domain aren’t the best arguments substantively, and so you have to allow some slack, but I worry that a lot of the new enthusiasts for industrial policy—I don’t think the Roosevelt Institute is thinking about this like I’m thinking about it and saying that they have to talk about jobs in order to make the case in the political domain. I think they believe a set of free-lunch economic propositions about creating more demand for particular sectors will create more total jobs that is a kind of progressive Lafferism—(laughter)—in the sense of things will pay for themselves and it will all work out. And so I’ve seen my place as being to resist that. 

But look, I was very strongly opposed—because I thought it was way excessive—to the scale of the original Biden stimulus; but I supported the CHIPS Act, I supported the IRA—though I would have changed, in both cases, some of the details. And I supported the infrastructure bill but, unlike some, I think they were considerably better in the form they were passed in than they were in the form they were originally proposed and advocated in, which would have been much larger, and more sweeping, and more interventionist. So I think it's about changing the grammar of the argument that you like rather than about ruling any kind of industrial policy off limits, and I think—you know, I thought it was a low moment of the Clinton administration when Ron Brown had a tense moment in the Mideast peace process, called Yasser Arafat to get him to buy AT&T for the phone system of the Palestinian areas. I thought that was a kind of reductio ad absurdum of a certain kind of use of U.S. diplomacy to serve economic objectives in support of the middle class. And I think when you start getting going on middle-class foreign policy, you have a tendency to get to that place. And so I have always been much more—much less comfortable than most of my friends with the phrase, “commercial diplomacy,” which is kind of about using U.S. leverage, you know, to get private equity firms—American private equity firms the deals they want in foreign countries relative to foreign private equity firms the deals in what they want. 

FROMAN: Let me touch on two issues quickly before we open it up—one on universities and one on AI. You’ve been pretty outspoken on what’s going on at Harvard and other universities. What is your guidance, recommendations to university presidents and administrations in how they deal with the environment on universities today? 

SUMMERS: Look, I think our universities are a staggering national asset, and the fact that people want to come from all over the world to them is an enormous advantage. When I was president of Harvard, I always used to tell the story of Yakovlev, who was Gorbachev’s ideological adviser and who formulated the concepts of perestroika and glasnost, and was asked where he got the idea for all of that, and said it was from his visit to Columbia in some exchange program in the late 1950s. And I tell that story, and I say that even though it has been subsequently pointed out to me that there were seven or eight people in that exchange program, and at least five of them were in the KGB. (Laughter.) And even so, I think we were ahead for having done that. So I think our universities are a huge and precious asset. 

That said, I think they are excessively run by their faculties who, for a variety of reasons, have developed a set of ideologies, and in simple terms, our universities are too much about social justice and too little about truth; too much about identity politics and too little about opportunity; too much about mutual esteem and insufficiently about excellence. 

At Harvard University, the most common grade by far—it’s given twice as commonly as any other single grade—is straight A. And the average grade point average is 3.75. There is one grade for the top half of students in a class, and nine grades for the bottom—uh-huh—half of the students, and I don’t see how you can be celebrating excellence rather than just promoting mutual self-esteem and doing that.  

And I think we have privileged comfort over controversy and vigorous argument, and shuttle off large parts of discussion. And I think we are paying a price for that, and I think the problems around anti-Semitism, which really shock me in terms of how pervasive it has seemed to be, are much more manifestations of the things I just talked about than as thought of as a separate problem. And I think it is up to trustees and strong leaders to reset the course because I don’t think it’s the right thing in a free society for government to start trying to regulate the content of what happens in universities, and I hope they will, but I’m not—and I think over time it will happen, but I’m not hugely encouraged by where things are right now. 

FROMAN: Last question—you are on the board of OpenAI. You are right at the center of debate between innovation and safety. How big of a deal is AI—is it likely to be for our economy, productivity? 

SUMMERS: There are the people who—there are people who know they don’t know, and there are the people who don’t know they don’t know, and I’m in the first category. I know I don’t know. 

That said, here’s a very crude, ninety-second version of the economic history of humanity. From the beginning of people until 3000 BC, there was no trend in standards of living. From 3000 BC to 1500, standards of living rose by an average of three or four basis points a year, so 3 percent a century. From 1500 to 1800, they rose by twenty or twenty-five basis points a century, and from some point in the 1800s till now, at the cutting edge, they’ve risen at 200 basis points a year. So you have seen a number of discontinuous movements of us stepping up.  

Here’s what’s potentially special about AI. The wheel was great, but if you had a wheel, it didn’t make more better wheels. Electricity was phenomenal, but if you had electricity, it didn’t make more better electricity. There’s the possibility that AI will make more better smarter AI. So I think that there’s a 50 percent chance that history books in 2300, when they write about the period from 2020 to 2040, will treat everything that happened in Ukraine, everything that happened in the Middle East, and even what happened in China as subsidiary stories relative to the groundwork that was being laid for artificial intelligence and the way in which that was changing traditional notions of scarcity. 

I don’t think it’s going to happen in the next three years. I think that the people who think that we’re going to—that the economy is all going to get bailed out in a cyclical sense by AI are probably wrong. Sebastian is the big expert, and he can correct me, but Greenspan was right to recognize the productivity acceleration in 1995, but it actually wasn’t from the internet; it was from a set of IT investments that had been carried on in the ’80s prior to that point. And so I think the idea that AI is going to produce seismic productivity and progress in the next few years is unlikely, but I think over the medium to long term, this is potentially hugely profound, and I think it has very profound impacts for America’s place in the world. And when I think about economics and national security—before I get to any of the stuff about resilience that we were talking about before—there are two other places that I would go. One is across the board technological leadership driving scientific progress because I think—I’ll just state the conclusion and if somebody wants to ask more I can defend it—I think my grandmother, between 1900 and 1974, which was her lifespan, saw three or four times as much fundamental change driven by technology as I have since I was born in 1954. But I think there is a reasonable chance that my granddaughter is going to have a set of experiences more like my grandmother than like I had, and that is, I think, the defining issue of our time. 

And I think the other thing that’s profoundly important for our national security is our capacity to mobilize resources around our national security. And the fact that we are now a country—with everything that’s going on with China, Middle East, Russia, whatever—that we’re now a country that is projecting declining defense spending in real terms over the next five years, and that we are a country where there is a universal, bipartisan agreement that there will be no mobilization of revenue from anyone outside of the top 1 percent of the population. That is what I think is—that and technology are the two aspects of economics that I think are most serious for our national security. 

FROMAN: Well, I think Larry has put a few ideas on the table that might spur some questions—(laughter)—so let’s open it up. Hamid? 

Q: Hamid Beglari, Redbird Capital. Nice to see you, Larry. 

So I want to pick up on the AI theme. As you know better than anyone, many people at the forefront of AI research believe that, in the next one or two generations of AI—five to ten years—the next five to ten years—machines will be able to do things that human—to do 99 percent of the things that human beings can do but better than 99 percent of human beings. There’s a spectrum of opinion, but that’s—and the time—people are notoriously bad at predicting the timing of these things because they are exponential, but let’s take that as a given. 

From the perspective of an economist and the implications this will have on human work and socio-economic policy, how should one be thinking about this? Thank you. 

SUMMERS: Look, in a way, AI is a little bit like the economics of Qatar. What are the economics of a country like Qatar? There is basically immense wealth. It is essentially available for free from selling the natural resources. And so it’s basically possible for everybody to live a really terrific life, a really comfortable, good life without work. But it’s complicated to imagine a society, after all of human experience, where people don’t have vocation, and there is an enormous set of issues around the property rights and the distribution of entitlements around that great wealth creation. And so, potentially, is it better to have huge—a huge wealth sink that will produce a hundred thousand dollars per person of resources for every member of your population, or not to have it? It sure seems like it’s better to have it. But the fact that it’s a high-class problem doesn’t mean it isn’t a problem and doesn’t pose all kinds of challenges.  

And so I think that is the basic way to think about this; that the ability to have huge amounts of output without painful toil is a vast, positive thing, but vast, positive things, when they come quickly especially, have to be managed with a great deal of care and thought, and it is going to be complicated to find the institutions that will—that will do it. And it’s even complicated to know who exactly should be empowered, and who exactly should be heeded. 

You know, I have been learning a certain amount about the world’s experience with things nuclear, and it’s an infinitely complicated subject that I don’t presume to be able to more than scratch the surface. It’s not my impression that if authority had been ceded to all the smartest physicists who gathered in Los Alamos that the outcome for the world would have been particularly favorable. It kind of would have been spreading around the technology for everybody to use, and having some gathering of some version of a more academic version of the Bilderberg Group said to be in charge, and I kind of don’t think that would have worked out very well. 

And so I think we kind of need to recognize that this can’t be left to the AI geniuses. On the other hand, if you had General Groves and you didn’t have Oppenheimer, that probably wouldn’t have been so good, either. And so I think it’s a hugely complicated set of problems, and I don’t think the United States—let alone organizing the world—I don’t think the United States government is remotely set up to deal with these issues in a thoughtful way, you know?  

A question I’ve asked—and it’s an exaggeration, and somebody who is here will correct me and say I’m being narrow-minded and unfair to some initiative that they were involved in, and so I’m just going to stipulate that—I’m just going to stipulate that what I am saying is more colorful than true, but I think it leaves a useful impression. 

In the history of the city of Washington, the anti-trust authorities who deal with big tech companies have never met with the national security authorities who deal with big tech companies, except in the context of a cocktail party. They’ve never met in the context of the U.S. government formulating its policy, and that’s obviously problematic if we’re going to get this kind of stuff right. 

FROMAN: Esther Dyson. 

Q: Thank you. Esther Dyson, Wellville, and also host of PC Forum where you spoke forty years ago with Sheryl Sandberg. 

The question is why did you join the board of OpenAI? How do you hope for Sam Altman to regain the trust of everybody? And how can we stop AI from being misused by predatory business models? 

SUMMERS: I joined because I thought I could make a positive contribution in what was a very complicated situation with respect to a very important institution, and I’ll stay on the board for as long as I continue to—continue to believe that. 

I don’t think this is the—this is not the forum for discussing particular business institutions, much less particular individuals, but I wouldn’t want to associate myself with some of the implications in the way that you phrased your questions. 

Of all the things that I’m worried about having to do with AI—the wrong people using it for the wrong purposes, the capacity to produce fakes of various kinds, and undermine communication and dialogue—I would say predatory business models strike me as being one of the lesser kinds of concerns and, you know, those of us who have responsibilities as fiduciaries of these institutions have obligations to law and to the broader society. And OpenAI, I think, has the quite substantial strength that it has a defined mission to which its board and its management are held responsible that is not the maximization of shareholder value. And I think that’s a very important thing, and I would be surprised if there were not various changes in evolutions over time, but I would be even more surprised if the mission became entirely redefined in terms of maximizing shareholder value. 

Q: (Off mic.) 

FROMAN: We’re going to go on to the next one. Let’s go to the virtual audience for the next question. 

OPERATOR: We will take our next question from Rafael Reif. 

Q: Larry, it has been a pleasure to listen to you. Thank you for doing this. 

Larry, if China were to blockade Taiwan within five years, and if that will limit the U.S. access to TSMC chips, what would be the impact to the U.S. economy? 

SUMMERS: I’ll give you a hint. 

FROMAN: Here’s a crystal ball. 

SUMMERS: It wouldn’t be good. (Laughter.) 

Look, I think—I think it is a(n) immensely problematic issue the degree of the world’s dependence on Taiwan, and chips, and the reason why I’m very supportive of the inclination behind the CHIPS Act, and the reason why my best guess is that we should and we will have a CHIPS Act II within the next three years that will be substantially larger than the first CHIPS Act is that I think the issue you raised—whether it’s a Chinese attack, or whether it’s something else—of vulnerability is very large. And I think—so I think it’s one of the major vulnerabilities that we have. 

That said—and that—what I just said is the main thing I think—but just, as everybody thinks about it, I would say that I have, as a(n) economist, a greater skepticism of all arguments of the form we have single points of dependence, we will run out of things, and we are immensely vulnerable. And it will be the end of the world because things depend on a supply of X. 

And I would cite as evidence for that kind of view three different kinds of experience. One, how wrong all the limits to growth stuff from the 1970s turned out to be.  

Two, it has been, and we—there have been problems with it, difficulties in the enforcement. There were lots of things wrong. But it has been the mission of the United States to deny critical inputs to the Soviet Union, to Russia, for the last N years, to reduce the ruble to rubble, and that has been like the major objective of U.S. and European policy, and yeah, they haven’t had everybody who is involved, but you kind of would have thought that when we cut off the export of everything critical because it was hugely important for us because they had become our sworn enemy, that we could have done more damage to their economy than the Asian financial crisis did to the economy of Korea. You would have thought that, and I actually would have thought that. But I was wrong—but I would have been wrong. 

And the third example I would give—and this is a more controversial example, and Adam Tooze would quibble with what I’m—maybe more than quibble with what I’m about to say, but it was a major piece of American policy during the Second World War to do strategic bombing of Germany, and a huge amount of thought went into ideas like ball bearings are central to manufacturing. There were only three ball bearing plants in Germany. If we can knock them out, we can destroy their industrial policy. And we had a ton of ideas like that, and that was a central mission of the Air Force. 

I have two connections with this. My mother was a summer intern at the OSS during the Second World War, and my Harvard colleague, John Kenneth Galbraith, the first important thing he did in government was a thorough review of the efficacy of strategic bombing of Germany after the Second World War. And he found we had done remarkably little damage to their production capacity because economies have much more capacity to rewire themselves than people think they will. So yes, and yes, I would do the CHIPS Act, and yes, it is an immense problem. But just in general, I would urge everybody that there is a little more capacity to reprogram and rearrange in economies than people thinking in advance tend to think there will be. 

FROMAN: This gentleman here, third row. 

Q: Willem Buiter, CFR. It’s a real pleasure to be here. 

I was struck by the relative optimism of your presentation. In my view, the political constraints on sensible policies are overwhelming. The political system in this country is failing. We have polarization, policy paralysis. We see a political system that gives a presidential choice between two unacceptable and unfit candidates. We are going to see, I think, in a short time—two to five years—growth of protectionism, failure to deal with immigration, and in all likelihood sovereign debt default and/or the end to federal dependence and high inflation. Really, this country is in a political mess. Thank you.  

SUMMERS: So, Willem, I think I’ve been reasonably widely quoted as saying that the election of President Trump would produce the mother of all stagflations for the kinds of reasons that you just suggested, and I can tell you that my friends in the Biden administration don’t regard the errors in my commentary of being excessive flattery of their policies. So I hope I have not overdone it on the Pollyanna side, and I have a degree of apprehension about the upcoming election, unlike the apprehension I had, you know—I never said—I’ve got a pretty long track record—I never said that if Ronald Reagan was elected it was going to wreck America. I never said that if George W. Bush was elected it was going to wreck America. So I do think that this is a quite unique moment of apprehension. 

On the other hand—I think this is important to say, as well—I think there is a reasonable reading of American history that a great strength of America is a kind of resilience that comes from the power of self-denying prophecy. We get ourselves into places where things seem terrible, and everybody expresses huge alarm, and some set of dynamics set in, and the problems have a way of receding. And I’m old enough to remember when the president of the United States declared a crisis of the national spirit, and when the phrase “evil of two lessers” was first coined to refer to a presidential election. 

I’m old enough to remember, as a kid, a time when the president of the United States couldn’t speak in public in the United States outside of a military base, and after that, we had a vast ethical scandal, and the Council on Foreign Relations’ own Henry Kissinger felt we had to craft a foreign policy that was acutely aware of the way the balance of power was moving against us. 

I’m not old enough to remember the McCarthy period, but I’m old—but I’m smart enough to have read about all the people who told Franklin Roosevelt that democracy was unable to deal with the depression, and he had to declare dictatorial powers. And so I am very mindful of the risks in the situation, but I also think it is a mistake to lose sight of our country’s capacity for self-repair. And I think it is a mistake to overly conflate America with American politics. And I think that something that deserves commentary—and there’s both good and bad aspects of this that does not get enough commentary—is when I left Washington in the year 2000, after the Clinton administration, I felt proud of having been part of the fact that I think the Clinton administration did a very good job of, all things considered, building a bridge to the twenty-first century. 

But if you had said to me, in 2000, that in 2024 the share of U.S. GDP and global GDP would be where it is, or even more, that the share of the market value of U.S. companies would be far greater twenty-five years later than it was in 2024, I would have regarded that as close to inconceivable. And that’s a tribute to a lot of positive things that happen in this country that give our country a lot of strengths.  

So I take a kind of mixed view of all this alarm. I am scared by it, but I’d be even more scared by its absence. And, you know, I read in the New York Times—I don’t know if it’s true, but it was claimed in the New York Times—that not a single CEO of a Fortune 100 company had contributed to Donald Trump’s presidential campaign. And if that’s true, that is really a quite encouraging thing—(laughter)—about the people who have a kind of immediate economic self-interest. And so I want to be careful of an excessively enthusiastic embrace of pessimism, lest it become a kind of self-fulfilling fatalism. 

FROMAN: The Council on Foreign Relations, of course, is a non-partisan institution. (Laughter.) We’re delighted to welcome people with a wide range of views here. 

I have felt so privileged to have worked for Larry, with Larry, to have followed his writings and commentary. I learn so much every time I have an opportunity to interact with him. 

Please join me in thanking Larry for being with us. (Applause.) 


This is an uncorrected transcript. 

Top Stories on CFR



The closely watched elections on July 28 will determine whether incumbent President Nicolás Maduro wins a third term or allows a democratic transition.

International Law

The high court’s decision could allow future U.S. presidents to commit grave abuses of power with impunity, with serious implications for U.S. foreign policy and national security.