Will America Ride the Next Wave of Innovation and Growth?

Lindsay: This is Lindsay Shingler, the host of Talking Policy. Welcome to our third episode in the new miniseries on Technology and Global Security in the 21st Century.

Nicolas: Hello, my name is Nicolas Wittstock, and I will be your guest host for today’s episode of Talking Policy.

In the past, economic growth has been characterized by periods of spurts and stagnations that are directly linked to major technological revolutions which fundamentally reshape not just what and how things are produced, but also reshape social institutions and the politics of societies. We now appear to stand at an inflection point of a new potential wave of rapid economic growth based on technologies like AI, cheap renewable energy, and biotechnology. However, this will not be an automatic process, and policy is going to determine who can leverage these technologies productively, and what the distributional consequences might be. I am joined today by Professor Mark Schwartz, professor emeritus at the University of Virginia, to discuss how this might play out.

Hello, Professor Mark Schwartz.

Mark: Hi, Nic. Thanks for having me here.

Nicolas: Yeah, it’s lovely to have you on. I want to talk to you today about waves of technological change and, really, about whether the current moment constitutes what people have called economic growth waves, or Schumpeterian growth waves, centered around a set of novel technologies.

But I also really want to get your sense about whether the U.S. is actually currently in a position to take advantage of this new wave of technological change in economic growth, if that is really what we’re currently finding ourselves in. So for some context, could you explain how Schumpeterian theorists of economic growth think about growth waves and about technological change?

Mark: Sure. So let’s start first with how mainstream economics think about growth, which is that it tends to be very high-level, in the sense of assuming factors of production are all the same, that economies naturally tend towards some equilibrium state, and that government intervention into the market typically is deleterious.

What Schumpeter said, in response to that emerging consensus in the 1930s, was that this was completely wrong for all sorts of reasons. That growth tended to happen in big spurts, that the economy was never in equilibrium, and that the key to these big spurts of growth was a package of innovations and changes that overcame limitations to growth caused by the exhaustion of the resources that had generated growth in the prior wave. And these new innovations were always a package of a new energy source, a new general purpose technology to manufacture things, new commodities, in the sense of merchandise [like] new consumer goods or new capital goods, a new way of organizing production different from the technology itself, so think new ways of organizing factories, and then he was a little bit iffy on the last two things that people now put into this paradigm, which is a new form of corporate governance, so how do we organize corporations, and the general management of the macro economy, how do we make supply and demand match?

So that’s the basic Schumpeterian package. Growth is uneven, it emerges from innovations that overcome resource limitations where resources are not simply physical, but also social and financial, and that almost always involve this package of six new things.

Nicolas: So I think it might make sense to start by defining what you mean when we talk about general purpose technologies. I think that’s a term that many people have probably heard before in the context of economic growth. But I think it’s useful to try to get a sense of what makes a technology a general purpose technology.

Mark: So, a general purpose technology is one that is, again, helping to overcome some limitation from the prior growth wave, and also is pervasive across the new growth wave. So in our most recent experience, which we can call the information communication technology, or ICT, plus biotech 1.0 growth wave—where biotech 1.0 is the Cohen-Boyer recombinant DNA process—the two key general purpose technologies were the invention of the semiconductor, which eventually allowed digitalization of everything and enabled a huge step up in automation, which you now find everywhere.

So to give some specificity to that, what we think of now as the modern flat semiconductor chip was invented in the early 1960s. And in the early 1960s, you would only find chips in a very small number of expensive applications: missiles to kill Russians and Chinese with nuclear weapons; rockets to go to the moon; some bank software; and eventually, by 1964, the IBM 360 computer.

Today, if we were doing this visually instead of audio, I’d be holding up my cell phone and saying, “Now there are chips everywhere.” You carry them around in your pocket, they’re in your coffee machine, your car is basically a computer that happens to have four wheels and an internal combustion engine in it. So you can see that chips are everywhere in consumer goods, and chips are everywhere in production processes.

Likewise, in the pharmaceutical and agricultural sectors, the first wave of biotech, biotech 1.0, became a general purpose technology for creating new kinds of seeds, like bio-engineered corn that would produce its own insecticide, or soybeans that could resist herbicides. And also of course, a huge number of drugs, most visibly and perhaps most important, human insulin as opposed to animal-derived insulin.

For those of you who really want to get into the weeds with biotech, any drug whose name ends in -MAB, that stands for monoclonal antibody, that’s a bioengineered drug. And most cancer treatments today are some unintelligible sequence of letters that end with MAB. So that’s a general purpose technology. It’s pervasive. It makes it possible to make new things, and again, it overcomes a limitation.

Nicolas: Let’s talk about limitations in a second. So you mentioned that there are generally six buckets, or six different parts of the bundle that typically describe these growth waves, right? We’ve talked about general purpose technologies being exceptionally important. What are the other ones that describe this last growth wave that we have experienced, and are they all equally important?

Mark: So, better to go back one step and note that Schumpeter and neo-Schumpeterians don’t think growth waves are automatic, and in particular that’s because some of these changes, indeed perhaps the most important changes, are social in nature. When Schumpeterians talk about quote-unquote “production processes,” what they’re actually talking about is, how is management controlling labor? And there are enormous political struggles, enormous factory floor struggles around that.

And then second, when we talk about managing supply and demand, these are also enormous political struggles because they affect the distribution of income. So growth waves are not automatic, and so the instantiation of those six factors can definitely be more or less and can definitely produce more or less growth depending on how well things fit together.

So to pick up one example, if you had asked me in 1976, “What is going to be the energy source of the new Schumpeterian growth wave?” I would have probably said, “Oh, it’s pretty obvious it’s gonna be nuclear power.” And then of course we had Three Mile Island, Chernobyl, and nuclear power died for political reasons. And we also had massive inflation, relatively speaking, which raised capital costs and made nuclear power uneconomical, particularly in the context of the changed political situation, which required a lot more regulation.

So you can see, in the ICT growth wave, a kind of missing factor. ICT is enormously electricity dependent. But what we got was a rebooting of fossil fuels, and then likewise, at the sort of more general political level, Reagan replaces Carter, or we can say the Republican Party, which represents fossil fuel interests, replaced the emerging Atari Democrats, which is a label your audience may not have heard, but was the tech-oriented wing of the Democratic Party, Atari being one of the original video game companies. And the Carter administration, you could say, was moving towards a nuclear and more electrified future. Carter famously put solar panels on the roof of the White House. The Reagan administration reverses all of that.

And so a package doesn’t necessarily have to have all these elements in line, but if you have most of them, you’ll get growth.

Nicolas: Yeah, so maybe another way to think about this, or to make this a little bit more easily accessible, would be to talk through an example of a growth wave that maybe combines these six factors, or most of these factors, most productively. And then, to get to your concern about limitations.

Because an important part of this entire model, I suppose, is the assumption that these growth waves are endogenously sort of exhausting themselves after a while, right? They’re not infinite. And maybe to try to understand why they ultimately run into barriers of further growth.

Mark: Sure. So, the one sentence here is all these new growth waves take advantage of some new, relatively easily available and cheap resource. And when they use that resource up, then you get a growth slowdown, and that’s the endogenous decay of the growth wave. So a good example of that would be, for your Gen Z or maybe very late millennial listeners, my parents’, your great grandparents’ growth wave, which was the automobile and petroleum growth wave of 1910 to, let’s say, late 1970s, early 1980s. What was the basis for that growth wave? Well, there was a new cheap energy source, which was petroleum. There was a new general purpose technology, which was the continuous flow assembly line production system.

And I want to say a few words about that, because people tend to think about the assembly line, and focus too much on the machines or focus too much on the workers, who are turned into machines. But the secret sauce of continuous flow assembly line production is logistics. Assembly lines cannot work unless every part arrives at the right place, at the right time, in the right quantities, because you cannot build a car, if: here’s the car body coming down the assembly line and… “Oh yeah, we’re missing steering wheels, just put them on later,” or “We’re missing valves for the engines,” Oh, just put them on later? No, you have to take the whole engine apart to put the valves back in.

Nicolas: The highly choreographed ballet, yeah.

Mark: Yes, exactly.

So, continuous flow assembly line production labor control was this effort to turn humans into biological robots. So if you’re actually working on the assembly line, the typical task is I have 30 seconds to mount a wheel assembly on an axle, tighten four or five nuts, and then another wheel, the second wheel on that side of the car, and then here comes another car. So the robotization of humans.

The new commodities were obviously cars, but also a whole range of internal combustion engine, petroleum-fueled equipment, eventually aircraft, but also lawnmowers, and then things that were run by electricity like refrigerators, dryers, washers, et cetera. So a whole range of consumer goods.

Corporate governance was the emergence of multi-divisional corporations that eventually turned into conglomerates. Think GE. Think also General Motors, which at one time was making not just cars, but also refrigerators, aircraft, and a few other things.

And then in terms of balancing supply and demand, to go back actually to the most important thing, you have the emergence, because of political struggles and the Second World War, of what we can think of as Keynesian demand management and welfare states. And what those did was they stabilized people’s incomes, and that mattered because a lot of these new products were expensive. You couldn’t buy them on a week’s wages. Bread, which is produced by continuous flow production, you can obviously buy out of a week’s wages, it doesn’t have to be particularly stable. But if you’re going to borrow money to buy a car, or borrow money to buy a house, you’re looking at a two-to-six-year time horizon for payments for a car. You’re looking at, in America, a 30-year time horizon for payments for a house. You need income stability, and state intervention in the labor market—which legalized unions and, in some cases, imposed contracts on business and labor outside the U.S.—stabilized wages.

Okay, now let’s go back and look at each of these pieces and we can talk about why things break down. Well, if everything’s built on cheap oil, what happens when everyone has a car and a lawnmower, et cetera, which is basically what happens by the end of the 1960s? And the answer is you run out of oil, or you run out of cheap oil.

If assembly lines require humans to be robots, what happens when the Depression and World War II generation of workers, who had experienced enormous hardship, disappear, and you get a new generation of workers that grow up in relative prosperity and tight labor markets, and they don’t wanna put up with this bullshit about speeding up the assembly line, or I guess people call it exploitation. What happens is they go on strike.

So oil prices go up; male assembly line workers go on strike. And then again, some of the, the secret sauce or hidden stuff here is that male unionized workers’ power in the labor market was partly a function of women being excluded from work. So what happens when the docile housewives of the 1950s, who again generationally experienced the hardships of depression and war, are being replaced by a generation of women with different expectations, who do not want to be sitting at home cooking meals and waiting for the husband to come home while they change diapers?

So you have three revolts that undo this growth wave, and they all emerge out of the exhaustion of the prior resource, which was cheap oil, cheap male labor, and docile housewives.

Nicolas: I really love that you chose this example because I think, even for casual observers of politics and economics, I think this is an example that really hits home in the sense that there’s currently [an] enormous amount of romanticism attached to that period of American economic history.

And I think the one big question here is: you described that there are endogenous sources of slowdown or decay of the functionality of this growth model. Is it possible to retroactively go back to it, or are there other reasons that that might not be possible?

Mark: We’re never going back to that.

Nicolas: Okay. And are the reasons for that only social, and the ones that you’ve described, or are there other reasons on top of those?

Mark: So it’s a combination of social and then, you could say, technical legacies. The original assembly line model involved dumb machines. These are pre-information technology machines, because there was minimal automation relative to what we have now. It required a lot of hands to put these products together, which is why we had tight labor markets, and so the bulk of the male population was involved in this kind of assembly line production, which was the basis for the solidarity that gave you strong unions and a high level of unionization in the economy.

And we’re not going back to that simply because most production is automated now. What used to take, in the extreme example, 50,000 workers at Ford River Rouge to produce about a quarter of a million cars a year, is now being done by between one to two thousand workers directly, and then maybe another four or five thousand workers indirectly. So the demand for labor there is simply not going to exist. We’ve automated away what were, in fact, a lot of really bad jobs. Again, go back to that guy who’s on the assembly line, putting a wheel assembly on the car, one wheel every 30 seconds or every 45 seconds, for eight hours a day.

Nobody wants those jobs. This is the fallacy behind Trump and Vance saying, “We’re gonna bring these factory jobs back to America, and we’re gonna make iPhones.” People don’t want those jobs. They sucked. And I can tell you that from my own experience, and the experience of parts of my family that worked in those kinds of factories. I’ll say it again, they sucked. They were, at the time, great in terms of earnings, but they sucked. So we’re not going back there.

Nicolas: Yeah. I think the other important point here is also that, around a specific set of mass consumption goods, there’s also an issue of market satiation, right?

So if you pile everybody into producing cars or whatever, I don’t think there’s gonna be sufficient demand for those goods anymore, right? So, yeah, you gotta switch to some other line of production in some way.

Mark: Yes. Saturation of the market, which cuts into margins and has some other effects is another piece of the exhaustion puzzle.

Nicolas: Let’s do one quick aside here.

These growth waves aren’t only relevant in terms of domestic or even international economic production, but they also influence international relations, right? Like, the relative power of different states in certain ways. Is that the case?

Mark: Sure. There’s a very deep connection here. So, if we just go back to the question of new general purpose technology, in almost all cases in the 20th century, these technologies emerge from defense-related or geopolitical struggle-related state funding of technologies where they’re trying to get some kind of technological edge, which means they want something new that relies on available resources.

And if we go back and say, where does the ICT general purpose technologies come from? I already hinted at that, but almost every technology in the ICT wave comes out of mostly U.S. Defense Department spending: hard drives, digital signal processing, touch screens, flash memory, GPS, the whole package of the cell phone, and I should also say cheaper chips through better semiconductor fabrication facilities.

All of these things come out of defense related concerns, and the combination of a big economy that can fund innovation across a broad spectrum of technologies, and therefore have not just the technologies themselves, but also a big enough economy to have a significant military, matters for the propagation of these technologies.

Denmark invented some versions of continuous flow assembly line, and here I use the word “production” loosely, but Denmark had an incredibly highly organized disassembly system to turn pigs into exportable meat. And that was one reason why Ford established its first assembly factory in Copenhagen back in the 1920s. Sorry to my Danish listeners and friends, but Denmark was never gonna be the kind of world power it was in the 1500s, okay? I mean, it’s just too small.

So you have, to go back to this automobile era, coming out of World War I, states very concerned with the ability to fight total war, which means they need a big economic base, they need machines, they need transportation. They’re all putting money into better aircraft engines, better aircraft designs, they’re putting money into subsidizing the expansion of automobile production. They don’t all get it right.

And by the time you get down to the actual war that they all anticipate will happen, you end up with a situation in which the biggest economy of the world is biggest because it has mastered and generalized continuous flow assembly line production. And that economy, of course, is the U.S. And the U.S. economy in the Second World War will be the only one that can sustain a truly global war. The U.S. is fighting in every major theater of the world, or providing food, ammunition, vehicles, and other supplies to the parts of the world where it’s not directly fighting.

So these growth waves, in terms of the innovations, are coming out of state funding. If we turn to labor relations and corporate organization, that’s also affected, and then of course affects, geopolitics. Because in wartime you need stability. You have government intervention to generate peace between labor and management, which almost always involves some accommodation of labor. And you have the transmission of these new ways of producing things from leading firms to laggard firms.

What that ends up meaning is that in the post-war period, you’ll have a dominant economy that exports its model to the rest of the world, partly because everyone else looks at it and says, “Wow, this is great.” Lots of prestige. But also they look at it and say, “We can improve our lives.” You can’t underestimate the psychological significance of American troops giving away food in war-torn Europe, in terms of people’s mental space in Europe after the war, right? That this is an economy that’s so rich that the average soldier can give away food.

So there’s an enormous prestige, but also prestige and practical element to innovating that leading sector. And so geopolitically, you get dominance because you possess that leading sector. It’s not just raw power; it’s also prestige emulation.

Nicolas: Yeah. I mean, one story my grandmother from Germany told her whole life was American GIs giving away chocolate to little children. So that was definitely something that impressed her very dearly.

In your writing, you have suggested that the most recent growth wave based on ICTs and biotechnology is also sort of exhausted at the current moment, but we’re also potentially in the midst of a new Schumpeterian growth wave with a new set of general purpose technologies. Most notably, you know, AI; more advanced semiconductors, I suppose; clean energy as a new cheap energy source. And of course, you know, we’re currently in this moment where the question really is who is going to be the leading economy surrounding these new sets of potential GPTs, right? And who is going to be the one that masters these in a way that gives them the status of being the leading economy, with all the geopolitical and geoeconomic “side effects,” I suppose, that that will have.

How do we tell, I suppose, whether or not we are indeed in the midst of a new wave like this right now?

Mark: Well, the way you tell is you ask people like me who have an answer. So, for your listeners, Yussef Robinson and I have an overly optimistic piece in American Affairs back in 2024, so they can go find a long form answer there.

The short form answer is the way we can tell that there is an inflection point occurring at the moment is that the ICT era is clearly hitting exhaustion on a number of critical indicators. So if we roll the clock back and say, okay, what were those six things for the ICT era? The cheap energy source, as I noted, was compromised, probably should have been nuclear, but turned into a rebooting of fossil fuels and importantly natural gas. As a resource, that’s not exhausted in a physical sense, but our ability to use it in the sense of carbon budget clearly is exhausted, and the resource is actually more expensive in the 2000s than it was in the 1990s.

In terms of general purpose technology, semiconductor production has become increasingly expensive. The tool to make semiconductors used to be a $100 million dollar to $1 billion investment. Now it’s a $20 billion investment to make a fab. And critically, because of the rising complexity of chips and therefore the increasing cost of the fab, the cost of building a transistor on a semiconductor chip has started to rise. So we’re gonna get more and more compute, which is what AI needs, but it looks like it’s gonna be also more and more expensive. That’s how you know that there’s exhaustion.

In terms of consumer goods, again, I would hold up my cell phone and much as you said about cars, what happens when everyone owns a car? We can say, well, what happens when everyone owns a smartphone? And smartphone sales have been stagnant since about 2018. Okay, they’re newer, they’re nicer, they do a few more things. But the big revolutionary step that the iPhone represented, we are not seeing anything like that. And the kinds of enormous growth that that set off in terms of sales is finished. Everyone who wants a smartphone and can afford it has one at the moment.

In terms of labor control, the current struggles in Silicon Valley, in other workplaces, and the post-COVID bump up in wages, I think indicate limits to the “gigification” of production. People are increasingly unwilling to put up with that. That doesn’t always produce a positive or progressive outcome. But definitely we are entering an era in which acceptance of worsening labor relations, worsening share of labor in GDP is ending.

In terms of corporate organization, the vulnerabilities created by outsourcing production have become increasingly clear and we can see some counter trends. So if you look at the car industry, which matters because electric vehicles are plausibly one of the new commodities that will drive growth, what we see is the most successful electric vehicle producer in the world is totally integrated. BYD is vertically integrated, so they’re reversing the outsourcing. And even if you look at a company like Apple, which still is tremendously outsourced in terms of the components going into an iPhone, what you see is an increasing verticalization. Things that used to be made by suppliers, like some of the memory, the modem, used to be discrete components that other companies made and then would be assembled in China. But now, Apple is designing chips that incorporate those components.

And then macroeconomically, the ICT era before COVID was delivering less and less investment and less and less growth. And that’s connected to this outsourcing model. Put in one sentence, you had a handful of companies that got all the profits and did very little investment, relatively speaking, and a lot of companies that got very few profits but needed money for investment. That produced this perverse situation in which the big tech companies were effectively operating a kind of bank in which they would re-lend their profits back to the suppliers. Apple famously loaned Corning, which makes the glass that goes on the cell phone, $200 million to boot up its next generation of Gorilla Glass about eight years ago. And that has attracted a lot of anger in the corporate world and also from antitrust authorities.

So on all these dimensions, you see exhaustion. Now to go to your question of what happens next, the answer is it’s really hard to tell, because again, there’s no automatic renewal of the growth wave. But what you can see are the kinds of things you mentioned, which are, okay, we can see a new energy source coming in. It’s photovoltaic renewable energy, and you can say there’s a clear winner here, which is China as a producer of those goods. Whether they can be profitable producers [is a] different and very important question. We can see new general purpose technologies. Here you can see what I call biotech 2.0, which is biotech based on CRISPR-Cas technologies, which allow you to do much more precise genetic engineering, affecting a whole range of sectors. So not just pharmaceuticals, but again, agriculture and, critically, materials. In terms of new commodities, obviously we can see things like artificial fibers already being sold. Electric vehicles [are] already a mass market commodity.

Where things get fuzzy are the three important things. Labor control, a little bit of corporate organization, and how do we manage supply and demand? And then how does that affect the bigger question of geopolitics? Up until the end of 2024, I would’ve said that probably we would see a rebooting of American hegemony based on Biden administration subsidies for the whole range of new technologies, Biden administration antitrust activity, and Biden administration intervention in labor markets.

The election in 2024, of course, reversed all those things. So I’d say we’re back now to an era of uncertainty. The Chinese have many, many strengths. Most obviously, they actually make things. The U.S. economy has also actually many, many strengths, but all of them are being rapidly degraded by the Trump administration, particularly in the area of science. So things are a bit more up in the air in both senses, which is to say we don’t know who will take the lead, and we also don’t know actually whether there will be a new growth wave emerging, [because] the political conditions aren’t there.

Nicolas: Mm-hmm. So what would the future look like in ten years if it doesn’t?

Mark: Right. So I’ll give you a few scenarios.

So scenario one, which is the worst scenario, is the U.S. experiences a period of civil disorder, with chaos around the 2026 election and the Trump administration declining to visibly support allies in different parts of the world. And the Chinese say, “Okay, our moment is here. We’ve been planning to invade Taiwan anyway around 2026-2027. Let’s do it.” And you get a complete fracturing of the world economy. Important point here being that an enormous amount of chip production capacity is located in Taiwan. Those factories are incredibly delicate. They almost certainly will be damaged in any war. And even though a new growth wave in some sense replaces the older one as the source of growth, it doesn’t replace the older one as a part of the economy, right?

Mark: You still need cars, and dishwashers, et cetera. So in that scenario, you have financial collapse, as the U.S. tells China, “Sorry about those $3 trillion of treasury bonds you hold, but we’re just deleting those entries at the Federal Reserve. What are you gonna do about it?” And, well… that’s a bad scenario.

Nicolas: Yeah, let’s not end on that.

Mark: Yeah. So, you know, the sort of positive scenario is the U.S. judicial and legislative branches, put shackles on Trump. Funding is restored. Elections proceed normally. And there is definitely a loss of momentum in the tech space, but with the restoration of funding and subsidies for rollout, we get a renewal of growth. And at the same time, the Chinese begin shifting purchasing power out of the hands of state-owned enterprises towards the average Chinese person, and they begin consuming more of what they produce. So trade imbalances go down, debts can be serviced by developing countries. That’s the “everything’s all right, things are great” scenario. I wouldn’t reorient your portfolio towards that.

And then the sort of, you know, most likely middle scenario is we enter something like the late 1920s, early 1930s, where there’s not catastrophic deglobalization, but there’s a steady erosion of trade relative to global GDP. There is more state control over financial flows towards friends, and we get the emergence of not the kind of self-contained imperial blocs that we had in the thirties, but a softer version of that, because trade is much more complex today. But basically a set of more self-contained regional economies in which the unit of account for invoicing trade is the currency of the dominant power. Basically a scenario of deglobalization and stagflation, slower growth, higher rates of inflation as production gets shifted back to the region, with lower levels of productivity initially, with supply chain disruptions. So a kind of milder version of the COVID shock. That’s where I would put my money. But in fact, actually, what I’ve done is it’s all in gold bars in a bunker somewhere located away from my house, guarded by my trusty security team.

Nicolas: Seems very reasonable, yeah.

Okay, my final question would be this. What kind of institutional, or let’s maybe call it state capacity, do you require to create some of the social and political bargains that you described in the beginning to take full advantage of the new growth waves? What would be the kind of institutions that would be required to take more productive advantage of those in the United States, or in some other place?

Mark: Yeah, typically what we see is that you have firms often allied with workers in those firms, but firms in what would be the new emerging sectors, which have already come into being. They’re not babies, but we could say maybe metaphorically they’re teenagers, right? But teenagers that are exhibiting a higher rate of growth and higher levels of profitability than the older adult industries.

They not only lobby politicians but of course fund them. And critically what they begin to do is take control over critical state institutions where control is not necessarily instrumental control. But, as Gramsci would put it, organic intellectuals associated with these firms become the chief civil servant, in the U.S. it would be the secretary of these departments.

So, I mean, again, historically in the U.S., American manufacturing exports were promoted by the Department of Commerce, which was created at the behest of [the] manufacturing industry, staffed by intellectuals affiliated with those firms. That’s the process of creating state capacity. You need both sides here.

I actually don’t like the phrase “state capacity,” even though I use it all the time, because it’s really the fusion of public and private power. That’s what we see creating the capacity to push through the funding and regulatory space, the changes in property rights that make it possible to have growth. All of that requires some degree of conscious, stable policymaking on the part of state actors, which means they need to be talking to people who are actually in the industry. And what we see today, of course, in the Trump administration, is a complete disconnection and a shift away from responsible long-term policymaking, towards, “How much money can we steal right now?” And by the way, in China, it’s not like things are particularly different in the sense that Xi Jinping, and party elites around him, are doubling down on expanding manufacturing, even though it’s very obvious that the world cannot absorb more Chinese-manufactured exports.

Nicolas: Professor Mark Schwartz, this has been a really enlightening conversation. I’ve learned a lot, and I’m sure our listeners are going to learn just as much. Thank you so much.

Mark: Thank you, Nic.