Ian Fletcher describes the fundamentals of industrial policy, how economic theorists ignore the practice of successful businesses and successfully industrializing nations, and why the US has the institutions it needs for strong industrial policy.
Transcript
Intro (00:00)
Wolf: Ian Fletcher, you wrote the book on industrial policy. I know this has been something I’ve been interested in for a long time. You wrote for us for Palladium. In Palladium 2, you wrote a great article, “The Economic Foundations of Industrial Policy.” That one was quite influential on our thinking, and we found it very theoretically enlightening.
So, yeah, I wanted to bring you on and talk about industrial policy as a subject, especially how do you make a strong economy is really the question, I think, underlying this. How do you actually take a country like the United States that has been maybe declining, especially in the industrial area, and bring it back to economic dynamism? So yeah, thanks for coming on.
Returns to scale and natural oligopolies (00:58)
Ian: Thanks for having me, Wolf. What you’re basically talking about here is you’re talking about the long-overdue death of a myth that this country has lived with for a very long time. Which is that the free market will take care of everything in terms of being a prosperous nation. Or a free market with a couple of exceptions, like okay, we accept Keynesianism in a recession. We accept the need for some social and physical infrastructure.
We accept the need for aid to poor people, whatever. But the productive core of the economy, where wealth is actually generated as opposed to used for something, that that is and ought to be, has to be, a free market, that nothing else is really going to work. Now, the US is visibly being outcompeted by a country I don’t have to name—that thinks basically the opposite, and they have very much a directed capitalist model. They have even a state capitalist model. A lot of their—economy works that way.
So, the question is, what’s really going on? And what my coauthor, Marc Fasteau, and I discovered when we sat down to write this book—which we actually started to write just before Trump was elected the first time. And we discovered that when you look at the theoretical underpinnings of how free markets are supposed to work—and you bring into the picture certain facts that are not exactly a secret but have kind of been ignored by neoclassical economics because they’re difficult to theorize, they’re difficult to fit into the overall model—of general equilibrium, you discover that there are very good fundamental theoretical reasons why the free market is not necessarily optimal all the time. Now, what you’re basically talking about here, for those of you listening who have a taste for economic abstraction, the phenomenon of increasing returns or returns to scale. Now, the reason that kicks you out of the free market model is, we’re talking very theoretically here, but it’ll make sense.
If you have some product where the more you make, the lower your cost is, well then whoever’s the largest producer has the lowest cost, and they out-compete everybody else. And if you have that dynamic in very strict linear form, then you end up with one producer for the whole world or the whole country. Google’s probably the best example. Full disclosure, I briefly worked there a long time ago. So, you have a search engine.
The efficient thing is whoever’s got the more traffic is going to have a better search engine. They’re going to have lower costs. They’re going to out-compete whoever their rivals are. They’re going to get more traffic. It ramifies on itself.
So, what happens there is even if you start out with a free market, you end up, just by efficiency, by the free market alone, you end up with a monopoly, and that monopoly is not only your outcome, it’s also the most efficient thing. Because if it’s just a matter of scale, if there are no other factors, then one producer is the right solution there. Now, most of the economy isn’t that extreme, but what we discover is that when you look at the parts of the economy that are the most profitable, that are technologically decisive, there’s a huge know-how component. And whenever you have a huge know-how component, you don’t get a monopoly necessarily, but you certainly get an oligopoly. There are really only two serious providers of smartphones.
There’s a limited number of companies that can make a jet engine, and a whole slew of industries are oligopolistic. Now, as soon as you bring that into the picture, what that means, particularly in a global environment where you’re competing against other nations that are trying to make their companies the most successful, is you have a global competition among nations to be the host of these global monopoly industries. It matters whether your country has industries like automobiles, semiconductors, aircraft, things like that, which is why countries compete to host these industries. Now, the interesting thing here is that, and we talk about this in the book, when you have these industries, they don’t just have scale. They have a lot of other important characteristics.
For example, as soon as you have scale, you have oligopoly. Well, then you get pricing power which is anytime you have anything that interferes with the pure functioning of a market, you get pricing power. Sometimes—that’s as simple as just being the only coffee shop on the block. But the interesting kind of pricing power comes from being one of, say, three companies in the United States that can make automobiles, which at one point was basically true, three and a half maybe. Or a limited number of companies that could make the high-end computer chips.
Now, when that happens, you get excess returns. You get excess returns in these industries to capital. And you usually get excess returns to labor. So these industries are lucrative. It’s not an accident that everybody wants to have them.
And they have other characteristics as well. For example, these kind of industries, we call them advantageous industries in the book, they tend to compete not on pure price. They’re not producing a commodity where your profits get squeezed down. You’re competing on technological sophistication. You’re competing on sophisticated understanding of buyer needs.
You’re competing on vendor financing. You’re competing on the synergies of your product with other things. And that brings me to another topic, which is synergy, which is another characteristic of what we call advantageous industries. For example, this all goes back a long way. When you had steel first made in this country—cheap with the Bessemer process, it was originally done to produce railroad rails.
It was eventually adapted to produce steel frames for skyscrapers, which totally changed the urban geography—of the United States. So that’s a synergy, and the reason that’s very important is industries that have synergies are catalysts for economy-wide growth. Because any given industry, no matter how good it is, no matter how efficient it is, it’s going to be limited by the fact there’s going to be a limit to how much demand there is by consumers for whatever it is they make. But when you have synergies and you create new industries off of the initial industry, that process doesn’t stop. Now, to give you one example, you have the jet engine, okay?
Which is invented for military use in World War II. It’s invented originally in the UK—and Germany. And when you have jet engines, you can build high-capacity passenger jets, particularly when you have something called the high-bypass jet engine—which was also invented in the military. So you can have the revolution in cheap air travel. So you do that, and then the long-run result of that is completely unrelated and frivolous-sounding things like Disney World, which you can’t have if you don’t have airliners, which you can’t have if you don’t have high-bypass jet engines—which you can’t have if you don’t have jet engines.
Natural rivalries for advantageous industries (09:05)
Ian: So there are good reasons why the world is in a state of rivalry to be the host of these key industries. And what it means is there’s a case for subsidizing if you’re an economic predator like China, say. Subsidizing this industry so that your version of this industry can take over the whole of global demand. Because once you’ve got it, it’s incredibly lucrative to have it. Now, granted, I’m speaking in caricatures here.
So any economically sophisticated viewer will understand there’s 1,000 exceptions here. But the baseline fact that we’re dealing with in this book is the fact of rivalry to host industries. Now, more technically, this is actually a rivalry to host economic activities because—one of the things people point out is any given industry, like, say, smartphones, the final assembly of smartphones, such as is done in Foxconn City in Shenzhen, is a low-wage occupation. It’s not particularly lucrative. But manufacturing of the chips that go into them, manufacturing of the displays, manufacturing of the glass, these are sophisticated economic activities.
So—given the complexity of modern supply chains, you’re often not necessarily fighting over an industry per se. You’re fighting over parts of the supply chains, what’s called an economic activity.
Wolf: Right, the particular parts of the process that are the most technologically intensive have the most increasing returns, the most pricing power, therefore.
Ian: Yeah. That’s what you’re doing. Now, there are a number of reasons why this gets you outside the basic logic of free markets that we know and love. One is simply the fact that an oligopoly is not a free market. Yes, there’s competition—but it’s not free market competition.
That’s why you have pricing power, which you have rent-seeking. Dynamic rent-seeking is when you seek rents by trying to have a technological advantage. In the standard economic story, which is naturally true in this instance, rent-seeking is bad if you’re just seeking some privilege from the government or some obstruction—where you can get an excess return. That’s static rent-seeking. Dynamic rent-seeking is very different.
It’s not only lucrative for you, it’s advantageous for the economy as a whole because it drives technological advance. But free market economics has long had a difficult time with the fact that this doesn’t happen in a free market. It happens when somebody’s figured out a way to escape the tyranny of the free market where you don’t make any money. Now, this is why the kind of thinking that we espouse and that our book is based on, the real home of this is not in econ departments, it’s in business schools. Now, that should be—
Wolf: Right. They understand this stuff perfectly.
Ian: Yeah, that’s a real clue. It’s as if you had an argument between physicists and electrical engineers about the behavior of electrons or something, and the physicists are telling you one thing, and the electrical engineers are saying, “Look, if you assume that, you can’t actually build anything. If you actually want to build useful electrical machinery, you have to assume something slightly different.” So anyway, the kind of thinking that is in our book is exotic and revolutionary and, frankly, regarded as stupid in the mainstream economics profession—in the United States, with very, very few exceptions. But in business schools, it’s basically that’s the model that they have.
They have to. They have to think in terms of a world where some people are winners and some people are losers.
Why economic theory overlooked successful industrial policy (13:11)
Wolf: So yeah, I want to talk about those two ideas of the free market. I guess, so yeah, I was listening to what you’re saying here, and oh, there’s an oligopoly as a result of the free market. It’s not quite a free market outcome, though. In business circles, this is obvious. Everyone knows this.
Of course, this is the game you’re playing. Sometimes you get a huge amount of investment, speaking of subsidies for a country to get an industry. Uber or something might put an enormous amount of effort into capturing that activity, that market demand—and then ride on having the moat on it. So this is all very well known in business. But in economics, both in free trade and then in national economics within a country, there’s this very smoothed-out model where there’s perfect competition, and it’s all comparative advantage, and everyone gets the best outcome by the free market mechanism.
But why is it that they persist on that model, despite it being not actually the practical model, either by rational states or by rational business actors? Why is it that the economists got into this idea of that particular version of the free market, that very smoothed-out thing that doesn’t include these factors?
Ian: Well, there’s a bunch of reasons. The first reason, I think, is that there was this misplaced turn in economics after Second World War—where people really thought you had to make economics a science. Because there’d been economics before then. You’d had enormously successful capitalist economies around the world, and you’d had enormously successful reforms of the defects of capitalist economies. You had the antitrust era under Teddy Roosevelt.
You had the New Deal. So we had relatively sophisticated economic thinking for practical purposes. But economists had physics envy, is the best I can put it. They wanted to be able to reduce this to some absolute and precise science, where once you’ve got a certain objective body of knowledge, that just is how the world works, and you know, and anybody who disagrees, you can just mathematically prove to them that they’re just not right. So one of the things that happened is it led to a turn of excessive mathematization—of economics.
And one of the foundations of this was Paul Samuelson, who’s generally regarded as the greatest post-World War II American economist. In his doctoral dissertation, which I believe came out in 1946 at MIT—he figured out a way to mathematize the foundations of economics in a way that nobody had really thought of before. And before he died, he actually wrote an article warning people about how they’d misused this, and he actually wrote me a nice letter once. But the point here is that mathematizing the conceptual foundations of economics and mathematizing the entire structure—are two very different things. The problem is that if you want to build a complete mathematical model of the economy that’s going to tell you what’s going to happen given data about your initial conditions, the way you can model a system in classical Newtonian mechanics—you’re going to need theories that give you closed-form solutions.
That is to say, given the following set of facts, we know that the following dynamics will operate, and you’re going to end up looking like this. And that’s just what we can predict. Now, the problem is that the kind of thinking that’s embodied in our book and in business schools and in the policies of a lot of other countries, most notably in East Asia, involves something called multiple equilibria. Now, the easiest way to explain that is if you go back to our example of Google as a search engine. If you start off with two different search engines, say Google and Gaggle, that have roughly the same or exactly the same technology—which one is going to win is going to depend on a lot of contingent events that affect the path that their competitive activity follows.
And the same thing is true if you look at the evolution of, say, the auto industry or the computer industry is notorious for this. All the might have beens and could have beens in the tech industry of people say IBM could have owned the world if they just understood the situation—a bit differently in the late ‘70s, or Xerox could have been Apple or whatever. The point is that when you have all these completely contingent events—you can mathematically say a lot about a lot. I’m not disparaging the use of mathematics when it’s an appropriate tool. And my—original background was in the IT business, so I’m not a sissy about exact technical disciplines.
But the problem is the only way to have an overall model where you just keep piling up your model of this on top of this, on top of this, until you reach general equilibrium, where you kind of solve the whole economy as a kind of giant set of simultaneous equations, is you have to have these closed-form solutions where you can’t have phenomena like increasing returns which give rise to multiple equilibria. Which means that you can’t have some sort of general equilibrium answer that tells you what’s going to happen. So, there was a huge mathematical bias against acknowledging the kind of realities. Now, some mainstream economists have openly acknowledged this. Paul Krugman wrote that economics took the path of least resistance.
Intellectual, least mathematical path of least resistance. This is not a secret, if you dig into it. But you have this huge problem of the misunderstood idea that economics should be a science at all. I’m very dubious of the very concept of social science. You can’t do—controlled experiments on social phenomena to anything like the extent—you would need to build complete models.
It doesn’t mean you can’t have useful knowledge of these things, but I don’t think science is the right paradigm for that knowledge.
Wolf: And even the comparison to physics is kind of strange. The physics envy within the social sciences. But somehow, it’s never quite doing what the physicists do either. Because what you’ve just described, this process of making simplifying assumptions at every step until you can solve the whole thing as a simultaneous set of equations, is more like, in physics, the analogy would be we linearize everything until we can just solve the universe as a single set of equations. And you wouldn’t get a good result from that.
You wouldn’t expect to get a good result from that.
Ian: Oh, sure. One of the problems with this attempt to make economics a science is they didn’t understand how the science they were trying to imitate, namely physics, works. And I also don’t understand, frankly, why did you pick physics? Is—chemistry less science? Is biology, which is a much better if you want to have a controlling scientific metaphor for what you do as an economist, I think biology would probably be a much better science to pick.
And you can’t tell me that biology is unscientific. That’s absurd.
Wolf: Yeah, but it might just be the prestige of physics following the atomic bomb. Right? In that era.
Ian: Well, David Teece at Berkeley, who’s a great economist at the Berkeley Business School, I met with him last week, and he actually said exactly that. He said, “I think,” because he’s working on this problem at a much higher level than I am, because he’s actually a very esteemed professor over there. He says he thinks it was because of the atomic bomb. That the Manhattan Project just dazzled everybody. It’s like, holy cow, physics just solved the biggest problem ever of—defeating these two horrible hostile powers and then deterring this other horrible, hostile power.
If physics can do that, don’t we want to be like physics?
Why Ian wrote the book (22:17)
Wolf: Right. And so we ended up with this kind of spherical cow economics for almost half a century. And so, I guess that brings us to the conception of this book. Somehow you got onto this problem it sounds like in the early 2010s, mid-2010s, realizing that you wrote the book previous to this one, “Why Free Trade Doesn’t Work.” And I’m curious, what were those conversations like?
Where did that come from? How did you get onto the idea that this had to be done, that you should be the one to do it?
Ian: Well, I certainly didn’t want to. I wrote “Free Trade Doesn’t Work,” which I sat down to write in 2007. Came out in 2010. I got curious about the whole free trade controversy, and I went on amazon.com to look for a book where someone was going to lay out, okay, this is the case against free trade, because you’re constantly hearing the case for it, for it, for it, for it, for it. And it seemed to be creating all these problems.
And I wasn’t sure at that point, but I went looking for the book that will just give me the straight-up case why this is bad policy. And the book didn’t exist, so I basically wrote the first book, “Free Trade Doesn’t Work,” so that I could read it. And when I started, there were fewer than half a dozen intelligent people in the English-speaking world I could even talk to about why free trade was wrong. It was so much the orthodoxy. It was so much, if you don’t believe this, you just don’t understand basic logic.
of economics. So I wrote it because at that point, I don’t think I had the right to call myself an economist. I’ve been working on this stuff for nearly 20 years now, so I’d probably call myself an economist today. But at that point, I was just an IT consultant who was crazy enough to say something no economist would say. Because if I had had a PhD in economics, I never would have gone down that—path because I would have had just so much intellectual investment.
The time, the years you spend mastering all these models, it’s unthinkable on a purely personal level—admitting that they’re basically useless—for the real world. That would have been a gut-wrenching trauma. But because I didn’t know anything, I was able to start from scratch and arrive at what I think I can now probably say is a reasonably vindicated view of the issue—that it doesn’t work.
Wolf: And it just came out of that curiosity of, you wanted to have the full understanding. You were not hearing the other side of the question at all. So you had to get the case.
Ian: Well, I was lucky. I knew somebody who was at a think tank that no longer exists in Washington called the US Business and Industry Council. They gave me some sponsorship to write the book, which was kind of a crazy move on their part—but they bet on my ability to produce something worthwhile, and I had some other philanthropic support from people I was able to get interested in the—project because I was the only person interested in doing that at that time. When I was writing this book, honestly, I was terrified that somebody much more important than me would come out with the same book before I could get it finished. I was terrified that Paul Krugman would suddenly come out with a book six months before—
Wolf: And scoop you
Ian: I was finished. Yeah, and just steal my thunder and make me look like just a complete derivative waste of time. But it didn’t happen. It’s only much later that people got interested in, “Oh, so maybe the free trade model isn’t right.” People are starting to realize that now.
But—
Wolf: Yeah, it’s funny how these things go. It’s almost within economics, people assume that the space is more explored than it actually is, not just academically, but in the market. Like, “Oh, there’s no product for that. That must be because there’s something wrong with that idea,” or whatever. But sometimes it’s just no one’s done it.
And so many people are so conventional, whole fields can just be in lockstep like that for a long time.
How much state capacity does it take to do better than free trade? (26:46)
Ian: Yeah. Academic groupthink is definitely an issue, but what’s, I think, more fundamental is just the fact that when you scrutinize the theories in favor of free trade, I’m not even talking about the critique here. When you actually look at what the theories in favor of free trade say, when you look at people who say, “Okay, Ricardian comparative advantage is all you need to know on this subject. It simply answers the complete question.” When, as I’ve explained, Ricardian comparative advantage is true in a very narrow set of idealized circumstances, and it’s a handy mathematical gadget for figuring out what your benefits from trade are in certain circumstances.
It is not, and interestingly was not intended to be by David Ricardo, who invented this, he wrote as much, was not intended to be a complete model of the trading economy that applies under all circumstances. That’s the guy who invented it. Said that. I can document that. So you have those factors.
When you look at those models, the classical models, one of the things you find is they don’t, even on their own account, actually prove that free trade is best. When you go and read the people who are defending free trade, what their case usually comes down to is, “Okay, I admit of some exceptions, but I’ll bet you that no government could ever competently engage in the interventions required to take advantage of these exceptions and produce a better result.” So what they’re—
Wolf: It’s just axiomatic.
Ian: I wouldn’t say the problem is it’s axiomatic. The problem is that they’re actually, even the classical model, has a lot less logic in it than you think it has. Because when you get down to it, all they’re really saying is not that free trade is necessarily best, but I bet you can’t do any better with an intervention, is what they’re really saying, is that—
Wolf: That they’re saying it’s hard to do better.
Ian: Yeah, they’re saying it’s too hard to come up with an intervention that would be better. It’s too hard to know when you should do that. It’s too hard to have the state competency to implement it. It’s too hard for the political stars to align correctly so that you’re not just engaging in somebody’s rent-seeking. That’s really the bottom line that I ran into again and again and again.
Because the hard mathematical shell is there to impress people. It doesn’t, even on its own terms, prove as much as a lot of people seem to think it does.
Wolf: Right. Well, maybe we should talk about that aspect then, just this problem of the actual competency required to put this stuff into practice effectively. Obviously, China has done a very effective job at grabbing a lot of the world’s industries. But in a democracy like the United States, very mature, lots of political interests going around, lots of regulatory capture or lobbying this way and that. Is there maybe something to that, that despite the kind of theoretical possibility, it’s actually too difficult for the government to undertake this project?
I’d be curious for your thought on that. Is it something that is actually within our power, and these people who are making those arguments just didn’t believe that it was possible, but they were incorrect? Or yeah, how do you see the actual level of difficulty in our actually existing political system?
The dollar is overvalued (30:47)
Ian: Oh, I definitely think that doing what needs to be done is within the competency of the US government and the US political system. Because you got to remember, you don’t have to do this successfully, which means getting a much better result than we have now. It doesn’t mean doing industrial policy perfectly. Governments don’t do anything perfectly. But to get to a result that would be trillions of dollars better than we have now, you can make a fairly short list of things that we should do that we demonstrably have done things similar enough to that in the past.
One profoundly non-obvious example, important example, is the US right now has a massively overvalued currency. We have a currency that’s probably overvalued something 12% to 14%. There are different calculations. That’s not an observable. And our average real tariff is a lot less than people think.
It’s only about 10% or 11%. So the US right now doesn’t have a net positive tariff at all because we have an overvalued currency—which also handicaps our exports, which handicaps the competitiveness of American firms abroad. So if you want to get serious about resolving our trade situation, right now, we have a trillion-dollar deficit in manufactured goods, so you can’t re-industrialize the US if you don’t bring that demand home. Okay? So if you want to do that, tariffs are not going to do that.
Tariffs are great for controlling what you buy from abroad as opposed to make here, and to a lesser extent, controlling where you buy from abroad. You can’t shift the overall trade deficit down with a tariff of any plausible scale. But you can do that if you bring down the value of the US dollar, which was done before. It was done in the mid-1980s with the Plaza Accord in the Reagan administration when they were facing a lot of upset about the trade situation in Congress at the time. So what it would take to do that is you would need some kind of moderate variable tax on incoming flows of capital into the United States.
Because the only way the US can run a trade deficit is we have to give foreigners who send us goods and services, we have to give them something in return because we’re not trading with Santa Claus. And what we give them is financial assets. That is to say, we go into debt—to foreign countries. We sell them shares of existing assets, T-bills, stock in Microsoft. You name it.
So the entire deficit, which means the principal constraint on the re-industrialization of the US, is only possible because you have these massive ongoing flows of foreign capital. And the US is gradually buying up. The US is gradually being sold off.
Wolf: Yeah. If you take the kind of outside model on the whole thing, the government prints a bunch of money, the money goes overseas to buy foreign goods, the foreign goods come back, but then the money also comes back to purchase US real estate, US treasuries, US equities. And so the net result is we’re selling ownership in the American economy.
Ian: Yeah. You can’t get something for nothing here. People tell me it doesn’t matter. I’m like, “Okay. Who’s the Santa—in this scheme?”
It’s a profoundly corrupt way of living if you carry it on—this long. Now, the problem with a moderate variable tax on incoming flows of capital, which is not that hard to implement, I was party to putting legislation together that was introduced in the Senate in 2019, the Baldwin-Hawley—Bill. That bill was partly written by the organization Coalition for a Prosperous America, that I’m on the advisory board of. The methods by which you would do that, they exist legally. They’re within the competency of, depending on what agency you had supervise, it could be Treasury.
Our preference was the Fed. Make it part of the Fed. The Fed has an existing triple mandate for full employment, price stability, and moderate long-term interest rates, usually referred to as a dual mandate because people forget about the third part. But we would add a mandate for balanced trade by means of balancing the dollar to it. I’m happy to debate this, but as far as I can tell, it’s pretty clear that you could do this.
It’s not some strange moonshot—of weird ideas.
Wolf: Legally and institutionally, it’s probably not very difficult.
Ian: Other countries do things like this on a much smaller scale all the time. The US even did something like this in the ‘60s with something called the Interest Equalization Tax, was brought in under Kennedy when we were trying to hold the dollar up to honor—our Bretton Woods commitments. My point is, in terms of state capacity, you can do this. The problem here is that Wall Street would hate it because their business model, simpliciter, is moving capital anywhere in the world to get the best return. That’s what they do.
It’s what they say they do. That’s an accurate description. So anytime you put a tax in place that says, well, you’re not going to be able to do exactly what you want, or if you do, you’re going to have to pay for it, which means you make less profit, they’re axiomatically going to hate it. So you have to overrule them as an interest group. There’s a lesser pushback from people who fear that bringing the dollar down would be inflationary.
We’ve modeled this, and if you bring the dollar down gradually over, say, five years, the inflationary impact is not zero, but it’s very, very small. It’s a lot less than other policies in the US right now that cause inflation. So there’s no free lunch—but this lunch is well worth the real price it would actually cost.
Wolf: So the bottleneck at this point is just the politics of convincing enough of the coalition to override some of these special interests? Is that how you’re seeing it?
Ian: Well, if you’re talking about right now, during the Trump administration, I have to assume the decision is being made by the president. And the interesting thing is that you can document the fact that all three of his top economic advisors, that is Howard Lutnick at the Commerce Department, Scott Bessent at the Treasury Department—and Steve Miran, chairman of the Council of Economic Advisors, they’re all on record of having written about this problem, the overvalued dollar, and in one way or another, there are different flavors of it, what you could do about it. Conceptually, as far as I can tell, they all agree with me. So I have to assume, and I’m not a psychoanalyst or a tea leaf reader, I’m not going to tell you why President Trump has made the decision not to do this, but I think it’s pretty clear that in this administration, the decision would be his.
Our industrial policy is ad hoc (37:53)
Wolf: Right. And it just comes down to that. And so I guess let’s talk about then the bigger prospects of how you’ve seen things moving on this in the whole industrial policy set of questions. I think you mentioned in the book that industrial policy is coming back as a topic. Certainly, a few years ago when the book was published, we were seeing that, or when you were putting this together at least, there were people, Biden’s administration was talking about industrial policy.
Elizabeth Warren was talking about industrial policy. There were new magazines and projects and think tanks kind of talking about industrial policy. A lot of people were thinking about this topic. And so have you seen that kind of getting traction, besides the decision of the president or particular officials? Have you seen the broader culture of understanding and intention within the political class changing on these fronts?
Ian: Oh, sure. When we started writing the book, the idea was so far out of the Overton window, most people weren’t even against it. It’s definitely an idea. People know what it is now. People have opinions of it.
If you look at the very establishment intellectual institutions, that is leaving aside actual implementation of policy for a minute. But if you look at institutions like Foreign Affairs is the magazine that basically ratifies what respectable establishment opinion is once an idea has appeared there. They are pretty much endorsing a watered-down version—of our thesis. They certainly wouldn’t go as far as we would go. They talk about Hamiltonian statecraft, which is basically coming back to the idea that the US is in an economic contest with a rival superpower, and just as we had a contest against Marxism, it’s relatively easy to win because just Marxism is so dysfunctional.
We now have an economic contest between this thing we call free market capitalism, which sort of is, and then the state capitalist model that China embodies, and—they recognize that. The establishment has cottoned onto this, for sure. The way they seem to be responding is the way that the powers that be always respond in the face of a threatening innovation they can’t get rid of, which is they figure, “What is the smallest change that we can make—that will absorb this challenge to the system and enable us to preserve as much of business as usual as we can?” Which is why you’ve seen things like the Trump administration has been very aggressive on industrial policy for semiconductors, and they’re kind of following that up with industrial policy for rare earths. This stuff is being pushed.
In terms of a policy for the whole of the economy, in terms of looking systematically at what industries need what kind of support—to thrive in the current international economic environment, that’s definitely not there. I don’t know if you saw this, but when the Trump administration killed most of the EV subsidies that Biden had put in place in the Inflation Reduction Act, that caused in excess of $50 billion in write-downs in the American auto industry. That’s not some estimated number from some economist. Those are actual SEC numbers go into annual reports where you go to jail if you lie about it. Those are hard dollar numbers from actual capitalist business corporations that the administration had no qualms about basically hanging out to dry because they don’t like green energy.
Wolf: And how do you read that? What’s the implication of them doing that kind of thing—
Ian: Well, I think in that partic—
Wolf: for industrial policy?
Ian: Sorry?
Wolf: For industrial policy.
Ian: Sorry. I think in that particular case, it’s quite clear that the administration has, for whatever reasons I shall not speculate on they really don’t like green energy, with the exception of nuclear, which I sometimes think might just be because nuclear power annoys those liberal hippie dips. I mean, nuclear power—is politically incorrect, so it’s the only green energy that the Trump administration likes. And also interestingly, just to get some subtlety here, geothermal is also popular, and I’ve been told the reason for this is that the technology you use to drill geothermal wells is pretty much identical to what you use to drill oil wells. So everybody in the petroleum business who drills for oil is really gung ho about this because it just gets them more—
Wolf: They can reuse the stuff
Ian: customers. Yeah. It’s basically drilling difficult deep holes for hard rocks in the ground. That’s what they do, and that’s what—geothermal is. So that, to some extent, is a special case because they have a particular animus against a particular technological project of decarbonizing our energy system.
The larger concern is the lack of systematic thinking. The fact that it’s all a bunch of impulsive, ad hoc responses to particular problems that, for one reason or another, the administration or people close to them have grabbed a hold of and they realize are important.
You have things like shipbuilding is apparently... They’re talking about shipbuilding, that’s going to be—a big thing. Thus far, we really don’t know how serious that’s going to be because the United States is absolutely nowhere in ships. It’s quite astonishing. The US builds something like one tenth of 1%—of the world’s ocean-going ships, which is—
Wolf: That’s insane
Ian: really quite shocking. We probably produce a larger percentage of the world’s pogo sticks for all I know.
Wolf: Right. And I guess this gets to this, I don’t know, extended ecosystem of difficulties for these kinds of policies. We’ve mentioned all these special interests and political issues around how things are perceived ideologically. There’s also, you mentioned rare earths. Rare earths really makes me think about the regulation issues around that, right?
It can be very polluting, or it can be perceived as very polluting, or there’s a lot of, I understand, regulatory barriers to a lot of these industries. How do you see that fitting into the industrial policy picture? A lot of this isn’t really about the economics even at some level, in the sense of whether the industry has the capital to get going or whether the government is pushing on things financially, but just are they allowed to do it? Is it efficient to do it under the current regulatory setup or those kinds of things. I guess I’m most curious about the, let’s say, non-economic barriers in business, the regulatory barriers.
How do you see those fitting in?
Ian: Well, regulation almost by definition is the part that government can have something to do with. Yes, granted, in some circumstances you have government directly developing technology or taking an equity stake, but for most of the economy, and I am by no means a fan of promiscuous interventionism everywhere because you think intervention is just something you want to do. I do get the Hayekian argument, I really do. But take one example. I was meeting with someone yesterday who works on the hyperscale build-out of AI data centers.
Datacenters (46:13)
Ian: And the constraints that they’re running to, one of them is local zoning from municipalities that don’t—want these giant facilities. It’s a myth that they consume all that much water—but it’s not a myth that they consume a ton of electricity. We have these giant facilities, each of which contains a few dozen jobs because it’s a data center, it’s not a—
Wolf: Right. They’re not doing that much
Ian: factory. And we’ve seen things that were reasonably anticipatable, like Governor Hochul in New York having a statewide moratorium. But then the shocker for me was when Governor Greg Abbott in Texas—did the same thing, because in recent decades, Texas has been the place where you can get—
Wolf: You can build stuff
Ian: you can get things built. Right. And then I asked this guy I was talking to, “What has the administration been able to do about this?” And the answer is, well, really not a lot because we don’t, unlike some countries, have very much national land use planning—in this country. We have the Department of the Interior for forested mountains and stuff like that.
But for urban land, the federal footprint on land use planning is very, very small. The ability of the federal government to do something about this huge backlash against data centers—is very small. And my question is, at what point do they just start bribing rural areas to accept these? So that’s a kind of constraint where the federal government has not been able to do anything. The part where the federal government probably would have an easier time if they would but try is to do with something called the interconnect for electric power.
The issue with data centers is not just that they consume a lot of electricity. The issue is that to even get hooked onto the electric grid as a major user often takes years, and that’s something called the interconnect queue. And interconnect queues in this country are mainly administered on a lottery basis, so you can’t just buy your way—
Wolf: Oh, wow
Ian: to the head of the queue. Now, that’s a relatively technocratic problem where you could hypothetically step in—with a fix. Maybe they do that tomorrow, but up to this point, they haven’t. And in case you hadn’t heard, the Stargate project, the big AI initiative, the administration has basically gone belly up. They got some things built, but they abandoned most of their sites, most of which I believe were then picked up by Microsoft.
So, a sporadic and spotty effort, even for things that get the administration’s attention.
Wolf: Right. This reminds me a lot of the California high-speed rail problem. When they tried to build the high-speed rail, one of the big things they ran into was just every little municipality they had to go through had all kinds of its own process, and barriers, and veto power, and so on. So these things that seem like obvious things that would make the economy stronger or otherwise sort of serve the industrial goals that the government even wants to do, has enormous funding to do, and is willing to put weight onto it. We’re still running into these problems where we can’t actually do it.
In favor of incrementalism (50:10)
Wolf: It’s a difficult problem, this whole atmosphere, not only of the political problems that prevent the government from trying to take action, but then all the little organizational problems once they do try to take action. And in the successful cases of industrial policy, I’m thinking of Japan or China to some extent, often there’s whole government departments that have quite a lot of power to organize these things. Or maybe it has to be done in conjunction with all these non-governmental parts coordinating the sort of non-governmental elite interests in civil society and so on that could influence that political climate. And I wonder about the prospects of that kind of thing. Not just from the officials using their established powers to get these things done, but from the broader ecosystem of the extended government reorganizing itself or being reorganized by someone to make this kind of thing really happen.
That feels like it’s this incredibly difficult task, but of course, it’s been done under FDR and so on in past times, and people have said that maybe now we need things like that. We’re sort of overdue for that kind of cleanup in that system. But I’m curious about whether you think there is stuff like that needed for industrial policy, or is that too far off and you think it can be done with existing mechanisms pushing on the margin despite all these problems?
Ian: I and my coauthor are very much in the second school of thought. You’re absolutely correct that the East Asian societies above all have much more centralized governments where they can make these kind of highly centralized decisions on behalf of the economic interests of the society as a whole. And that does seem to work for them. Leaving aside any issues of whether it would be a good or bad thing—there’s no way that’s going to happen in the United States. You’d be talking about a political earthquake to motivate change on that scale of such incredible size that I just don’t find that plausible.
More importantly, I don’t think you have to. I gave currency as an example of something that’s well within state capacity. If you look at the fact that the Biden administration did correctly observe that electric vehicles are not just about climate change. I don’t care. Even if you don’t believe in climate change—electrification is the direction that the auto industry is going globally.
So if you want the US to have a competitive electric vehicle industry in a world where it’s competing with subsidized players around the world, you have to give your electric vehicle industry here a boost. You can’t just wait for the market to catch up or you lose.
Wolf: Right. You have to invest in it. You can’t just be scared.
Ian: You have to. It’s about the next generation of the entire auto industry, which is huge and important. Yes, it has this environmental aspect, which is important, but even if you left that out of the picture, it was a policy that made sense, and they did do it. Now you have a different administration gets elected, and they canceled most of that. We have a political process in this country.
People make these decisions politically in a democracy. What else would you expect? That doesn’t mean it can’t be done. It means we have to make the right decisions. In the long run, that’s true of absolutely everything in government.
The hidden American industrial policy of the late 20th century (54:19)
Ian: One of the reasons why the US got so stuck on the free market idea for so many decades is basically that we cheated. We told ourselves we were living in a free market completely free market economy, and it performed well. In point of fact, we actually did a lot of industrial policy behind the scenes, only we didn’t call it that. So we talked ourselves into believing in an economic model that was not actually what we were doing. I’ll give you a couple of examples.
I’ve been banging on about currency because it’s so important, even though the topic tends to make people’s eyes glaze over. From the end of World War II to 1971, you have the Bretton Woods—managed currencies, fixed exchange rate system. Now, that was created to solve related problems that had occurred in the 1930s, but thanks to that system, it was almost impossible for national currencies to get so far out of alignment that nations could run huge trade imbalances. And if you look at the trade imbalances in the world in that era that people fussed over, they’re a rounding error compared to what we deal with every day. So that was a huge industrial policy no one acknowledges as such.
Or if you look at state-supportive technology, which is one of the other big aspects of industrial policy we haven’t touched on as much as perhaps we should. Where do the most important technologies come from? Well, there’s this myth that they get invented in some dude’s garage in Palo Alto, and he makes a billion dollars. Look, God love those people, but the most important technologies of the post-World War II era, things like semiconductors.
The transistor is not created in a free market environment. It’s created in a state-sanctioned monopoly called—AT&T, which had Bell Labs. And for decades, they had this sanction from the government to be a state-regulated monopoly that had this incredible technology research arm that produced... I have a list in the book somewhere, but the obvious one is you have the transistor, you have the laser, you have all sorts of other stuff like sound movies. You have an unbelievable list of technologies came out of this one company, which was not state-owned, but it sure as hell wasn’t a free market.
And the reason—that they had the money to do this incredibly long-horizon research was totally antithetical to free markets. For a start, they had a guaranteed market for whatever they invented, not because they earned it in a competitive marketplace, but because the federal government guaranteed that they would be the phone company. So if they spend billions of dollars inventing lasers based on the idea that one day they’re going to be useful for long-distance telephone communication, which is true—they can do that because they have the long time horizon, which is why so much fundamental technological research did come out of companies that, for a brief or long period of time, had a monopoly or quasi-monopoly position. You probably know about Xerox PARC—which invented the mouse, the touch screen, all this great stuff. Xerox didn’t know what to do with it, but they were another company that had this incredibly profitable near monopoly on making Xerox machines.
And you have the David Sarnoff center of RCA, which develops color television, things like that, when RCA was this incredibly formidable—company, which it no longer is. Radio Corporation of America. It was set up by the US Navy—after World War I so the Brits wouldn’t own all the important radio tech. So if you look at other military technologies, jet engines, obvious, rocketry, satellites, all these technologies were not originally created for profit. They were created to defend us from the Soviet Union.
And the lesson here is not necessarily that we want to become Prussia—and militarize our economy, because it’s not necessarily the military that’s going to be fruitful for the next generation of technologies. But you can’t develop, or you often don’t develop the most important technologies on a straight-up market for profit basis. Now, let me be clear, sometimes you do. The last two major technologies that were developed on a for-profit, free market basis were the electric grid and the automobile. And the reason that was possible is that intermediate stages of those technologies were monetizable.
Before Henry Ford created the Model T, cars were expensive, specialized toys. Well, there was a market for that, so you could make money selling them and develop the technology that was eventually dumbed down and cheaped up to make the Model T, and then you have the mass production car. Electric power was useful even before you didn’t have a nationwide grid, so you can develop the technology step by step until you get to the full-scale thing. Now, the example today that comes to mind is AI. Because although there is a big governmental footprint in the development of AI, and I don’t want to disparage it, the key innovations like the paper, “Attention Is All You Need,” which came out of Google.
Wolf: Yeah, it’s Google.
Ian: The reason that they’re able to not just do the theoretical research behind generative AI—but actually implement it is because the intermediate stages are useful. I believe that for quite a few years, the single biggest use of AI was just the recommendation algorithm in YouTube. This comes out of some history of that I read. I could have this wrong, but the point is that... And that’s, of course, immensely valuable because YouTube is a going concern.
So if you can have a better recommendation algorithm that uses AI to figure out what video the person wants to watch next, and they do watch the video, that’s worth a ton of money. So anyway, the fact that AI is monetizable in the intermediate steps between zero and AGI—is the reason you can do this in the private sector. But—
Wolf: Not every technology
Ian: not every technology historically works that way. There’s no guarantee the next technology will work that way. You have to have strong support for both approaches. Otherwise, you’re going to get technologically surpassed by some society that is willing to put down the money. For example, you may have heard recently, I forget the gentleman’s name, but there’s a professor at UC Berkeley working on new materials who—just announced that he’s decamping to China, where I assume the government has given him a blank check to develop new materials.
That’s a huge field. Advanced materials. Japan exports more every year by dollar value of advanced materials than they do of cars, but nobody knows about this—because advanced materials are not sold directly to consumers. You buy a carbon fiber tennis racket, okay, you know it’s got advanced materials in it. But for the most part, this is a huge industry that most people are not aware of—and the US is not stepping up to the plate to make sure that we’re number one in this, and China is.
Wolf: So there’s these existing industries like advanced materials, things like carbon fiber manufacture, where I think it’s very well covered by the theory that you’ve laid out. But there’s, I guess also implicitly there, and what we’ve just been discussing, is the development of new technologies, maybe even new industries, new economic activities, that we haven’t identified yet as being valuable. For example, the laser or artificial intelligence or the internet or the various things that have ended up being incredibly valuable.
The missing intermediate steps in industrial development (1:03:33)
Wolf: With the concept of the advantageous economic activity or the advantageous industry, I think there’s at least a clear guide of where government attention and support should be flowing in terms of developing those existing industries, capturing those existing industries. Do you have guiding theory for how we should be investing in things like actual new technology? Is it just spray money everywhere, just support for R&D generally, or through academia, or through things like DARPA, or the sort of XPRIZEs and challenges? Like they just had that Heavy Lift Challenge earlier this week, I think, or last week. What kinds of mechanisms are best in your eyes, and what’s the theory behind that?
How do we know where to put the support?
Ian: I got an answer in two parts. The first part is, paradoxically, figuring out what the technologies of the future are going to be—is actually not that hard. There are a lot of people putting out lists. Robots, you think that’s going to be a big technology of the future? Yes or no?
Genetic engineering, yes or no? Fusion power, yes or no? Electrically powered automobiles, yes or no? Electrically powered aircraft one day—yes or no? It’s actually not that hard to peer into the crystal ball and see what technologies are going to become important.
The White House Office of Science and Technology Policy puts out an official list. There are a bunch of other institutions who do. That is not actually the hard part of the problem. The hard part of the problem consists in the fact that the path from pure scientific discovery, where you come across a law of nature that makes something possible—and standing up a commercially viable industry that actually sells products to customers and makes a profit for its investors and pays wages to its employees. That path is very complicated, and there are a lot of steps in that path that are not well taken care of in this country.
For example—we have very strong pure science capability support by the federal government. You have National Science Foundation, you have NIH. Peer-reviewed science is a well-understood thing, and there are problems with it, but for the most part, we know how to administer that and get good results out of it. Turning a ready-for-the-market technology into a company is also something that Americans are very good at. There are a lot of entrepreneurs out there.
There are a lot of venture capitalists willing to fund them. There are a lot of engineers willing to work at startups. People understand that model. People like that model. People believe in it.
Everybody wants to do it. So it happens. The difficulties are mostly in the steps between pure science and usable technology. Because a lot of technologies go through many, many steps where you’re not talking about pure science, that is discovery of laws of nature. And you’re not talking about something you can actually sell to a customer.
And the US is not that good at funding that intermediate step. People have this idea, which goes back to Vannevar Bush and the Endless Frontier report just after World War II. That the public sector should fund science, because science is a public good, and the private sector should invent and make things, because that’s what inventors and entrepreneurs do. That’s too simplistic. That’s leaving out this giant missing middle.
Now, what we have in that missing middle is a strange patchwork of compromises. For example, the military, and I mentioned this before, is probably the single biggest institution in this country that pays for developing technology. Not researching science, not discovering laws of nature that make a stealth aircraft possible, but figuring out how to actually make the materials that you put on the surface of an airplane—that will do that. And they did this, as I mentioned, for jet engines and for aircraft generally, for satellites, for rockets, for a whole bunch of other stuff. I really hate it, by the way, when people use Tang and Teflon as the example of governmentally supported innovations, because those things are just not that important.
Okay? No. Jet passenger planes and the internet, okay?
Wolf: Yeah. Those bigger things.
Ian: So part of how we get around this missing middle problem is we’ve had the military do it. Now, for many decades, we also had the phone company—
Wolf: Yeah, Bell Labs
Ian: do it. So it was the military and the phone company were our technology development institutions. NIH does it for pharmaceuticals. They do it reasonably well. They’re actually one of the better functioning parts of the pipeline.
But when you have some possible technology that you can look at the science and say, “Okay, bet you could make one of these, and if you did, it’d be really great. You could do all sorts of things with it.”
When you have a technology like that, that isn’t pharmaceuticals, and it doesn’t have obvious military implications, it can be very hard in this country to find support for these non-monetizable steps of the technology path.
There’s fragmentary institutions here and there that will do a bit of this. But, for the most part, the US has not systematically implemented across the board support of non-commercializable steps in the technology development path. We haven’t even really, as a country, gotten our heads around that concept. It’s a very important concept. The average voter should have it in their heads that technology doesn’t grow on trees.
Some of it is going to be done for a profit. A lot of it, before you get to the better mousetrap, you have to make the better cheese, and to make the better cheese, you need the better this, and nobody will eat it. And so these steps in the technology development path that are neglected, that’s a big part of industrial policy this country could do. There are models for doing it. They certainly exist.
Wolf: Mm-hmm. What are the best models of other countries doing it or having done it in the past? How do you think that should actually be done? Is that a new office of technology development, or is it some other kind of initiative with the universities? Or how do you think would be the best model to actually do that?
Ian: Well, around the end of the Obama administration, when a gentleman named Ron Bloom, an investment banker, was finishing up the auto bailout—he asked President Obama, kind of as his parting shot, to establish something that was then called the National Network for Manufacturing Innovation. It’s now called Manufacturing USA. And it’s modeled on an institution in Germany called the Fraunhofer-Gesellschaft. Which basically, it’s a network of institutions. Germany has about 100.
Each institution is focused on one technology, is partly funded by government and partly funded by industry. So you have an institution that’s focused on, say, there’s one that we have in the US now. We have a small network with just over a dozen institutions. So you have one, for example, I think it’s in North Carolina, focused on power electronics, which is one of those obscure but desperately important things for the electronic systems you need for controlling and managing large amounts of electric power. Apparently, that’s a field that’s in a state of innovation and flux.
And what you do is you have part support from the US government and part support from industry. And what they do is companies can go to these institutes in—Germany for example, and some company says, this is probably a terrible example in terms of the actual engineering, so please if you know the technical details of this, don’t write in and make fun of me. But as an example, suppose you had some company who says, “We have this great idea. What if you had ceramic-coated titanium ball bearings?” “Nobody’s ever done that, though we know what titanium is, and we know what ceramic is. But we think that under some application of high temperature or high pressure or whatever, that the concept of a ceramic-coated titanium ball bearing might be just the greatest thing ever. Now, we don’t want to devote a ton of our money to researching this for a couple of reasons. One, if this turns out to be a great idea, you can patent the particular details of what we then manufacture and our process. You can’t patent the concept.”
Right? “So we’re willing to put some money into this, but there really does need to be some public sector support for this idea because if it works, the benefits are going to be generalized—to the whole of the economy or the whole of the industry, not just us.” And then Fraunhofer-Gesellschaft has engineering labs where they can do that kind of thing. And we have Manufacturing USA in this country, which desperately needs to be bigger and needs to be better supported by the federal government. It has an annual budget of $300 million a year, which is chump change on the federal level.
But that’s one model for how you go in and support the missing middle of technology development. There are other cases. For example, the National Renewable Energy Laboratory in—Golden, Colorado, which just renamed itself the National Laboratory of the Rockies in order to escape the ire of the administration. I know, I know. I was talking to some people from there a couple of weeks ago.
Anyway, NREL was responsible for a huge chunk of the improvement in the efficiency of photovoltaic cells. The Department of Energy had a project called SunShot. Okay? It’s a play on words on Moonshot, where I think it was, was it double the efficiency of solar cells? I think it was double.
It might’ve been something different. So they did that. It was so successful, it finished three years early, and they doubled down on it with, I think it’s called, what? SunShot 2030. They did it again, and lo and behold, government scientists can develop useful technologies.
It’s not beyond their capability. It’s not a waste because once you develop the tech, then anybody could use it, so it benefits the whole economy. It’s a reasonable idea that developing technology, for some technologies, not all, but for a great many technologies, it’s a reasonable responsibility of the federal government. There’s a track record of its being done successfully. We have institutions, albeit far too small, that have done it already.
How to make sure domestic R&D benefits domestic industry (1:15:55)
Wolf: So in the case of solar technology in particular, to really take advantage of that, you would also want to actually control the solar industry domestically. Right? If you make the solar industry more valuable, you’d want to actually get a slice of that. But I understand most solar panels are manufactured in China. China has this enormous global market share in solar technology.
So how much do things like those kinds of innovations end up just being that that inadvertently is a research lab for Chinese manufacturing as opposed to something that actually benefits the American economy? How do you see that problem playing out? Is this something we need to focus on building up the actual domestic industry first and invest in the things where we actually have industry, or is it somehow that we benefit despite not having the direct industry?
Ian: I’m glad you asked because, forgive me for blowing our own horn for a minute, but Coalition for a Prosperous America, the organization I’m on the advisory board of, has for years been the biggest lobby in the United States for appropriate tariff protection of the US solar industry for precisely the reasons you’re talking about. And an interesting paradox is that lately, the Trump administration, which does not generically appear to like the idea of green energy, has actually gotten pretty good on putting in place the tariff and other measures that are going to be needed to rebuild not just solar panel assembly, not just photovoltaic manufacturing, but the entire supply chain. They’re actually starting to get reasonably serious about that. So yes, you do have to do these things because it’s not beneficial to create a technology only to have it commercialized in another country. So this is why I tell people all the different parts of industrial policy fit together.
You might think—that tariff policy has nothing to do with scientific research. Well, no, it does because if you want your science to pay off economically, you need the technology research built on it, and then you need some way to ensure that your domestic industry is competitive, not just doesn’t have its market taken away from it by foreign competitors who burn coal to run photovoltaic plants full of slave labor. But you then also, as I mentioned, have to have a competitive currency. So all the parts fit together. If you try and do industrial policy as a bunch of isolated one-off initiatives, you’re going to get some successes when the problem is obvious and the solution is a narrow package of policies.
So many of these problems, you won’t. You don’t necessarily have to have policy intervention everywhere. As I said, look, I’ve read Hayek. I understand that the default assumption of free society is the government leaves things alone. But you do always have to be asking the question when you look at an industry, even if your default policy assumption is to do nothing, you have to ask the question: Does this industry need industrial policy interventions to thrive in this country?
Because increasingly the answer is yes.
Wolf: Right. There’s this integrated domain of industrial strategy and industrial policy that ties together all these different seemingly disconnected elements that work together to support the most advantageous industries and make sure your country stays economically strong.
Ian: Yes, and you don’t have to put it all in one government department. You don’t have to have—a federal department of industrial policy or an industrial policy czar or some office bossing everybody about. You can have a decentralized process, but you still need everybody, every decentralized player has to be thinking systemically. And I’ll give you one example of that, that’s actually specific to our book. As I’ve been promoting this book, trying to get people to read it, the real breakthrough in getting people to read the book was when we discovered that the number of people in this country who are interested in industrial policy itself is not zero, but it’s pretty small.
The number of people who are interested in something that they think industrial policy could get them, that they want, is huge. You could be talking about anybody from people in the Air Force who are worried about drone industry, to people in green energy who want a green energy industry, to people worried about imported generic drugs poison people. There’s just a vast range of people, and I’m astonished every day—I end up interacting with somebody who I discover has some entirely new angle on something they specialize in, they want, where industrial policy is interesting to them because it could help them to get this thing.
Wolf: And it ties it all together between all these different little things that people are focused on, all these little—struggles that we’re having all over the economy. They’re solved, this very central problem of can we actually maintain overall industrial economy?
Ian: Well, that’s why I think people need to think systematically, particularly—in a free society with decentralized decision-making because if the decisions are going to be decentralized, people need to see the whole picture so they can see where they fit in it. If nobody’s going to be telling them what to do, they need to know.
Wolf: Right. Well, I think that’s a good place to leave it. Thank you very much, Ian. This has been a great discussion. I really enjoy your work.
I think it’s very important. Definitely, we do have to be thinking systematically about how the economy actually works and how we can make it run better.
Ian: Well, thanks, Wolf. Thanks for having me on your show.
Wolf: All right. Until next time.