Build or Die! US Industrial Policy with Rob Atkinson

Feb 07, 2021

How should the US frame industrial-policy grand strategy in light of China's rise? What can America expect from its allies on tech policy? Also, how does the power-trader paradigm in Albert Hirschman's analysis of twentieth-century Kaiser and Nazi trade policy help explain China today? I had Rob Atkinson, president of the ITIF (Information Technology and Innovation Foundation), a nonpartisan DC-based think tank, on the ChinaTalk podcast last week to discuss.

Why Care?

Jordan: Why is it important that America lead in these industries?

Rob: For a couple of main reasons. Look at the UK: they used to lead back in that old wave of technologies, and they lost out to Germany and Japan, and then the US. The result has been that it's very difficult for them to be competitive. If a country can't be globally competitive, its currency level is going to fall dramatically.

In the case of the US, that means that in twenty years we're going to have to pay a lot more for our imports, our standard of living will go down. Secondly, these industries are propulsive industries that create good jobs. Jobs in the tech industry pay about 75 or 80 percent more than average jobs.

The third reason really is national power. Countries that have strong IT industries have more power nationally, and that's important for all sorts of diplomatic reasons and certainly for military ones. Let's look at what's called the Third Offset Strategy in the Department of Defense. The first strategy was related to nuclear weapons back in the fifties; the second concerned things like stealth. The third one states that we have to be technologically more advanced than our adversaries because our adversaries oftentimes are bigger than we are. Certainly, that's the case in China. IT is critical to us being able to continue to have a military lead over China. That is an important thing for us to have––hopefully we'll never have to use it––but as a deterrence. If the Chinese government knows that the US military has significantly stronger force capabilities than it does, it's going to be much more hesitant to do anything in Taiwan.

The Origins of US Industrial Policy

Jordan: I grew up in a kind of unique era in US history where this industrial policy stuff of pushing forward America's IT national leadership was not at the forefront of governments' minds. In one of your papers, you take this trend all the way back to the Civil War, talking about how the North's armaments regime to fight the South is actually where you start the clock on this.

Rob, let's do a brief history starting, say, at 1861 and through the 1980s and the Reagan administration.

Rob: Oh, sure. I was watching the inaugural ceremonies the other day, and I was so thrilled when Vice President Harris mentioned Lincoln as one of the important figures in US history. She talked about Lincoln enabling the construction of the transcontinental railroad and the land grant universities.

For these reasons, Lincoln was one of the most important industrial policy presidents we had. The transcontinental railroad linked together our economy; it helped build the railroad industry and the navel manufacturers. And obviously, the land grant college has played a critical role in enabling technical education for manufacturers. [Rob recommended this economic history of the US off the air.]

But actually, I'd go back even farther. The War of 1812 was actually critical to American industrial policy. We entered that war, and we were so badly prepared technologically––the British out-gunned us, out-shipped us, everything. And so Congress put together a crash program on military technology. This is partly where the Springfield armory came from, where we created interchangeable parts for rifles, but also shipbuilding technology and the like.

Of course, all this goes back to Hamilton too. That's something people often forget. In the history of the US, everybody seems to think that all the founders were anti-Federalists and Jeffersonians, and they forget half of the founders were actually Hamiltonians who wanted a strong industrial policy.

So anyway, then you jump up to Lincoln, then Roosevelt. The United States at one time had the most powerful and largest radio and color television industry in the world: RCA, Radio Corporation of America. People forget that it was actually created by Franklin Roosevelt in the Wilson administration because Marconi (this was the Huawei of the time) had all the patents, Italian and British combined. And our navy was saying we cannot afford to have this; it's too much of a national security risk. So, unlike the Chinese, we bought out the rights to the Marconi pads––we paid the company Marconi, and then we built our own industry, our own firm, and it came to be called RCA.

We did many things like that. But where we really got into industrial policy was after World War II: we created essentially the best and most effective industrial policy program ever in mankind's history. It was basically all built around confronting the Soviets. For example, in the 1960s, the US government spent more on R&D than every other company and every other government combined outside of the US, funding things like DARPA, DOE, DOD, NSF. That gave us a huge lead.

Jordan: Make the case that America's rise to technological dominance from, say, the 1930s to the end of the Cold War was a function of design as opposed to random institutional endowments that the United States had.

Rob: One of the problems with the industrial policy debate is that a lot of conservative economists, but some moderate economists as well, say, "We don't want Soviet-style planning," as if that were an alternative. That was never the alternative in the United States. The alternative was laissez-faire, where the government didn't do much, but there was an active state that supports the private sector to drive innovation. If we hadn't had that state, I don't think we would have had a semiconductor industry the way we do now.

We certainly wouldn't have had the internet. The internet was built out of defense funding. We wouldn't have had GPS. We wouldn't have had computer interfaces the way they are. Just go down the list: it was government funding that played such a central role in all of that.

We're blessed because we've got fantastic entrepreneurs, we got the Steve Jobs of the world who do great work, but they're building on a base that government helped create.

Is a Neoclassical Economic Policy Incompatible with Strong Industrial Policy?

Jordan: You have a quote in one of your papers that in Reagan's 1987 State of the Union, he went up in front of Congress and said, "We will guarantee that government does everything possible to promote America's ability to compete." That framework, coupled with all the trade actions with Japan and SEMATECH and this big R&D push he did, really belies the narrative or received wisdom that a lot of folks have of Reagan's being a cut-taxes and shrink-government person. He did that on a lot of other fronts, but his philosophy was on board with this role that you think the government should play when it comes to promoting national IT.

Rob: Reagan was a huge supporter of creating the research and development tax credit. Then the first Bush administration actually supported the creation of what's called the Manufacturing Extension Partnership. So we have a system of over 60 centers that go out and help small manufacturers. You mentioned SEMATECH––there's a whole set of things like that that we created. We changed the law to allow pension funds to invest in venture capital. There are all these things we did, and Reagan understood that competitiveness is essential. Unfortunately, we don't see that so much today.

Jordan: I want to talk a little bit about the reframing of this debate towards the more neoclassical "let the markets do what they want" viewpoint that dominated through Clinton and Bush, and pretty much Obama, and maybe even the first two years of Trump. I'm trying to understand where I got this idea, because this is something that I very much bought into as an obnoxious high-school debater, and I feel like just trying to pinpoint it.

It was definitely just in the air, but also The Economist and their op-eds, they sounded so smart and sure of themselves, and this was the view that they took: whenever government tried to push any industry, it was always a boondoggle in the making.

Rob, what were the structural things that changed (aside from the Cold War ending) that led the United States to focus away from the more industrial-policy-forward outlook on treating strategic industries?

Rob: The end of the Cold War was a critical factor, there's no question about that. We had a national purpose and were willing to do things to achieve that, which included industrial policy. But also one of the things that happened in the late seventies was the rise of supply-side economics, the overthrowing of the Keynesian regime.

In Keynesianism, at least there was a recognized role for government. And then that morphed into essentially the rise of neoclassical economics, for which The Economist magazine is the Bible. People read it and think they're getting the truth. Neoclassical economics became so strong, it became what John Williamson called "the Washington consensus," and under the Washington consensus, the government role is very minimal.

One of my favorite quotes comes from Alan Blinder's quote, when he was the head of the Council of Economic Advisers under Bill Clinton. He said, "There's nothing the government can do to raise the rate of economic growth." Paul Krugman still pontificates today without knowing what he's talking about; he said, "competitiveness is a dangerous obsession." Boeing may compete with Airbus, but we don't compete with Europe.

Most economists have at least have the pretense of having some evidence for their claims, but when it comes to this question of industrial policy, it is completely evidence-free. They don't cite any studies.

Here's my favorite of all time. Larry Summers—Obama's guy, economics czar—wrote a big piece back in the late seventies, where he did a DRI econometric model that modeled the economy first with an investment tax credit for investing in machines, and then one without an investment tax credit. What he found is that not having an investment tax credit actually lowered GDP growth and lowered investment. Duh. So what did he recommend? Getting rid of the investment tax credit because it was industrial policy.

Jordan: Can you talk about the appeal of a neoclassical strategy when it comes to this kind of thing?

Rob: There are two parts of that: the appeal to the priesthood and the appeal to the layperson. The economics profession really fundamentally shifted to becoming a mathematical one. There's a great quote by Ken Boulding: "Mathematics brought rigor to economics. It also brought mortis." In other words, rigor mortis.

Neoclassical economists simply have no way of modeling industrial policy. They have no way of modeling innovation. They really don't. And they will acknowledge that. And rather than say, "Our models don't work," they say, "Let's put it outside the model." That's why Paul Romer's work was called "endogenous growth theory," bringing innovation back into the model. But by and large, economists don't understand and don't even talk about innovation.

One of the best-selling macroeconomic textbooks by Greg Mankiw, who was in the Bush administration, had no discussion of innovation except for one page. And it was a discussion of an ice-cream maker.

So that's number one. The economists simply can't model industrial policy; they have this really religious view that markets always get it right.

Jordan: I heard an interview the other day with Paul Romer, and he said he was so depressed that his profession wasn't going to take endogenous growth theory seriously that he just left and went to some education software start-up in the nineties or early two-thousands. He was just sick of fighting these fights.

Rob: Exactly. I feel bad for Paul because he did fantastic work on that and he was ultimately just rejected and ignored. If you have that view, you're automatically outside the club, and you probably won't make it as a high-level economist.

Then the question is: why did policymakers buy into this? To some extent, it favored the interest of business for a long time. You look at a group like the National Association of Manufacturers (NAM).

We helped the Obama administration launch a program––we did a big event on it and a report––and it ended up becoming the Manufacturing USA Institutes. At the time, we were working with Congress to see if we could get the program encoded into law, and the NAM rejected it. They lobbied against a program to help US manufacturers. And why was that? It was partly just the ideological function of them wanting to be left alone. But generally speaking, it was about the fact that they would much rather have a government that does nothing and leaves them alone than one that's involved with them and helps them.

Jordan: Because fundamentally (and you make this point a lot in your work), what this new technology does, when done right, is create creative destruction. You have new firms getting created and the old ones that are on top now have to either evolve or slowly die.

If you're sitting there, it's 1992, America is dominating every single industry you can think of, and you're a leading company, it's not all that much in your interest to rock the boat and introduce new variables, which may end up forcing your business to do things it's not particularly comfortable with.

Rob: Absolutely. The other big question is, why did Reagan support this? Partly because the capitalists of the United States (in other words, business leaders) supported an industrial policy because they were shut out of the Japanese market. The Japanese market was huge––the second-largest market in the world, super fast-growing. They wanted into that market. The Japanese wouldn't let us, and so their interest, as capitalists and as Americans, aligned.

That vanished by and large once China came onto the scene because of China's strategy. They looked at what Japan did and they said, "We're not going to make that mistake. We're going to get the American companies, the American capitalist to invest in China." Therefore, American companies were pretty happy. They thought, "We don't need an industrial strategy, we're doing fine." But the point of an industrial strategy is not to help companies per se; it's to help the US economy, the US system.

Once companies had a different interest potentially, then you saw that shift. The reason it's shifting back the other way now––look at the semiconductor industry today, where they're really threatened by Chinese semiconductor policy, and that's why they're now supporting this new CHIPS Act to help them domestically.

Digital Realpolitik

Jordan: So Rob, why is "digital realpolitik" the framework that the United States should take into 2021?

Rob: Well for starters, it's not just digital. It should be related to all sectors of economic policy.

Going back to the Truman administration all the way to the present day, the view of foreign policy was that we were willing to sacrifice our economic interest in order to maintain hegemony and influence. Part of that meant we were willing to actually help countries industrialize at our cost––at our expense! "As long as you support us on this North Korea thing, we're going to let you off the hook on this trade thing that's hurting American businesses," we'd say.

That continues today. There are folks in the Biden administration who still believe that it's more important for us to be the global leader, even if it means sacrificing our own interest. That's frankly what Obama did to some extent with climate change. He let China get away with certain things because he thought he needed China to help out on climate.

What we need is a realpolitik approach, which is to say, we're going to put our interests first. That's not to say we're going to ignore global interests and our global responsibility as leaders of the free world, but we're not going to subsume our economic interests the way we've done for sixty or seventy years.

Balancing Openness and Supporting the Middle Class

Jordan: Can you talk a bit about the openness principle that dominated a lot of Obama-era thinking on this topic, and how that doesn't quite rise to meet the challenge of today?

Rob: There are kind of two elements to this: advanced technology broadly speaking, and then digital technologies and the internet. We issued a report last week, where I was arguing that if you look at what the Obama administration strategy was––particularly under Secretary Clinton––it was really that the United States should support what's called an open internet: free speech, etc. And we spent a lot of political capital trying to make that happen. The problem with that is that it doesn't really do us a lot of good at the end of the day. We're doing it because we're a good country and we want to support human rights, but it doesn't help us.

There are two problems with this approach, firstly that it didn't achieve very much. Most dictatorial, authoritarian countries are not going to change. They're going to continue to block political speech that they don't approve of, imprison activists on the internet, etc. We can say we don't like it, but it's not a lot we can do about it at the end of the day. More importantly, it took a lot of political capital, which we don't really have enough of. We ought to be using it to fight for our core digital interests.

Jordan: It's interesting to watch the likes of Jake Sullivan and other economic advisers because when they urge putting the American middle class back in the center of US trade policy, it's really a riff on this. Instead of just lowering tariffs in the name of neoclassical openness, you actually need to lean more on making sure that the US sees the benefit rather than just favoring countries that we respect.

Rob: One thing that worries me about that new "doctrine" of trade/foreign policy putting the middle-class first is this: if we don't have strong multinational corporations that are gaining market share––not domestically but overseas––that hurts the middle class. And I worry that this middle-class framing that the Biden administration has doesn't really fully reflect that.

A case in point is intellectual property. When other countries force US companies to give away their intellectual property, like in drugs or Hollywood, the narrative from a lot of the progressive is, "Well, it doesn't affect the middle class, it affects the capitalist class." That ignores the fact that when you look at who's employed in the movie industry, it's mostly middle-class technician workers who put up the lights and do the recordings and all those things.

I think the middle-class framing is okay as far as it goes, but what it really needs to emphasize is that the United States much more robust competitiveness. If we don't have that, there's not much we can do for the middle class.

Political Economy and National Industrial Competition

Jordan: And then you get to get back to a tax policy, looking at all the favors we're doing for nominally US-based companies, but how much of that is being reinvested here versus, like, Ireland.

In one of your recent papers, you walked through some not ideal scenarios that this new era of national industrial competition could bring us towards. What I found interesting is the emphasis you put on the EU as a key swing player. Why don't you walk us through your framework of EU regulatory imperialism, and the extent to which Brussels is willing to follow through on this doctrine, and how the next couple of decades may play out.

Rob: There are really three governing doctrines or principles around the digital economy. One of them is China, which is obviously authoritarian both economically and politically. It is a control state, and frankly they've done pretty well when it comes to building digital champions. Then there's the European approach, which really hasn't worked. They don't have hardly any digital companies, but they keep thinking they're going to get some. That's fundamentally based on what's called the precautionary principle.

European policy is focused on breaking up big tech companies; it's focused on incredible regulation on privacy––way more than you need to protect people's privacy. It sees AI as a threat; it wants to put all of these barriers in front of it. The problem is, if they want to do that in their own twenty-seven countries, so be it. The problem is that they have what I call EU regulatory imperialism. They're importing it around the world.

I was down in Brazil a while back, and they were basically just adopting almost carte blanche what the Europeans were doing because they thought that was the right model. And it's not the right model. It's a model for slow digital growth at best. Squashing innovators.

I was talking to one company in Silicon Valley, a really cool company, and I asked if they were going over to Europe, and they said, "No way. It's impossible with the privacy rules and the AI rules. Maybe in five years when we get bigger and we can afford a lot more legal help."

The US approach is really the superior approach, which is a light touch on regulation and an emphasis on industrial policy around tech. We have to make sure the European approach isn't the dominant approach in the Western world.

Can Albert Hirschman’s Analysis of 20th Century German Trade Policy Teach Us about China?

Jordan: It's really interesting watching China in the past two months start to incorporate a few more hints from the Europeans when it comes to this sort of thing: pushing forward more anti-monopoly policy and copying word-for-word a lot of European digital privacy law into Chinese regulations. It remains to be seen how much that will hamper growth.

There's also the alternative that doing anti-monopoly stuff opens up room for creative destruction. Watching the feedback loop between the United States, the EU, and China is going to be fascinating to follow over the next few years.

Before we get to what the United States can do to elicit some happier outcomes, let's take a step back and talk some Albert Hirschman, my favorite twentieth-century political economist. What can we learn from Hirschman's analysis of Kaiser and Nazi trade policy?

Rob: I recently read the recent biography of Hirschman called The Worldly Philosopher by Jeremy Adelman, and that brought me to a book Hirschmann wrote on foreign trade and power. It's actually his first book, National Power and the Structure of Foreign Trade, published in 1945, which he wrote when he was at Berkeley. It was really eye-opening. He was writing about German trade policy, really through three regimes––the Kaiser regime, the republic that lasted a short while in the twenties, then all the way through Hitler. Almost everything you said, if you put the word "China" in there, you've got the exact same analysis.

At the time, the way people thought about trade (and they often still do) is you've got free trading countries and free traders, and you've got protectionist countries and protectionists. Hirschman said, no, there's a third category: power traders. They use trade to exert their national power. They do that in a lot of different ways, but ultimately it is a strategic approach to hurt and penalize other countries and industries that threaten them and also to make sure that other countries are weak and dependent on them. That's what China's doing today. So to me, it was really an eye-opening book.

Hirschman was a translator at the Nuremberg Trials when this book was released.

Jordan: You write that "Hirschman's key insight was that some countries focus not on maximizing free trade or even protecting their industries, but on changing the relative power of nations through trade to achieve global power. Germany's policies and programs were designed not only to advance its own economic and military power, but also to degrade its adversaries' economies, even if that imposed costs on their own economy relative to a free-trade regime." How is that framework not addressed by the current global trade system under the WTO?

Rob: What's interesting is that the tactics that they use back then are the same tactics China uses now. I guarantee that Chinese economic policymakers have read Hirschman because they're so well-read, so astute ... they know history. For example, in the twenties, attorney general Mitchell Palmer wrote, "the German trade power tactics included destroying business competitors by state aid, cartel, combination, dumping, bribery, theft of patents, espionage, and propaganda." That's exactly what we're talking about with China today. China is essentially saying that they're willing to give up some short-term economic gains.

If you look at the Made in China 2025 strategy or the Strategic and Emerging Industries program, there are incredibly expensive programs for what they get. No rational country would do what they're doing, because it just costs so much; it takes so much to get the little that they do. The fact of the matter is they have a lot, so for them, getting a little is a big deal.

They're doing that entirely for economic power. If they wanted to raise their GDP, they wouldn't be doing that, they'd be figuring out ways to get their retail, wholesale, and insurance sectors more efficient, they'd be opening up to banks from the west, etc. That's not what they're doing. What they're doing is trying to gain power, because they know they have to be an advanced-technology powerhouse.

They also know they have to degrade our power; that's the other key thing. They're not just saying that they want to get better. They want to get better relative to us, and that means degrading US capabilities.

The WTO and Its Shortcomings

Jordan: You write that "the WTO was created with the idea that countries are either free traders or misguided protectionists," and that the available remedies for the hijinks of early twentieth-century Germany and present-day China don't really get resolved by things that the WTO can do.

Rob: Well obviously there was no WTO in the first half of the twentieth century, so regulation was more difficult. Hirschman had this sort of almost idealistic view of a kind of WTO-like entity. The problem is that the WTO is essentially toothless when it comes to most of the Chinese practices.

I was a co-chair of the US-China Innovation Expert Group under the Obama administration. I always remember during one of our first meetings in Beijing, we had a luncheon with a bunch of government officials, and I was talking to an official from the National Development Reform Commission (NDRC), and I said, "Weren't you guys worried about joining the WTO?" And he said, "No, no—just the opposite. The WTO gave us a get-out-of-jail-free card. We knew the WTO wasn't going to be able to penalize us." And the reason is that most of what China does is not written down all on paper. It's not put through an administrative procedures act that you can judicially go against in the WTO. That's one of the big limitations of the WTO.

The second big limitation is that the WTO is a political body, just as the WHO is a political body. As Mark Wu has written at Harvard Law School, "We've got to give China a few wins here and there." That's not how that process should work. The WTO has not been able to stop China or even slow them down in terms of their unfair power tactics.

Jordan: I want to read another Hirschman quote: "Between those who ignore the danger of external economic relations becoming an instrument of national power aims and those who see the danger but try to remedy it by the defensive and offensive weapons of economic nationalism, a place should be indeed left to those who, faced with danger, refuse to allow the policy of either the ostrich or of gribouille [someone who naively sees the adversary as innocent]."

Rob: If you look at the US approach, that's what we've done. A lot of the Obama trade officials were in the latter category: "Oh, China's not a problem," or, "China's going to become like us," or, "who cares whether we lose furniture"... they just wanted to wish it away and not see it as a problem.

And then you had Trump, which at the end of the day was a waste of four years when it comes to China. We didn't get anything for it. The only thing Trump did, which I give him credit for, is he raised the issue. He made it much clearer what the threat was. But in terms of his taking any meaningful steps to do anything about it, he probably set us back because he alienated our allies.

The answer is we've got to do as Hirschman said. We've got to focus on what we can do domestically and then strategically focus with allies (and other parties, and the WTO) on where you can limit China internationally.

Jordan: DATO: Democratically-Aligned Trade Organization. A few episodes ago, I did "NATO for trade," but I like that this idea has now evolved to the point that it has its own acronym. What's the idea behind this, and what can the United States realistically expect like-minded countries to go along with in order to take the action that Hirschman recommends?

Rob: The WTO is not really designed for power trade. If you exert some sort of economic attack on the foreign industry, by the time you get to the WTO, the corpse would come to the hearing and say, "I'm dead. Thank you." In other words, you can't wait so long for many of these things because it takes so long, and by that point, the US industry is dead.

We saw that, for example, in solar panels. By the time this whole process finally worked out, it was too late. What China is doing is using power to impose harm. That's a NATO issue. If Russians attack us, we can't spend a year figuring out what we're going to do; we can't go to the UN. We've got to respond immediately. And that's the idea of a DATO, a NATO for trade, as you put it.

If you look at Chinese wolf-warrior diplomacy and what they're doing to Australia right now ... they don't like what Australia has said––not what it did––and so they're going to punish them.

Australia said, "Maybe China is responsible for the coronavirus." Yeah, no kidding. Or they express that they don't like what China's doing around human rights. So China decides they're going to punish them, and they've done it through trade-policy punishment.

Australia is standing alone. It's ridiculous. We're just going to let Australia stand alone and take that punishment. We need some kind of alliance that says, "If you're going to punish one of us, you're going to punish all of us, and we're all going to react at once." That would slow the Chinese down significantly. They can't go around picking countries off individually, threatening them, bullying them to get what they want, because they know that all of our allies would come together and say, "No." We don't have that right now.

The EU’s Crucial Role

Jordan: The key swing player again comes down to the EU, and the question that arises in your writing is to what extent should the United States be willing to compromise on some of these EU policies, which would likely lead to less dynamism domestically, but would allow the US and the EU (and Australia, South Korea, Vietnam) to be able to have this stronger response in the face of these aggressive power-trade-type activities.

Rob: I consider the EU very important to that just because they're so big and they're a market-based democracy. Increasingly, almost as every week goes by, I'm more and more skeptical. I think the fact that the EU signed that investment deal with China was a slap in the face to the United States. Are you kidding? First of all, they're going to get screwed by that. There's no question that's a bad deal for them, because China won't have to live up to their end, and the EU will.

By the way, when we talk about Europe, we're really talking about Germany. That's the linchpin. The Germans seemed to be wanting to be a giant Switzerland and not take their global responsibility seriously other than making as much money as possible in the short run.

So I don't know whether we could get Europe to step up to their responsibilities, but if we can, it would be great. And if we do it, I wouldn't want to give up much. The Europeans would probably insist the United States assume their privacy policies, and that would be a disaster. It would lead to less innovation, you'd have to pay for websites, etc.

We have to be careful. I think we should do something along the lines of what Prime Minister Boris Johnson proposed, which is a D-10 coalition: commonwealth countries, India maybe, Japan. We'll start with the countries that have some fortitude, that are willing to stand up and say something. Then if Europe wants to come along, we'll help them. But I'm increasingly skeptical that Europe is going to be able to do it, although I hope they will.

Jordan: Before we move on, what do you say we do a little teaser to get the kids excited about Hirschman. Aside from his early book you mentioned, why do you think his work is still relevant today?

Rob: Hirschman had this long career and wrote many books on a variety of different topics. He wrote a book later called The Strategy of Economic Development, about how developing countries should grow their economies.

There's a joke: Two economists are walking down the street. One of them sees a twenty-dollar bill and walks right by it. The other guy says, "Why didn't you pick that up?" And the economist says, "If it was really there, somebody would have already picked it up." In other words, they live in a magical, almost theoretical world. Hirschman started all of his work from what was actually going on the ground––what was happening. Institutions were the key to his insights and analysis. Economists don't understand or don't think about institutions. Economists are, at the end of the day, glorified analysts of markets that have some prices to them. That's just a small component of what an economy is. Economies are much more about institutions, and Hirschman was really the guy who thought about institutions so much better than any other economist. That's why I like Hirschman so much.

Jordan: He also has a crazy biography: born in 1915 to a Jewish family in Berlin; decides to fight on behalf of the Spanish Republic in the civil war; flees to the US and ends up translating for the Nuremberg trial; lives in Latin America for over a decade.

The other thing that I think is good in our era is that he does not waste words. A lot of his books are 150 or 200 pages. They're the kind of books in which every paragraph teaches you something. You don't want to skim it because you're worried that you'll miss a little gem. There's also a kind of sense of humor in a fair amount of his work, which is something that's also very hard to find when you're talking about someone who's doing an excellent job at analyzing the ways institutions function. And English is like his fourth language! Just not fair.

The Importance of Focusing on Sectors

Jordan: Let's turn back to what national industrial policy should look like. What's the distinction between supporting factor conditions and key industries and technologies, and where do you strike the balance between investing in the two?

Rob: The big limitation of neoclassical or conventional economists is that they're indifferent to what's called sectoral mix. Michael Boskin famously said, "potato chips, computer chips, what's the difference?" The very first and most important principle for an industrial strategy is that you have to be focused on the sectoral composition of your economy.

In other words, would we care if the US economy ended up simply selling waste paper and plastic and tourism to China and importing robots and 5G equipment and semiconductors? Damn right we would care. The problem is a lot of economists don't care.

A colleague of mine once told me he was talking to the head of a big think tank in Washington, and he asked him how much manufacturing the US could lose and still be okay, and the guy said, "All of it. It doesn't make any difference." You can't have that attitude.

That's the first thing. We need to focus on sectors. Then the question is, what are the most important factors for each sector in terms of competing? You're writing something on semiconductors. We need to understand the semiconductor industry better. What are the key things that'll make us competitive? In some industries, that might be better skills and more early-stage R&D and a better tax credit. In other industries, it might be a more effective trade policy, or it might be a public-private partnership like SEMATECH. At the end of the day, it's boiling this all down to look at sectors, and then, in particular, how technology policy can affect their viability in the US.

Jordan: One of the surprising things that I've learned in my research in the past six months on this is the dearth of knowledge within the US government that doesn't come filtered through lobbyists. Why is this the case that there's a shocking paucity of expertise on these particular key strategic industries,

Rob: You're really not overstating it. It's almost mind-boggling how little expertise there is these days, and I think there are several reasons for that. One reason is, where would you learn that? You get all these young folks coming into the Biden administration coming from Harvard or Princeton or Yale ... they don't teach this kind of stuff there; it doesn't get taught. Even if you're in a public-policy program or an economics program, a foreign affairs program, international relations, you don't learn this. So that's number one.

Secondly, hardly any think tanks in Washington write about this stuff now, though increasingly they are because it's becoming cool.

Jordan: We're beginning to see the seeds of interest in this stuff, with the likes of Marco Rubio putting out a big bill, the CHIPS act and whatnot. But you see that there aren't grizzled civil-service veterans who can advise on this stuff. Instead, the United States is kind of reliant on corporations and outside expertise, when the whole idea of this industrial policy is not to necessarily support the incumbents, but to create better paradigms for domestic growth. It's unfortunate, and there needs to be an effort to better understand these questions.

Rob: Very much. And I'm more optimistic than I was five or ten years ago. You mentioned Senator Rubio, but there are multiple folks in the House and the Senate now. The House China Caucus had a really excellent report that talked about industrial policy and industrial strategy. Senator Todd Young, a Republican from Indiana, co-sponsored a bill with Senator Schumer on what's called the Endless Frontier Act: $100 billion to core technologies like quantum and AI. So some of it is happening, and they're talking to folks who know about industrial policy.

We're moving in the right direction. How much the Biden administration will really focus on this remains to be seen, because most of their priorities now appear to be social-policy oriented.

Big Is Beautiful and Winning Over the Left

Jordan: Let's talk about that for a little. You have some takes about how you're worried that the left wing of the Democratic Party is going to hijack this momentum and put it towards uses that won't end up making America an internationally competitive industrial machine. What are the traps that more left-wing democrats may fall into?

Rob: There are really four traps: The first is that a lot of the progressive wing of the Democratic party is really enamored with mom-and-pop small businesses. They don't think big companies are all that important. And yet, as Mike Lind and I wrote in a book a couple of years ago called Big Is Beautiful: Debunking the Myth of Small Business, that's just simply wrong. Big companies pay higher wages; they're more unionized; they spend more money on environmental protection. But the progressive left is increasingly demonizing large companies. You cannot be internationally competitive unless you have large companies as well as small and medium.

The second is they really want to import a Euro-style regulation system. Sure. We've long supported a national-privacy bill; for example, we're not libertarians, we're not anti-regulation, but you must do it the right way. They don't want to do it the right way.

Third, they increasingly want to have everything seen through the label of race or racial justice. Again, that's not to say that there aren't issues there, but when you spend all your time there, then increasingly big technology companies are going to be attacked for not employing enough racial minorities. Frankly, there are very few in the pipeline. If you want to deal with that issue, which we should, it's a pipeline issue.

The fourth is that the progressive left really wants to have an industrial policy, but for a lot of them, it would be only focused on green industry, on climate. That's a mistake. We have a clean-energy innovation program. We believe in it deeply. But you can't build a globally competitive international economy only around clean technology. You have to have aerospace, robots, semiconductors, the whole nine yards. So I worry that the whole thing is going to be hijacked. Everything is going to have to be related to clean tech.

What does the US Need to Do?

Jordan: Does the United States need a national-technology council on the level of the NSC or DPC, and why does the current state of Office of Science and Technology Policy (OSTP) just lead to it getting bullied around?

Rob: There are two things we need. First off, we need an agency––a national technology agency––partly, just as you've mentioned, for the research and analysis. The government has no analytical capability to figure this stuff out. You need an agency that has some of those capabilities, as well as some funding capabilities.

Within the White House, Biden has proposed bumping up the OSTP to cabinet-level status. That'll help, but it's not everything. At the end of the day, OSTP is 90 percent science and 10 percent technology. Zero percent is innovation and competitiveness. Maybe the NSC they could do that, but it's not really their thing.

So yes, I think a national technology competitors council would be a very good entity to add to the mix within the White House.

Jordan: Any last words, Rob?

Rob: I think the fact that we're doing this interview the day after President Biden's swearing-in, and just a really great day, I think the really critical question in the United States in the next four years is whether Jake Sullivan and other members of the cabinet will really take this seriously. I hope they do, my fingers are crossed.

If we don't do what we need to do, the window is closing quite fast. We have ten more years left, I believe. If we don't get it right, then we should just assume that we're going to be a second-rate power technologically and eventually militarily. We've got a decade to get it right, and I hope the Biden team really focuses on that.