Like The Hub?
Join our community.

Robert Atkinson discusses the value of big business and why we need to embrace an industrial policy

Podcast & Video

Today’s Hub Dialogue is with the founder and president of the U.S.-based Information Technology & Innovation Foundation, Robert Atkinson. Atkinson, who was born in Calgary, is one of Washington’s leading policy thinkers on innovation and technology, including how these issues interact with geopolitics and political economy. 

In 2018, he co-authored a book with Michael Lind entitled Big is Beautiful: Debunking the Myth of Small Business. He recently published a paper with the Macdonald-Laurier Institute entitled Big is Beautiful: Strengthening Growth and Competitiveness in the Canadian Economy which draws on some of the book’s insights and analysis for the Canadian context. 

This conversation has been revised and edited for length and clarity.

The case for big business

SEAN SPEER: I’m honoured to speak with Robert Atkinson, whose thinking and writing on innovation, technology, and industrial policy have come to shape my own work. Thanks for joining us, Rob.

ROBERT ATKINSON: Thanks for having me. 

SEAN SPEER: In your book with Michael Lind and your recent paper for the Macdonald-Laurier Institute you take on what, in an age of populism, is a pretty controversial idea. That is the case that big business is important to our economy and society, and something that ought to be valued by policymakers and the general public. Could you unpack why large corporations are actually crucial to innovation, productivity, and the general health of a domestic economy?

ROBERT ATKINSON: Well, thank you, Sean, it’s pleasure to be with you. 

If you think about all the goals that most of us, including progressives, assert, like better working conditions, more equity for disadvantaged groups, higher wages, cleaner environment, etc., on average in the U.S. and Canada, big businesses outperform small businesses on these fronts. 

These are averages—so there will be outliers on both sides—but the averages speak to the benefit of large corporations. In Canada, large companies pay about 42 to 50 percent higher in wages; they’re more likely to employ women; they have much better benefits; they have much better working conditions; workers are more satisfied working at large companies; they have higher productivity. From the book that Mike and I did, they also spend more on the environment; they spend more on social responsibility than mid-sized companies and about the same as small firms; they’re more unionized in Canada and the U.S. 

You go down the line and the evidence is pretty striking. Big companies are the way we achieve progressive ends, and the idea that progressives are opposed to them is, frankly, bizarre.

SEAN SPEER: One of the challenges that we face in Canada, which is a bit counterintuitive in light of our conversation, is that we actually don’t have enough large firms. What factors explain why we don’t have many Canadian businesses scale into large firms? 

ROBERT ATKINSON: There are two big reasons why scale matters so much more now. One is that information technology enables scale, and I don’t just mean internet platform companies. I also mean other parts of the economy like banking and insurance—although Canada has big banks, it underperforms in most other technology-enabled sectors. When you get that scale, you get much better productivity and lower costs, and IT has both enabled and led to even greater returns to scale as a result. 

Big companies are the way we achieve progressive ends.

But the other key factor is global competitiveness. It’s really hard to be globally competitive unless you have scale. The only alternative is to be a really niche player where there just isn’t a giant market and there aren’t very many players. Canada has a hard time with that because even though you used to have NAFTA and now we have the USMCA, Canadian firms still don’t take advantage of North American integration to scale as quickly as American firms do. Remember the Canadian market is approximately about 10 to 12 percent of the U.S. market. It’s hard for firms to scale merely serving the domestic market because there’s just not that much market to build from.

SEAN SPEER: In your work and in this conversation, you’ve made the case for why big businesses are important to the economy. What would you say, though, to those who might argue that the same companies seem to be increasingly prone to what one might describe as “woke capitalism”? This is something, of course, that you’re seeing a lot of criticism in the U.S. and elsewhere. 

Is it something that we ought to be concerned about? And if so, is there a role for public policy to address what is perceived as a perverse combination of market power on one hand, and then an institutionalized ideological worldview, on the other hand?

ROBERT ATKINSON: I think the way to think about that is, what’s the alternative? Should we break up big companies, because conservatives think they’re too woke? That’s called cutting off your nose to spite your face. We’d be less competitive and we’d all have lower living standards. 

The reason companies do that, to the extent they do, is because they think it’s in their interests. So, to the extent conservatives don’t like it, they should do exactly what progressives do, which is to mobilize against companies that are too woke. 

Why do these big companies go out of their way to align themselves with some really left-wing causes? It’s because they think that they’re going to get all the market benefits, and they’re now going to get a pass from possible left-wing critics. The progressives won’t attack them. They’ll attack somebody else. And the rest of the public, the rest of the consumers, will ignore it.

I don’t think there’s a role for government to say that companies cannot be woke or something. But I do think if conservatives or centre-right people don’t like it, then they should mobilize against it. I know one organization in the U.S. for instance that is trying to do that. They are trying to raise awareness for people on the right or the centre-right to say, “Hey, there are these companies that appear to be pretty liberal. Maybe you should spend your consumer dollars with somebody else.” That strikes me as a market response to the perception that some companies are becoming too institutionally left-wing. 

A Canadian DARPA

SEAN SPEER: Rob, in addition to publishing this book on the role of big business in the economy and society in Canada and the U.S., you’re also a leading thinker and scholar on innovation and industrial policy. I’d be remiss if I didn’t take a chance to ask you a couple of questions along those lines. 

In the last Canadian election campaign, we had competing promises from the two major parties, the Liberals and Conservatives, to establish a Canadian version of DARPA. What advice or thoughts might you have for Canadian policymakers? What are the lessons to be learned in terms of standing up a Canadian version of DARPA?

ROBERT ATKINSON: What that argument is about is not necessarily whether you go with a specific DARPA model or not, but it’s really that the government should have a stronger role in supporting applied research that’s oriented to the critical industries and technologies that Canada has to be strong in to survive in this world. That’s really what the debate is about. 

Once you’re on the other side of that Rubicon, then the question is much more about whether it’s a DARPA model or more what we used to have in the U.S. called the Advanced Technology Program. You get into very narrow technocratic issues about design, governance, and so on. 

But, overall, the reason why DARPA succeeded, or one of the multiple reasons, is because DARPA oftentimes has a customer at the end of the pipeline. The agency would help with these firms or these network of firms with these particular technologies, with the idea that they would then transition into a product that would be bought by the Department of Defense, and eventually a lot of other buyers as well. 

I think the question for Canada is whether defence procurement in Canada is big enough to enable that process. That would be the key question I would look at if I was a policymaker advising the government. If not, what other areas might be better fits for this applied research model in Canada? We have ARPA-E in the U.S. for energy innovation; President Biden wants to create ARPA-H for health. We have HSARPA for homeland security. Everybody loves ARPA models. Boris Johnson wants to do something like that in the U.K. They should because it is a good idea. But the key is to think about the end-user. 

The other thing about DARPA, which is important and a lot of people don’t understand, is that DARPA is not just about the money. It’s also about the signaling. It’s about that these companies can then say, “We’ve been vetted by these engineers and other people at DARPA.” They can go out into the private venture market and access private capital because of my DARPA involvement. It’s more than just money.

SEAN SPEER: One issue that has been raised in the context of this policy promise to establish a DARPA-like organization or agency in Canada, is whether public policy ought to be prioritizing incremental innovation, or if it should tilt in favor of radical innovation. Do you have any advice on how policymakers should think about the balance between an innovation policy framework oriented towards incremental innovation and one that pays more attention to radical innovation?

ROBERT ATKINSON: There are several different debates as you say. One is the debate about whether you should be putting money into basic science or a little bit later in the process—namely, applied or late applied? A lot of people say, “Well, basic science has a bigger benefit because it’s contributing to more breakthroughs.” I don’t buy that anymore. I think the notion that we could just invest in basic science is wrong. Most of the benefits of that spillover to other countries, particularly for a small country like Canada. That’s why a lot of our major competitors, like Korea and Taiwan, are investing in applied research where they can capture more of the benefits. 

I think the notion that we could just invest in basic science is wrong.

The fact that U.S. companies—I don’t know the data on Canadian companies but I assume it’s similar to U.S. companies—are spending less on breakthrough innovations reflects a market failure. First of all, you’re taking a big risk. And secondly, it might not happen for 10 years, and yet the stock market is saying, “We want returns right now.” 

So, I’d be tilting towards more radical innovations, and really helping Canadian companies do that because I think that’s where the biggest market failure is right now. But, again, this comes with the caveat that policymakers need to keep the ultimate goal in mind, which is how do you get commercialization of technology in the Canadian marketplace to create strong companies that are going to employ high wage workers in Canada in a sustainable way? That’s the lens through which these decisions about DARPA and radical innovation need to be viewed. 

The case for industrial policy

SEAN SPEER: You’ve been, in your thinking and writing, an advocate for what one might describe as an industrial policy for some time. If you’re in the game long enough, you can start to see your ideas gain resonance. Last year at the National Conservatism conference, something like three-quarters of the attendees endorsed the principle of industrial policy. We’re now seeing U.S. senators, including on the left and the right, talking about the need for industrial policy. 

First of all, why do you think an industrial policy framework is especially important today? Secondly, speak to its critics. How can you design an industrial policy that mitigates some of the political economy risks that have bedeviled a more activist economic policy in the past?

ROBERT ATKINSON: So, on the first point, at the broadest level, assuming you’re for growth, which you can’t even assume that anymore because there’s a lot of progressives who now reject growth, but assuming you’re for growth, then growth policy, which is the standard frame people like, is indifferent between traded and non-traded sectors to grow the economy. If you go down a level to the question of competitiveness, then you get into competitiveness policy which tends to be about general, neutral levers of competitiveness. But even that’s not enough. You have to go down another level more to focus on sectors and technologies that are high-value and strategic, and that’s where you get into industrial policy. 

The reason for an industrial policy is because the idea that somehow the U.S. and Canada can be indifferent and just say “It really doesn’t matter whether the U.S. has the capability to make semiconductors; we’ll just buy them from somewhere else,” is highly risky. Or that the U.S. and Canada don’t have to worry about drug production. That’s really risky too. Imagine if we’re in a conflict with China, even if it’s a heightened Cold War, and they invade Taiwan, we are now incredibly dependent. You just cannot do that anymore. That worked in this vision of global harmony but that’s not the world we live in. So, we have to think about that and how best to protect our interests. 

There really are some strategic sectors that we cannot give up. We can do that through allies, so maybe Canada focuses on one or two things and we focus on six or seven. That’s great. But as allies, we have to think about that. There’s room for more regionalized or even value-based supply chain arrangements. 

But let me come to your second question. What are the challenges in an industrial policy framework politically? Well, I think if you look at what we’re doing now, we don’t have a formal industrial policy and it’s politicized because there’s no guidance to it. There’s no analytical frame that says, “This is why you want to do this and not do that.” So, it comes all in terms of lobbying power and pork-barrel politics. So, I’d argue that an industrial policy framework—especially if you tie it to national security needs—can actually mitigate the tendency towards politicization. Of course, everything is politicized because everything is politics, but I think national security technology, if you put it through that lens, the risk of politicization there is less. That’s also the case if you design the programs themselves in a way that’s insulated. 

A good example of that is when I used to be at the National Institute of Standards and Technology, and it ran a program called the Advanced Technology Group. It was designed in a way where the decisions were not made by politicians. It was all peer-reviewed. You had to have industry match, and the industry had skin in the game. All the evaluations the program sent—it really wasn’t politicized, and it did a pretty good job. It’s something we’re going to have to do, and we’re going to have to figure out how to do it well. Some other countries have done it well. Some countries have done it badly because they haven’t set those programs up to be successful or it’s been too politicized.

SEAN SPEER: Well, it’s been a fascinating conversation. Your recent book and your MLI paper are both called Big is Beautiful. We’re grateful for the chance to pick your brain on these issues. We’d love to have you back, particularly as we learn more about the federal government’s plan to bring expression to this commitment for a new Canadian Advanced Research Projects Agency sometime in 2022. Thank you so much.

ROBERT ATKINSON: Thank you. I’d be happy to talk again. Thanks for having me.