AI and the Euro-Area Economy | Steven J. Davis, Philip Lane | Hoover Institution

Steven J. Davis (0:10)

Joining me now is Philip Lane. He's chief economist and member of the executive board of the European Central Bank.

Philip, thanks so much for joining us. It's great to have you.

Philip Lane (0:22)

I look forward to the conversation. Great to be here.

Steven J. Davis (0:24)

Okay.

So I wanted to chat with you about AI and how it's rolling out in Europe. And let me start with a slightly impertinent question, which is, why is Europe so far behind the United States in AI technologies and in the rollout of AI-related infrastructure?

Philip Lane (0:44)

So I think the consensus over the last 20 years is the productivity gap between the US and Europe is really coming from the tech sector. So that was there before AI.

And then AI is really amplifying that basic differential. And then I think when people look at that, of course, there's a lot of agglomeration externalities behind the centrality of Silicon Valley. And I think when you look at the Draghi report, other reports, we know some of the differences. Europe does have a single market, but the reality is across the 27 member states, there's a lot of differentiation, there's a lot of local regulation. So if you like, the scale economy is very important in AI.

And when you are confronted with 27 national markets in Europe, things will go more slowly, you know, less amplified than in the US economy. Okay.

Steven J. Davis (1:41)

So that makes sense. But I want to drill in on parts of it. The part about Silicon Valley is special and there's nothing quite like it in Europe. And that's why you see this concentration of say, the foundation models in the US rather than Europe. That all makes perfect sense. And I get your point about fragmentation too.

But as you described, Europe's been lagging behind the United States and the adoption of information technologies for a long time. It seems to me it's more than just the fragmentation because a lot of the implementation has to occur at the local level, at a firm level and so on. And Europe's behind in that respect too.

Philip Lane (2:20)

Well, I'm sure the measures will differ, but what I would say is we conduct ourselves another surveys. I do think outside of the tech sector, the differences are not so big. So we are seeing firms doing a lot of implementation, if you like, type investment. We're seeing a lot of workers reporting that they're using AI on a daily basis in the workplace. So I'm not so sure because remember, at the firm level, there's very big incentives to explore AI, adopt AI.

Europe has labor shortages.

Europe has a highly educated, there's a big high education level in the workforce. So when you look outside of the tech sector to the using sectors, I'm not so sure the differences are so big, except going back to this basic issue of scale, that essentially, if you have a big non-tech firm in America, who can roll out essentially an AI policy across all of the states, across many locations, there are fixed costs involved. And in Europe, directionally, I think it's similar, but maybe it's just more muted because of these scale issues.

Steven J. Davis (3:32)

The point that on the implementation side, Europe may not be that far behind the United States, outside the tech sector itself. So I take that point, but still, there are some big countries in Europe, Germany, big economy, France, Spain. They're big enough to take advantage of a lot of these scale economies. And I want to recount to you something that a partner to VC firm in Silicon Valley said to me.

The conversation was about European startups. He says, yeah, we look at European startups. And whenever we find a promising one, the first thing we try to do is get them to relocate to the United States. He went on to say, well, the regulatory environment is just much less friendly to scaling up. And it was a little distinct from your point about fragmentation.

Philip Lane (4:15)

But maybe this goes back to essentially, once you have a leading country like the US.

So I think the rate of firm creation in Europe in tech is similar. Lots of people set up a new firm. Yeah. And then, if you like, once it becomes a little bit successful, and then the question is to go to the next stage.

It is a pretty solid equilibrium that a lot of these firms do end up relocating to the US. But let me emphasize here, Ricardo Rice at the LSE has been working on this recently. There's some nuances there about moving the leadership to the US, maybe having still a lot of engineers across the European cities.