Economic Roundtable: Structural Fallouts From the Iran Conflict

Our Global Chief Economist Seth Carpenter concludes the two-part discussion with chief regional economists Michael Gapen, Jens Eisenschmidt and Chetan Ahya on the second order effects of the energy shock from tensions in the Middle East.Read more insights from Morgan Stanley.----- Transcript -----Seth Carpenter: Welcome to Thoughts in the Market. I'm Seth Carpenter, Morgan Stanley's Global Chief Economist and Head of Macro Research. And once again, I am joined by Morgan Stanley's chief regional economists: Michael Gapen, Chief U.S. Economist, Chetan Ahya, the Chief Asia Economist, and Jens Eisenschmidt, our Chief Europe Economist. Yesterday we focused on the immediate impact of the Iran conflict, how the energy shock is feeding through into inflation, and, as a result, shaping central bank decisions across the U.S., Europe, and Asia.Today we're going to go a level deeper and talk about some structural issues in the global economy. It's Wednesday, April 15th at 10am in New York. Jens Eisenschmidt: And 3pm in London. Chetan Ahya: And 10pm in Hong Kong. Seth Carpenter: So, even as we're waiting to see whether or not oil prices stabilize following a temporary ceasefire – or not – the broader effects are still working their way through the global economy. Labor markets, supply chains, and then, of course, back to the more longer-term structural themes like AI driven growth. So, the question, I think, has to be: what does this shock mean, if anything, for the next phase of global growth? And does it reshape it? Does it change it, or do we just wait for things to go through? Mike, let me come to you first. One risk that we've been focusing on is whether this kind of shock really changes some of the structural positives in the U.S. economy. The U.S. has been, I would say, outperforming in lots of ways. We've had this AI driven CapEx cycle. We've had rising productivity; we've had strong consumer spending. What are you seeing in the data about those more structural trends? Michael Gapen: I think what we're seeing in the data right now is evidence that oil is not disrupting the positive structural trends in the U.S. I think AI CapEx spending is largely orthogonal to what we've seen so far. It doesn't mean that we can't see negative effects, particularly if oil rises to say $150 a barrel or more where we think you might see significant demand destruction. But with oil where it is right now, I would say the evidence is it will probably weigh on consumption. Gasoline prices are higher. It's going to squeeze lower- and middle-income households that way. But so far, the labor market appears to be holding up. And business spending around CapEx seems to be holding up. And the productivity story remains in place. So right now, I'd say this is more of a break on consumer spending, maybe a modest headwind. But not an outright hard stop. And I think those positive structural elements and AI-related CapEx spending are going to stay with us in 2026. Seth Carpenter: I hear in your answer part of what for me is always the most uncomfortable part of these conversations. Where I have to come back to say, ‘But of course it depends on how things evolve…' Michael Gapen: Of course, It depends… Seth Carpenter: So, then let me push you on AI specifically. You and your team have published a few pieces recently about AI. How AI is affecting the labor market, and maybe some hints as to how AI is likely to affect the labor market. So how should we think about that? Michael Gapen: While it's still too early, I think, to draw firm conclusions, Seth, we do find that there's some evidence that AI is pushing unemployment rates higher in specific occupations that are exposed to task replacement. So, what we did do is we broke down the data by occupation, and it's clear that the unemployment rate has been rising. But that's just a general feature of the economy at this point in time. Over the last 18 to 24 months, the unemployment rate has gone higher. So, what we did is a second-round effort at kind of controlling for cyclicality. And when you control for those, we do find evidence that the unemployment rate for occupations that have high exposure to AI is higher than you would expect, given the cyclical performance of the economy. But the effect is really small. It's maybe about 1/10th on the unemployment rate. So, I don't want to be too Pollyannish and say, ‘Oh, there's no evidence here that AI is disrupting the labor market.’ We'd say that there is some evidence there. But, so far, it's mild and it's modest. It's a little more micro than it is macro. So, we'll see how this evolves. But that would be our initial conclusion so far. Seth Carpenter: So, Mike, that's super helpful. When I think about the AI investment cycle, though, I have to come back to Asia because a lot of the AI supply chain is there in Asia, especially with semiconductors and others. But there's lots of supply chain ...