Bloomberg Surveillance TV: October 13th, 2026

Bloomberg Audio Studios.

Podcasts Radio News.

This is the Bloomberg Surveillance Podcast. I'm Jonathan Farrow, along with Lisa Abramowitz and Anne-Marie Hordern. Join us each day for insight from the best in markets, economics and geopolitics. From our global headquarters in New York City, we are live on Bloomberg Television weekday mornings from 6 to 9 a.m. Eastern.

Subscribe to the podcast on Apple, Spotify or anywhere else you listen. And as always, on the Bloomberg Terminal and the Bloomberg Business App. We begin this hour with stocks mixed as the banks set the tone for earnings season. Mike Wilson of Morgan Stanley writing, look to financials relative performance and bond volatility for signs the worst is over.

Mike joins us now for more. Mike, good morning. Good to see you.

Good morning.

Are you seeing signs the worst is over?

Yeah, well, we've seen the breadth. I mean, everybody's talked about this now, but I want to go back to June. That's really when this correction started, I would say, in the broader equity complex, and it was an internal correction at first. We call it the transition to mid-cycle. Some of the early-cycle stocks have really been hammered, like semiconductors, which now are having a second leg lower. Financials is another early-cycle group. So the market's on to this. I mean, it understands that financial conditions have really been tightening quite significantly since and the weaker parts of the market have underperformed.

So I like that. And by the way, the market understands all this, so it's digesting. The question that we don't know the answer to yet is, do we need a final kind of catch down in the S & P 500 to kind of finish this correction? And I would say it's sort of 50-50, but that would be something we would really embrace because we don't see any issues with growth.

What would suggest that we need that final catch down?

Well, as we wrote in our notes this week, I mean, the two things I'm really focused on, I'm focused on a lot of things, but those are two that stick out to me, is financials relative performance, because that's a leading indicator, and then bond volatility, because equity vol really has been suppressed, hasn't really caught up. So maybe we just need equity vol to catch up to where bond volatility has been, and then that's your final sort of clean out. I mean, I'm a trader by trade, right? So, I mean, this is how I look at the markets.

I'm probably different than some strategists. I'm kind of in the weeds on this. And so I like what I see. I like that setup. But I'm not convinced that it's time to really just come in with a big hand and say, OK, everything's clear.

I'm picking up on a couple of words you used for markets are digesting this. Some people might say markets haven't been digesting what's happening in the bond market. You push back against that beneath the surface. Where do you see that process going on?

Well, I mean, small cap stocks, low quality stocks, which has been our call since June. The early cycle groups have really underperformed. And even with the AI complex, I think this is healthy. We've gone from sort of the enablers outperforming to more of the adopters. And we are seeing that in our work, that companies are now... using this technology to expand margins. So that, to me, is encouraging.

That doesn't mean we're not going to have bumps along the way. And most recently, we've had more concerns about AI safety or some of the nimbyism concerns around the election. And let's not forget, we still have the midterm election in three weeks, and there's all this anxiety around that. And markets tend to get the most anxious before the results.

So I actually view the midterm elections, no matter who wins or what the outcome is, as a clearing event so the market can kind of look forward.

You don't think it's a catalyst for more potential policy volatility?

I don't because I don't think that there's going to be that big of a sweep where any real policy changes occur.

I'm talking about the executive branch, not Congress.

Well, I mean, that's definitely not going to happen because, I mean, and that's the thing. Even if the Democrats were to sweep Congress on both houses, I mean, Trump has veto power. And so it's going to be really hard to change policy in the short term.

You mentioned there could potentially be a catalyst for a catch-down. What would that be? Could it be something political like the midterms?

Well, I think it's happening. The market, again, is this cauldron of digestion. And so it's happening right now. And this is why people have gravitated to the S & P 500, because it is the highest quality index in the world. And so there's a reason for why this is all happening. Typically, okay, when you get these corrections that are, the breadth is this bad, you eventually get to the generals, as they call them, or the highest quality parts of the market. I like to use the analogy of a flood coming in. You go to the high ground, and eventually the flood gets to the top room, and then those come down. But to me, that would be positive. I mean, I would embrace that, because... Once again, I don't see any real concern about growth. We don't

have a growth issue from an earnings standpoint, I don't think, over the next six or 12 months.

Is a growth scare emerging in some of the parts of the market that might tap the brakes on economic activity?

Yeah, but that's what we want. I mean, the economy has been booming. I think we grew 8.5% in the third quarter nominally.

The problem is that things are too hard.

Yeah, they ran it too hot. That was our call last year. The market figured that out. The multiples went up. Now the multiples come down because financial conditions are tightening. And this is probably a good time to talk about what's going on with the Fed. So Warsh, I think, has had this view that I want the market to do the tightening for me. So what has he done?

They've raised rates 25 basis points. He hasn't told us what they're going to do. He's allowed the market to go to where the neutral rate is, but it's done it in a pretty orderly way. Now, Bonvol, you could say, has gotten a little disorderly, but not in a way where it's going to cause a growth problem. The last thing I'll say about this is for all the consternation around treasuries, Your friends in Europe, I think, have a bigger issue. So it's like the old hiker analogy, right?

You've only got to be faster than the slowest hiker in the market. And so I'm not worried about the U.S. fiscal sustainability issue when you have other problems around the world that are much more acute, number one. And I still think the U.S. economy, if we choose, okay, has the most flexibility in funding our deficit.

We heard something similar from John Gray. earlier on this morning.

Yes, absolutely.

But when you're talking about it, it sounds like, well, the U.S. is good because everyone else is bad right now. That's a reason why you potentially like America.

Well, what else are you going to buy? And that was my next question for George Sarvalos of Deutsche Bank. He's making the point that maybe people won't buy treasuries. What are you going to buy? Going to buy Europe right now? European debt with the difficulties that some of those countries have got? That's a big ask.

No, absolutely not, especially as they have a presidential election. And we've seen this time and time again with a place like France. They cannot get their house in order when it comes to fiscal discipline.

For the Federal Reserve, you mentioned it, Mike. We've heard repeatedly on this program that interest rates don't matter to the tech trade because so much of this stuff is rate insensitive. The stuff started to get me a little bit nervous. And Julian Emanuel, I've ever called, was talking about that earlier on this morning. Where do you come down on that? How rate sensitive is this build out?

Well, I think it's actually starting to happen now. So remember, it's not just about the rate level of treasuries, but actually credit spreads are now widening in some of these hyperscaler bonds. And so that's more of a supply issue, in my view, than a true credit issue, because these data centers haven't even been stood up yet. So it can't be a credit issue yet. because they're not even making interest payments. Now, there are some that are weaker than others. But my point is that a combination of rates and spreads widening, just like the Fed chair said, I'm going to let the market go to where it needs to go to calm things down. And that has so far been a pretty orderly process. We have seen

supply slow down a little bit because of that. So that's a good outcome for now. I anticipate there's going to be weak links, though. There will be credit issues within the build-out of this incredible CapEx cycle, but not to the point where it stops it. I mean, that's not where we are right now.

As you begin to firm up and form bigger ideas for next year, are you thinking more about the other parts of the market away from that tech trade?

I'm thinking about more of it as both. So, once again, I want to go back to what I was saying earlier. I think next year is about the adopters. So in other words, this technology is now at a point where it's actually useful. A year ago, we were talking about, oh, let's say ICAPEX. There's like three stocks that were kind of benefiting from a spending standpoint. And most of us were like fumbling around going like, I don't get it.

What is this thing? And all, of course, the tech savvy people were like, yeah, you don't understand. And they were right. And now we're seeing applications being built for dummies like me. You say, oh, I get it, whether it's Muse or Claude or whatever these things are. They're usable in a way where the average person can say, now I can make my life more productive. So that's what we think next year is going to be about.

And there's going to be more transitions. And that's our job is to figure out where these pockets of opportunity are going to be.

Where does that leave chips?

Well, I think we talked about this three months ago. We thought they were going to have a pretty good correction. They have because they are still somewhat of a commodity, and they are now going through that. And we have this analog with silver stocks, which people kind of laughed at three or four months ago. It's right on track.

So I think these stocks look like they're going to now make a new low. That also is part of the adjustment process. Let's not forget that they're also early cycle stocks. group like financials are an early cycle group so there's two things going on here one you have the capex cycle transitioning from enablers to adopters but then you also have this early to mid cycle transition from the economic standpoint which is classic you know sell like semi sell financials in the short term but now that's pretty well advanced okay so we're like we're getting to the point now where these things are probably starting to get attractive.

Again what's the mid cycle sweet spot Well, it's.

Really a factor. It's quality. So it works across sector. So even within semis, you're seeing the quality names do better than the low-quality names. Or within financials, you're seeing the higher-quality names do better than the lower-quality financials. So it's a factor call more than a sector call. I would say, you know, health care, industrial. And cap goods is an area we've focused on lately because the revisions there are still quite good. And those tend to be high ROIC businesses with very good backlog. So that's the first place we start looking back to an area that's been beat up. Financials is another one, quite frankly. The larger cap financials, which are down 20%, 30%. I mean, these are starting to show some value.

Stay with us. More Bloomberg Surveillance coming up after this. Chip stocks falling on risks of growing competition from China. The SOX Semi Index now on its longest losing streak since late July. Sarah Kunz of Clio Capital writing, everyone seems to be looking for a sell signal and maybe creating one. one out of nothing. We don't have shocking China news. We have linear expected progress and an incredibly jittery market. Sarah joins us now for more. Sarah, good morning.

Give us your response then. You alluded to it, your reaction to the potential competition coming out of China.

This competition isn't potential, right? We have known this. Remember the deep-seek freakout? That feels like it was a long time ago now. And this is exactly what we expected, right? We knew, we've known for years that one of the reasons that China likes to get their hands on the NVIDIA chips... is because they like a good reverse engineer.

And so none of this is surprising. It's coming about when it's been expected, maybe even a little bit slower. So the fact that it's immediately triggering a sell-off across Asia, potentially spreading into the U.S., that's not a great sign, right? These are not investors with diamond hands.

The scarcity issue, Sarah, I think that's the issue and it speaks to the jitteriness, jitteriness, jitteriness, whatever, that we're seeing in the market this morning. Sarah, the scarcity factor, there is a risk that you just switch to abundance at some point because everyone's doing the same thing. I can't think of a big tech company that isn't building their own chips right now. Sarah, can you? Isn't that the issue here? Isn't that why people are a little bit nervous?

So I've been saying for a while that I think NVIDIA in particular has a couple year head start. And this was, you know, two, three year head start. And this was like a year ago. And so, yeah, over time, that head start will narrow and there will be other good chips. And then the other thing to keep in mind is there's a huge focus right now to figure out how do we use less compute, a.k.a. how do we need less powerful chips, right? And so it's not surprising that that gap is narrowing. Now, there are a lot of other names that are not going to have as much of a.

Problem, right?

There are other pieces of the supply chain that are just going to be harder to replace, whether that's a Taiwan semiconductor or some of the other pieces along the way where there aren't as many competitors that seem like they're kind of a viable near-term, hey, we're going to launch something amazing.

Sarah, what's making me jittery this morning, I'm not sure if you saw it, but overnight, Anthropa came out with a blog post basically talking about how AI models are capable of self-replicating without access to a data center. And they can basically hijack your everyday uses if you have a smart oven or a nest in your home. What is going on with continuous warnings week after week for a company that is also preparing to IPO?

I mean, I think some of it is certainly the attention. Some of it is the regulatory capture they seem to want. Some of it's liability, right? If someone warns you repeatedly, hey, this thing is dangerous, it's a lot harder for you to turn around and sue them and say, I didn't know this was dangerous, right? And when you look at sort of the metas of the world, these landmark settlements that a lot of companies have had to pay out over very different kinds of safety issues, it makes sense to get in front of that. And then I think in particular at Anthropic, there are a lot of true believers who, you know, they're trying to convince the Pope that AI is alive, right? They see something there that

is probably not there and they want to sort of profitize and get more people on board to understand what they see as the potential upsides and downsides.

Do you think they're going to be forced to start talking about this even more differently following the midterms?

Because this is a huge election issue.

I think eventually we are going to see somebody like a hugging face say, hey, guys, it's your AI.

You owe us for this hack.

And I think that that is going to massively change kind of everything because we do not know what kind of damages a judge will award for something like a giant hack.

Sarah, I've been confused by this in the last couple of months, and I certainly don't pretend to be an expert on these things at all. I ask people like you about them to try and get some insight as to what's actually happening. The way these frontier labs talk about some of these issues, they talk about it as if these models have spontaneously taken on human traits of deception, and they don't seem to reflect on potential failures in engineering that got to this point. Sarah, isn't it an engineering failure to have models capable of doing these things?

It's a little bit of an engineering failure.

It's also, you know, a little bit how much of it is by design on some level, right? If you're building a cybersecurity AI, you might want to have some of these capabilities. It has generally played in their favor over the last few years to be able to point to their AI and say, wow, this is so powerful.

Right.

And so there's a question of how much of it is a failure, how much of it, what do we call it, a feature or a bug, right, in the tech industry. And there's a question of that. And then there's also a question of how soon do they think they will be held liable, right? We know that social media for decades now has gotten away with saying it's just our platform, it's not our fault. And so if that same sort of rationale holds with these AI companies, they're not really incentivized to put guardrails on. And then where there is certainly an engineering failure is on the other side. A lot of these hacks aren't happening because the AI is sort of all-knowing, all-seeing. They just go to billions of websites and brute force password is password,

and sometimes it works. And so that's a failure on the side of the sites getting hacked because they're not.

Doing a good job. Stay with us. More Bloomberg surveillance coming up after this. Under surveillance this morning, lingering fears of an AI bubble.

If that revenue increase that's expected to come in the trillions of dollars as a result of AI is not arriving quick enough, then that will certainly have implications for a lot of valuations today. The real nightmare scenario, if we just think into 2027, is now that the interest rate sensitive sectors of the economy begin to slow down.

And at the same time, AI begins to slow down.

So here's the latest this morning. Chip stocks falling for a second day after a tech breakthrough in China. The fears of growing competition adding worries to one of the most powerful AI trades of the year. George Sanovellos of Deutsche Bank writing, the biggest systemic risk in the market next year is something going wrong in the AI ecosystem.

Concentration risk is immense and it's dollar negative and very bond positive. George joins us now for more. George, welcome. Let's talk about that bet into 27. How are you thinking about things?

So I do think in 2027, it's all going to be about the return on AI. There's obviously a lot of themes in the market, the energy story, the situation in France and Europe.

But if we're thinking.

About the single most important driver of the market this year, it's been all these upgrades to growth expectations and CapEx, which in large part are relating to AI. So I think the next 12 to 18 months are all going to be about how the market assesses the CapEx payoff. And are we seeing that work out or could you potentially be in for downside surprises? And I think that's really going to be at the core of the market discussion next year.

Obviously, in your seat, you're exploring the relationship between that and the U.S.

Dollar.

George, how much has that theme been sucking capital into the United States? How dollar positive has it been up until now?

So, it's a great question.

And I think in the absence of AI, the dollar would have been a lot weaker. We've obviously seen this very large repricing in yield. How that's reflected in the balance of payments is a very large slowdown in fixed income inflow into the U.S.

Foreigners do not want to buy treasuries.

But of course, that has been compensated for by an even bigger increase in inflows into U.S. equity and capital markets. to finance the AI build-out. So I do think this completely changes the behavior of the dollar. If you're in a situation where something unwinds in the whole AI ecosystem.

The dollar would be a lot more vulnerable.

And as a result, it's no longer, I don't think it would exhibit the safe haven behavior we're used to seeing.

But something a lot more correlated to equities.

And this was a very similar dynamic in 2001 and 2002 as the dot-com bubble was unwinding.

George, do you expect those themes to be exacerbated by what might happen to rate differentials too?

I think they would be exacerbated in the sense that the U.S.

Fiscal position is a lot worse if you compare it to 2001 and 2002.

So, just to be very clear, it's not our baseline expectation.

But if we're thinking about the distribution of risks in the market, for example, the market is very focused on what's going on in France and energy in the near term. But I think structurally, the biggest asymmetric risk that's not priced is is this unwind of AI.

And to your question, John, the U.S.

Has become very, very leveraged and exposed to AI to finance its external deficit and, by extension, its fiscal deficit as well.

So, George, I didn't expect you to allude to that because initially in the quote I read, you said it might be bond positive. Now I'm wondering whether it is indeed bond positive if it goes the other way.

Well, I think you're going to enter– if the AI story is unwinding, you're going to enter into this debate of how much yield is potentially going to be influenced by effectively a weaker economic situation, potentially requiring even wider fiscal deficits against obviously a very large.

Repricing in Fed funds.

At the moment, the market is pricing the inflationary consequences of AI and not the opposite. So I think there will be a big debate in terms of what dominates. We probably have a much stronger view on the shape of the curve. So in an environment where AI is unwinding, I think you would expect to see very, very substantial steepening of the curve and potentially an increase in term premium in the background, as ultimately the U.S.

's ability to attract capital, I think, would be worse.

That's where I want to go, George.

Yeah.

Let's just sit on that. It's really important. So the front end's taken care of. I get it. If it's a risk to the economy, the Fed's not going to be hiking as much. The front end is anchored. The long end is interesting to me. You mentioned what might happen to the fiscal position. Are you suggesting, George, that if there's a hiccup in the AI trade in 27, the fiscal position comes into question a little bit more and we can't attract the capital at the long end of the curve in the U.S., even in an economic downturn?

So we have to think about why is the market not potentially penalizing treasuries even more. If you look at the shifting yields globally, it's been relatively symmetric, even though the U.S.

Is a much worse fiscal position.

And I think the reason is the market is expecting the ultimate payoff of AI to come.

And what would that mean?

It would mean high growth in the U.S., higher ability to pay off deficits. So, if we're in a situation where for some reason that gets revised lower, whether it's lower adoption rates or some other shift in technology, we didn't discuss politics at all. But of course, there's a very large discussion around regulation and we have the midterms coming in.

So, it could be any trigger.

But if that happens, by extension, the fiscal deficit numbers, the debt projections in the U.S.

Would look worse.

And of course, it also depends on what the U.S. fiscal reaction would be.

It may be that the U.S.

Decides to do nothing, but it may be the government decides to step in to manage some of the associated disruption. So, I do think it's quite ambivalent what the ultimate impact would.

Be on term premium.

This is the Bloomberg Surveillance Podcast, bringing you the best in markets, economics and geopolitics. You can watch the show live on Bloomberg TV, weekday mornings from 6am to 9am Eastern. Subscribe to the podcast on Apple, Spotify or anywhere else you listen. And as always, on the Bloomberg Terminal and the Bloomberg Business Hour.

Bloomberg Surveillance TV: October 13th, 2026

Kapitelklicken springt direkt hin

Shownotes

Transkriptvom Podcast bereitgestellt · Zeitmarken springen hin