KATHY JONES: I'm Kathy Jones.

LIZ ANN SONDERS: And I'm Liz Ann Sonders.

KATHY: And this is On Investing, an original podcast from Charles Schwab. Each week, we analyze what's happening in the markets and discuss how it might affect your investments.

Well, hi, Liz Ann. So last week, we started out talking about tariffs. But I think this week we have to talk about inflation. We just got the Consumer Price Index (CPI), which was not great news for the markets. What was your takeaway relative to the equity market?

LIZ ANN: Yeah, I mean, it was not good news, and the equity market certainly reflected that. You and I are taping this still in the a.m. hours. So a lot can happen through the course of a trading day. But it is the case that in the past, when we've had disappointing CPI reports and the market has opened down, there's less of a tendency for it to recover throughout the day. So, you know, again, we're taping this still in the a.m. But all four of the main reports that come out of CPI: so the headline reading year-over-year, the headline reading month-over-month, the core reading, which is excluding food and energy, both month-over-month and year-over-year, all were hotter than expectations. So it certainly changed quickly perspective around what the Fed is going to do in terms of market expectations, pushed a possibility of another rate cut way to the end of the year. It certainly takes rate cuts off the table anytime soon. What's also interesting, and this is, I think, something that will continue to be done by economists and something we'll continue to look at is breaking CPI into the services component and the goods component because it's the goods component that tends to be impacted most by tariffs. 

And there would have been good news in this report were it not for the pickup on the goods side because the services components actually trended down a little bit. You didn't get another big spike in rents. It wasn't explained by a spike in medical costs, but you did see that goods piece pick up. And I think these reports will increasingly be important, not just in general and for the perspective of the Fed, but as it relates to tariffs and how we're starting to see it infiltrate into the numbers, which is why even though we're taping this on CPI day, tomorrow we get the Producer Price Index.

And I think there might be some interesting nuggets in there that might reflect some of what we're starting to see on tariffs. What about you, Kathy?

KATHY: Yeah, you know, just anecdotally, the building where I live is looking at replacing all of the windows, which are like 50 years old. And one of the issues, they've been holding off because aluminum prices went up the last time around, and that fed into the cost. And now the quotes they're getting include a 25% increase due to tariffs on the cost. And you know, it's just going to ripple through unless something dramatically changes, those sorts of price increases are going to ripple through, and it's going to make it more difficult for consumers, particularly those looking for goods for those home projects or automobiles or you name it, you know, it's going to ripple through unless something changes. 

So not very optimistic on the inflation prospects, and I do see that in the bond market, of course, that market took it pretty hard because it really does tie the Fed's hands. So at this stage of the game, current inflation is running too hot and, the momentum is edging higher, not lower, which is a problem. And then on top of it, you have the prospect of tariffs possibly leading to price increases down the road. So I'm not surprised that the bond market is pretty much now down to maybe one rate cut way at the end of the year in December, which is as good as saying they're not expecting any rate cuts anytime soon or even this year. So it doesn't surprise me. And I think that what we are looking at, unless things change, and things are changing very rapidly out of Washington, but unless things change pretty quickly, I think we're looking at 10-year yields testing that 5% region again.

LIZ ANN: You know what's interesting too, Kathy, I just got this question this morning by email about … more of a comment than a question, I should say. And it was why anybody is suggesting that tariffs can explain part of the hotter inflation—or that they will—clearly weren't paying attention to what was going on in 2018 when there was no inflation that erupted when we were in the midst of the trade war. And my reply back was, "Well, they were much smaller. It was pretty much just targeting China." They were a lower percentage, and, that said, if you took PCE (the Personal Consumption Expenditures) Index, which is the Fed's preferred measure for inflation, and you looked at just the overall change, yes, that generally continued to trend lower, but if you pull out the goods on which tariffs were applied, the inflation rate on those went straight up. It just wasn't enough to bring overall inflation up. Plus, the backdrop in 2018 from a macro perspective was very different than where we are right now. But to suggest that tariffs didn't play a part, they did in the goods that had tariffs applied to them. With the washing machine story often being the proxy and the way to describe what happened with tariffs in terms of an increase in prices, a decline in sales. So there was an effect in 2018. It was just a bit more minor because the scope and the percentage of tariffs were fairly low compared to what we're experiencing now.

KATHY: Yeah, there are a lot of carve outs as well in subsidies that were paid. So offsetting factors, and you know, I think a lot of people aren't very closely in touch with the agricultural community, but those tariffs hurt a lot because they lost market share, and the U.S. agricultural industry worked for decades to get access to the Chinese market for soybeans and corn and wheat and sorghum and all these crops. And then the tariffs kind of said, "Oh well, you're not a reliable supplier. We'll go elsewhere." And they never fully recovered from that. They recovered some. But China buys a lot now from Argentina and Brazil and other countries because it turned out that we were not seen as reliable suppliers. And I think the agricultural community is facing this again now. 

LIZ ANN: They're not only facing it with tariffs, but the immigration piece of it too.

KATHY: Yeah, exactly. There are consequences for these things. And, you know, we'll just have to see whether this is more bark than bite, as some people think, or whether, you know, this is going to have a lasting impact. But for right now, you know, the market's focus is on current inflation, and it's running too hot. And that means the Fed's on hold. And that means I think we're looking at higher yields for a while.

Speaking of China, our guest today is coming to us from Hong Kong, and he has some really interesting insights on all this news about tariffs and technology related to trade conflicts like DeepSeek and AI. So can you tell us more about him?

LIZ ANN: Yeah, this is an important topic, and it did sort of lose the front-page status in light of a lot of this tariff news, but I think what the news out of DeepSeek, a quant hedge fund, small hedge fund in China, figuring out how to create a large language model, basically a competitor to ChatGPT at a fraction of the cost that is represented—the expenditures by many of the major players in the United States—is fascinating. And now as you relate it to tariffs as well, it's important because when you think about the tariffs on imports from China, but also some of what the retaliation announcements have been by China, and my guess is that we're not going to see that dissipate anytime soon.

It's important in the context, not just of AI specifically, but the tech sector more broadly. If you look at all 11 sectors in the S&P 500®, the tech sector by far has the highest percentage of top-line growth or sales and revenues that come specifically from China. So I think this will continue to be an important story, especially in a backdrop like we're seeing this year where, if you look at the Magnificent Seven group of stocks that are sort of considered AI or AI-related basket, much more dispersion in terms of returns. And I think this is also a year, so far, where the actual fundamentals of companies are driving stock prices versus, say, a year like last year where a lot of it was broader monetary-policy driven or valuation expansion. 

So I think this will continue to be an important topic, and it is one of the things that I spent some time talking to with our guest on this episode. It's also the case, by the way, that the sit down that I had with our guest, it'll be a week prior to when this episode drops. So you might hear a reference to the Super Bowl and think, "Boy, that's in the rear view mirror." Just there was a week lag. So keep that in mind. 

So our guest is Louis-Vincent Gave. He is the founding partner and chief executive officer of Gavekal Research. Louis received his bachelor's degree from Duke University, studied Mandarin at Nanjing University. Louis joined the French army, where he served as a second lieutenant in a mountain infantry battalion. Then after he left the army, he joined Paribas, where he worked as a financial analyst, first in Paris, then in Hong Kong. He left Paribas in 1998 to launch Gavekal with his father, Charles Gave, and Anatole Kaletsky. The idea at the time was that Asia was set to become an ever more important factor in global growth and that consequently Gavekal needed to offer its clients more information, more ideas relating to Asia. 

Louis is also the author of seven books, the latest being Avoiding the Punch: Investing in Uncertain Times, which reviews how to build a portfolio at a time of rising geostrategic strife. Obviously very topical. Louis speaks English and French. I don't speak French. I speak a little bit, but not enough. We therefore conducted the interview in English, and he is joining us today from Hong Kong. 

LIZ ANN: Well, Louie, I'm so thrilled to have you at least on camera, although our listeners can't see you on camera. We have been a consumer of Gavekal's research for a long time. I'm an avid reader of everything you and your colleagues put out. So it's a real treat, especially given that … I'm taping this on Friday. You're taping this on Saturday, given the time difference coming from Hong Kong. So a double thanks for the time of day or morning that we're doing this for you.

LOUIS: Are you kidding? It's such an honor, and I'm delighted to be spending this time with you. And thanks so much for these kind words. It really means a lot, especially coming from you.

LIZ ANN: Well, thank you. Thank you. So a couple of recent reports that really resonated with me, and the good news for our listeners is that we will post links in the show notes to these couple of reports that I want to start with and reference, and the first one you wrote about a month ago, called "Building Narratives Around Obvious Outliers." And I think one of the things you put right up front that I think is a great sort of starter part of the conversation here is, "Learning to stay humble since no two cycles are ever the same. In any given year, we are offered the opportunity to say, 'See, here is something we have never seen before.'" And then you went on to highlight some of the incredibly unique things that have happened in the past year. So with that as a first toss in a very broad sense, what has stood out to you most in terms of how unique this past year has been?

LOUIS: The most obvious thing, the question, I think, that every investor has to confront is both the level of concentration, of course, in the U.S. equity market, number one. Not only is it a concentration by market cap, but the concentration of the sectors really, to have all top 10 stocks essentially be derivative of tech one way or another is somewhat unique. To have them all go up double digits or triple digits in a given year is somewhat unique. Especially to have them go up so strongly when the rest of the world really didn't do all that much is somewhat unique. So you know, I think if you look at last year, the numbers are in the report. I'm quoting from the top of my head. But let's say your large-cap tech in the U.S., I think, was up 35% last year. Small-cap tech was down 1%.

If you look at the top 10 stocks in the U.S., like I said, they were all up double or triple digits. You'd look at the top 10 stocks in Europe, you only had two that were up. Eight were actually down. So these kinds of divergences are pretty unprecedented. And then you have to get your head around it. Now, the narrative in the market is that we are living through an era of U.S. exceptionalism, that these enormous companies, the Microsofts and the Alphabets and the Facebooks, etc., are now so big as to dominate capitalism not just today but for years to come. Now in a sense this is a little counterintuitive because the whole history of capitalism tells us that the bigger you get, usually the less efficient you become, you know, as you get big, you get fat, you get lazy, you get stupid. And this is why if you look through history, on any 10-year period, the top 10 stocks are never the same. It sort of always evolves. You might have one or two that stay in cycle after cycle, and Microsoft is a perfect example of this. It's been in the top 10 now for 30 or 40 years, but it's more the exception that confirms the rule. So it's an unprecedented time where all of a sudden we decide big is efficient, and it's going to stay like this forever because if it doesn't stay like this for at least the next decade, then these valuations don't make sense.

LIZ ANN: And of course it also has meant quite a chasm between—however you define it, index-wise or factor-wise—between growth and value. So talk a little bit about that too, how that broadens out to that relationship and how unique the divergence has been.

LOUIS: Oh yes, absolutely. Sorry, it wasn't just big versus small. I should have mentioned, as you pointed out, growth versus value, which has been also on an unprecedented divergence. And here I think we can look back to 2007, 2008, which was really the start of this U.S. growth bull market that we seem to be in. From 2007 to today, you've had only two years in which U.S. growth stocks have underperformed value stocks in the rest of the world. It's been an unprecedented run. And I would say up until recently, it was behaving as a fairly typical bull market, where 2007 to 2015, you have pretty steady outperformance. 2016, a bit of consolidation, a little mild underperformance. 2017 to 2020, the market's outperformance accelerates. And then 2020 and '21, it goes parabolic. It goes the U.S. growth outperformance in 2021 was through the roof. At that point you're like, "OK, this is a typical bull market, sort of steady Eddie, accelerates and then goes into a blow-off phase." And then '22, big correction. And at that point you think, "OK, this has behaved like every bull market pretty much that I've ever seen in my career." 

What was very unique in the past two years is, after this correction, a huge rally in growth stocks again, that not only took out the previous highs but just blew past them. But in that growth bull market, it was extremely, extremely concentrated around a handful of stocks. And it was really driven by all the excitement around artificial intelligence. I think what we've seen in the past couple of years, late '22, you see the release of ChatGPT, everybody realizes, "OK, artificial intelligence is going to change the world. And to get a seat at the artificial intelligence table, you need to be able to commit to 50, 70, 100 billion dollars of capital spending every year." And so you look around the room and you think, "OK, well, there's only a handful of companies that can sit at that table." Not even Tencent, not even Alibaba. These guys are out. They can't play at the big boys' poker table. And so these handful of companies go absolutely through the roof because again all of a sudden we have a perception shift, I think, in capitalism, where well before being big means that, over time, lower and lower marginal returns on invested capital because that's what happens when you're big. All of a sudden there's this perception, with AI, you're going to get higher returns on invested capital, and it's going to accumulate into just a handful of names.

And so you have this huge bull market driven by, you know, as we said, you know, 10 names for all intents and purposes. But this is where we might be living a change right now, because this whole belief that if you don't spend $100 billion, you don't get a seat at the poker table, that belief has just been blown out of the water by what's happened in China in the past three weeks.

LIZ ANN: And I have a bunch of questions that I want to ask about DeepSeek and AI because it's so top of mind. But let's put that aside for just a couple of minutes because I wanted to also mention some of the other unique aspects of the last year that you wrote about—the fourth consecutive negative year for long bonds, the fact that you saw unbelievable growth in the crypto space in terms of market cap, the incredible rally in Chinese bonds, and then also gold's outperformance. So you don't have to take them all individually. But it wasn't just within the U.S. equity asset class that a lot of unique things happened. It was across the spectrum of asset classes.

LOUIS: Absolutely. And to be honest, the things you listed, I think, are partly tied to actually what's been happening in China. And for me, one of my starting points, when I look around the world, is one of the biggest anomalies in our global economy today is the size of the Chinese trade surplus. Up until three, four years ago, the Chinese trade surplus was $200 or $300 billion a year, which was already massive.

Like, to be very clear, $300 billion trade surplus is already as big as any trade surplus the world had ever seen. Today, the Chinese trade surplus is probably running right now on a pace of $1.1 trillion. Trillion with a T. It's just gargantuan. Logically, what should be happening is that the guys making shoes for Nike or the guys making the new cars that are being sold all over the emerging markets, as they make money, the money should be coming back into China. It should be pushing up the value of the renminbi. It should be pushing up the value of local asset prices—i.e., the guy making money decides to buy a new house or decides to buy stocks or whatever, it should be pushing up local asset prices, and with that you should be getting a boost in Chinese consumption, and then the trade balance in China would come back down. 

This is how the system sort of mends itself normally. What's happening today, I think, is that both the consumer and the entrepreneur's confidence in China has been completely, completely destroyed. It's been shot to pieces. It's been shot to pieces because of domestic policies. It's been shot to pieces because of international policies. You know, if you're making shoes for Nike, and all you hear is that tariffs are going to be 60% next month, it doesn't make you want to take much risk with whatever money you're earning. And so all this money that these Chinese entrepreneurs are earning, it's either, to the extent that they can, they stick it abroad. So you look at Hong Kong, bank deposits in Hong Kong in U.S. dollars have grown by $220 billion, billion with a B, in the past 18 months. I mean, it's enormous. 

So either it gets stuck in Hong Kong bank accounts, and from there, maybe it goes into crypto. Maybe it goes into Mag 7 stocks. Maybe it just stays there in cash. That's one part where the money goes. But if you're Chinese, you have really only four choices with what you can do with your money. You can put it at the bank, and if there's no demand for credit, which is the case in China today, then the bank has nothing to do but buy bonds. So bond yields go to new lows. You can buy gold, and they're definitely doing a lot of that because when you're afraid for the future, gold seems like a good idea, right? It's like, "I don't know what the future is made of, but gold protects my purchasing power, so I'm going to buy that." You can buy local equities, or you can buy local real estate. In essence, when you're Chinese, those are your choices. And because they've lacked confidence in recent years, they've bought gold and bank deposits, which eventually go into bonds. But this is where also this year we might be starting to see a shift. Chinese equities are starting to do better, and things like—you mentioned DeepSeek—things like the announcement of DeepSeek all of a sudden give Chinese people confidence that "You know what, U.S. was trying to trip us up technologically, but maybe we can do this on our own. Maybe we're not going to be constrained by the America's essential tech war" because that's what it's been. If you start to see the Chinese bull market, which is starting to me to look like more and more plausible, if Chinese equities start to pick up, that could be a bearish scenario, of course, for Chinese bonds because then money leaves bonds for equities, of course, but even potentially for gold.

LIZ ANN: So you mentioned DeepSeek again …

LOUIS: Sorry …

LIZ ANN: No, no no, believe me, other than tariff news, which I'd love your thoughts on that, too, I think, sort of the stocks, the companies, the people that were hit by the DeepSeek news, I don't know, maybe they're happy about the tariff news supplanting that on the front pages. It wasn't all that long ago that we got that news. 

And the other report that you wrote even more recently that we will link to was at the end of January—it was titled "Another Sputnik Moment." And I want to actually quote from that report at the very end because I think it frames the conversation around this and your thoughts on it. And it was that "DeepSeek destroys the idea that although having a few tech titans controlling the broader tech ecosystem might be socially pernicious, at least it makes for technological progress and ensures that the U.S. remains technologically dominant. In this latest Sputnik moment, the tech war between China and the United States, between open source and closed system, and between capital-light and capital-intensive models, has just taken a very interesting turn." So expand on that a little bit.

LOUIS: Yeah, absolutely. So look, I think there are three massive consequences from the DeepSeek announcement now. Very briefly, in case people don't know what we're talking about, China, three weeks ago, released an open-sourced artificial intelligence model and claimed that this model was built for a small fraction of the cost that the ChatGPT was built for or any of the other models that have been put out there. So I think, you know, of course we can debate whether it really costs that little, etc. But the first point I'd make is that, in my career—and I've been in China for 30 years—when China enters an industry, you can kiss your margins goodbye, and you'll never see them again. So I think that the perception in the tech industry was "We're going to spend billions and billions of dollars, and with these billions of dollars we're going to build a moat for ourselves, and we're going to basically build ourselves a cashflow stream that will last for years to come in artificial intelligence." And that belief is now destroyed. 

Again, when China enters an industry, you can kiss your cash flows goodbye. They go down to zero. The way China does capitalism, I'm sure your viewers will have seen the movies The Hunger Games, or some of them will have seen The Hunger Games. This is how China does capitalism. Lots of contestants and a lot of dead bodies along the way. So that's the first point I'd make. 

The second point I'd make is that up until the release of DeepSeek, I think the perception, both in China and in the United States and all over the world, was that in 2018, the U.S. shifted from a trade war to a tech war with China. You know, when the U.S. said, "You're not allowed to have semi-conductors anymore, and we're going to constrain your growth by making sure that you're technologically constrained." And China said, "Fine, then I'm going to spend a lot of money and build my own tech ecosystem."

And essentially, you know, the U.S. said, "Good luck to you" and sort of laughed. And up until 30 minutes ago, the perception was, you know, the U.S. is still far, far ahead in this sort of technological race, and China is nowhere in artificial intelligence. And the only question is whether it will be Microsoft or Facebook or Amazon or Google who wins the artificial intelligence race.

And now all of a sudden, we find out that, literally a Chinese quant hedge fund enters the race out of nowhere with a budget of, you know, two shoe strings and some bubble gum. And then behind DeepSeek, here comes Alibaba saying, "Hey, by the way, my model is better than DeepSeek's." And then here comes ByteDance, the owner of TikTok saying, "By the way, my model is better than Alibaba's."

So now not only are we seeing the typical Chinese thing where you commodify everything and you crush everybody's margins, but you also blow a huge hole in the belief that China was constrained technologically. Because suddenly it looks like that's just not the case. So this DeepSeek moment, you've seen a number of ideas that I think people really held onto that have just been shattered. 

The first one is the idea that you could spend your way into building yourself a monopoly, which was essentially what Microsoft and Facebook, etc., were trying to do. So that belief is gone now. The second belief was that China was far behind technologically. That belief is also gone. 

And I think the third belief now that we have to grapple with was that, you know, there was this belief that all these big boys were going to spend tens if not hundreds of billions of dollars, and therefore semiconductors for a long while were not going to be a cyclical industry. It was just going to be structural spending because historically semiconductors is amongst the most cyclical of industries. You know, it's very high capital spending. And if you do your capex at the wrong time, then you're out of luck. Well, this belief that semiconductors are going to be structural, not cyclical, for the coming years, led semiconductors … historically, semiconductors trade between three and six times sales. You look at the semiconductor index, it trades between three and six times sales. Usually at six times sales, you want to start lightening up. This cycle, we've gone from six to nine, just powered through. And now we're rolling over. So I think we're also seeing with the artificial intelligence story, a huge hole in the belief that semiconductor spending is going to be structural, not cyclical. Turns out that maybe it's not different this time. Maybe it's just like it's always been, and semiconductors remain cyclical.

LIZ ANN: Yeah, and you wrote about in the context of the spend of billions of dollars gave out this impression that these were very big moats and that that would protect some of the early leaders here. Now that that is breaking down much more quickly than a lot of people thought, another thing you commented on, but I'd love further thoughts, is we've obviously already seen some cannibalization here in terms of what DeepSeek was able to do. Does that start to happen among the U.S. leaders in the AI space?

LOUIS: Yeah, well, look, I think the first thing, of course, you have to think through is if you're one of those U.S. leaders is your whole projection on revenues that you were hoping to get out of artificial intelligence models, you can kiss those goodbye. Today DeepSeek is as performant as any of these models, if not more, and the very high end of DeepSeek is going to cost you 6,000 USD a year. So that's the first problem. I think the second problem that DeepSeek does for these guys is DeepSeek is an open-source model. So everybody points out, "Oh but if you go into DeepSeek, and you look at Tiananmen Massacre, it doesn't give you anything." That's because you're on the online version, so it sends you to their Beijing servers where they have to comply with the local rules. 

But if you download DeepSeek onto a PC that is hosted in the United States, then it's open source, and you can put whatever criteria you want on it, and then you get all the answers you want on Tiananmen. So I highlight this because the whole open source nature of it, remember all these guys, ChatGPT, Facebook, etc. They were all doing it closed source, you know, "We're building this, and we're keeping it for ourselves," which, you know, sort of betrays the whole spirit of technology, really. You know, the technology grows by sharing as many ideas as possible. Technology grows by being open. The more knowledge you share, the better the knowledge gets. 

It's really ironic in a sense that the closed source was in the U.S. and the open source was in China. I think five years ago, you would have bet the opposite, given the cultures, given the political systems on top of it, etc. So it's almost as if all these guys came out and beat China at their own game. Now, everybody's like, "Oh but this is the CCP[DH1] [MB2] [1] behind DeepSeek, etc. I don't believe that for a second. This was released by a Chinese quant hedge fund. Now, anybody who knows China a little bit knows that the government hates hedge funds. If this had been released by, say, Beijing University or by Huawei, it would definitely have the government's fingerprints all over it. But a quant hedge fund, in Hongzhou, a secondary city, you know, this came from the bottom up. And in a sense, it's extremely exciting because it tells us that, you know, we actually don't live in a world where you need to spend a hundred billion dollars to be competitive. And that a few smart guys sitting in a secondary city in Hongzhou have a shot, have a shot at building something great, which means that a few smart guys perhaps sitting in Kansas City, or in Paris, or in Mumbai, or in Rio also have a shot at building something great, which is encouraging.

LIZ ANN: Not to mention this has to be a benefit to a much broader swath of industries and companies that have been trying to figure out how to bring AI into their businesses and, you know, all else equal, this ostensibly lowers the cost significantly. So you know, it's been one of our theses that we were in a shift from sole focus on the enablers, the infrastructure of AI, to the actual users of AI. This, I would think, speeds up that process and provides a much lower cost to bring it into your workplace, whatever that is.

LOUIS: No doubt about it. You know, 100%. You know, if you're Charles Schwab, you're probably thinking, "Oh, OK, what do we do with AI, et cetera?" And you probably had Microsoft coming to see you and say, "Hey, we can take care of your AI, and it's going to cost you 100 million bucks a year." Or I'm picking a number out of thin air, but something like that. And now you're thinking, "Oh, hold on. Maybe we talk to these guys." Now, of course, in this tech war that started in 2018, initially the move was we're going to prevent China from benefiting from Western technology so that they can't develop. And now the response in the Western world, now that China says, "OK, well, we've built our own technology, and it's just as good, and it's, you know, 1% of the cost of the other guy," our response is: "Block it."

Like literally in the U.S. Senate, it's, "Uh-oh, no, we can't have that." And you already have senators putting in bills saying that anybody who uses DeepSeek should go to jail. Again, it's so counterintuitive to what you would have expected just five or 10 years ago, where you would have expected the U.S. to be the open system and China to be the closed one, but invariably, it seems to be going the other way.

Now you could say these rules aren't going to get adopted. It's just, you know, lobbyists trying to do their thing. But it's going to be fascinating. Now, the good news is today Charles Schwab may feel very reluctant to have a Chinese AI for a number of reasons. And I get that, but it means that tomorrow you'll probably get an Indian AI or maybe a French one, or again, maybe a few smart guys in Kansas City come up with one.

LIZ ANN: You keep saying Kansas City. Are you, is that your, that, that's, I didn't know if that was your rooting interest for the Super Bowl.

LOUIS: No, no, no, I didn't even think of that. 

LIZ ANN: I'm wearing green. I'm wearing green. And that is because I'm an Eagles fan. So I know we just went completely on a tangent, but why not? Why not have fun?

LOUIS: No, I just picked it because it's in middle of the country. OK. Sorry. I should have said Philadelphia. I'll say Philadelphia next time. I just picked Kansas because it's in the middle of the country. 

LIZ ANN: You're right. You know, sometimes the word "bubble" gets tossed around a little bit too frequently, but you have written about, not so much "AI is a bubble," more of the "Could it be a bubble?" What are the characteristics thereof? And I know one of the kind of existential questions that you asked in this report was, "OK, if AI has stalled, number one, what takes it over?" And then what are your thoughts on how that would work its way into market behavior? Do you think it would just alleviate the concentration problem and shift flows and interest into so many other areas of the market that are not tied into this AI narrative? Or do you think it would potentially be something significant enough that it brings on the next full bear market?

LOUIS: Important question, and I wish I had a yes or no answer for you, but I think when you study bubbles, first, there's two kinds of bubbles. There's bubbles in productive assets. So for example, railways in the United States in the 19th century or, you know, internet broadband at the end of the last century. And then there's how they get financed. Are they financed through debt, or are they financed through equity? Now, the worst kind of bubble is a bubble in unproductive assets, i.e., tulips or land in central Tokyo that is financed by debt. Because when that blows up, you don't have anything new. You have just as much land in central Tokyo as you did before. The best kind of bubble is in productive assets funded by equity. Because here the only problem becomes that basically you just put the wrong price on assets, and you got to reprice. 

Now, it strikes me that if we have had a bubble in AI, it's probably in a productive asset. You've had a lot of capital spending done on the premise—it's very much like the internet of the late '90s—on the premise that we're going to get these great returns very quickly. Now, it turns out that, you know, perhaps that was wrong, and so assets have to reprice. But you know, if tomorrow Nvidia goes down 50% or 80%, does it really matter? I'm not sure it does because I don't think there's a ton of debt attached to this. Now, in the bubble of the late '90s, the TMT bubble—and yes, sorry, TMT stands for Technology, Media, and Telecoms—you had a lot of debt that was hidden in the system. Sun Microsystems, Lucent, Nortel, they all did tons of vendor financing so that when the cycles turned around, not only were they hit on their income statements because all of a sudden there was less demand, but their balance sheets also got destroyed because they didn't get paid for the previous sales that they'd made. 

So what always kills you, I think, in a bubble is the hidden leverage. Same story in 2007, 2008. The banks had issued a lot of mortgages and had essentially written insurance against these mortgages saying, "Hey, if these go bad, we'll take them back on our balance sheets." And so before you knew it, the bank balance sheets just blew up. And that was the hidden leverage. I don't think today we have that. I really don't.

I'm looking for it because when you have big enthusiasm, big bull markets somehow, you know leverage sneaks in, and it gets hidden, and it's sneaky that way. But I'm not sure where the hidden leverage is. So I do think we have a bubble in a lot of things like, you know, a lot of silly prices in crypto, a lot of silly prices on the stock market, valuations that are too stretched, but if it blows up, I'm not sure it blows up our economy You'll just have assets that change hands from weak hands to strong hands, and that'll be that. So I'm not too worried that a correction in, say, the Nvidia share price or the Google share price or the Microsoft share price will implode our economic systems. I'm really not. What I would say is, and to answer your question, is the market is still obviously in a mood of looking for the new things. And I think there's two potential big new things.

One of them, there's more and more press, more and more excitement around autonomous driving. And I think we are getting close to a point where autonomous driving is going to be a thing. I mean, here in China, it is becoming a thing. There's a number of cities now with full-on autonomous driving. And I think you have something like 6 million people in the U.S. that drive for a living.

So it's something with potential big social repercussions and potential big productivity gains for the economy. So that's one. And I think the other, you mentioned tariffs. You know, Marco Rubio came out in his very first interview and said, "Look, the unipolar moment is over. Now the U.S. just has to take care of itself, and others have to take care of themselves." And that's pretty much the pretty clear message from the Trump administration. And that message has been delivered in such a blunt way that I think if you're in Europe today, if you're in Canada, you're looking at a lot of the infrastructure spending that you should have been doing for the past 10 years. And whether that be in defense, whether that be building your own power grid, whether that be building your own pipelines and your own LNG[2] terminals, etc., stuff that you were kind of depending on the U.S. for.

You're like, "OK, I'm not going to depend on the U.S. for this anymore." So if I'm Canada, you know, I'm now thinking, "I don't want to send my oil at an underpriced value to the U.S. anymore or in my natural gas. I'm going to build LNG terminals in Eastern Canada. I'm going to build the ports and the pipelines to British Columbia to sell my oil to Asia." Because the reality is if I only have one client, I actually don't have a client. I have a boss.

And so I think you're going to see a massive infrastructure spending boom in Canada. I think you're going to see a massive infrastructure spending boom in Europe. And you know, when you say infrastructure spending boom, invariably that means commodities. And what's interesting, I mean, so all this is when I look at the commodity indices, they're all starting to break out on the upside. So everybody today is bearish on commodities, the premise that, "Oh China's over, it's not doing well, etc." But I think they're underestimating the amount of infrastructure spending you're going to get all across Latin America, all across the rest of Asia, and across the non-U.S. developed world.

LIZ ANN: I could do this for another hour. 

LOUIS: I'm sorry. I'm so long-winded.

LIZ ANN: No, I'm sure our listeners could do this for another hour. I've sat here with all my bullets that I did in preparation for this. And you literally and figuratively checked every single box. So other than didn't know whether we were going to have time to dive more into the crypto space.

I'm kind of glad that we didn't. It's not my bailiwick. So I think you touched on it. But I love how you tackled that last part because with all these disruptions, even potential bubbles, you know, born out of those are new opportunities and positive stories. And so I think it was good that we ended on a positive note. But man, we covered a lot.

I am so appreciative. Now I'm even more fascinated by the work that I have read again, just in in advance of this. And thank you, of course, for allowing our listeners to have access to some of the work that I mentioned and that we talked about today. So we appreciate that. But I'm thrilled you joined us at what is the very wee hours of the morning in your neck of the world.

LOUIS: Thank you very much.

KATHY: So Liz Ann, last week we got the first jobs report of the year. And now we have news that the Department of Government Efficiency, otherwise known as DOGE, is planning to lay off thousands or tens of thousands of federal employees and close down some departments and agencies. And when I do the math, I say there's approximately 3 million federal government employees. Say they each make $100,000 a year on average. And this is including people who are making a lot of money, and that pulls up the average, which is probably closer to $50,000 to $75,000 for most. But when you take those numbers, and then you say, well, if we cut 10% or 20% of them, you're not coming up with a lot of money relative to the $2 trillion in cuts that we're looking for. And arguably, you're making the government less efficient, not more efficient, in terms of carrying out its responsibilities, particularly around regulation. People want less regulation. They want it to be more efficient. And if you don't have people to help process the forms and all that, where are you going to go with this?

So I have my doubts about taking the approach of just cutting people as a way to become more efficient. But I do wonder what your take is and how that affects the unemployment numbers nationally.

LIZ ANN: Well, I think it's probably, all else equal, could cause a little bit of a move up in the unemployment rate, even if the numbers in the aggregate are not significant. But I think we also have to think about this in the context of the immigration side of things as well, because in the last four and a half years, all of the growth in the U.S. labor force has been in foreign born.

I've said that before and then been criticized. "That's not true! It's not all illegals!" I said, "That's not what I said. I said 'foreign born,' many of whom came here legally." But you combine what's happening with the federal employees and losing their jobs, that that adds to unemployment all else equal, and then the pressure down on the labor force where all of the labor-force growth in the last four-plus years has been foreign born. And I think it has the potential to sort of wreak havoc with a lot of the employment-related numbers, including unemployment claims. And I know the buzz that I'm hearing from inside Washington on both sides of the aisle is it just feels very chaotic right now, and trying to gauge what the actual impacts are either on—to your point, Kathy—trying to do the math associated with "How much does this actually save?" versus "What is the impact on employment numbers?" So we've been joking often throughout the course of this year that if you wanted to emoji the year, the symbol would be the "I have no idea," the shrug-the-shoulders emoji. 

I did have an interesting conversation with a client at an event who said, "Well, if we did fire all the federal employees, isn't that good in the sense that all the people looking for work could now apply for those jobs?" And that was one of those where I just really had no idea how to answer that question because it didn't make any sense. So it's incredible what the varying perspectives are on what this means.

KATHY: Yeah, I get all the time is some sort of comment about "All the jobs created have been government jobs," which is absolutely factually incorrect. And I think people throw in health care and health care services for some reason, assuming that that category is government supported, which it's not. So I think there is a misperception that the size of the labor force relative to the growth in the economy has been very much concentrated in the federal government. And that's not true. That's actually been very consistent for decades relative to the size of the labor force. We have seen some growth in state and local government. And it is one of the things that puzzles me, too, about this approach, because if you are trying to focus on areas where you want to cut regulation, local government is often responsible for a lot of the regulation. So focusing on the federal government, we all want to get rid of waste and fraud and abuse. And I'm sure there's some capacity there. But I think focusing just on the federal government misses probably a lot of where things could change. But, you know, we'll see. It's early days and a lot to absorb. It's drinking from the fire hose, as they say, every day.

So Liz Ann, key indicators that you'll be watching next week. We've got a lot of data coming out.

LIZ ANN: We do have a lot of data, and we get a lot of housing-related data, especially with some of the swings that we're seeing in longer-term yields that can have implications for both the data but also sentiment. So we get the National Association of Home Builders Housing Market Index, which is basically a builder sentiment index that comes out. And that can have some volatility based on perceptions, which includes around things like mortgage rates and implications for sales. So that's an example of some of the soft data that we often talk about, the survey-based data, in contrast to some of the hard data. And we do get some hard data on housing next week as well. We get housing starts and building permits and existing home sales. We get a couple of regional Fed components. We get claims, obviously that's every week. 

The Leading Economic Index comes out next week. That's trying to find some stability, but at a still fairly depressed level. We get S&P Global's version of Purchasing Managers Index, is not maybe as widely followed as the Institute for Supply Management version of PMIs, but increasingly capturing some attention, especially if it shows something maybe different from what was shown in ISM. And then at the end of the week, we get University of Michigan consumer sentiment, which includes inflation expectations, which is something that the Fed keeps an eye on.

So what's on your radar? I saw you, our listeners didn't see you, roll your eyes at that. So I'm curious as to "Is there some skepticism in terms of the inflation expectations coming out of that?" I'm guessing it might be given the partisan divide that happens there.

KATHY: Yeah, yeah, I used to love that survey. The reason I loved it because, you know, if you work with data, you've got decades and decades worth of regular monthly data to parse. And you know, it's almost like baseball statistics. You know, you just got so much to deal with. It's a lot of fun. But in recent years, it has been, as you mentioned, what we've seen is this huge partisan divide has driven those expectations. So it's lost validity in my book because you're not really getting this kind of true reading on what the consensus in aggregate expectations are. So unfortunately, it's like many things these days, we're just divided by our own perceptions of reality right now. 

LIZ ANN: I think we get the Fed minutes, too, right?

KATHY: We get the FOMC minutes. I don't know how surprising they are going to be, or if there'll be much data in there that we can glean any insights from, simply because it's pretty clear the Fed's on hold. We know what they're watching, and it's always interesting to just see what the highlights were, what was the discussion about. Were a few people on this side and many people on that side, that kind of thing. So always interesting reading, but, I don't know, probably overshadowed by other developments. 

LIZ ANN: So packed episode, and that's it for us this week. Thanks as always for listening, and you can keep up with us in real time on social media. I'm @LizAnnSonders on X and LinkedIn. I'm nowhere else at this point. I've had lots of imposters and scams, so make sure you're following the real me. And in terms of our written reports and videos and all the places where we put lots of charts and graphs, you can find that at schwab.com/learn.

KATHY: And I'm @KathyJones—that's Kathy with a K—on X and LinkedIn. And we are truly grateful for everyone who's taken time to leave us a rating or review on Apple Podcasts or feedback on Spotify or a comment on X. Thank you for taking the time to say thanks. And please consider telling a friend about the show. We'll be back next week with a new episode.

For important disclosures, see the show notes or visit schwab.com/OnInvesting, where you can also find the transcript.

[1] Chinese Communist Party

[2] Liquid Natural Gas