Full Transcript: Weekly Wrap on Bubbles, Monopolies, and AI

Insights from Jeremy Grantham, Andy Constan, Edward Chancellor, and Marc Rubinstein

Jack: Welcome to the Excess Returns Weekly Wrap. I’m Jack Forehand, joined as always by my good friend Matt Zeigler. Matt, we—

Matt: Got some pretty good clips this time. I mean, I just can’t believe Jeremy Grantham. I feel like we say this every week right now. This is like a murderer’s row of holy crap. I’m going through this and I’m going, “I am embarrassed by how many cool people we’ve had on Excess Returns this week.”

Jack: Yeah. Grantham’s somebody we’ve— when you start a podcast, he’s one of the people you’re like, “I wanna get him on eventually.” And we’re six years in now and we got him, so that was kinda fun. He was a really nice guy. It was a lot of fun to do that one. But even in addition to that, we got all kinds of stuff. We launched a new show with Andy Constan this week. Andy’s show—

Matt: Is killer. This is so cool. I’m so excited he’s doing this. And that was— what an inaugural episode. I mean, this is really cool.

Jack: Yeah. For anybody who wants to subscribe, it’s called First Principles with Andy Constan. It’s on our YouTube channel. It’s also on any podcast platform out there. But what was really cool about that episode is Andy has seen a lot of these bubbles firsthand. He’s been investing, and he was with Bridgewater, he was with Brevan Howard. He’s been at all the big names. And just the stories he has and the details behind what went on, which we’ll see in some of the clips here, that I just didn’t know. It was really cool just to walk him through. We were gonna do an episode on both bubbles and what to do about bubbles, and we decided to turn it into a two-part series ‘cause there was so much in the first part just about his experience in the bubble. So we’re doing the rest, the other one next week.

Matt: Andy in teacher mode, this is my favorite thing every time he’s part of an interview or something else, is when he goes into this teacher explainer from experience mode. So the idea that we’re getting him to do this, and yes, give me three episodes of the bubble conversation, ‘cause this is so good.

Jack: We’ll make it as many episodes as it takes, I guess. And then on top of that, we’ve got Edward Chancellor, who’s, I think, the world’s foremost authority on capital cycles. At least that’s how Kai Wu, who did the interview— this was Kai’s show— explained him. And then we’ve also got Mark Rubenstein, who knows the inner workings of the financial system like better than anybody I’ve ever seen. That was a really great episode.

Matt: Chancellor’s amazing, so I’m super excited that Kai sat down with him because the history lessons that he pulls from— he’s up there with a Grant Williams or somebody like that, where it’s just like they will tell you about a corner of history that explains something from a perspective where you’re like, “I had no idea.” And now I am so excited to talk about railroads and canals and stuff like this, and I wanna dive into the history. And we will talk about railroads and canals in a little bit. Yeah, we will. And same with Mark. Net interest is one of the most reliably fascinating Substacks. It’s something that I feel like I shouldn’t be interested in. And then he’s another one who will explain the plumbing— just the seat that he’s had. The front row seat to covering the financials industry for like 30 years means much like Constan with bubbles, he just has this perspective on everything, down to how payments work, where there’s competition with credit cards, at which scale. Is it national? Is it global? How to think about all this stuff. And nobody has financials knowledge like him. It’s amazing.

Jack: So let’s get into it. We’re gonna start with Jeremy Grantham. And everybody kinda knows there’s these been these big companies that have done a lot better than you’d expect big companies to do. They’ve defied what Michael Mauboussin would call the base rates on this stuff. And there’s been a bunch of explanations, but one of them we haven’t talked about a lot in the podcast is this idea of monopoly. So here’s Jeremy talking about that.

Jeremy: History was pretty clear. Asset classes mean revert. Sectors within an asset, like small versus large, they mean revert. And even companies, in the end, mean revert. And then to be more detailed, mean reversion at the corporate level says if you make abnormal profits, you will receive competition. If you make obscene profits, you’ll get ferocious competition.

If you’re having a slump and you’re not making much money at all, or even losses, you will frighten away capital and you’ll have no competition at all. And eventually, as the market slowly grows or rapidly grows, you will reach a period of shortage, and everything will recycle and your stock will go up, your profit margins will go up.

And there was a very clear history of that happening. And in recent years, one has to say, that is not as clear as it used to be. The bottom 90% of the market seems to fairly clearly still mean revert, and yet you have a novel emergence of a kind of elite, a few handfuls of stocks that seem to have gone from strength to strength.

And you have to ask the question, why isn’t the money flooding in to compete these ridiculous returns and drive them down? And one is the winner-take-all nature of software. The guys who get there first have such an advantage, it’s hard to break in. The other is the attitude of the government, of the administration, to monopolies.

And you can look back over this interesting different phase. And of course, this time is different. You look back to about 2000 and you say, “In what way is it different?” And one of the ways, at the top of the list, is a steady concentration in every industry. There are fewer companies and bigger companies and more dominant companies in every industry.

In most of them, it’s not that big a deal. In some of them, it’s massive. And what did the Justice Department, et cetera, do about this? And the answer is, uniquely in this time period, nothing at all. So companies were able to quickly develop not just domestic monopolies, but global monopolies. The Mag Seven were not universally seven global monopolies, but there were several in that group that were.

And there is no better way to make money than to have a near monopoly or a complete monopoly, to be a price setter. So ask yourself— are the Mag 7 setting prices? And the answer is mostly yes. They fulfill the characteristic of a workable, profitable monopoly, and they’ve been tolerated. In other eras, the government would get in there and say, “Nah, Standard Oil is too big. We’re gonna break it up into nine pieces,” and so on. And that’s pretty effective. It’s pretty effective at certainly breaking the monopoly. Then you have more competition from the pieces and the prices tend to be lower and the competition tends to be higher. The characteristic, the unfortunate characteristic of increased monopoly is that the growth rate of the system tends to slow down.

So the profit margins of the monopolists go up, the share going to workers goes down, inequality increases, but the growth rate of the GDP tends to slow down. So a lot of people think because there’s so much profit being made by the top 20 firms, that somehow everything in the garden is great and growth rate must be higher. No, it isn’t. Growth rate in GDP is actually slower for the last 20 years than it was for the prior 20 and the prior 20 to that.

Jack: Yeah, I mean, I’ve kinda thought about this the other way. I’ve thought about the nature of these businesses and how they compound and how that’s led them to be what they’ve become. But there’s this other idea that, you know, they have, to some degree, there’s been some monopolistic-type behavior here. And Jeremy’s argument here is, you know, maybe the regulators should’ve been a little more involved in this than they were. I don’t know if I agree or disagree with it, but I thought it was an interesting take. It’s one I hadn’t heard a lot.

Matt: I think it’s a fascinating take, not just because regulator involvement and how individual sectors mean revert versus the whole system mean reverts— there’s just all these fascinating layers with who’s participating in what level. But yeah, this is a really interesting counterexample that I don’t really hear anybody else talking about. And especially framing— he broadens the definition of monopoly from the way that I would normally think about it.

Jack: Yeah, what do you do? I mean, technology leads to monopolies to some degree. Was Google Search a monopoly? Not anymore. But was Google Search a monopoly for a long time? I mean, yeah, it was. They got to that point because they built a great search engine, and they outperformed everybody else. But at a certain point, it’s just so hard to figure out what to do about this stuff. I don’t know if I have a view that the government should’ve stepped in and done something about that or if I have a view, let the market work itself out. And we’re kinda seeing the other side of that now because Google Search monopoly— the government didn’t do anything about it, but the market is now doing something about it with AI. So to some degree, letting the market do its thing does its thing over time. But Jeremy’s point is also true, which is, we did have a high level of monopoly in certain areas, and that certainly was part of why we haven’t had mean reversion.

Matt: Yeah, the idea that the technology becomes the new baseline, and then that basically destroys the monopoly on its own from market forces— I don’t think you can say that reliably, but it is a very interesting thought exercise to think about how that draws competition in, pursuing the profits or looking for a way to crack that. But also, it just becomes a thing that everybody uses ‘cause nobody has done business in the last 10 years without relying on Google Search for something. And that’s a really interesting point to think of, especially as we try to play AI forward.

Jack: So our next clip is Andy Constan. And people love to say, “We’re in a bubble. We’re not in a bubble.” But what you don’t see a lot of is the mechanics of what goes on behind the bubbles, the stages of a bubble, what history, how the bubble has played out in various historical bubbles. And so here’s Andy. This is a kind of a long clip, but it’s well worth it. Here’s Andy talking about the phases of a bubble.

Andy: These types of environments typically start with something new. And something new in the internet boom and, if we’re in a bubble today, the AI boom, was technology, was some new thing. And you can look back to, and again, before my time, you can look back to a variety of industrial revolution technological advancements. You can look to China, where they made a huge productivity move, bringing people from the farms to the factories. You can look at major productivity changes as it tends to lead to some sort of bubble-like equity outcome. So there’s a new thing that’s technology. In 1982 through ‘87, the new technology— it wasn’t really new technology. We just had ended a major inflationary episode. We deregulated— the United States deregulated the financial industry, in particular the savings and loan industry. There was a small technology advancement, which was the invention of Lotus 1-2-3, which allowed people to easily scenario-analyze companies. And there was the innovation of Mike Milken in terms of creating a market for high-yield debt.

And that kicked off the thing that was really new to the markets, and that was the LBO. And so when I think of the 1987 crash, I think it was impacted by lots and lots of things, and all bubbles have lots of things going on in them. But in 1987, that bubble was driven a lot by a trend toward the LBO. We know what kicked off the something new in ‘95. In 2005 through 2008, where you had the housing boom, we had a period of time where globalization had essentially ended inflation. And with the end of inflation, financial conditions could be left very accommodative with no risk of inflation. And that created a levering up in banks and in the housing market.

So the new thing was the end of inflation and globalization. And that was a driver for what ultimately turned into a bubble. And by the way, this is what I think. I could be wrong. This is just how I’m thinking through these things. Now, as I said, ZIRP and QE drove a bubble in bonds, which ultimately peaked when the economy was shut down during COVID.

And then today we have the ChatGPT moment, which I don’t remember what you thought about it, but I thought on January 10th of 2023, when Microsoft made its investment in OpenAI— for many of us, we’d been playing with the first public version of ChatGPT. A new version had just come on. And for any of us who have done any sort of statistical analysis through their careers, there’s been a slow burn of regressions leading to neural networks, leading to machine learning, all happening as compute power increased. That’s been a forty-year slow burn in terms of what ultimately inflected with that pretty much one-off event when the AI trade has been one direction since then, basically.

And so I like to think of those as the precursor to the bubble behavior, which is either a significant regulatory change, a significant easing, or a significant technological development. And so, or lastly, a significant exogenous event. The bond bubble would not have occurred without COVID.

And then you have escalation events, and that happens along the path of that framework, and that’s when you go into a bubble. And for me, those things were just an explosion of deals in 1987. In ‘98, the Long-Term Capital easing ramped and escalated the tech bubble. In 2005, financial engineering, in particular tranched CDOs, tranched mortgage product, doubled, tripled X the leverage, squared the leverage, whatever you might wanna call it, and escalated the housing bubble. Obviously, the pandemic itself was the final thing that caused the bond bubble to go parabolic. And then, as I said, I think we saw some unnecessary easing of financial conditions.

Today, we had a super hot inflation print. It’s been, I don’t know, 62 months since inflation is above target. And in 2023, and even in late 2022, before this whole AI trade got started, the central banks, in particular the Fed, eased to deal with financial stability around the banking crisis, the small banking crisis we saw in the spring of 2023. And they gave up on their inflation mandate, and that escalated this thing. So those are the things. There’s the root conditions, which don’t have to be a bubble, but root conditions can become a bubble. Then there’s the escalation events, and then there’s the peaking, and I think we’re in that phase right now. We’re in the peaking phase. Now, how long that can last? Quite some time.

Jack: So this idea: something new, escalation events, the peaking phase. And you can watch the full episode to see we went through that exact thing for all of the different things Andy has seen that he considered bubbles in his career. And it’s just— you do see these common things. And it’s something new— doesn’t have to be technology, as he pointed out. It can be something else. But you usually get something new, you get these things that escalate the bubble, and then you get this peaking phase.

Matt: I love the idea that you need the fertile soil for this thing to grow in, and that becomes something that’s really hard to understand. But it’s something where you can sort of understand in the new phase what’s taking root and what that is, and then how deep those roots are gonna go and how much fertilizer proverbially is being thrown on this thing that determines the bubble. So it’s an interesting construct to kind of check back in with the earlier steps as you watch the thing advance.

Jack: Yeah, and he used the word bubble regime, which I think is so important to use because—

Matt: So important.

Jack: He talked a lot about this idea. This is not like, “Oh, we figured out the stages of the bubble, so now we know where we are in the bubble, and we’re gonna predict when it’s gonna end.” It’s none of those things. It’s basically like, in general, this is the way they work. When you’re in a bubble regime— and we’re gonna talk in the next episode about how to invest in a bubble regime— you’re in a regime, but you’re not in something you can time. You’re not in something you’re gonna be like, “Oh, here’s the end coming, so I’m gonna short the bubble or something.” That’s not what he was getting at.

Matt: Yeah, that idea of how it’s not that it’s uninvestable, but it’s untimeable again. So humble yourself, realize what this is, and then understand that you may not— you can experience a cycle super, super fast or super, super slow, and you have to take that part out. You just have to accept that these bubble dynamics are at play, and that opens up a whole other way of thinking. It’s crazy how interconnected these four conversations this week are, ‘cause I feel like they all approach this from a different angle.

Jack: Yeah, and we’re kind of with the next one, we’re gonna get into some other stuff related to bubbles and new technology revolutions. And this is Edward Chancellor talking about this idea that people tend to overstate both demand and profits on the accounting side when we get into one of these phases.

Edward: In these tech booms, people overestimate demand. And I think going back to the railways in 1845, the CapEx spending at that time would have required, within a specific window— if someone’s got to crunch the numbers— that would have required passenger rail traffic to increase by threefold over the next five years. And given that there were a fair number of railways already by that time in the UK, it wasn’t gonna happen. During the dot-com bubble, there was this sort of urban legend going around that data traffic was doubling every two months, when in fact it was only doubling every six months.

And this little factoid actually originated with some company that was later taken over by WorldCom, which later went bust. And it was cited everywhere— all the brokers picked it up, even the US government picked it up. So everyone believed it. But in fact we actually have data. There’s this guy I know called Andrew Odlyzko, who was at the time at Bell Labs, and in 2000, just around the time that the tech bubble was peaking, he put out a paper giving the true demand growth. And no one— the accurate data was available in real time and no one paid attention to it. And the upshot was WorldCom went bust and a host of those other so-called alternative telecoms carriers, altnets, went bust. And there was massive overcapacity in fiber optic cable and all the telecoms equipment suppliers like Nortel and Ericsson and Lucent took big hits.

And actually there was a massive decline in profitability. So one of the features of the CapEx booms is that they actually produce profits because if someone invests and the other person, the buyer, doesn’t actually immediately depreciate what they’ve acquired, then aggregate profits rise.

And so what you see in the late 1990s going into 2000 is a massive surge in reported profitability. And then because that capital turns out to be misallocated, new CapEx is immediately curtailed, and then you have to depreciate past CapEx. And so you have a collapse in profitability.

And we’re seeing something very similar today in that the depreciation schedules for these AI chips, GPUs, has been extended— and I think you probably know better than I, but I think from roughly an average of three to three and a half years to six, six and a half years. And I understand that— because if you buy a GPU and keep it in a warehouse because you haven’t actually built your data center yet, you don’t actually start depreciating the GPU until it’s actually in the warehouse. But there is a sort of technological depreciation that is going on even before you actually start using the chip.

So we’ll see. But the market is being driven, as far as I see, by a strong economy on the back of a lot of CapEx and very strong earnings growth. But those are contingent on the investment turning out to be profitable and the demand being there.

Jack: So this is interesting from the perspective of— there were a couple of things in this. The one that was most interesting to me is we’ve had this situation due to intangible assets where we’ve effectively been understating the profits of technology companies for a very, very long time. Because when you invest in an intangible versus a tangible asset, we expense it, it brings down profits. And this clip was the first time I’d thought about the idea that when you have a massive CapEx cycle, you kind of do the reverse, right? Because you’re not taking depreciation right away, so you’re actually overstating the profits at the beginning of this. And eventually depreciation will kick in. But I hadn’t heard it presented that way, and it just led me to think, this is the opposite of what we’ve been seeing with these companies for like a decade here.

Matt: It’s really weird. And it makes it really weird to play the AI cycle out forward. Because then to your point, there’s all these other weird knock-on effects that are occurring because of this. We’re seeing— you’ve been following this Utah data center build-out thing between Mr. Wonderful and whatever else. Have you seen this in the industry?

Jack: Not a lot. I’ve heard about it peripherally, but not as much as you have probably.

Matt: Yeah, not enough. And so there were some Gallup numbers that people are citing in some reviews. So there’s a lot of pushback against this phase of the build-out. And it’s interesting because of the AI super cycle, or whatever we want to think about it— if that bubble’s gonna keep on blowing, we need more data centers, and we need more CapEx to be expensing and go through this. But now all of a sudden there’s this Gallup poll that’s like seven out of 10 people would rather live next to a nuclear power plant than a data center. And this is the bubble regime at play. Because if there’s a political will and a push for this other stuff, I don’t think the demand for AI goes away because of all the business adoption and the new way that we’re considering this tool becoming like baseline again. So these counter trends really kind of help frame out to me just how long and how unpredictable this cycle could be, ‘cause it could pop tomorrow, or we could be six years forward and going, “Well, we just bottlenecked this stage of the growth, but since the demand didn’t go away, we had to build those data centers, I don’t know, in outer space or something.”

Jack: And that sort of indirectly plays to what he was saying at the beginning, which is this idea that people tend to overstate the demand at the beginning of a bubble, which is very true. And I’ve been thinking about that a lot because on one hand that’s been true in all these bubbles. We tend to overstate what the eventual demand’s gonna be. But we seem to have limitless demand right now, and so part of you says AI is different— this is intelligence, we have limitless demand. And the other part of me says, “Well, that’s exactly what everybody else was saying in all these other bubbles.” So I don’t know the answer to that, but it’s just a really interesting thing back and forth right now.

Matt: Yeah, you don’t raise money in any of these businesses without being optimistic about the future. And definitionally, you have to be over-optimistic about the future to get the commitment. I think the question here is how transformative is the baseline technology as it gets adopted into all these different businesses. And even if we restrain some of that demand by holding back on some of the supply or the build-out, maybe that just extends how far we have to go with this because it also could extend the optimism. But there’s lots of problems with this too. But I’m fascinated by this applied to right now.

Jack: Yeah, and the more I talk to small business people, the more I realize there are a lot of people out here who are using this technology in many, many innovative ways. And maybe your average person still is not, but it just makes me think maybe the demand is gonna be stronger than it has been in the past. This demand seems to be like it’s gonna be going on for a very long time, ‘cause this can have transformational effects on almost any business.

Matt: Yeah. And think about the adoption cycle of the fax machine versus the internet and email. Just the advantages that email had over the fax machine for sending information, and then internet access and download speeds and all the other stuff as it moves it forward. Think about how long that grew and became embedded, again, as this baseline technology. So it’s really interesting to think about this, if this demand isn’t gonna go away, or if this is the way this is going to be— there’s still a lot of strength and desire to do this, and it’s immediately applicable by everything from the diner down the street and how they’re ordering materials for their supply chain, all the way up to a JPMorgan or whatever else. So this baseline technology being adopted— it doesn’t really feel like that part of the demand is just gonna go away. It’s kind of insatiable until we’ve established that new norm, and that feels like it’s a way off.

Jack: I’m just happy the fax machine is mostly gone because I freaking hated that thing. Nothing ever good happened with the fax machine. It was always broken. It’s like you’re just waiting for the paper to come through. I’m just happy it’s done.

Matt: I’m dealing with a nightmare estate situation with a transfer agent, and almost all of it has been able to be accomplished via attaching documents via email. And I’m just waiting. I know the fax request is still gonna come. Because it’s such a painful process that I know it’s waiting for me in there. But every time I correspond with these people and we move the ball the next step forward, I’m like, “I know you’re gonna ask me to fax something.” It’s just a matter of time. I’m so scarred, a traumatized child.

Jack: At least they have eFax now, so that could be done electronically as well. Exactly. So getting into the next clip, we’ve talked about private credit a lot with a lot of different people on the podcast, but Mark had a unique take and kind of a unique inside look at what’s going on. So here’s Marc Rubenstein talking about private credit.

Marc: It’s reminiscent, isn’t it, of subprime being contained. But they know that, so they are not going to be making as bold a statement, having not done the work. And I think on this occasion, what they are saying is that the amount of money ultimately that’s invested in private credit on terms that limit redemptions is tiny in the context of the overall financial system.

And the fact that these gates are in place, that these redemption limits are in place, it creates headline risk. It creates reputation risk for the providers, and we can talk a little bit about that. Potential liability risk for the providers, we can talk about that. But it serves a purpose, which is unlike deposits, you cannot get a run on the private credit firm, and therefore the risk is largely mitigated.

Now if the holders of these private credit funds are institutions, then who cares? Right? They are big enough to read the small print. They are big enough to withstand redemption gates. The question— what kind of makes it a bit more topical over the past couple of years is that a lot of the holders of those funds are increasingly retail investors. At least affluent investors. No longer just high net worth, but increasingly mass affluent investors, who in many cases have been put into these funds by advisors, and there are various incentive structures around that, who probably should have read the small print, but more likely outsourced it to the financial advisor, and have ended up locked in these funds that they now may want to be getting out of.

And although it kind of feels watertight from a legal perspective, you can see— I mean, we’ll go on and talk about Blue Owl, which was one of the first private credit firms to put up gates, but you can see the impact it’s had on their share price and on their reputation, and on the financial flexibility of their owners, of their founders, who in at least two cases put up stock in Blue Owl as collateral for loans. Stock price collapsed. That created some problems for them. So there are knock-on effects. They’re not systemic. They’re not the sorts of things that the Fed should get involved in, but they do create questions. And at the margins they represent risks.

Jack: So I’ll throw this one to you because you did the interview. What were your most interesting takes on this?

Matt: I think— so Mark’s got the most credible, nuanced view on this private credit thing that I’ve seen that’s not— there’s the whole you have to— where you stand is a function of where you sit thing. So when you read the KKR or the Apollo or any of the private credit providers, when you read their self-assessment of the system and why they’re not gonna cause the next GFC, you have to kind of read that with the expectation of you’re paid to have this opinion. You have to have it. This is like the people raising money for AI data centers. You have to be over-optimistic or else nobody’s gonna give you money. So you better turn up the charisma a little bit too much so that you can get to your goal, so you can maybe stem these redemptions, stem some of this other stress. What Mark’s saying, though, is not on the opposite side either, where this is the next global financial crisis or this is a disaster.

So thinking in layers of where are the various bottlenecks in these redemption pieces where it could get somebody into trouble or it might be a moot point. So thinking through, does this roll up to some insurance companies? Does this roll up to somebody else? And he seems to think that we can navigate through this without a giant GFC-sized disaster.

However, the contingency is— and this is whatever the reverse bubble or the bubble popping scenario is— is if something else is going wrong at the same time as you’re in this redemption gating problem, where there’s either a combination of fraud in some of the cases we’ve seen so far, or there’s people who need money and now there’s a regulatory intervention. That can turn into something else that’s worth watching. It needs something else that it would have to combine with it. So that assessment of private credit feels like one of the most genuine, we-don’t-really-know-what’s-gonna-happen, but here’s the things to watch for that I’ve seen anybody come up with outside of the major private credit pushing firms.

Jack: Yeah. Me and my good friend Claude had to have a little battle about that episode because Claude was basically like, “We’re doing the over-the-top private credit title and thumbnail for this,” and I’m like, “No, we’re not. We don’t do that.” But that’s what you see on YouTube all the time. Everybody is selling into private credit is the next catastrophe that’ll destroy the economy, and there’s a middle ground, which is there’s problems in private credit. It probably won’t bring down the financial system. That middle ground does exist. Now, that’s not a good YouTube title— “There’s problems in private credit, and it will probably not bring down the financial system.” But nonetheless, that probably is the truth here.

Matt: And it’s fascinating when listening to Ben Hunt and talking to him— we’ve had him on multiple times talking about this specific topic. He’s looking at what are the knock-on effects inside of other industries, and then how can this turn into a problem? Again, not necessarily GFC size in proportion, but from the context of what do all these middle market companies do who have become dependent on private credit for financing in the last ten years? And Mark goes into great depth with this in the interview, where he talks about the evolution of private credit as an industry, post-regulatory change in the GFC, so that this cottage industry springs up and grows over time, and now you have these floating rate loans where you apply some leverage on the top, and here’s all the various structures and all the ways that businesses and entities and now regular people can access this stuff.

And he’s like, “This is still solving a lending problem.” And one of Ben Hunt’s questions is effectively, “Okay, so what happens to the middle market companies who depend on this for financing if all of a sudden they can’t get financing? How do we understand what the knock-on effect of that is? Where does that consumption go? What happens to a supply chain if they get pulled out of it?” And that’s the part here. The nuance is in the combinatorial effect of some of these things, and if something else is going wrong. Nothing else goes wrong, this probably just slowly rotates its way out of being an issue, and the industry is fine. Something dreadful goes wrong or six things happen at once, now we could have a big problem and a lot of headaches and a lot of frustrated people and probably insurance companies.

Jack: So we’re back to Jeremy Grantham again, and we asked him about the impact of AI and particularly the impact of AI on margins and on companies across the economy, and here’s what Jeremy said.

Jeremy: When a new technology comes through, the early adopters often make considerably more profit than normal. Their profit margins become wider than normal. The medium adopters maybe have a slight edge. The late adopters bring the market in to balance again. And if I go back and look at prior cycles, what I have to conclude is that when the smoke clears, any new technology is merely a cost of doing business.

Let me focus on the asset management business, which I know a lot about. I remember not so long ago when we had to dig deep and buy our first prime computer, a mini computer. It filled the room the size of the one I’m in. It generated enormous heat and it processed the data pretty quickly, and we had a competitive advantage for two or three years until everybody gritted their teeth and paid up for their mini computers.

And then it became a cost of doing business. Everyone in the investment management business had a computer, right? I can assure you everybody was not having higher profit margins than they used to. The profit margins settled back down to normal. The return on capital, which is the central driver in capitalism— how much money do you make on your investments— has been remarkably stable for a couple of hundred years.

And after the computers had all been bought, all you got was a modest return on the cost of the computer. That was it. You had to put more assets into the game of money management. You made the same average typical return on it, and the early adopter thing was gone. Well, think of AI. The guys who adopt it professionally first have a huge advantage.

But fast-forward a few years, everybody is doing AI, aren’t they? Everybody is paying for the service that they need. It is now in that point in the future the cost of doing business. It will not move aggregate profit margins or aggregate profits notably higher than they are typically. Only in the early adoption phase, where we are now, does that effect occur.

Once everybody has settled in, this is just another cost of doing business. It’s obvious. I think it’s incontrovertible, but you would never guess it from the conversations of the day. There are many other problems and possibilities with AI that we haven’t talked about, but that’s the one I think I understand the most.

Jack: So this is interesting because it is what’s happened with some of these past things. If you end up with some sort of technology or some sort of product or something that gives you an edge, you could have a short-term edge, but over time it becomes— I think he called it a cost of doing business, which is once everybody gets it, then basically everything kinda comes back down to where you were. And that is, I think, a fair take on how this might play out.

Matt: I think this is great. I love that he connects this to basically the product adoption life cycle because I always think of everything in terms of who are the early adopters, who are the mid-cycle adopters— mm-hmm— who are the late adopters, and then where are we in that process? And him mapping this adoption cycle onto profits and who gets the benefit totally checks out. I’ve also been— I think I’ve mentioned it here before— Dennis E. Taylor, the Bobiverse series. I’m caught up on the current iteration of this. So in this series, there’s a lot of AI, it takes place in the future. But one of the things that keeps happening is you develop a new technology, then you have a slight edge until you discover a more superior technology, which is like running into a wall. It’s over and over again: you have something, you have a temporary edge, it slowly depletes as everybody adopts it, and then everybody runs into a future wall.

And I keep looking at this and thinking, “That’s a great metaphor too,” ‘cause even when the late adopters finally catch up, now the whole system is just waiting for the next disruption. And that’s where we’re gonna end up here. I’m convinced.

Jack: I think it’s kinda the same thing with data sources and alpha in our industry. Absolutely. The idea that if one hedge fund has pictures of the Walmart parking lot from satellites or whatever, they get an edge. But then eventually the person who has the data on the pictures from the Walmart parking lot sells it to every single hedge fund, and we all end up in the same place, but now we’re just paying more money because we’ve got the cost of the pictures of the Walmart parking lot.

Matt: Yeah. And we’re not even getting good Walmartian material out of it. Nothing else. Yeah, don’t send me to Reddit for my Walmartians. Use that data for sources of good, please, hedge fund data analysts.

Jack: So this next clip, we’re back to Andy Constan, and this is what I was referring to earlier, which is his unique takes having been inside of some of these things. So here’s him talking about Long-Term Capital and how that served as fuel for the bubble in the late ‘90s.

Andy: There are a couple of things that happen during a bubble. You always look for what I call contagions that could either cause the bubble to extend or are consistent with the sort of post-bubble world. Long-Term Capital had— it’s interesting.

One of the contagions you can have in a bubble is those who are fighting the bubble being bankrupted. But generally, those don’t have meaningful contagions because whatever they have to dump, they dump, and whatever unwinds. There’s so much liquidity around, there’s so much available capital around that losses can be absorbed by the system.

So during the period of time when a bubble is inflating, you rarely have contagion. So I don’t think the Long-Term Capital thing was caused by the stock market rally. There’s some tweaky little stuff about their vol position that probably had some impact, but it’s not really there.

Long-Term Capital was over-levered in primarily fixed income instruments and got a margin call. The problem is that the central bank massively overreacted. What they did is they arranged for the entire fund— which by the way, the numbers are laughably small how small they are right now— they forced 13 banks or 11 banks called the consortium to come up with $1.3 billion. That’s B for billion, not T for trillion. That’s crazy. Nothing. $1.3 billion to buy the positions that Long-Term Capital had and assume their positions. So it was nothing. But they still cut significant interest rates significantly to make sure this didn’t become a financial crisis. It wasn’t gonna become a financial crisis. It was taken care of. That was that.

But they still did these— not only did they cut, but they did surprise cuts. And this is from the same guy who, 40% ago, had used the term irrational exuberance to describe the stock market. He was cutting into a stock market that was up 40% from when he made those comments. I think that was the number. And so that was like adding rocket fuel to the bubble.

Jack: This is one of the things— Matt, I was saying— I didn’t know this exactly. Obviously Long-Term Capital was a major issue. First of all, the idea of how big it was is insane. It was nothing. What was it? 1.3 billion. It’s this massive bailout, and then it’s just some crazy low number relative to what we think about today.

Matt: I remember I went in and reread the book. What’s the title of the Long-Term Capital Management book? You know what I’m talking about?

Jack: Yeah. Smartest Guys in the— Smartest Guys in the Room or something? Or is that a different book? I forget. I’m trying to remember if that’s the Enron book or the—

Matt: Uh, maybe that’s the Enron book.

Jack: Yeah, maybe. I don’t remember then.

Matt: Whatever the Long-Term Capital Management book is, which is exceptional and great. And I remember I reread it either just before the pandemic or during the pandemic as a fun thing— you’re moving a bookshelf, and you’re like, “Oh, I love this book. I should check it out again.” And I got to that part on the dollar amount, and I was like, “This is awful, how laughably small this feels.” Because this is not that long ago. Even my student loans when I was in college, from a similar era, I look at and I’m like, those feel more terrifying than the bailout of Long-Term Capital Management.

Jack: I mean, Nvidia has that hiding in his couch cushions basically these days. It would be like, “Oh yeah, no problem.” Wouldn’t even notice it’s gone.

Matt: Every one of the Mag Seven stocks basically has that in their couch cushions right now. Yeah. And yet this almost destabilized the global financial system.

Jack: It is pretty crazy. And Andy in the clip talks about the idea that he didn’t think they needed to be bailed out. But what was the more interesting part is we were pretty far into the bubble at that point, and this was fuel. So we had a series of cuts because of this, pretty far into the situation, and Andy’s argument was this made the bubble be a lot bigger than it otherwise would’ve been.

Matt: This linkage between contagions— and this is where it was really interesting having Mark next to Andy, playing back through some of these clips, and we’re gonna get more into some of Mark’s stuff in a minute. But it’s this idea— Mark talked about the extended cycle that we’ve been on basically since the GFC, and we haven’t had a real recession. COVID, yes, to a degree, but a real recession or a real credit boom-bust cycle basically since the GFC. We’ve had some slowdowns. We’ve had an earnings recession. We’ve had whatever you wanna label the pandemic in the sense of what happened. But it’s like we’ve just engineered new ways to do different forms of stimulus all along the way that at some point it’s gotta create a problem.

And I know— make your YouTube titles out of this, make your thumbnails— at some point, the crisis is coming. It’s the worst crisis ever. It’s the greater Great Depression. I don’t know. But this idea of contagion will infect these things all the way along, and we will do our best job from a policymaker standpoint and whatever else to try to fight it in the other direction. And it doesn’t feel good on the other end when you start to zoom out and think in these terms.

Jack: When Genius Failed, Matt. That’s the book. Thank you. Roger Lowenstein, I believe.

Matt: So—

Jack: It was Roger Lowenstein. So for whoever’s typing that comment right now and saying, “These two idiots don’t even know the name of that book,” you can stop typing the comment because we finally figured it out.

Matt: We’re still two idiots, but I appreciate— Yeah, no, that’s true.

Jack: So you can stop at the two idiots and then hit submit on the comment.

Matt: Hey, but you know, smartest guys in the room and geniuses— I see, it’s close.

Jack: Yeah, I was in the vicinity. Smart, genius, yeah. I mean, that could have worked as the title for that one.

Matt: It could have. It would’ve been an okay title.

Jack: So anyway, back to our clips here. This is what you alluded to earlier, which is Edward Chancellor studied all of these booms historically, and we’re in the 1850s right now, and we’re talking about railroads and canals. So here’s Edward talking about that.

Edward: Sometimes the new technology doesn’t attract too much attention in its early years. I’m thinking, for instance, of when the railways came to Britain in the 1820s. 1825 was the launch of the first passenger railway in the UK, which is called Stockton and Darlington. And the first few railway companies had dominant positions, no competition. They were pretty profitable, and their technology was proven. And then we had two successive waves of investment, one a decade later, sort of 1835, ‘36. That led to a boom in the stock market. A bit of overbuilding came down, but it wasn’t too much damage. The real problem, or mania, came in sort of 1843 to 1845.

And that is really the period in which there was a massive— there were launches of many, many railway schemes across Britain. And in terms of projected capital expenditure, I think it was running to about ten percent of UK GDP, so actually much higher than AI today.

And there were too many— not all the schemes actually went ahead, but the upshot was far too much duplicative investment. And you famously had, as I would say, three railway lines between London and Peterborough, which is in East Anglia, three railway lines between Leeds and Manchester.

And obviously, if you have three lines running between two places, they’re gonna be less profitable than if you have one line. And the upshot of that is that the railway index, I think, lost about sixty percent of its value. And ironically— I just looked at this the other day— canal stocks. Canals were the most obvious losers from the railway mania, and they did lose in the end. But actually, you did better investing in canal stocks between 1845 and 1850 than you did in railways.

Jack: I mean, this gets at the idea that a lot of people say, which is that there’s a difference between the performance of the companies and the actual technology, and you have to separate those two things. So he’s talking about this idea that the railroads were the new thing, but ultimately, if you had just bought the canal stocks, you would’ve ended up doing better than if you’d bought the railroads.

Matt: It’s so amazing, and I think with part of this too— were there a lot of railroad tracks around where you grew up? Did you grow up on the wrong side of the tracks?

Jack: Yeah, well, we have the railroad into New York City, so it’s a huge thing here. Tons and tons of people commuting every day.

Matt: So here’s the interesting thing to me— you grew up on commuter rail lines, and I grew up on all industrial rail lines.

Jack: Yeah, we don’t have any of that.

Matt: Yeah, so railroads to me— and it’s crazy, even where I live now, I’m not far from basically an industrial rail line that connects a bunch of plants and other things together. So in coal country in northeastern Pennsylvania, it was all about the way that you got— it’s like, okay, the coal comes out of the mountain here or out of the hole in the ground. Then you take that coal, and you put it on here, and you take it over there to sort and filter. Then you take it, all the way down to Bethlehem and Allentown, where you’re making steel, and then the steel’s coming back up to basically get on the train in Scranton and go out across the country to build it.

So rail for transportation was a part of it, but the real moneymaker was the supply chain expansion between these factories. Saw the same thing in Connecticut. Used to live— there’s a whole canal system between Hartford and Springfield that’s fascinating. I lived on that for a while. And it’s just thinking about how these canals ran, and they connected all these mills for all these industrial uses. You don’t have to do the same amount of maintenance on the canal. It’s really interesting to think about the companies who did this, how you would’ve invested in it, what those returns would’ve been, and how— back to the table stakes idea— there are companies in 2026 who are still using the rail for transporting stuff between shipping facilities, like blocks from my house. Twice a day they go by. It’s the worst train if you’re trying to get somewhere and you have to wait because of the amount of those giant storage container things that go by— it’s a 15-minute wait for that train to pass you by.

Jack: What’s funny is with the commuter rail growing up, railroad crossings aren’t really a thing. It goes over the road or whatever. And then when I would go down to Georgia growing up, you’d be like, “Holy crap, when is this train gonna stop coming by?” It’s just going. It’s just going forever. You’re sitting there for 15 minutes. You’re like, “How can they even pull this thing with an engine— this thing so long?”

Matt: It’s amazing to think about, and it’s really— I love zooming out the way that he zooms out here to think about them as businesses, how they were financed, and then why different aspects of basically moving stuff around would mean totally different things, and people in different locations would have a totally different understanding of it. It’s a really cool piece of history.

Jack: So our next clip is back to Marc Rubenstein, and you asked him at the end— I put this in ‘cause I thought it was good— you kind of asked him at the end to sum this up. What is his take based on the overall situation, and what is he positive about and what is he negative about? So here’s Mark talking about that.

Marc: But the market overall is confusing, right? And people I speak to— you know, I regularly speak to people still managing money, and many of them say that they don’t recall a time when they’ve been as confused as currently. Against a tide of bad news— this is all quite well-rehearsed, but against a tide of bad news, the market just powers ahead.

And many explain it away through market structure, role of passive, for example, more retail investors. Obviously AI has a huge influence on market internals right now, and allocations. So it’s just really confusing. Against that, financials become kind of quite simple. So against that, financials— they’re never a safe haven because that playbook is still there to be pulled down.

And Europe in particular, I would say. You know, Europe— we’ve seen this huge divergence between US and Europe over many years. Like a lot of trends, so many trends have accelerated over the past few years. We spoke about this in the context of private credit growth, but just the AI trade, US versus Europe— anything that was tootling along in a linear fashion on a chart has just tipped up over the past couple of years. And US versus Europe is another example of that.

And Europe’s not as bad as people think. We’ve spoken about Revolut. Probably they’ll list in the US. Doesn’t really matter where they’ll list. DeepMind is a UK company listed in the US, inside of Google Alphabet. We talked about Jane Street. One of its largest competitors flies under the radar, a company called XTX, which is London-based. Revolut is another. Arm is another, could be one of the biggest beneficiaries of a change. I mean, it’s again traded in the US, could be a beneficiary of change in free float rules that NASDAQ is introducing for its indices recently. So a lot of innovation in Europe. It’s not that bad. And so yeah, I would say long short, Europe versus the US, financials versus the market.

Matt: A piece that I just feel like people don’t know about Mark, and there’s a great— we did an Intentional Investor interview where I got to talk about a lot of these career milestones with him. That’s worth checking out. If this intrigues you, you should check it out. Because he talks about— he ends up running a financials-focused long-short book through the GFC. Through the worst— the global financial crisis— he’s running a long-short strategy through it, and they survive and do pretty well. So he does okay for himself through this period, including the recovery on the other side, which is really interesting to talk about. So what he was pointing out is there’s a whole bunch of things in— so number one, this idea— did you— this is, I think, in the longer interview, where he’s basically like, “Growth is the enemy of financials.”

Jack: Yes. Which is— yeah, growth is bad. That’s my thumbnail, actually, that I’ve got up right now.

Matt: Oh my God. What an amazing idea in the era of scale that growth is bad and it’s actually a red flag when you’re looking at these companies. Because you don’t wanna see them trying to grow too much because that always begets problems when you’re in financial services. You want more of a utility actor than somebody who’s chasing growth. So inside of all that, he’s looking at all the things that have sort of quietly been built at lower valuations that are doing interesting global stuff, not only outside of the US but in Europe, that are plugged into these systems.

So a more stable growth path, a much lower valuation, and a whole different opportunity set that people don’t seem to understand the size of. His story about Revolut name-checking XTX, the Jane Street competitor, a bunch of these entities, DeepMind— outside of the US, not a take I was expecting, and very, very intriguing. I’m very excited to run this past some other international investors and just say, “Do you agree with this? Or is there a hole here that you see?” ‘Cause I’m intrigued. The pixie dust is on me.

Jack: So as we wrap up, we’re back to Grantham for our last clip, and he has this— GMO has this thing called the bubble detector, which we talked about in a different clip, but I hadn’t heard this other thing he has in terms of this thing that’s happened at the top of every bubble before it’s popped. So here’s Jeremy talking about that.

Jeremy: We had every condition of a bubble in place, including my very favorite rule that I could or could not describe to you, which was flashing brightly. And that rule only occurs at the top of 1929, the Nifty50 of ‘72, and the tech bubble of 2000. It has only occurred three times in history, and the fourth time was late 2021.

And that is when the market leaders of the second to last year— where they’re all up 70, 80%— the market leaders start to go down as the market, the broad market, the S&P, continues to go up. So in 1929, the S&P was up 40%, but the S&P had an index there called the low priced index, which were $5 flaky fallen angels typically.

And they had been up 80% in 1928, and come January ‘29, they started to go down. And before the market broke in October, they were down 40%. They were actually down 40% with the S&P up over 30. Oh my. That was what I call the biggest scream from the stomach of the market in history, and nothing like that happens again until 1972, where you get a significant, but only a faint echo in comparison. The S&P goes up 17 in 1972, and the average of the S&P goes down 17— so that I can remember it forever.

And nothing like that happens again until 2000. As the tech bubble begins to break, growth stocks go down 50%. The S&P continues to go up. The balance goes up another 14%, so that you have the same level on the S&P in September 2000 that you had at the growth stock peak of March 2000.

Why does that happen? I think it’s because the players— Mr. Prince once said, “If the music’s playing, I’ve got to keep dancing.” And that is ultimately an important statement of how the professional money management business works. You cannot fight a major bull market. It is ruinously unprofitable. It’s not optimal behavior at all. So the big companies never fight the market and never tell you to get your ass out of the market, and they never have, they never will. If you are waiting, dear listener, to be told to abandon ship by the Goldman Sachses and the JP Morgans, you will have a long wait. They have never told you, they never will tell you, because it’s simply bad business for them.

They do very nicely being bullish all the time and trying to be a little quicker and slicker on the execution on the way up and the way down, and it works very well. Thank you.

Jack: Yeah, so I mean, I probably knew this, but I had never thought about it this way. This idea that obviously the highest beta names start to roll over way before the market is. So his detector is the high beta names are rolling over, or the names that have been leading up the bubble. But the other important part is the market is still going up at this point. And he saw that in all— and you definitely saw that in 2021. And he talks about 2021— it’s funny, it’s somewhere else in the interview— but he talks about 2021 and how AI ruined his perfectly good bear market when it came in 2022. Because he was all excited about his bear market, he was right about it, and then AI had to come and destroy the whole thing.

Matt: Yeah. Not saying I loved hearing about him getting his heart broken by his precious bear market getting interrupted, but it is really interesting to think about— back to the bubble regime idea— you can kinda have multiple bubbles existing and new fertile soil being created all at the same time. And here’s Jeremy Grantham also pointing out— he gets one good pot shot in at the major banks, at JP Morgan and Goldman or whatever in this, and you gotta love that coming from him. Yeah, they’re never gonna tell you at the top. The people who are in the business of raising money and financing these things and being over-optimistic are always looking for a new place to raise money and be over-optimistic and do this stuff.

Jack: They’re gonna tell you the opposite at the top, right? They’re basically gonna be like, “Yeah, we got all these amazing opportunities.”

Matt: Yeah. So you have to act in the construct of knowing new bubbles could be starting every day. There’s new fertile soil. There’s people who are gonna try to chase this stuff, and it makes a whole different system to navigate around. The humility in here is really interesting to me.

Jack: And it’s interesting ‘cause I asked him, “Are we seeing any of that today?” And he said, “Absolutely not.” But what was interesting is before we had Claude Mythos, we were sort of seeing some of that. Some of those high beta names were starting to roll over relative to the market. You’d seen value come back. You had seen a lot of the types of things he was talking about. The market was still going up, but some of these leading names were underperforming. And then you had good earnings from semiconductors and all that stuff, and that’s completely reversed now. So he’s right— right now we’re not seeing any signs of this.

Matt: It’s really interesting to see him think through this in real time. That was a really cool question.

Jack: So that wraps it up. We had some really good stuff today, so hopefully everybody enjoyed it. Matt, I’ll let you take us out.

Matt: Well, first off, whoever has to dust those shelves behind Jeremy Grantham— God be with you. All those wood slats and those angles— I’m glad that’s not my job in the episode. They were perfect though, weren’t they?

Jack: They were, yeah.

Matt: They were immaculate. I’m pretty sure I paused and zoomed in on a screenshot just to see. But yeah, I was almost like— “Is this a fake backdrop?” Because this is a well-made game. If we had a backdrop-off—

Jack: Between our various guests, I mean, they would be right up there, right? GMO?

Matt: Well, and let’s just say, Chancellor, best chair in any podcast— like, what was he sitting on? The hair coming off of that thing. I don’t know if it was like a bear rug. Like some kind of— I had a dog who had weird hair like that for a long time. Best chair I think I’ve seen in any podcast— just the most questionable amounts of fur coming out from behind it.

Jack: We’re gonna see, now we’re gonna have to do backdrop awards, ‘cause Mauboussin wins most appropriate for the guest because literally piles of books on the floor behind him, just going up. So now we’re gonna have to do a whole backdrop rule.

Matt: Please. I wanna do the Backdrop Awards. The Backdrop Awards on Saturday. Literally nobody would watch us doing this for an hour in an episode. I don’t care. But we would argue it anyway. We’ll put it somewhere else. Right. I wanna do the honorary Backdrop Awards. Yes. That’s what I want. All right, all right. Enough shenanigans. Excess Returns on Substack. Make sure you go there. We’re doing top lessons from these investors. There’s show transcripts. Train your LLMs on this website. We are building this as a resource for us to use. It’s right there for you to use too. Come subscribe. Check out what we’re putting out alongside every single one of these episodes. If you’re watching this on YouTube or your podcast app, you know what to do. Like, comment, subscribe, all the things below, and we are out.