We are excited to announce the launch of a new podcast, The Intangible Economy with Kaiwu. AI and the broader technology revolution are changing how we live, work, and create value. In each episode, Kai will sit down with investors, researchers, and other experts to discuss how innovation and other intangible forces, such as brands, human capital, and network effects, are transforming markets and investment outcomes. In this episode, Kai sits down with Edward Chancellor, one of the world's leading experts in capital cycles. They discuss lessons from past capital cycles and how they can be applied to the current AI cycle. If you would like to continue receiving new episodes of The Intangible Economy, you can subscribe on all major podcast platforms using the links in this episode description. Thank you for listening. We hope you enjoy the new show. So one of the features of the cat-hacks, booms is that they actually produce profits, because if someone invests in the other person, the buyer doesn't actually meet it, they'll appreciate what they've acquired. Then aggregate profits rise. They have no trouble investing tons of money during these technology transitions. But they do have a trouble spotting the winners. So then the question is, can you get a investment in the potential as in the other type of physical investment? The city of the city is pretty obviously yes. Hi everyone. Welcome to The Intangible Economy. Our guest today is Edward Chancellor, who is a financial historian, journalist, and investment strategist. He is the best-selling author of Devil Take the High Most, a History of Financial Speculation, one of the classic books on investment bubbles. At least in my opinion, he's also the world's foremost authority on capital cycles, having written two books on the topic. In addition, he's the author of The Price of Time, the Real Story of Interest, and most recently helped our former boss, Jeremy Grantham, write his autobiography. Ed, welcome to the show. I see you, Kai. What you did mention is that you came to GMOs, my analyst back in 2008. So we came back to the financial crisis together, don't we? Yeah, and the other piece being that to add to your many accomplishments, helping advise me on my economics thesis on credit cycles and bubbles at Harvard. And if I'm allowed to play our trumpet together, you remember how we were working in early 2009, and I got you to do a piece of research that showed that quality stocks, that GMO was heavily invested in at the time, had tended to deliver author or high performance during boom period, but during bust periods, simply because they had a low beta to the market, in which case the arts was that when you wanted to get out of, when you thought the bust was coming to an end, you wanted to get out of quality as quickly as possible, which was actually, we did have a research in February or nine, just in a couple of weeks for the market turned. It, that was, and I credit you because you actually redefined quality in the various markets that we looked at historically as volatility or low-volved stocks, because we didn't have data for, you know, through quality back then. Still that. Now, when I think of sort of good pieces of research that I'd been involved in in my life, I'd say that was definitely one of the top five pieces. Yeah, very timely. And yeah, I was definitely a pleasure working with you at GMO. And that's why I'm so excited to have you on this podcast today to go through some of your research and your work. Okay, let's, let's, let's be bad. Let's start. Okay. So the topic I wanted to start with is, I guess, the topic of the day. US big tech companies are investing trillions of dollars into AI data centers. It would have set to perhaps be the largest investment infrastructure boom of history. Setting aside any of you on the technology itself, what does the history of capital cycles from the real Roy Mania to the Japan bubble to the dot-com boom in the late '90s teach us about how the current cycle may play out? Well, it's hard to leave the efficacy of the technologies. We're going to have to get back to that in a minute. But the general principle looking at past technology booms. And as you know, I wrote a paper with generally grant them on this in February this year, which is available on the GMO website. But the general pitch is is a new technology arrives. People get very excited about the impact of that new technology. Sometimes, sometimes the new technology doesn't attract too much attention in its early years. I think you've been 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 a dominant positions, their competitions. They were pretty profitable and their technology was proven. And then we had two successive ways of investment. One decade later, sort of 1835, '36, that led to boom stock market, a bit of a building came down, but it wasn't too much damage. The real problem, or mania, came in 1843 to 1845. And that is really the period in which there was a massive, there were launch of many, many railway schemes across Britain. And in terms of capital employ, or projected capital expenditure, I think it was running to about 10% of UK GDP, certainly not much higher than AI today. And there were too many as a result, not all the schemes got actually went ahead, but the upshot was far too much duplicative investment. And you're famously, as I was saying, three railway lines between London and Pete Proge, at least Anglia, three railway lines between Leeds and Manchester. And obviously, if you have three lines running between three places, there could be less profitable than if you have one line. And the upshot of that is that if the railway index, I think, lost about 60% of its value, and ironically, I just looked at this the other day, you know, Canal Stops, Canal Stops, when there's obvious losers from the railway mania. And they did lose in the end, but actually you did better, investing in Canal Stops to 1845 and 1850, but you did in railways because Canal Stops, you'd be beaten down a bit with railways still have plenty to full. Now look, in the very long run, in a very long run that's saying, 20-30 years, that investment was pretty benign for the economy. And you know, although the railways were never quite profitable, they'd been in that early years, it didn't. It was good for the economy. And if you got into railway stocks in 1851, it's bubble of burst, it was perfectly fine. Now then, you know, we have other sort of new technologies that are much more competition. I mean, you just think, you know, motorcars in the late 19th century, I think the US has something like 2,000 motor car companies. And the upshow of that was, you know, we did General Motors, which was a winner, and had to be recapitized twice. And actually, GM was really a sort of roll-up of failing car companies. And Henry Ford, I think, he only, his, or motor company, wasn't listed, but Ford only succeeded on his third attempt. It was usually over, you know, over-invested aircraft was much the same. I think, I mean, Warren Buffett, police patent, there were actually three aircraft companies in Nebraska, believe it or in Omaha, Nebraska. That, you know, again, far too many aircraft companies founded.
huge, huge, very lost making. And that, you know, that person has really continued you, obviously, into the.com era, and where, you know, again, massive cat bags. And that's interesting, what's interesting about the, you know, what we use to call the TMT balls, technology, medium telecommunications. That, what's interesting is that the capital is largely down by telecoms companies. And they, they were, they did the heavy, they had the spending, but they really want the winners. And so one of the points they made in a, in a recent column in mind was actually markets of rich, you know, they have no trouble, you know, investing tons of money during these technology transitions, but they, they, they do have a trouble spotting the, the winners and then the end, in the cat, and then, you know, then as you know, after, after the.com bubble, but us, you know, Nasdaq uses 85%, 95%. You know, he, sorry, 70, 78, 79% of its value. Amazon goes down more than 90%. So even though Amazon emerges as an eventual winner, there is a massive, you know, massive loss. And it's not, it's dangerous to say in hindsight that you can spot Amazon. And as the winner and get its, you know, because there could be, you know, there were other businesses around, WEDVAN, this and that, you know, that the might of, the might of actually taking the prime place. So anyhow, so that's the general picture. The general picture is a new technology. Everyone gets very excited about it. A huge amount of investment. Investors tend to anticipate the profits flowing more quickly than they can, than they actually tonight, to be the case. And there, you know, there is, there has always been, I think there's always been a shake out of it. Possibly the one exception is the, is the arrival of the telephone, another is a revolutionary technology where the, the bell telephone systems began to insert itself and got a pretty early monopoly. Telegraph also moved to, a bit early, moved to the monopoly very quickly. So, you know, you could, you could say that if you've got a revolutionary new, new technology and for reasons of inherent monopolistic, what do you want to point out, it's sort of environment, there is case that the new technology can arrive without, you know, without a huge, a, a very massive, but if the barriers to entry are relatively low and therefore you can have more than one dominant plan, then you're like, historic, you're likely to get, you, you, you, you have had over investment. And that's all we know, we know, we can't say that the, you know, the future will resemble the path, but it should sort of guide our judgements, say, at least and then, you know, then we can, then we'll concess current circumstances and say, you need to what extent it is, is this time different or not? Right. So, if I'm hearing you correctly, you are saying that, you know, over the course of history, there have been many successive waves of capital cycles, which generally start due to the advent of a new technology, which, according to tracks, invest in capital. In a few cases, such as the, the, the telephone, the industry has consolidated into monopoly pretty early and has been relatively stable, but the, the vast majority of instances end up with an influx of capital and a fragmented market structure whereby profits are competed away to zero and, and that leaves investors in these companies with, you know, pretty, pretty rough outcome. Well, and so then I'll add something. You remember one of the key, um, precepts of the, of the capital cycle theory is that, um, it draws on the prisoners dilemma, um, being theoretical problem, uh, which is that, um, you know, in the business dilemma, you've got two prisoners and, and the questions, you know, do, do they, um, do they keep quiet and, and both serve as sort of moderate sentence? Or do they both rat on each other and, um, get very long sentences in other words, it's, it's suboptimal to rat. Uh, but, um, but the way the game, you know, the business dilemma game structured is, is, there's an inducement for both to rat. It doesn't, it doesn't work by not, you don't benefit by not ratting. And, and, and I think it's the same, you know, you know, the capital cycle during the investment, a new investment, you see other people going into this, uh, into the new area. And if you didn't get in, there's a possibility that they might come out, you know, dominant monopoly and crush you. And, um, so you go in, but if you go in, you may come for, you know, I, the end result may be suboptimal for, let's say, the investment, what is all it? It may, it may actually, you may actually hang on in there. So, it's not necessarily, you know, maybe a sort of agency problem, but it's not necessarily irrational, um, for the individual. And I think that, you know, with AI, you know, there's one narrative. I don't know, I mean, I don't know whether you wanted to, you know, how robust it is. But one narrative is that, um, when, when Open AI came out with his, was, you know, chat LGBT in, in November 22, um, Microsoft saw this as a way to team up with Open AI to break open the Google search monopoly. And then, you know, Google and started, you know, tell it out to get, you know, um, respond. And then, of course, you know, then everyone else by standing by the way, so I'd say, you know, this is a great technology. We want to bid us the action to, so of course, of course, we're going to get a huge cat page. Yeah, I mean, it's interesting to think about how the different big tech companies have responded differently to the, to the game theory, right? Because we have Apple, of course, of the famous example of a company that has largely abstained from the arms race and it's kind of waiting to see how things shake out. Whereas amongst the hyperscalers, you see ever increasing investment in quotes, um, from CEOs being like, I'd rather go bankrupt than lose this race, right? So there's a clearly a, um, you know, this, this dynamic is clearly happening amongst at least some subset. Maybe your point is just you only need a few players to be bought into the idea in order to drive the entire cycle. Yeah, and you may have been pulling this case more cases than I am, but you know, it's not just, as you know, it's not just the way you know, the top, you know, the largest hack US tech companies doing it, but there's a whole load of, of, um, of other competition, you've got that, what does it keep seeking China? You've got to, I mean, they have listed a few, um, Chinese AI companies recently and then you've got, you know, what, you know, vast amount of, you know, a VC capital, going into the same space. So the, the, the tech, you know, the big tech campaigns into AI gets most of the, uh, most, most coverage, I, I'm spazin in, in, in, in absolute terms, uh, relatively, it's, it's, it's the largest section, but there's a lot of other, uh, investment going on at the same time. Right. So I guess the key question then is, so we know that there's a lot of supply coming online, a lot of investment coming into the AI sector. I guess the question goes, uh, it is ultimately, will there be enough demand, right? Will there be enough demand to meet supply? Um, how should we think about this side of the equation? So, um, it, it, it, it's often the case, what has been the case in, in these tech groups that people, um, overestimate demand. And I think in back the railways in the, in the, in the 1845, the capex spending at that time would have required, you know, within a spec, this, if someone's further crunched on this, it would have required, um, if passenger rail traffic to increase by threefold over the next five years. And given that, you know, that there were fair number railways already by that time, and the UK wasn't going to happen. During the, um, during the dot com bubble, I, there was this sort of, I put, I put, I put an legend going around that, um, that, that, that data traffic was doubling every, um, every two months, when in fact he was any doubling every. six months and you know that this little factoid you made it actually originated in there with some company that was at least taken over by WorldComb which laid a web bust and it was signed actually everywhere you know or in media sale all the brokers pitched up even as you know US government picked up said one believed it but in fact we actually am data this as guy i know called Andrew of the leak who was at the time at bell laughs and in in in two thousand see just right rather the time that the tech bubble was eking he put out a you know a paper second giving you the tree and the melon growth and and you know and no one said that the the actual you know accurate data was available in real time and um no I'll be attention to it and the actual was your welcome when bus and the host of those other um a secret alternative telecoms carries all nets when bus and there was an massive over capacity in fiber optic cable and all the um in a telecoms equipment supplies like you know Norton Erickson and and and loosened it took took big hits um and actually you know there was a massive decline in you know in in in in in profitability it's in one of the features of the of the cathex booms is is that they as you know they they actually produce profits because if someone invests and the other person the buyer doesn't actually immediately depreciate what they're uh what what they've acquired then um they um then aggregate profits rise and say well you see in um you know in in the late 1990s going into two thousand massive search and reported profitability and then because that capital turns out to be misallocated then you have then new cathex isn't immediately curtailed and then you have to depreciate past cathex and so you have a collapse in profitability and something you know we're seeing something very similar today and that as as you know the depreciation schedules for these um AI chips if you use has been extended and um I think you probably may better know but I think from some roughly an average of three to three and a half years to six six and a half years and I understand that they because if you buy a GPU and you know keep it in a warehouse because you haven't actually um um um built your your data center yet uh you did actually start depreciating the GPU until it's actually in the warehouse but there is a sort of technological depreciation that is going along going on you know even before um you actually start using the chip um so yeah so um we'll see but yeah you know the market is being driven as far as I see and by um you know strong economy on the back of a lot of cathex and very strong earnings growth but you know there there's a contingent on the investment turning out to be profitable and the demand being there now can I just say going back to on the demand question and you look you and by all means concentrate me if you have different view is there is this there is this view put out that you know that AI is who we're on the we're on the cusp of um you know official general intelligence even singularity and that uh this um you know that AI whatever that might be is it's just you know it will be able to do you know almost every sort of concealable uh human function apart from you know physical you know physical uh plumbing um and you know I I know I'm very very skeptical of that you because and because I don't I mean when I if I was talking about AI as a large language models large language models work through sort of inference through probabilities of you know what is the most likely next taken to the idea I can't see how AI how that it can need to regenerate the original thinking or activity and then be as you know that it's you know sometimes it's going to make you know the wrong selection um and then for all you get the the hallucinations and I think the hallucinations because these machines aren't sort of genuine reasoning models um there's a uh are inherent to the uh to the large language model technology and if that's case then if that's the case you know I really don't think that the demand that is muted apparently it is going to be there and I I don't know if you pick this up do you see that last week or so a story about some uh company that provided software for car rental businesses do you see this I don't know they used you know everyone's wanting the Claude the new Claude vibe vibe coding the use cord be uh update their their software and it wiped out their entire customer database and it took some others sort of quirky things too so stayed up the day the whole business crashed and the companies that we're using as far as Einstein the company you know the car rental companies were using the software were suddenly frozen um um the a couple of months gave the the winner of the of the rose society's Faraday prize for the top scientific prize I call Michael Woodridge um gave gave gave the Faraday lecture which I think is worth listening to he he's making these sort of points that I'd made and that was I've just taken like points from head and and he he then thinks it might be you know what what he calls that a hint and bug moment you know when the hint and bug moment was I didn't know when it was in 1937 one one of these uh being in zeppelin everyone was excited by these zeppelin's uh um uh uh and and and the hint bug that didn't you know blew up somewhere over some US airfields they want to have second thoughts about the technology so I I I I mean I may have this wrong but I as far as I can understand the hairpacks into large language models is not you know about simply improving search and helping with research but actually is posted on you know a great um you know breakthrough in and into a genetic AI and um as I said well we see I mean it's not yeah you know it's not uncommon for investors to or you know I mean investors believe in in companies and in markets to to imagine that technology is more advanced than it actually is I mean you think but lot of those you know go back to the dot com bubbles some of his businesses would have been viable had we had the um faster internet connection but a time people were on by large working on sort of you know incredibly slow dialogue connections that the businesses that would have been viable ten years later were not viable then said yeah I didn't know what's going to happen to you know AI development maybe someone fix you and merges a um you know the large language model technology with it um I think you know these say the reinforcement learning technology model and then maybe they can harmonize and get something much better but but at the moment they again seem to be quite that so ed so you believe that the you believe that they're that these models can be error for on and that will limit the total addressable market of these models I guess where I would I think an interesting analogy would be that of like self-driving cars right so we know for example that wemo is not perfect but that and its error is not is not zero percent but so so too are keeping drivers imperfect right and so one of the interesting phenomenons we've seen is that for whatever reason politicians and and drivers have a bias against AI whereby you know the the tolerance threshold so much lower that if there's one big you know kind of headline about a bad wemo accident suddenly you know the the interest in in the technology you know go what wanes significantly and so what with your hint and burn moment is that kind of what you're saying here that we all know that AI is not perfect we all know it's a it's a nascent technology um the AI bulls will say it's going to and be getting better over time.
This is the worst of the LVB. I mean, it's kind of hard to argue with that. How much better it will become? We don't know. But you're saying that there is a kind of tail risk for the AI thesis around if just one really bad thing happens that could potentially put a pause on AI development. - No, I'm being a bit more adamant. [LAUGHTER] They know that if given the problems of hallucinations and these people who test models for hallucinations, I looked at one report from I think October of last year at the family. The best performance, 2% hallucinations. And I think 20th performer was out of there. It might have been 20 cents, but I have to say 20. You're not going to-- there are certain activities where you cannot tolerate that degree of error. And you won't-- you can't. Yeah, I say-- well, you see what I mean. There are certain areas where you can tolerate a certain amount of error. Right. You know what I think? I think what that means is that certain things you would not want to give AI. Like, you don't want to make AI in charge of the nuclear weapons, right? But in the case of doing R&D and doing research into a completely green field technology where it's like the alternative is to do nothing, then I think the bar is much lower, right? So then the question becomes, what percentage of economic activity falls to the category 1 versus the category 2? Because that, as I said, will constrain the total addressable market of a technology like this. And what I guess is being implied is that when the demand forecasts are being written by the AI companies, it's to an extent baking in both categories. And you're saying maybe only one category is actually available in which case the potential demand curve is a little bit less enticing than is currently being pressed by the market. Yeah, or put it simply, the ruling principle of Manloid Park is fake it till you make it. And the amount of hype in that you know, or need to say by definition, all these technology minions have large doses of hype. And you know, that's not in many cases have hype over the long run is, you know, it's vacated. But there's never really never, ever, ever been as much hype. As we see around AI and Frank, if you take, you know, you take the efficacy of the amount of hype relative to the proven efficacy of the technology, the ratio is just more extreme than anything before. And I think Thomas Edison, he hyped to know the, it can just above technology before he'd redeaked, fixed the problem. And it worked out in the end then. But you know, and I think what's happened is that in these expected markets, we've lived through in recent years, the reality distortion field that Steve Jobs upsurized it and then you, you know, must take it into the public markets, you're Tesla, all, almost just, you know, get to do something or about to achieve something when you happen to achieve it. And that in itself can become some extent self-fulfilling process in, in most cases, because it gives you the capital to then make the development. But everyone is doing it. And that, and the theories and the, and this huge range of great competition. And I think that that just increases the prospect of it coming under. So I want to like kind of take a flip side of this. So it sounds like on the, you can feel for the interject if you disagree, but it sounds like on the kind of AI investment, boom, you are, you know, negative that you kind of see a repeat of history, these historical booms and busts in the making and, you know, that would generally make you bearish on, you know, I would assume all the companies in the investment supply value chain or are the particular pockets, you know, that you find to be, you know, areas where you think that the process would be less dim. And look, I mean, we all have our own sort of specialities. So now, I mean, it perhaps it's some people look at markets where the investors or strategists can cover many different bases. And I'm not saying I'm solely a capital cycle investor, but if you are a capital cycle investor, you know, you're sort of your principal bank, just that step back whenever you see a surge of investment across many companies. You capital cycle investor can, if you, they can rationalize investing in a company that is itself investing a huge amount, providing the industry or sector that they're operating in, it is not investing massively too. But when a whole sector is plunging in, then you know, the capital cycle investor, it prefers to step back. And I, you know, yes, you know, the moment you follow the you've got a lot of the picks and shovels makers of the, or AI have all been void out in the crap, you know, the grand companies and the cornering, a lot of the same companies that actually were involved in the, in the TMT boomer now come back. And I, I, I, I, I'm my own, some, you know, copper stock and, and copper, copper is, as you know, all set up benefit sheets, I'm sort of unwitting at benefit ship. And the reason I invested in that copper stuff is not because I anticipate the AI thing, but because I thought that, um, that, that there've been a number of investment in copper and, and so you never watch, this is one of those, you know, very big investment things where everyone really gets pulled up in it where they like it or not. Right. So you actually wrote a piece recently for our, readers, where you said markets have poor scorecards for spouting AI losers. This is the piece on the anti-bubbles, right? And, and so maybe you could walk me through this because I think, you know, you talked in this piece about some cases where obviously the market got ahead of itself, saying, hey, this, they're going to be the long term winners and they were not. But there are also cases of the market pockets of the market perhaps today, even that, um, or trading kind of at disparate, depressed multiples on this idea that maybe they'll be disrupted by AI. So maybe walk me through this argument, which I think is pretty interesting. Okay. Well, um, first of all, as you may have, we used to have discussions about this at GMA that, you know, problems of shorting bubbles, yeah, because you can, you can, even if you're accurate, you can have massive drawdowns. Um, and this is probably not worth shorting a bubble, even if you have perfect foresight. Uh, then, and I hadn't, I think at a sort of, um, from a sort of pop down level, I think there is this, I think bubbles do have a sort of crowding out effect. The capital gets drawn into one sector and it gets pulled out, uh, of other parts of other parts of the market. And we saw that in the, um, in the, in the GANJ or the TNT bubble, uh, what, with what we'll call the old economy stocks, um, niccompties that were deemed to be, um, you know, were deemed not be, um, affected by, um, negatively affected by the, uh, arrive of the internet. They were, um, they, right, I represent. Uh, but the old economy stocks were businesses that were deemed not to partake in the, uh, in the new internet economy that were old economy that was soon back replaced. Now, those are economy stocks, uh, pulled off. And, um, that offered me very good. They're, you know, very good companies were selling cheap multiples, even while the market itself was on a high, uh, all time, uh, all time high. Now, and then you know, as you know, values, stocks, uh, uh, as a fact, uh, were very depressed. Small cat was very depressed. Emerging was very depressed. So, you know, there were, there were a whole load of anti bubbles in, um, by 2000. And those, those were areas where they offered, you know, really great investments. So you could make money in the anti bubbles and while the mass attack was using the 80% of its value. And then I like, you know, drew attention. I'm with you in that same piece to the, um,
the excitement over the energy transition. And if you remember, this the SPAC boom of 2021 was largely around stuff relating to electric vehicles and general all those sort of lighter companies and EVs and that's helpful. But in 2021, the traditional energy stocks were very beat and up. The energy sector went down to about 2% of the SP500 from its average level of the range, sort of 8%, 10% waiting and Tesla was worth more than the entire listed North American energy universe. And to me, it didn't take huge amount of analysis to see that this you know that the oil companies weren't going to disappear overnight, that their investments weren't going to be sort of so-called stranded assets and that EV growth was largely driven by subsidies and that so on. And so that was a really nice antibubble. You didn't have to short the SPACs, but you it had created a wonderful investment opportunity. I mean, friends, friend of mine, Johnathan Tapo, runs company called Brevard Capital. He was pointing out that there was this company called Garrett Motion that provides the turbos that goes on to internal combustion engines that had a 40% market share, 60% of incremental market share selling on 7 times P. I mean, it's in the stocks up 150% at last year or so. That's an antibubble. And you know, where are the antibubbles today? I mean, so Tapo shares the same skepticism about agentic AI, thinks that he, he hears stocks say some car auction business or you know, online travel it doesn't believe that these online businesses, whether in sort of, you know, whether real state market or travel, whether they're going to be disrupted. And he also mentioned, I think the London Stock Exchange Group that not only does the unique, does provide you the market for, you know, stock, the stock exchange, but also not a lot of other financial data that whatever. He doesn't believe that their, their, their modes are interrupted by or, or, or, or fatally compromised. And the market, I think market is revised its opinion somewhat, but definitely it took three months ago in the, you know, the market was, I think sort of overplay, but sort of selling everything. There was a sort of massive overblown set off. The, the only company, I, I saw that it, that really has been impacted, a genuinely impacted was one of these sort of California educational technology companies that provided sort of, um, essay, essay notes for lazy students and you know, stock, yeah, AI's, general place that business stocks down 95 percent. I've got no argument with that. Stage four, you know, that, that AI may make, you know, one, one or two errors in a student's essay, but that, you know, that beat, you know, we all make mistakes. It's not going to be the end of the world if it does. But you know, want AI to be booking you a ticket to London and then find that it's flying you to Timbuk too, you know, that, that is a real problem and unraveling that problem is going to be very costly for a travel company that went down that route. Got it. So, so you endeavor, believe that a lot of, what sounds like software companies or other companies that, you know, are selling down 50 plus percent on this idea that they will be disrupted by AI, that those are interesting places to at least look to the extent that, you know, you don't want a shorter bubble because as we know that can be quite challenging. So, where areas where investment has been crowded out, I mean, using potentially the perceived losers of the AI disruption, which, you know, as you write in your article very well, historically the market has a pretty bad scorecard when it comes to identifying who will actually in the long term the losers, not always the folks who initially sell off. Yeah, I mean, it requires a bit of, you know, a bit of analysis because what, you know, people who hold that view and also point out that software as a service companies, you know, were trading it with a tremendous bubble, you know, three or four years ago. And, you know, part of their selling off is really just to do with getting back to what was their value. And the other issue is that they have, you know, they love stock options. They pay out most of their earnings to their employees. So, you know, there's a business that everything else being equal, you want to be, you want to be pretty close to. And when I last looked, those, some of those companies were really not, you know, they may have, it may hold off a lot, but they weren't, they weren't, we weren't optically cheap. And I think once you take into base, into account the stock base compensation, they were probably, well, they were avoiding. So yeah, I think, you know, look for the anti bubble, but, you know, but, but the stupid time in the analysis. Right. Of course, because there will always, in these cases, also be companies that are truly being disrupted by technology and those will be zeros, right? You don't want to, you know, but be buying blockbuster into its downfall. Okay. So let's switch gears now from technology to other capital cycles. You know, when we were working together at GMO, I know you were spending a lot of time on, this was in the early 2010s, spending a lot of time on China. So I think this is a really interesting example of a case where, you know, I think the capital cycle did a good job kind of explaining what ultimately happened with China in terms of the fact that, you know, shareholders have not received a good return on their investment in Chinese stocks, despite a pretty robust economic growth. So maybe walk me through that episode and kind of, and then also bring me to the present and where we are now in terms of the capital cycle with China and other emerging markets, if you can. I wouldn't say you say investors and, well, despite robust economic growth, I'd say because, because of the growth, because the growth was faster than I knew was, was higher than the returns on, on capital. And then actually the companies that grow had to issue, had to raise more capital. So in fact, actually, if your returns are low and your growth and in an environment that are growing as a growing quote, it's actually particularly negative. And I think my lesson from China was this, you know, the lesson from China, so I said, I'm the same as this is, you don't want to, when you're looking at markets, you don't want to look at valuation alone, you want to look at valuation and, you know, returns on capital and the capital cycle. And, you know, China had, you know, was the largest investor, you know, went through the greatest investment boom ever. And, and they, and, and, and so just, as you say, you know, had to relatively strong economic crave tailed off a bit, you know, over the last few years, still raised strong growth and absolute miserable returns and, and, and the shareholders were, they are muted. I believe it's Chinese having tendency to, to add companies to the index at a very high valuation. And then, and then people who bought it at the high valuations, and the index gets diluted by these new issuances and, and, and, and, there's just a poor return. So I think the capital cycle, you know, has been, the capital cycle, NASA is very well, vindicated there. And, as you remember, we did a lot of work on the real estate. And, I, you know, while we've indicated it, yeah. I mean, yes and no. And, you know, the, I saw the other day a chart showing real, real Chinese house prices below where they were in, in 2010. And, you know, a lot of these big, you know, we used to be sure some of this, and, and, each Chinese real estate company is like, ever ground, you know, they, they've all gone. They, many of them have gone bust. And, it took, it took longer to play out than I expected. But, as you know, I have a long, we've been in this business, you know, if you, you might make a sort of, I'm not meant to say, he's signed observation and then be, you know, raised by how long it's taken to play out. So, yeah, I, I, I'm not for a Chinese estate. I knew some people think,
But I do some works with the marathon asset management. And they had machine markets people think that some of the Chinese real estate developers now, particularly in the tier one, tier two cities are in a relatively good state. I'm not following the Chinese story test not to tell you where we are now. I just from 15 years ago, I'd say that opposition have largely been indicated. - Got it. So here's another fun one. So this is one of your broiders columns. Entitled, big booze can sweat off. It's multi-year hangover. I'll let you tell the story, but basically in the COVID boom, people were stuck at home drinking booze. These stocks did well and then following the 2022 unwined they have since collapsed and price and evaluations have fallen as well. But you and your article from July last year argue that these companies are Lindy. So what does that mean? And kind of how are you thinking about these stocks today? - So the Lindy effect is a sort of joke. It's interesting. It was a notion that urban legend has derived from a Broadway cafe and someone at the bar saying, how long do you think this show is going to run for? And Bess says, the show is going to run for as long as the show has run. They know it's the show's been going for two weeks. It's the show's been going for one night. It will be for one more night then it'll be placed. If it's been going for five years, it'll run another five years. And then it turns out actually that's the sort of rule of thumb that was actually turned out to be, we can be at create. And what I was suggest is, it's probably the case for the spirits companies. I think it's been a particularly spirits because the companies that would beat and up our companies like Pané Ricca, Campari, they had show Remi-Country, Brandi sales, really took off to the pandemic. Because people were drinking and they were speculating. And they say, when they went hand in hand and they were starting, and apparently they laid it up that drinks cabinet. So by the end of the pandemic, there was the space of the drinks cabinet and they've been clearly drinking off the spirits and it says, but you could say that was a stock problem. And then the other issues is the GLP1s came along, these sort of weight loss drugs, and they apparently put people off alcohol. And then there might be an issue that, sort of younger generation is taking cash them in rather than drinking, I did it. But my guess is that these Brandi companies and spirit companies, they've been around for a couple of hundred years and you've got, you know, countries like Endergetting Rich, I don't really like to play the sort of emerging consumer demand story because it's often what the breakers do. But I just, you know, I think it's, it's obviously not hard and one can find the exception. But I think it's as a sort of good principle. It was. And where they are now, again, not following it particularly, I saw that the ad show recently stopped picked up because it had better than expected results, whereas Empire is still a little dags. I like, you know, I particularly like the story of a red, remeqon trip where you could, where the company was valued, less than the market value of the Brandi and his faults. And that sort of reminded me of, there was a story in Germany, German hyperinflation when the, when the Daimler-Benz company was worth less than the, the cars and the factory lots. And I thought that was, you know, a good story. I wouldn't surprise me at the GLP ones, you know, if that sort of, that's, it wouldn't surprise me at that. I had the, the, the diminishes somewhat over which years and eventually people will drink down there. We're all cold, you know, and they'll go, you know, they'll get back to the bottle. That's my, that, that I anyhow, that's what I sort of, there's another anti-bubble. Well, I guess one thing is that the, the AGI crowd, they don't like to drink. So maybe it's, maybe that's two sides of the same coin. And maybe it is, it is the anti-bubble. Yeah, I see. So that's probably why, if AGI, you know, were, and ready to take over, there'll be a lot of idle ads. So they might actually start drinking it, who is? Right, so maybe it's a head against AGI. We have nothing better to do. Right. Since the machines are doing all our work. Okay, so one question I had that's kind of more personal in based on my own curiosity is, you know, as you know, my, my area focuses often on intangible investments such as an R&D software advertising human capital. As you know, in the US, for example, in 1995, intangible, intangible investment award was about, you know, 12% of GDP. Since then, intangible investment has increased to what 16% of GDP, while intangible investment has fallen to 10% over the past couple of decades. And so capital cycle theory, of course, have originated, you know, many years ago, tends to be more focused on investment in physical capital. Right. And as you talk to it, the energy cycle, emerging markets, AI data centers. So my question is, you know, is it fair to assume that the capital cycle also applies to intangible capital? And if so, you know, where some notable examples that we can point to either in the past or today, where we may have seen capital cycles in, you know, mostly or purely intangible assets? - Well, I mean, capital is capital, isn't it? So I don't think, you know, if we believe in the concept of intangible capital, which I think we probably should do because the reason, you know, is really, is really due with the counting convention. Tell me if I'm largely to do with my conventions that we don't put, you know, that we tend to expense R and D and we don't tend to put brand values or, we brand values are on the book. And unless there has been acquisition, is that correct? - Yeah. - And then, yes. So then the question is, can you get a investment in intangible as in any other type of physical over investment? City arms pretty obviously, yes. Troubably, and you might be able to give me some better examples, but I remember that back in the early 1990s, huge excitement when Blacks, 18 and now Blacks, they Smith Klein, and having this great blockbuster, else drug called Zantak, that encouraged huge amount of investment in my big farmer in R and D that the cost of blockbuster drug development was stored over that period until the blockbuster drugs that they came out with were, you know, not, not, you know, didn't want to live in a decent tunnel on equity. So, you know, I think that's one that comes to mind. I think, can you name a, do you think of other intangible? I mean, as I say, I thought anything really, what we're seeing today, in the AI space we've been talking about, you've got, and you've got obviously huge physical capex in the data centers, and then a huge valuation placed on the AI scientists, you know, and metting out and I didn't hire people for, and right, you know, 100 million sellers. Is it their like, well, fairs or something? And, and, you know, that would look to be a case. I mean, there are the AI companies here, the Merrimerati, former Chief Technology Company or Open AI, Chief Technology Officer of Open AI. She left, what a couple of years ago, for my own company, they raised it, raised my evaluation of 12 billion dollars. Even they, she had, she said, "Well, I'm gonna tell investors what they're gonna put that, what the company was gonna do." So, that seems like an intangible.
"Rapital boom." I frankly, and then you know, yeah, okay, they go, they go, they go brand valuations. This is not so much Capex, but just brands getting over. I jump out in the late 1990s, New Buffett, who normally talks a lot of sets. I've got a bit carried away with the lights, Cape Cola and Gillette. We started calling them the Inervate Pulse, and they, I do think that either Cape Cola or Gillette's are attracted to you from out of Capex, but they definitely, or Capex competition, but they definitely, you know, went, you know, their intangible brands were definitely overvalued at that time and the lights of the Amiens, Gillette was eventually swallowed up by Proct and Gamble, but Cape Cola, I mean, had a very poor decade, after that. Yeah, I think that's right. You know, obviously, an asset, an asset, capital capital, whether it takes the form of a factory or, you know, the IP embedded in an NVIDIA GPU, which is, you know, this half the value of the actual data centers. Highway, it's embedded in the value added of the semiconductor. Kai, we've been talking about this software of the service. I've read, might as well draw it out, that there, I mean, I did write a piece back in 2022 about, you know, these software as a service business, where I am, you know, at certain valuations, attracting capital inslaves and several, I remember one of the capital cycle metrics is when, you know, you know, you're really in favor because someone creates an index and then update the index to prove how well it's historically performed. And then you can pretty much guess that from the main, the inception, the new index is going to do poorly. See, I think, yeah, this, you know, the SaaS bubble of 2020, is to another very good example of an intangible bubble. And another interesting kind of related topic is if you go follow human capital, just talent, where's, you know, the Harvard MBA indicator, where all these, where all the best and brightest going, right, it was big tech in the past cycle. Before that, if you remember, it was, you know, Goldman Sachs and Finance, right? So in the bubble that kind of picked, you know, 708, everyone wanted to go work for Wall Street until they didn't. And then it was tech. So, you know, maybe there's an interesting element here where it's, you know, the capital cycle has both kind of human capital elements as well as the actual physical accounting capital. Well, well, Harvard, I mean, it's just the, I mean, not just HBS, but the undergraduate, you know, the graduating body, then very, very reliable contrarian indicator. Well, I graduated from graduate, I joined Harvard undergrad and went into to work as a value investor in 2009. So I think, you know, investment at the bottom of the capital cycle. So the good idea of value investing at the time. We did that. All right. I got a couple more questions for you, Ed. So here's one that I like. So there's this idea in Silicon Valley that's popular today that bubbles may actually be productive, that they may actually be good for the economy long run. There's a book that's been making the rounds called Boom, Bubbles, and the End of Stagnation by Hobart and Heuber. So the idea is that even if these bubbles form and they eventually end in tears for investors, they're ultimately productive to the extent that they accelerate the development of a genuinely transformative technology. So in the sense, the argument will go even if AI ends up being a bubble, not that everyone there's conceding it, but to the extent it were, it would have still been a good thing to have happened. Do you find this argument at all compelling? We depend on what the respective ones looking at it from. I tend, you know, and I think you do too. We tend to view the wealth and the perspective of an investor, you know, who's trying to maximize his gains, minimize his risk. And I didn't, you know, from the perspective of society, yeah, if you can bring forward the technology, whether it's, you know, railways or cars or aircraft or internet telecommunications, or EVs or batteries or AI, that's all well and good. Sometimes, you know, sometimes it might not, the whole technology might be a sort of dead end, but perhaps not the end of the world. But still, yes, I think we talked, I mentioned the railroads earlier. They'd succinctly come in searches of CapEx, both in the US and UK as properly, I think probably good thing in the end for both economies. But really just, you know, really disastrous for the investors who partook in them. And so I think that's, you know, you have to bear that in mind. You're under, we're under no compulsion to make a loss-making investment for the benefit of society. The other issue is, and as I've already mentioned, investment, booms tend to lead to a misallocation of capital and over investment. And that tends to slow rather than increase economic growth. And even if the dot-com bubble, because of that fiber optic cable that was laid, mainly you could get, you could get netflix up and running and all of you video composing, whatever, that's all well and good. But actually remember that the downside of the dot-com bust with the market gate down, what was the S&P down, 40%, so I remember. You know, it required the, we didn't require, as a response, the federal self-flashed interest rates to 1% to fend off the inflation, and then we got a housing, we got a bus, say, in fact, the upshot, the way I see it is the upshot of the dot-com bust was the cable financial crisis, the upshot to the global financial crisis. With low interest rates and so on and so forth, is it collapsed in productivity growth. So we've got the new technology and that's all very well for companies that end up as winners and so on. But actually the very long tail of these bus can be in your big, quite severe. And so I think it's sort of irresponsible to argue that the bubbles are good for snacks. It's very tip, one of the things you see during bubble periods is a sort of light headiness. Hey, you know, my portfolio is up 50% there. What does it matter if there's a downside? But of course people don't feel like that when they're nursing sort of 90% loss on losses on yesterday's high-flowers. So, and this is a big, it what's going on in the AI space is big cat X boom as you know, taking place with very weak consumer confidence as they call K-shaped economy and problems happening elsewhere in the financial system, private credit and private equity and really a lot of legacy VC stuff that was badly invested. The temporary here too much about so the whole sort of private and you know, private alternative investment world is pretty so pretty bad place. So I would have thought that if and when this AI boom ends and turns to bus, a whole load of problems will emerge. And and and the value I'm as the valuations come down and if people just simply won't be saying that. You know, that type of poetry is a sort of giddy commentary that invariably accompanies the bubble. That's really interesting. So one point you made that that was kind of interesting in particular is this idea that the kind of like long-term consequence of the dot-com bust was you know, a fed mindset that led to ultra low interest rates for a long period of time, which you know, fed if you were, you know, and this is in your book The Price of Time, you know, kind of was the original sin behind a lot of the speculative excesses that came afterwards. Right? So that you know, something can happen and it looks fine, you can pass it over, but that's just selling the seeds for the next cycle. Now obviously rates have increased over the past few years, you know, from their from their COVID lows. Do we think that
that the regime has actually changed, or does the fact that we now are seeing a massive cab X boom in AI investment suggests that the money never left the system. - Well, I mean, I do like to hedge my bets, but I'd say a bit of bet, you know, that I think that the interest rate cycle had happened in 2021-22 for long-term yields, and historically, you know, those interest rate cycles had been to be multi-decay, you know, often, that's into 40 years. And if you remember, no, that could cheer me out as another thing, that we did a lot of work on, you know, which is like, why were our bond models so wrong? Because we always have to talk about mean reversions. And what we find is that these, these very long bond cycles, so I always say we can't, I always say you can't predict anything about long-term rates, because they're not mean reverting, like they like equities. But, you know, they do go on these very long cycles. I do think that we are, that we have entered into, you know, a prolonged period of an upward trend in long-term rates. However, there was clearly a lot of liquidity left over from that sort of COVID lockdown and quantitative easing sludge, you know, at any one, $8 trillion of money printed by the World Central Bank and, you know, a lot of that kind of sort of excessive things, a lot of it was still held on the Fed balance sheets. And that has been drawn down interest rates remain low-redited to nominal GDP growth. And so the question is, what's going to happen? Is interest rates going to go up? Or the World GDP going down, or perhaps, you know, possibly the worst, the worst, and GDP goes down and interest rates go up? So, I think there was some liquidity, a lot of liquidity left over from that period. You know, the, it's always difficult to know what's going on in the plumbing of the financial system, but, yeah, there was perhaps more liquidity than I expected, left over. And there are, as you know, these, you know, problems in private credit, which, three, just the financing arm of private equity that is still updating away, and I expect that will continue to be the case. And the real estate market is, I think, in America, it's definitely more of that, doesn't it? Because, as prices really haven't come down enough to, you know, in response to the shift of interest rates. So, that seems to be, the fact that, you know, it's interest, I haven't done work on it, but I think it's an interesting place, US, residential estate, because there hasn't been much investment there recently. But, valuation seem, probably too high, relative to interest rate regimes. So, that's, I mean, that's either a threat or an opportunity. So, it's not, I think it's not good. It's sort of middle of the line. I'm sorry, good is gonna happen. It's not like bad is gonna happen. I'm not quite sure which. - Fair enough, yeah. Okay, so, I just got one more question for you, which is our standard closing question. What's the one thing you believe about investing that most of your peers would disagree with? - So, I guess, one area where I would, you know, differ from a lot of, you know, a lot of, you know, sophisticated investors, I suppose, is that I've always been a bit of a gold bug. And, you know, people, you know, like, you know, our old boss Jeremy Granfirm and, you know, the team had been, you know, what I was saying, I didn't like gold has no difference than this, that. And I, you know, I always, you know, I think I always felt that, I was really what? Yeah, I always got this at a vistic attraction to gold. And I think that it does, I think I do like gold's hedging aspect. I know it is difficult value, but I like the way that gold is an asset without a liability. And I think that in a world where equities, you know, taking the US, I've looked overvalued, and Bournem's might be on, you know, 30, 40 years, I would trend in valuations as well as, you know, really severe debt problem I'd have makes. I, I, I hold the view that having a decent allocation to gold is a good portfolio position. So, and that, you know, I mean, I, look, I mean, this could be right for you. The last 10 years actually having gold, you're having a sort of, it's gold, you know, it's been obviously a lot better than having bought and say, say, I might be, I might be tightening a, I'm sorry, it's sort of near the end of, near the end of its view, but I, I, I, I'm drawn to gold in a way that they have rich, you know, and persons, you know, they have reached TFA is not. So I think that's probably the, the one area where I might disagree. Interesting. Well, thank you, Ed. I know I've taken up a lot of your time. So really appreciate you coming on. Good guy in there. I'll see you in London soon. See you soon. Thank you for tuning into this episode. If you found this discussion interesting and valuable, please subscribe on your favorite audio platform or on YouTube. You can also follow all the podcasts in the excess returns network and excess returns pod.com. If you have any feedback or questions, you can contact us at excess returns
[email protected]. No information on this podcast should be construed as investment advice. Securities discussed in the podcast may be holdings of the firms of the hosts or their clients.
PodcastTranscript AI