Tim Graf (TG): This is Street Signals, a weekly conversation about markets and macro brought to you by State Street Markets. I'm your host, Tim Graf, head of macro strategy for Europe. Each week, we talk about the latest insights from our award-winning research, as well as the current thinking from our strategists, traders, business leaders, clients, and other experts from financial markets. If you listen to us and like what you're hearing, please subscribe, leave us a good review, get in touch, it all helps us to improve what we offer. With that, here's what's on our minds this week.

Matt Smith (MS): The more negative case is just simply that the AI boom begins to slow, CapEx slows, revenues don't come through as people expect. That begins to put some downward pressure on and quite optimistic earnings estimates.

If at the same time, the lagged effect of interest rates really begins to wear down this low-end consumer, you know, housing is already under significant pressure, these positive feedback loops that have ratcheted the economy up begin to work the other way around.

TG: The last week has reminded us of two pertinent facts in financial markets. First, trends are never one way, especially when they get into hyper-extended territory. But also, when in place, long-run momentum can be very, very hard to break. And so here we are again, after a tweet-induced day or two of volatility, we have equity markets back near all-time highs, and the resilience of tech stocks successfully passing yet another test. Even though some of the froth around stocks has been reduced, valuation concerns remain. And the search for value in markets has been a frustrating one. But as our guest this week points out, if you do it right, returns are still out there to be had, and there are long-term focused investors who still understand the appeal of a more patient, fundamental oriented process.

I got a chance to talk to Matt Smith this week. He's a value-focused portfolio manager at QSM Asset Management, a long-short equity hedge fund that's based here in the UK. Matt gave a really interesting guide on how to think about markets for the long run, as well as how he's thinking about stocks with traditional valuation measures back into nosebleed territory.

Thanks, Matt, really appreciate you joining this week. And it's great to meet you as well.

Just wanted to get things started with your background, learn a little bit about you, your career path, what has kind of driven you towards value investing in your career?

MS: I've done 30 years in the industry now, and of that, seven years was on the sales side. So, I worked in equity research at Credit Suisse and Deutsche Bank. But most of my career has been on the buy side. So, for the last 18 years, I've been a portfolio manager on a global equity long short fund. Most of the time, that was at Majedie. And then in 2024, a few of us left and we set up a new business called QSM.

In terms of what drew me to value investing, I think really people are, how people invest really comes down to how they're mentally and emotionally wired, I guess. So I've always been slightly contrarian and, you know, I like a bit of a bargain and not really a trend follower, I guess. So, I mean, that's really why fundamentally I got drawn to value investing. From our perspective, really what that means is it's just a function of the fact that people and markets tend to get overly emotional at certain points. And so what tends to happen is that companies become particularly cheap when trading has been difficult and they can become particularly expensive when they're coming off a really strong run. And so we just take the other side of that. I mean, it is just a function of trying to find these situations where everything has become just a bit too extended in either directions.

TG: Thinking about people you've worked with, but also, I mean, look, there are legends of value investing like Warren Buffett and Ben Graham. And I wanted to talk a little bit actually about the literature on that, how much you follow it. But people like that, is that who you kind of look to for your long-term inspiration?

Or are there people you've worked with over the years that you thought, or even not worked with, but kind of worked adjacent to in the industry who you thought, you know, that's a really good model to follow when it comes to this style of investing.

MS: The broad framework of value investing, there's sort of variations to it, but the broad framework of it hasn't really changed much in the last 20 to 30 years. I mean, and I think most people would agree that you're trying to find decent businesses at a point where they're trading at significant discounts to where they traded historically.

And I mean, I think there is a bit of a discussion about defining what a decent business is and within value investing, you know, I guess you can split it into people who just focus solely on the higher quality end of things and people all get into deep value. So there's, you know, there's a bit of a discussion around that. But I think the broad framework in terms of trying to find what is, in the long run, a good business trading at a substantial discount, I mean, it really is pretty straightforward, to be honest.

I think the difficulty really is rather than the intellectual framework of it, which I think most people could understand. It's the practical implementation of value that's quite difficult. You know, particularly in recent years, markets have become a bit more momentum orientated. And the flow of money that you've had into a combination of short, shorter term trading, momentum-orientated trading, you know, passive trading, out of active funds, all this sort of thing, that's made the implementation of value a bit more difficult. And the real driver of that is the fact that the trends tend to get a lot more extended.

I mean, in a way, you could say, well, you know, the rise of sort of algos and machine-based trading and all this sort of thing, that should make the market more efficient. And I guess it does make the market extremely efficient on a five-minute to one-day view. But arguably, the rise of the momentum, in particular momentum trading, it really leads to things being pushed potentially quite a long way from fair value.

The thing I think I've learned really over the years is it's more the, I guess, emotional behavior around that, in terms of the fact that it's sort of a given that you have to, everybody has to be diligent and you have to understand, well, not everyone has to be diligent, but to be successful, you have to be diligent. You have to have a very strong grip on the fundamentals. And if you've got a strong grip on the fundamentals, it helps you stick with positions because when prices are moving around, if you understand fundamentally what the business is worth, it's easier to look through those price moves.

But on top of that, it's just that it's really remaining very level-headed through the price swings and also just quite patient. I mean, I think in an era of sort of everybody's got attention deficit disorder to varying degrees in markets these days, I think. And it's the ability to look through the noise and the volatility, ride out some of the short-term pain in the view of generating some pretty decent long-term returns. So when I was working at Majedie, there was James Duffer and Chris Field, and they're very experienced managers. And I think it's that level-headedness and patience in volatile markets that's the difficult bit of value. It's easy to understand. It's a bit more difficult to do it practically.

TG: Actually, I think one of the key questions I had for you was, you mentioned the diligence and following fundamentals, but of course, you also alluded to the fact that momentum is driving markets. And it almost seems like people are being rewarded financially for not doing the due diligence and just following trends and attaching themselves to things that are going up, where obviously value investing is a little bit more thoughtful than that. I'm wondering how you talk to your investors about that, when that does seem to be the fad and growth over a value has been the theme that people have played out for the last 20 years and how you manage expectations around that, I guess.

MS: Well, I mean, that's the difficulty really is that at the most basic level, sort of value is deferred upside or deferred pleasure, if you like, whereas growth is just sort of instant gratification. I mean, that's really the trade-off. And where it fits in with investors, I guess, is that they, the people in invested in our fund, I think they understand what we're doing. We've made it quite clear that we take a three-to-five-year view. There's a long-term track record of 15, 16 years in the prior fund, and the people invested understand all that. So, I mean, we're lucky that there's still a few people like that in the market.

And I think there's a, I suppose, if you're running a multi-manager portfolio, or you're trying to put together some, a portfolio that's resilient to different outcomes, let's put it that way, then I guess you could say that there is an argument that you, especially where spreads are currently, in terms of the cheapness of cheap stocks relative to expensive stocks, that spread is historically wide. There's an argument that perhaps, even though it's not been particularly well looked upon over the last 20 years, that you should have something like that in your overall portfolio to give you some exposure if these trends begin to reverse after a very long run.

I suppose an example would be something like the bond market, wouldn't it, where bond yields peaked in the early 80s or whatever. And you really had a 40-year bull market in bonds, and it became axiomatic that yields were not going up. And there were lots of arguments why that demographics and all this sort of thing as to why yields could never go up. And then they turn, and then you've gone from 0 percent to 5 percent on the or 1 percent to 2 percent on the long end to 5 percent in sort of the blink of an eye in a way.

And so once the market ends up very tilted to one side, once the turn comes, it can be quite vicious, I guess. And so I think the attraction of a value-based fund is just that you're potentially getting exposure to that long-term turn in performance, that when it really kicks in, you're probably going to be too late, or you're going to miss quite a lot of upsides in quite a short period of time. So, I mean, people either subscribe to that view or they don't. I mean, I can understand why you wouldn't, because the 20-year performance of value has been pretty difficult.

TG: So yeah, I'm thinking about your ideas specifically that you have that you invest in or short sell ideas. You mentioned you had a three to five year view, but I'm just curious, is that kind of the time frame you expect your specific trades to work out, or do you put them on a little bit of a shorter leash, say if they're not working, or if they play out to your expectations?

How long does it take for a thesis that you have to typically play out? And again, how do you manage expectations around that, both for yourself as well as for your client?

MS: If you think about where the weight of money is currently, the weight of money is in time frames. I mean, I'll argue somewhere between a month and three months. I mean, there's a lot of money that is chasing one to three month returns, returns over results, and all this sort of thing. So, the market is just incredibly efficient at that time frame would be my view, because there's just so many people chasing it.

As you take your time frame longer out, there's less and less competition in terms of the fact that the market just isn't generally thinking at that level. So, the three-to-five-year view really is, we do quite a lot of get involved in a lot of restructuring stories. And I mean, a three to five year is a reasonable time frame for a company to be turned around. So, what we're trying to do is we get into a diversified portfolio of these long-term turnarounds. And then it's trying to manage the portfolio from A to B. And what that really involves is that, you have to have a long-term view in terms of what goes into the portfolio, but a shorter-term view in terms of trying to manage the momentum around that. And so that's a combination of position sizing and making sure that you've got various things firing at once.

So, you can't have the whole portfolio in quite early-stage ideas that, you know, are just incredibly volatile and haven't really gained traction. You sort of need a mix in the portfolio, where you've got some early-stage ideas that have come under a lot of pressure, but new management have just been put in, but it's not really gained traction. And some ideas that are quite a long way through the restructuring process, where management has really got a grip of it, earnings momentum has really began to kick in, price momentum is sort of decent. And so it's about balancing the portfolio in that way, because what you're trying to avoid is just massive drawdowns. You know, nobody wants massive drawdowns. You know, I mean, everybody's quite happy to take a long-term view, but between A and B, they don't want to be incurring significant losses. So, you have to try and manage momentum within the context of a sort of a long-term portfolio, if that makes sense.

TG: And I'm curious about the idea generation process, where you find these opportunities and particularly how you assess valuation, you know, what are the kind of your… the things you fall back on, you mentioned before that the fundamentals of value investing probably really haven't changed since Ben Graham probably in certain senses.

But what sort of things do you fall back on? And I'm particularly interested about what you mentioned about the various time cycles and whether that actually changes depending on where you assess the timeline to be for a particular opportunity.

MS: I guess a couple of the big ones would be things like screens. You can screen on pure valuation as in which sectors and stocks are particularly cheap relative to their history or the market. You can screen for stocks that are cheap, that have come under significant pressure pricewise. And then you can add filters to that, where what you're ideally trying to find is, say, a stock that has fallen is cheap, but is showing some signs of earnings momentum kicking in, or can define price momentum in lots of different ways. But I mean, it's really not that difficult to come up with a screen of stocks that might look interesting. They've gone down, they're now cheap, momentum appears to be picking up. And then the trick is trying to filter through that screen to get to the ideas.

Some of them will be the standard value trap that looks cheap, but stays cheap. And some of them are going to be the winners that are cheap, are going to become a lot less cheap and go up three, five, seven times or whatever. So I mean, obviously, there's some skill involved in that. And in terms of the metrics that we'd look at, if you split it into cyclicals versus defensives, on the cyclical side, things like EV to sales, really, I mean, I'd say that's probably, if you had to pick one metric, that's probably one of the best metrics. If you look at EV to sales relative to history, you compare that to where margins are relative to the history. And it's quite easy to interpret. If you can buy a stock on a very depressed EV to sales multiple where you think that margins have not been impaired relative to where they've been historically, then probably that company is pretty cheap. So, for more defensive companies, you can look at things like free cash flow yield relative to history would probably be one of the better ones. And then also depending on how asset-intensive something like price to book versus ROE. I mean, those three things are quite simple ways to look at it. They'd be the things that we'd focus on.

TG: Is there any element- I mean, you're very experienced at it. And I'm curious what sort of qualitative factors go into it based on kind of experience or, I don't know, call it gut feel, maybe.

But also not just that, but like listening to management calls, I mean, do you listen to earnings calls for your ideas, or not just for your ideas, but to kind of get a sense of whether the ideas you have are worth pursuing? Does that factor in as well?

MS: The earnings call transcript thing is quite interesting because if you, obviously like a company does an hour-long results call, and then you read, I don't know, three or four. Let's say you read five results notes about the results call. And you'll get often, I mean, not always, but often you'll have three or four completely different views on what, even down to differentials in terms of what the management actually said, not just the interpretation of it. And it's because of the fact that if you're an analyst on results day and you're listening to a results call, you're sort of trying to listen to the call and the trading desk is probably phoning you up and maybe some clients are e-mailing you and you're partially focused on it. And also, there's a big perception thing in terms of, I think everybody's guilty if you hear what you want to hear to a degree. You've got certain biases going into a call, you then hear what they're saying and you fit that around your biases.

So personally, I think it's quite important to actually listen to the results calls. And what I tend to do, we tend to go through it the day after in terms of once you've got a transcript and you can go through it in a lot of detail. And that way, I think sometimes you can get a quite different take as to what the management actually said compared to what the market slightly thinks the management said. And I think that there is an experience element that if you've been reading, you know, you've been going to results calls for whatever, 20, 30 years, then you've, you know, sometimes management saying things quite obliquely that actually are quite important and people aren't picking up quite so much. So yeah, I mean, I think that the results, going through results in detail is quite important for us, yeah.

TG: Do you have a blind spot that you're aware of that you need to work to kind of overcome throughout your career, just something that you know, okay, this is what I tend to do and I need to make sure I don't do that this time?

MS: Yeah, I mean, the two, I've got two sort of big blind spots if you like. And the first one is, I mean, just getting in too early. I mean, I've been talking about it for 15 years and I think we're probably getting a little bit better at it, but you know, it's still a problem in terms of the fact that I think sometimes things are, I mean, less so on the short side, the short side is difficult, but on the long side, you can often see how things are going to play out on a sort of a two- to three-year view.

And if I've managed to convince myself that I can see, you know, you've got new management team in, they're doing the right thing, the strategy is in place, you're beginning to see the industry, like let's say there's a supply problem in the industry, which has put pressure on prices, and you're beginning to see capacity coming out of the industry, all this sort of thing. Then I tend to think, you know, you can see, especially if there's a lot of upside, let's say you think that the stock could easily triple, then I tend to just sort of start getting involved, when in reality, it can take, I mean, especially now, it can take the market a long time to get engaged in a turnaround story.

And I think the European bank sector, I mean, I think it's just sort of a prime example of that, where, you know, I mean, they had a very difficult financial crisis back in sort of 2009 or whatever, you know, that they then spend, let's say, best part of 10 years sorting out their balance sheets. So at the same time, they're dramatically improving their cost basis. So, you've got just fundamentally, significantly improved businesses.

And it wasn't until the earnings momentum really, really kicked in, you know, as central banks started raising rates, that, you know, the market gets engaged. And then suddenly, the sector's sort of tripled and quadrupled in a short period of time. So, the problem is that that timing of, you know, exactly when is the market going to get on board with this?

I mean, we've got, I won't sort of go into the individual stocks, but, you know, we've got stocks where one of the things that was important was to de-leverage the balance sheet. So, you know, sometimes we'll buy businesses that have got quite a lot of gearing, where they're on a very low multiple because the market's worried about the gearing. And de-leveraging leads to a big re-rating of the stock because you reduce the financial risk.

But, you know, we've got companies where they've halved their debt burden and halved their net debt to EBITDA ratio, and there's been no impact on the multiple. So these things can just take quite a long time to play out. So, getting in early is one, and then just sometimes just cutting things a bit too quickly. And those two things in combination are a bit lethal to be honest. So you have to, you know, those are the two things that I find I have to work on most.

TG: Yeah, the second one is one of the more commonly noted ones. Yeah, you're not alone. Thinking about things on the short side, I wanted to know how your process is any different than what you do when you're looking for investment opportunities on the long side. I mean, as you alluded to before, it's a bit more difficult.

How does the process change when you're looking for short selling opportunities?

MS: Yeah, I mean, the big problem between longs and shorts is it's just this, your risk trade-offs are just completely different, aren't they? So, on a long, uncapped upside, downsides limited to you lose all your money. But so you can easily get to three, four, five to one risk reward trade-offs on stocks if you're right. Whereas on the shorts, your upside is you make 100 percent unless you keep scaling it up on the way down, obviously. But your downside is sort of uncapped. So fundamentally, you just have to be a lot more careful. And the problem with the short side is you do need to generally really get the timing right in terms of the fact that it's just very dangerous being short stocks that are being upgraded. I mean, that's just the fundamental reality. Really on the short side, the focus is on trying to ensure that we're in stocks where revenue and profit downgrade are going to kick in in reasonably short order.

That's the key. And then keeping them relatively small and extremely liquid. The slight weird thing when you look across the market through time is that the weight of short positions is very heavily biased to the worst performing stocks in the market, in some ways is obvious. In other ways, at certain points, you can properly be run over because if there's a large amount of short interest in an extremely cheap stock that's been through a period of, a long period of difficulty and is beginning to turn around, you can just get explosive moves to the upside. So, we don't get involved in any of that.

We're really focused on very highly valued stocks where everybody's really bullish, where we hope that we can find some sort of angle where they're going to get disappointed on the revenue and the profit outlook.

TG: Especially in these markets, is there any kind of day-to-day risk management that you do around that? Like very short term, I'm thinking literally day-to-day or within a week where you think, okay, maybe this idea is not necessarily going to work out in the short run, even if it is something I have high conviction in on that kind of longer-term view.

MS: To be honest, we're more driven by, I mean, rightly or wrongly, we're more driven by just the fundamentals, just moving in the right direction or not. I mean, I think there's lots of traders who are extremely good at short term trading, and I definitely would not put us in that bucket. So, what we try to do is you come up with your view on why this stock is cheap or expensive. And then the thing that will make us change our view is if the fundamentals are moving. I mean, obviously, if the price moves significantly against you, then it makes you reassess, are you missing something on the fundamental side? Unless the fundamentals are moving, we're probably not going to be doing much in terms of trading around positions. I mean, occasionally, if things have had a massive run-up or we might trim some, but really, it's more about just checking, are the fundamentals right or wrong?

TG: Okay, well, let's shift from the concepts to the application and talk about current markets. I mean, as a starting point, I never like to cite the Bank of America Fund Manager Survey because we are a competitor to it, and we do a lot of flow of funds work. But there was something that caught my eye last week as I was prepping for this, which is this notion that of the fund managers surveyed, it's something like 90, 91 percent see the equity market as overvalued. And now thinking about our own institutional investor flows, we see nothing but kind of active weight, off benchmark exposure inflows into the US equity market in particular, tech especially, we'll talk a little bit about that.

This disconnect between the two, how do you square that? How do you talk about something being overvalued and yet still keep allocating capital to it?

MS: It really is just this thing of agency risk, career risk type thing where, I mean, we're in a market that's done effectively bottom left to top right for sort of 16 years off the 2009 lows, isn't it? And in a quite spectacular way in terms of double digit returns sort of year in, year out with the odd drawdown. There's a Darwinian evolution where if you were of a bearish bent or you were sort of anti-US equities or this, I mean, you've basically been weeded out largely. You either have lost all the money that you manage or you've been fired or you've just given up. So there's a bit of a sort of an evolution within markets that leads to a concentration of money in the people who are naturally biased towards buying US equities. I mean, that is sort of one thing that's happened, I think, generally.

And then it's just this thing that, I just think it's true that if you're a fund manager, a lot of fund managers are worried about what returns they're generating and all this sort of thing. But really losing money or significantly underperforming rising markets is just career-wise a difficult proposition. People just think, well, the risk of me not being engaged with this market that seems to recover from every sell-off and does double-digit returns year in, year out is just too high. And I can see that it all looks a bit nuts, but everyone else is involved. And so, I can't not be. And then the sort of the extension of that is, I think there's a feeling that if you lose money when everyone else is losing money, then that's not so bad as losing money when everyone else is making money. There's just sort of big behavioral things going on. And it's just a function of the fact that you have had such a long run of such strong performance out of the US equity market.

TG: The last couple of weeks, you hear a lot of comparisons to 1999, 2000, and the bubble word gets thrown around. And there's cases that are made very convincingly for and against that, that were kind of in the baseball terminology, we're in the early innings or we're in the late innings.

Do you have a sense of how the equity market feels relative to some of those historic parallels at this particular time?

MS: I mean, I was working at CSFB in the height of the dot-com bubble in sort of 99 into 2000 and all that sort of stuff. And at the time, CSFB was, I mean, to a degree, it was at the center of it in that you had, there was a famous banker called Frank Quattrone who was running their tech business here.

TG: I was there as well, 2000 to 2008. So, yeah.

MS: Yeah. So, he was a character. And then, at the time, you saw that there was, you know, well, I guess one feature was, is you had from the banking side, you had a number of IPOs that were sort of questionable. And I guess the people would say, you are getting a surge in IPOs now, and there are some companies coming with relatively low levels of profitability, but you're not getting the surge of IPOs that you had in 2000. So, that's like the bull case is that, well, actually, in 2000, we had this surge in dodgy companies floating, and you haven't really seen activity move to that level this time around.

But I think the difference is that you could argue that a similar mechanic is going on behind the scenes. It's just all happening in VC funds and private equity. There's no real reason for those sorts of companies that used to have to float to get the cash or to get the funding to come to public equity markets anymore. They're really just raising money and staying private for lots of different reasons. So, I think that where in 2000, you saw the surge in IPOs, and people are saying, well, you're not seeing that now, so you don't really need to worry about it. I mean, I'm not sure that's actually true. You are seeing those companies being funded. It's just not being funded in the private markets.

In terms of the similarities, one of the things that I slightly vividly remember was in 1999, in the run up to the top, what would happen is these companies would announce that they were going to launch a website, and the shares would do 10 or 15 percent in a day. I mean, one of the ones that really struck me for some reason was there was a company called EMAP, which was like a UK-based magazine company, and they came out and said, we're going to launch a website, and the share price shot up. And 20 years later, the magazine industry has not been helped by the internet. And you see a similar thing now with, I mean, it's like Broadcom today, isn't it, with people just have to announce some deal with OpenAI, stock moves, huge moves at Oracle and AMD and all this sort of stuff. So, I'd say that the way that you're getting just, not just large percentage moves, but just absolutely enormous market cap moves.

I mean, like multiple hundred-billion-dollar market cap moves on deals that they may or may not come to fruition, put it that way. I mean, I guess that's a sign that people are just quite excited about the whole thing. So that would be, I guess, another thing, one thing that is pretty similar to what you saw in 2000. And then the other thing is just the crowding. I mean, I think that the two interesting things about 2000 was the crowding into certain ideas. So I mean, obviously, in the market at the moment, you've got this massive outperformance from the Mag 7 and anything sort of AI related. So, sort of power through to the chip companies in the lower restaurants and all this sort of thing. There's been a huge concentration in performance around that. And everything outside of that has sort of been left behind, partly because the economy is a bit more difficult.

But the market is very bifurcated. You're either sort of AI or you're not. And if you're anything to do with AI, you could argue there's not a huge amount of differentiation. It's like AI related, tick. You go in the right ETF, you get massive flows. The stocks perform really strongly. And that's very similar to the, you know, when you think back to 1999, I mean, that was basically all about TMT versus old economy. So you were either tech media telco or you were old economy. If you were tech media telco, the shares went exponential. If you were old economy, then there was almost no price that people were prepared to pay for those assets. And the second interesting point about that is that the market can just get things just unbelievably wrong in terms of if you think of TMT. So, in 2000, they bundled together tech media telcos. The market was sort of right about tech. I mean, you had a 50-60 percent drawdown in a couple of years afterwards. But on a 20-year view, the market was right about some of the tech companies, Amazon and all this sort of thing, Microsoft, etcetera. Even if you bought them at the peak, you've made money on a 20-year view.

But media and telcos, which was the other two-thirds of it, it's just been an absolute horror show. I mean, a lot of the media companies either don't exist or have been persistent and significant underperformers and sort of the same for telcos. So, I suppose the interesting thing is you've got a similar bifurcation that you had in 2000 in terms of market performance and attention and flows of funds and all that sort of thing. And then when you think back to 2000, it's pretty clear that the market's guesses about who are going to be the big winners out of this new technological change can just be off by 180 degrees basically.

TG: You wrote about the AI and AI adjacency, I don't know what you want to call it at this point. There's circularity to it both in terms of they invest in each other, they buy from each other. But you highlighted a lot of the risks in your most recent investor letter that I read. And I was wondering, that's an interesting framework actually to think about it in I think, in that there surely will be mega cap companies who survive, but then there may be some who struggle over the long run and will be the ones washed out.

How do you think about that in the current environment? Where do you see the massive risks accruing?

And which parts of the market, like you mentioned media and telco, which parts of the market are they going to be potentially in this cycle that maybe don't quite make it?

MS: You sort of got to ask yourself, why have they gone on such a cap ex bender? And obviously, the large driver of it is the fact that you've had an explosion in demand in terms of the fact that the AI models had a problem in the fact that they sort of hallucinated a lot. So, whatever the number was, 20 percent of answers could just be sort of invented. And so then the next step in the models was to move to these models that basically do iterative thinking. And as you got iterative thinking, the processing required to, which did improve hallucination significantly, maybe cut it by 90 percent, not to zero, but significantly. But that led to an explosion in tokens, and that led to an explosion in demand for compute. And so that's the sort of fundamental driver that's led to this pickup in CapEx. But when you look at Microsoft and CapEx is up, from 2019 to next year, it's going to be up about seven times. I mean, I think part of that as well is just reflects the fact that for all these big tech companies, it's potentially very positive would be their argument.

It's potentially also an existential risk in terms of the fact that it's creating this, just this sort of phase shift in the technology industry, and nobody really knows how it's going to work out. So I mean, if you think about, you know, OpenAI is one of Microsoft's biggest customers, then they don't really disclose how big it is, but OpenAI is now talking about shifting its compute to other people. I mean, what does that mean for Microsoft? And, you know, they're talking about becoming the operating system for the computing industry. And, you know, what does that mean for Microsoft? And then, you know, Google is, they really started the whole thing, but a bit slow off the start. And so that gave OpenAI a bit of a lead in the race. And then now people are worried about what does that mean for search? You know, is ChatGPT just going to eat into Google search revenues?

And obviously, sort of, Meta is worried, and that's why they're going out and trying to recruit all of these, you know, all the AI guys on, you know, US$100 million salaries and all this sort of stuff. I mean, I think the behavior illustrates the fact that deep down, they could be quite concerned about what this actually means for their business.

So, I mean, I think if you looked back at most technological revolutions, it's sort of relatively unusual for the winners of one to carry through to the winners of the other. I mean, you know, Microsoft was an unusual example in that they really managed to pull off the sort of PC, Windows through to cloud, and now they're trying to pull it off into AI, but it's unclear.

I guess the issue is, is you've got these enormously profitable, extremely highly valued, very large market cap companies who are facing fundamentally, potentially quite a disruptive environment. I mean, the irony is, I mean, if you think about Google, so Google, you know, rolled out Gemini AI into the search bar on, in Chrome browsers. So, you know, the Chrome browsers used by about three and a half billion people, I think. And the reason they've done that is because, well, I mean, who knows, but plausibly the reason they've done that is to head off the risk to the core search business. But at least if they're going to lose share in search, they'd rather lose it to their own AI service rather than open AI. But I mean, what does that mean for the sort of monetization of AI services?

If, I mean, it's a bit like, if you think back, you know, Zoom was a big thing, wasn't it? During the pandemic, it shot up a huge market cap. And then Microsoft bundles Teams into 365, and that's sort of the end of that. I mean, Zoom market cap is down sort of 80 percent or whatever. So there's just this issue that if the risk is so existential, that the companies are prepared to take a profit hit, the incumbents, if you like, are prepared to take a profit hit in order to bundle AI services for free, then what does that mean for the overall monetization of all of this just pan-galactic amounts of capex? These, I think, are the sort of the question marks hanging over it.
So, it's just the uncertainty as to how all this is going to get monetized. What does it mean for the incumbents? Are they going to be able to retain all of these legacy profit pools?

You know, I mean, another historical example is if you take Vodafone, and Vodafone went parabolic in 2000, and one of the things that everybody was talking about was the fact that 3G was coming along, everybody was going to be doing video calls, that was going to be a huge amount of revenue for Vodafone.

What happened to video calls? Well, video calls basically ended up being done for free because of the fact that over-the-top services like WhatsApp and everyone else came in, brushed aside the sort of Vodafone side of it, and then they're serving that up, and the revenue to Vodafone is effectively zero. I mean, you've got to pay for your data each year, but that was really no more expensive than you were paying for your phone calls originally.

So, when you have this disruption, these sort(s) of legacy profit pools can just come under pressure. So I mean, that is what we'd see as the main risk, you know. Are there risks within the investing world and particularly within AI and all of its adjacencies that you've mentioned to the macro economy?

TG: We typically don't necessarily think of stock market corrections if they do come as necessarily being bad for consumers because typically, your asset owner class has the lower marginal propensity to consume. But that has changed a little bit in that. I think it's now half of US consumption comes from something like the top decile of income earners or asset owners.

Do you think this is a little bit more nefarious potentially for the US economy if you do get a correction?

MS: If you just look at S&P market cap, I mean, it's now it's 1.9 times GDP. So, I mean, that's roughly almost twice what it is in 2000. Obviously, not all US equities are owned by US citizens. But I mean, broadly, every 10 percent move in the S&P is about an 18 percent swing of US GDP in terms of wealth effects.

So, if you think back from 2019 through to now, you've had about a 30, well, a bit over a US$30 trillion surge in market cap in the S&P. And I mean, that's roughly the same as current US GDP, which is also US$30 trillion. So ,there's a bit of an argument about how exactly do these wealth effects work. But effectively, that huge wealth effect, so US$30 trillion of extra market cap, 87 percent of all of the equities owned by US households are owned by the top 10 percent of households.

So really, that massive gain has accrued to quite a small part of the US population. And those guys, like you said, that top 10 percent is about 50 percent of US consumption. And then if you think about the bottom 50 percent of the income distribution, only own 1 percent of the total equities owned by households. So, they've got some exposure. It's not really anything compared to what the exposure that the top 10 percent have got.

One argument is that the US economy is just really lopsided because you've had these sort(s) of shocks thrown at it in terms of historic surge in Fed funds. And then you've got the tariff shock that's beginning to unfold at the moment. I mean, I think everybody has just been surprised at the resilience of it. But arguably, it's just simply the fact that there's been so much wealth created in the stock market that this has just supported this high-end spending, which has allowed the economy to sort of regain momentum and sort of power through the shocks. The risk is that what's really going on under the hood is you've got this sort of half of consumption coming from 10 percent of the population that is robust and tied into the equity market. And on the other side, you've got the other half of consumption that's related to the other 90 percent of the population. And for them, it's just a bit more of a difficult time because you've got a price level that is significantly higher because of the pandemic inflation, and you've got the tariffs coming through and you've got relatively high fed funds.

So they're struggling. And so that's why, I mean, there's been lots of companies that have been talking about this recently that they talk about something along the lines of the high-end consumer remains robust, but elements of the lower end consumer are under pressure. The issue is how does that sort of play out? The more negative case is just simply that the AI boom begins to slow. CAPEX slows. Revenues don't come through as people expect. That begins to put some downward pressure on and quite optimistic earnings estimates. If at the same time, the lagged effect of interest rate really begins to wear down this sort of low-end consumer. Housing is already under significant pressure. The low end is beginning to buckle.

These things like Tricolor, which was subprime auto, I mean, there's just sort of signs of cracks. If that begins to crack at the same time that the AI boom begins to come under pressure, then you could see S&P earnings forecasts begin to come under a bit of pressure. If they come under pressure at a point of historically high multiples, then obviously the downside can be quite significant for equity markets. And then what happens is these positive feedback loops that have ratcheted the economy up begin to work the other way around. So, if wealth effects start to move negative, high-end consumers maybe pull back a bit, that puts more of pressure on the economy, more downward pressure on earnings.

To a large extent, that's what happened in 2000s. You had a stock market shock that triggered wealth effects and a fall in spending in tech that then rippled through the economy. But this time around, the wealth effects could be bigger, given that the stock market is so much bigger relative to GDP than it was previously. So, I guess that's the downside risk.

TG: Thinking about it in terms of market regimes, earlier this year, I read Barton Biggs' old book called Hedgehogging, and it's not unique to him, but he divided the market cycles up into secular bull and bear markets versus kind of cyclical bear and bull markets. 2000 was the kicking off point for a secular bear market that lasted really through the financial crisis, I guess, till about 2013 was when we sort of broke out of that into a proper bull market that has now gone on for the last 12 years.

At the same time, we've had this evolution in investing where passive flows come in and money is stickier, it seems like, or at least there's more willingness to buy dips, I guess.

Is psychology around that so difficult to overcome that secular bear markets are, if not impossible, very difficult to see? Or could we be on the cusp of something like that related to some of the things you just highlighted?

MS: If you think back to 2007, then the accepted wisdom was that there had not, I think this is right, there had not been a nationwide decline in US house prices since the Great Depression. And so, the argument was, well, house prices just can't go down. They might be flat, but they're not going to go down. And if they go down, they're not going to go down very much. And so, that sort of fundamental belief led to just a whole lot of wacky stuff going on in terms of people gearing up exposure in loads of different ways to the housing market. And then when it turned out that the housing market could actually go down because there had been a historic supply surge and far too much leverage in the system, I mean, it got pretty messy, pretty quickly basically.

So, I think the problem is that if people become completely convinced that the only thing that can happen is an asset can go up, it almost breeds the conditions for that asset price to fall quite sharply. If you think about just the current structure of markets, I mean, just noddy things like the Economist had an article two weeks ago on why leveraged long investing was a good idea. And so they were using the example of some guy in the state who basically was running a portfolio with a 60 percent LTV.

So 60 percent LTV is fine when the market goes up. If the market goes down 40 percent, I mean, pretty much you're wiped out. So it's not so good. So, if you were trying to think about what the right level of loan to value for a leveraged long equity portfolio is, and you look back at the fact that, well, markets have fallen 40 percent twice in the last 25 years, then probably that's a pretty aggressive strategy to be running. And then the sort of surge in options trading and the fact that everybody now, it is just completely axiomatic that US equities just go up. And if they go down, they're not going to go down for very long and you just buy the dip. The more people believe that, the more risky the market environment becomes because of the fact that it just leads to excessive amounts of risk taking. And the problem is, is it will carry on till it doesn't. But when it stops, it could be a bit messier than people think.

And the problem is, is the starting point is, S&P to PEs in the top decile of their 20-year range. And that PEs at the top decile at the same time that margins are close to record. So, there's a wide amount of literature that's saying that the best guide to 10 year returns from equities is the starting PEs, basically. And if you, you know, on a one-year view, it doesn't mean anything. On a two-year view, it doesn't mean a whole lot either. But on a 10-year view, it's a key driver of return. So, if you're starting at a very elevated PEs, then your return expectation should be relatively low. But the issue is, is all these, you know, for a lot of investors at the moment, are they really expecting to get, say, 4 percent out of US equities? And probably not. And if that 4 percent comes with quite a lot of volatility, you know, nobody knows when it's going to happen, but there's a risk that people are going to have to fundamentally reassess what their return expectations are. And then if you think about that relative to bonds, you know, maybe bonds aren't quite ridiculous an idea everybody thinks currently, as they might seem.

TG: Well, Matt, I always like to bring things full circle. And I always like to finish with things you like and ideas that you have. And so thinking about valuation, thinking about all these long-term questions of valuation, as well as some of the short-term momentum factors you've highlighted as potential risks, or the sectors that you've highlighted as potential risks.

What do you like? What is good value here for someone who has to be invested sector-wise? I know you can't probably get into specific stocks or anything like that.

MS: Yeah. So sector-wise, it's a bit of a split. So anything that does badly when the Fed raises rates has been a bit of a horror show. And so they could plausibly be set to benefit if the Fed has to cut rates or continue cutting rates. And again, if you look back at 2000, one of the interesting things was that as the market collapsed, there were certain sectors that actually performed surprisingly well. I mean, there's a number of stocks that doubled as the market halved. And a lot of those stocks were in what we call early cycle sectors.

So that would be things like elements of consumer discretionary, particularly things related to housing at the moment, and also early cycle bits of materials. So things like chemical companies, packaging companies. So on the cyclical side, it's really sectors that are exposed to a decline in interest rates that we think are interesting because of the fact that they're very cheap, everybody hates them, there's large short positions. And generally, capacity is coming out of a lot of the markets that we're interested in because profits have been so difficult for so long. So that would be consumer discretionary and bits of materials look interesting.

And then on the defensive side, there's a bit of agreement, which I guess is sort of slightly concerning that healthcare is cheap, but healthcare does look particularly cheap at the moment. And especially if you can get, there's a number of turnaround stories that we're involved in, in healthcare, that I think the market's really underestimating the profitability that they can improve over the coming years.

But again, they'd be, if you thought that bond yields were a bit more likely to go down than up, then healthcare can work, and there's a lot of idiosyncratic things going on in that sector. And then the other one would be staples, where again, it's sort of flat on its back, a lot of short positions. Things have been quite tough, but the interesting thing about capitalism is that after tough periods, companies gradually drag their way, or good companies gradually drag their way out of it. So, there's a lot of rejigging of portfolios and reinvesting in pricing and streamlining business operations and all that sort of thing. So, it'd be those four sectors would be consumer discretionary bits of materials, healthcare and staples.

And then geographically, I mean, I suppose it could be a bit of home country bias, but UK domestic stocks have been absolutely dire and for sort of understandable reasons. But if you get anything other than a complete disaster out of the November budget, then you could be beginning to put a floor in under some of those stocks. So that would be one of the other things we're thinking about.

TG: Let's hope that is all true when it comes to the patriotic story and the patriotic case for UK stocks. Matt, you've taken us on a whirlwind tour all around the investing landscape. I just wanted to say thank you so much for your time. It's been a real education. I really appreciate you doing it this week.

MS: Thanks.

TG: Thanks for listening to Street Signals. Clients can find this podcast and all of our research at our web portal, Insights. There you'll be able to find all of our latest thinking on markets where we leverage our deep experience in research on investor behavior, inflation, media sentiment and risk. All of which goes into building an award-winning strategy product. And again, if you like what you've heard, please subscribe wherever you get your podcasts and leave us a review. We'll see you next time.