Capital Gearing Trust p.l.c (CGT) Earnings Call Transcript & Summary

Good morning, everyone, and welcome to CG Asset Management's Annual Investor Day. It's wonderful to see so many familiar faces today at Grocers' Hall. For those I haven't met yet, my name is Katie Forbes, and I'm Head of Investor Relations for CG Asset Management. As you may have noticed on your chairs, we brought back the stocks by popular demand. So consider this an early festive gift for you all. You'll also see a printed agenda on your seat. So the agenda is now actually inaccurate, but we will update you all as we go on. We've also noted down all our different open-ended funds. I think a lot of you are familiar with Capital Gearing Trust that we do operate -- we do manage 5 open-ended across the fixed income and multi-asset space. Today's sessions will run from 9:30 a.m. to 1:00 p.m., and we'll conclude with champagne, Canapes and mince pies for everyone. That's something to look forward to. This morning, we'll go through Peter Spiller to start, then Chris Clothier and then Alastair Laing. And our guest speaker today is Sir Ben Wallace, which we're very excited about. Once again, thank you for being here, and we hope you enjoy the event. And Peter, over to you.

Good morning, everyone. So I'm going to talk about investment trust and property. But before I do, I just want to make one observation, which is that when Trump came to Par, his big thing was tech and the tech sector has really soared. In Europe, the emphasis was on defense and defense stocks have been very strong. Here, the labor government has targeted housebuilding and renewable energy. Both have collapsed. So it's perhaps fortunate that they have not promoted investment trusts or property. So I've been involved with investment trusts for over half a century. And the only point of putting out the record of CGT here is to show what can be achieved by investing in investment trusts because all this record has substantially been built on the back of investment trusts. They've consistently outperformed open-ended funds, and they've been a much more interesting investment. So the narrative when I began my career, was that institutions had outsourced part of their fund management to trust. But at that stage, they were beginning to develop their own capacity for investing, particularly overseas and we're selling the trust, which they had used for that purpose. And that persistent selling undoubtedly helped create these large discounts. Discounts in those days were often in the 30s. And there was one group in particular, which was in the 40s, but that was a result of, frankly, criminal corporate governance. I mean that literally. We've come a long way since then in terms both of discount and corporate governance. Today, the only trusts that are at those elevated levels are ones which are controlled by a single individual or institution. It is plainly the duty of directors of such trusts to make sure that the minorities are not oppressed, but no such vigorous directors seem to be appointed. So more generally, though, discounts on equity trusts have moved to much lower levels. A turning point came with the new Millennium when it became much easier to buy back and issue shares. From then on, all discounts have been voluntary. They nevertheless persisted and bankers came up with all sorts of wheezes to issue stock in a high discount environment, including continuation votes and promises to protect certain discounts, normally 5%. A number of those promises were not kept, often citing size as an issue as the world had changed since the promise had been made. And this actually has proved to be a good source of return for us because we were able to generate good alpha by reminding boards of their commitments and encouraging them to keep to them, usually outside the press, I should add. More recently, issuance has largely been in alternatives, where all too often NAV stands for not actual value. Indeed, the last statement from the Chair of Aquila European Renewables, and I quote because it's extraordinary, distinguish between business as usual valuations and the -- and that's those based on their value as models and the actual much lower NAV as determined by a willing buyer and willing seller. Even worse, the fees in those cases were based on the model. So the background has been unfavorable for trusts. Outflows from U.K. assets have been persistent and been going on for a long time. And the extraordinary performance of the World Index because of the extraordinary performance of the Magnificent Seven has meant that actually, if you bought an ETF on the World Index, you probably did better than you did in most investment trusts. And -- because it's been so hard to match, people have been switching. Although we believe, as Alistair will get into later, the extreme valuation suggest that, that situation will not be sustained. And the regulatory environment has drained liquidity from all small companies, not just investment trust, but certainly including them. So that roughly those with a market cap of GBP 700 million or less have difficulty in establishing their viabilities. So the framework for looking at the sector today is influenced by 3 factors. One, there are in the area of 230 trusts that need to act to achieve viability. Secondly, we've got creditors in the form of Saba and the like, who will press for closing of discounts even if that means liquidation. And the third element is that directors are very well aware of the first 2. And also, it seems safe to say, have a greater commitment to their shareholders as the principal stakeholders in any trust. So some trusts have converted to an open-ended structure, the latest example being Smithson to the great advantage of our funds. Buybacks have become both more common and more written vigorous, sometimes at an annual rate of over 20% of the existing capital, which is pretty impressive. On occasion, even that fails to meet the discount targets. Rapid buybacks do not always mean rapid enough as exemplified by Smithson. So Boards have a number of choices. They can consolidate. And there have been some instances, not always a great success, and Chris Clothier will talk later about some of the pitfalls that can occur with consolidation. They can liquidate or open end. They can buy in stock aggressively. And rather surprisingly, we still see some directors saying that buybacks are not effective in reducing discounts. So the maths are very simple on this. Discounts get wider when the supply of stock is bigger than the demand. And -- so if the supply of stock is great and the demand is not there, including the buyback in sufficient quantity, then the discount will not be effective. But if providing this demand is greater than supply, then it does clearly work. They can introduce unconditional tenders or they can offer regular realization opportunities. And that works very well for trusts with illiquid assets. What doesn't seem to work is long-dated conditional tenders, reliance on improved performance and marketing, which I've heard for 50 years, never quite worked yet, but maybe it will in the future. And most of all, doing nothing until there's a crisis. Now I've left the best solution to last, the structure that Capital Gearing Trust itself has, a discharge control mechanism, that removes the threat of wider discounts and importantly, allows for issuance and growth and unlimited liquidity. With the following win from stamp duty who at least for 3 years announced in the budget, it would be good to see a new trust issued with that structure. So my vision for investment trust is that there will be fewer but much bigger equity funds with niche structures for niche and illiquid trusts. Overall, they can offer an excellent solution to the investment needs of retail investors. And they could easily, and I hope will be larger in aggregate than they are today, many few of them, of course. In the process, discounts have and will have tightened much further. And this offers good opportunity for Alpha for our funds. And I just put this up to show the scope. So the investable universe is GBP 216 billion, of which about 60% is equity investment trusts. Of that, GBP 52 billion is in funds which have commitments of one kind or another to discount levels. And GBP 20 billion have some kind of tender or exit mechanism. And just to put it in context, we obviously have a very negative view about equities. But nevertheless, our current equity exposure is about GBP 300 million. So there's plenty of scope to take advantage of these narrowing discounts. So turning to real estate briefly. We have, for a number of years, presented at this Investor Day reasons to be cautious on real assets, but certainly including real estate as an inflation hedge. This slide, I think, is absolutely compelling. So it shows the relationship of various asset classes to rising inflation correlation. And as you can see, apart from gold, it has been wholly negative. Index-linked have not been going for this period, by the way, so they're not on the chart. But it's negative for equities, but even more negative for real estate. And I'm going to look at one particular reason why long-term returns have been so disappointing in real estate is depreciation. So this building is where I first worked in the city. At least it's the site of where I first worked in the city because it was quite new when I was there, but it has been redeveloped twice since then. And that's generally in line with an office cycle of 25 to 30 years. So if the building and demolition costs are roughly 70% of the initial value, investment accounting would have knocked 2.8% per annum of the return. So 4.5% cap rate for super prime as we are today, becomes 1.7% return before costs. That compares with an index-linked yield of over 2.1% over the RPI and a financing cost using a marginal -- margin of 150 basis points over the 5-year SONIA of 4.4%. This looks very low given that rents historically over the last 40 years have nowhere near matched inflation. We took on our office in 2022. The rent we signed up to at that stage was broadly similar in nominal terms to the rent that had been paid on similar offices in 1989. So it does look as though offices, in particular, have quite a long way to go before the cap rates catch up with the changes in interest rates that we have seen since 2022. Now it's true that the implied office yields from share prices of real estate companies imply a yield of 6% to 7%. So that's far less concerning. Unfortunately, applying that sort of yield to the balance sheet suggests that the LTV rockets from less than 40% to over 50%, a challenging rate in this environment. And this depreciation, I think, explains the inability of a well-managed property company like Land Securities to match inflation. So the share price here is indicated by the pink line and the RPI, U.K. RPI by the dark line is not encouraging. That goes back -- that's in case you can't read it, over the last 37 years. So we apply different rates of depreciation to different kinds of buildings. But our models overall show much lower returns than most analysts. And you will not be surprised to learn that although we can find some interesting opportunities, one of them represented here today, there are few. So we started off investing in investment trusts, and they still offer the best opportunities. I'm going to hand over to Chris.

I thought that applause was for me, but I doubt it was for Peter. Good morning, everybody. Thank you so much for coming. So I did have a presentation where I was going to talk about our investment process and those sorts of things. But then the whole HICL-TRIG thing happened, and I thought it'd be much more fun to talk about that instead. So 2 weeks ago, yesterday, HICL proposed a merger with TRIG. But before I get into the details of that, I probably need to give you a little bit of background. And let's start by what those funny collections of letters actually mean. Apparently, a picture is worth a thousand words. And so on that basis, this slide is the best part of an assay. So on the left, you have HICL, and they own a portfolio of infrastructure assets. And so sort of in round numbers, a significant chunk of the portfolio is in public-private partnerships, PPPs. These are concessions where the private sector builds as it might be a hospital or a road and then operates them for the government and then eventually hands them back. These concessions were largely taken out in the early 2000s, typically for 20 to 30 years. And so they're all coming towards the end of their lives. The other part of HICL's portfolio is long duration, typically regulated assets. So these would be things like water companies, electricity transmission, the high-speed rail link down to Dover, that sort of stuff. On the right, you have an image of what TRIG does. It owns renewable infrastructure assets, so chiefly, wind farms and solar farms here in the U.K. and Continental Europe. And that's the first bit of context. Second bit of context is that infrastructure has been a pretty grim place to be in recent years. Now why is that? None of this will come as a huge surprise. Of course, over the last few years, we've seen interest rates rise quite markedly from the COVID lows. And that meant that investors all of a sudden were able to earn satisfactory returns from a portfolio of, say, gilts. And so they were tempted to sell their infrastructure assets. As Peter very rightly put it, the supply of shares in the investment trust market in the short term is fixed. When demand falls away, something has to take up the slack and that's something is price. Right. That is enough of the context. Let's get on to the transaction. So on Monday, the 17th of November, a piece of news hit the wires which was that these 2 investment companies were going to merge. And this was a transaction that we absolutely hated for 3 reasons. Let's start with valuation. HICL was going to massively overpay for TRIG's assets. The chart here shows the relative discounts to NAV of their portfolios on the Friday night before this transaction was announced. I don't need to look at the detail. The key point is that the market was placing an 11% greater discount to NAV on TRIG's portfolio than it was on HICL's. And that was an assessment that, broadly speaking, we agreed with. However, the proposed transaction was done on completely different terms. It was going to be struck on an NAV for NAV basis. So what this meant in practice was that the Board of HICL were forcing the shareholders of HICL to purchase a pool of assets that in most cases, a lot of those shareholders had decided that they didn't want to own in any event and that they were going to force them to pay a higher price than they could have bought those assets for themselves if they've just gone out into the market to buy the shares. So that was our first issue. The second issue was the terms of the deal. So TRIG shareholders were being offered GBP 100 million of cash out. HICL shareholders were being offered no such thing. This meant 2 things. First, it meant that the price that HICL shareholders were going to effectively pay for TRIG's assets was even higher than would otherwise have been the case. And second, that the combined entity would now be more leveraged and have less financial flexibility than it would otherwise have done. The voting, that was also asymmetric. TRIG shareholders required a 75% threshold to approve the deal. HICL shareholders only 50% despite the fact that this transaction fundamentally changed the nature of both trusts by roughly equal amounts. And third, the clear beneficiary in all of this was the manager. Both investment trusts are managed by the same company, InfraRed Capital Partners. First of all, they managed to avoid a continuation vote in TRIG that is coming up next year and where there are some questions as to whether or not such continuation vote would pass. And second, they were effectively creating a poison pill, both in respect of the management contract, but also in respect of the combined investment trust itself. And the reason for that is that by merging these 2 very diverse businesses, you're creating such a hodgepodge of assets that very few, if any, management firms would be able to manage such a diverse pool of assets and very few, if any, potential bidders might emerge for the combined entity. So the results was that the day 1 share price, the HICL share price went down by 6.5% and TRIG went up by 5.5%. The market concluded roughly as we did, that this represented a straight transfer of value from the shareholders of HICL to the shareholders of TRIG, on top of which, in aggregate, it also destroyed GBP 50 million of market value. The Board, however, said that this was absolutely fine and that the share price reaction was the share price reaction and the market would eventually come to its census. So then the Board, as you might imagine, went about setting out their rationale for the transaction. But what was extraordinary about it was that the 3 key rationales that they put forward failed to stand up to any kind of scrutiny whatsoever. So first, the Board said that the increased scale of this vehicle would attract new buyers, which presumably would result in a re-rating of the shares. And while that is probably true to some extent, what the Board failed to think about was the very large number of sellers that they would also create in the process. Now the first group of sellers were perhaps the shareholders of HICL who had chosen to own HICL and didn't want to own the assets of TRIG. By contrast, there would also be shareholders of TRIG who had chosen to own TRIG and didn't want to own the assets of HICL, and that both of these groups of shareholders would perhaps sell some or all of their stakes on the far side of the transaction. Second was an institutional shareholder in Australia that we spoke to who said that he would have to sell his entire stake in the combined company because it no longer fit within his institutional mandate. And then closer to home, there were the infrastructure funds here in the U.K. that invest in listed infrastructure stocks. Now as an example, one of these funds might well have, say, a 7% stake in HICL and a 7% stake in TRIG, 7% of the NAV of their fund, reflecting the fact that these are 2 of the largest infrastructure stocks in the U.K. market. So on a pro forma combined basis, that would represent 14% of their funds. And that would blow straight through their UCITS single stock limit of 10%, requiring them to sell at least 4% to come back in line with the regulations. So that was the Board's claim number one. Number two was that the transaction would enable the reorientation of the portfolio. So the Board of HICL has long made it clear that they wanted to reorient the portfolio to long-duration permanent assets and away from PPPs that are essentially short duration and self-liquidating. And that was a strategy that I think, by and large, the market and indeed, we were supportive of. However, this transaction did precisely the opposite. So the vast majority, about 75% of the net present value of TRIG's cash flows is made up of fixed rate subsidies that essentially roll off over the next 10 years. So financially speaking, they are functionally very, very similar to the PPP portfolio that the Board has said was the thing that they were trying to get rid of. And third, the Board claimed that this would result in a higher dividend. Now there is nothing wrong with a rising dividend, especially where that dividend is supported by underlying earnings growth. However, there is a great deal wrong in our view, with artificially juicing a dividend by purchasing a low-quality, high-yielding earnings stream in order to raise the overall dividend. And indeed, we can't think of any examples of a business that has achieved a rerating of its shares by pursuing such a strategy. But the strangest thing of all about this was that there was something that was in the grasp of the Board that would address all 3 of these objectives that they were trying to meet. And that is capital allocation. And let me just explain. So should the Board use the cash flows from the maturing assets and from additional sales, which they had demonstrated they were able to achieve and to deploy that money into buybacks, they would have met all 3 of these objectives. So let's look at each of them in turn. The first is they would have created a new buyer for the shares in the company in the form of themselves, and that would have led to a re-rating. The second is that they would have naturally reoriented the portfolio because by disposing of the PPP assets or allowing them to roll off and then using the cash flows from that to buy back shares, all that would then be left of the company was these longer duration assets that they were seeking to reorient the portfolio towards. And finally, if you buy back shares on a discount, you generate immediate earnings per share growth. And where earnings per share goes, so too the dividend can ultimately follow. So what did we do? We started by publishing a letter, and we got it up by lunchtime on the day of the announcement. We wanted to take control of the narrative, and we then spoke with the Chairman that day, and we asked him to withdraw his proposals, and he told us that he would not. We then carried out an extensive consultation. We tried to speak to as many shareholders we could and substantially all of the top 20, we spoke to shareholders across 3 continents. And through that process, we built a coalition of the unwilling as we refer to ourselves. And so we built a group of in excess of 10% of the register who were prepared to speak out publicly against the transaction and about 20% of the register who supported us publicly and privately. We also engaged with the press. And this was an incredibly effective because it meant that we were able to take our story, take our message to smaller institutional shareholders who we didn't know and take our message out to private shareholders who obviously we weren't able to contact. In a similar vein, we made use of social media. This had a very similar effect and enabled us to have conversations with shareholders, very productive conversations with shareholders that we wouldn't otherwise have done. We launched a website, and we managed to get that up within 48 hours of conceiving of the idea, concernshiclshareholders.co.uk. I'm always absolutely astonished by what CG Asset Management is able to achieve given that we are, as you know, a very small firm with limited resources. But this one really took the biscuit. And so to that end, I would like to take my hat off to Chris Taylor over in the corner there, Katie just there; and [indiscernible], must be here somewhere. There he is at the back, who took the lead on this and did a fantastic job as you can -- I'm sure you will agree. It was very elegant as well as being highly informative. Thank you very much, everybody. We then published a letter jointly with our coalition of the unwilling and sent that to the Chair. And we followed it up with another letter that was signed by over 100 private individuals, which was roughly half of the private individuals who have gotten in touch with us to share their concerns about the transaction. Where are we now? And so let's bring it up to the present day. So yesterday will be the day that is emblazed in my mind as being the day that the Board of HICL withdrew these proposals. However, yesterday will be very much a different day in the minds of Katie because yesterday was the day that her boyfriend, Joshua proposed her. And I think I'm delighted to say that she accepted. Can I suggest you all go up and congratulate her. Can I also suggest that you don't look at least directly at the ring on her finger because you'll probably be blinded by the refraction of all of the light of the universe that sparkles off it. Actually, and since we're on the subject of milestones, we had another incredibly exciting milestone in CG this year, which was the birth of Hassan's first child. So Sophia joined us April in April. He's nodding. So our babies under management has increased by some quite considerable percentage this year. Congratulations, Hassan. Anyway, sorry, where was that? Yes. So yesterday, the Board of HICL said that they announced that they were withdrawing their proposals. And obviously, we're very pleased with that. It is a good first step, but it's not quite the end of the story. There are a lot of issues in the investment trust sector, particularly within the alternative space at the moment, and we applaud Boards who grapple with these issues and try and take hard and brave decisions. However, we cannot applaud Boards for being indiscriminate in the actions that they take, far better in our view to do nothing at all than do something simply for the sake of being seen to do something. Just because this deal was a deal that the Board thought that they could do does not mean that it was a deal that they should do. So I'm afraid that resulting from this, questions remain about the judgment of the Board and the independence of the Board from the manager. And what about us? Why was it that we put so much time and effort into reversing this transaction? Well, as I said earlier, we judge this transaction to be a straightforward transfer of value from the shareholders of HICL, that is all of you, our clients in this room, to the shareholders of TRIC. And that was something that we were not prepared to sit idly by and watch happen. Now this was an example of us deploying our efforts defensively on your behalf in order to protect value. But we can use exactly the same techniques, and we do use exactly the same techniques and other techniques in order to unlock or create value on your behalf. This is essentially one of the things that you pay us for. The investment trust market is a funny corner of financial markets, but it's one that we have, thanks to Peter and his 50-year history, deep expertise and experience of. And so that is what we will do to try and deliver additional returns to you. Thank you.

Great. Well, now for something different, no more trusts. So the 30th of November 2022, that's almost exactly 3 years ago, AI had its sputnik moment, by which I mean it just exploded from a kind of powerful but niche technology into a cultural and definitely financial phenomenon. And that, of course, was with the launch of ChatGPT by OpenAI. And I've referenced sputnik there, but I think financial historians and general commentators have talked about a lot of other similar phenomena as they try and grapple with what is going on and how the world might be changing for us as consumers. And Neil Ferguson, for example, his favorite analogy is the railway mania of the 1840s. He may well be right, and I've read a bit about it, but not actually being there, I cannot confirm the texture of how it felt. But I was around in the other of the more -- the cited historical references, which was, of course, the late 1990s and the emergence of the Internet and digital telephony. Not only was I around, but it was a pretty big event for me. For context, I actually graduated from university in the year 2000. I knew quite a lot about history, and I knew precisely nothing about nothing else. But I did know that the Internet was the future, an extraordinarily powerful emerging technology. And I knew that because the narrative was everywhere. It was all around me, and it was incredibly persuasive. So I just thought what better way of emerging -- kind of immersing myself in that. Then by my first job, which was going across the Atlantic to join a venture capital fund. I tried to find a photograph of around that time, but I couldn't find one, but who needs photographs anymore because ChatGPT can just create it for you. So here I am in New York in 2000, big smile on my face. I'm about to invent the future. Now for those of you who remember the event, you may question whether turning up -- at what turned out to be quite an insubstantial venture capital firm in June 2000 was actually a good idea. And without going into all of the details about what unfolded over the next 6 months. Suffice to say, by 2001, I was going back over the Atlantic, leaving my glitzy dreams of New York to return to Edinburgh, start trying to learn something about something and starting an accountancy qualification. And while I was sitting there in a grim windowless room by Haymarket Station trying to sort my debits from my credits, mostly successfully, but not always, one of the classic quotes of the dot-com bust came out. It's classic for so many reasons, but not least because it really emphasizes the emotional nature that the kind of -- mental kind of disharmony that comes from an incredibly powerful narrative that is deeply believed in, that unravels so quickly. And for those of you, just to set a bit of context, Sun Microsystems was the open AI of the dot-com era. It created high-end computers, high-end servers, and it created and owned the Java programming language. At its peak, it had a valuation of about $200 billion. But by 2002, that share price had fallen by more than 90% by more than 95%, I think. So it had been a pretty rough ride for shareholders, and they were frankly furious. They were putting their concerns to Scott McNealy, but in a way that only a pugnacious U.S. CEO could do, he completely turned the tables. He essentially said Sun Microsystems was never the issue. The issue was you. The issue was investors. And he had this amazing quote, which I'm not going to read the whole of. But essentially, he said, look, in 2000, this company was trading at 10x revenues. And he summarized by saying, do you realize how ridiculous those basic assumptions are? What were you thinking? Now this quote may be familiar to a few of these -- sorry, to a few of you, possibly from the time, but also for the fact that Chris put this up in our Investor Day in 2021. And he put it up there. So I'm just stealing his work here, but he put it up there because he was revealing the patented CG Asset Management, what were you thinking index. And this index seeks to measure the portion of the S&P 500 that's trading at greater than 10x revenues. That level that Scott McNeely could, after the event, identify as absolutely crazy. Now I'm going to ask all of you this question. I'm not going to ask you to shout it out, but please try and come up with a number. And obviously, that's quite difficult in abstract, so I'm going to give you a few framing points. Firstly, as best we can tell, kind of long-run average is about 5%, a pretty small number of truly extraordinary companies with great growth prospects. When Chris came up with this chart in 2021, some of you may remember it was at that point, high peak of 20%. But there may be a little clue here that the scale on this axis goes up to 30%. Well, pick your number, and here's the big reveal. So I'm sure plenty of you in the room picked 30%. But I'm afraid for those of you who did, you're wrong because, of course, I set the axis. The actual number is literally off the charts in this case at 33%. So this is obviously only one prism of looking at the world and maybe the most extreme one. But there is something extraordinary about this environment. And we've got to ask why is it now that such a substantial portion of the S&P trades at these levels that history so rarely supports. And many of you already know the answer. It's a small number of extremely large and powerful companies that are broadly sit at the very heart of the AI infrastructure rollout. The list here, you can see is NVIDIA, Microsoft, Apple, Meta, Broadcom, Tesla and some others, and those others include Oracle and Palantir and a few companies like that. You'll note that this list does not include Alphabet, which when I looked this morning, was trading at 9.7x revenues, and it doesn't include Amazon, which has a very large but low-margin consumer business. But undoubtedly, a lot of the value placed on Amazon actually comes from its cloud computing and AI piece. So in many ways, 33% actually underestimates the scale of what's going on here. By the way, I have absolutely no idea. I have no theory of mind why Tesla is in there. So I'm not going to talk at all about Tesla. But all these other ones, these are the royalty of AI. But as many have pointed out, it remains fascinating to me that these companies are valued at 10x revenues now, not least because these companies that are collectively amongst the most powerful corporations ever built are starting to turn on each other. Clearly, NVIDIA, Broadcom and most recently, Alphabet are all chip companies that are directly competing with each other. And I would note that historically, semiconductors are never a market that has sustained long-term monopoly positions. This might be changing, but historically, the periods of monopoly have been relatively short. And then you have the digital -- the classic digital monopolist, Microsoft, Apple, Meta, and I'd include Alphabet in there. And these companies emerged at least as our corporate methology goes. They emerged from garages on the West Coast or Meta, in Meta's case, a dorm room in Harvard, revolving around a beautiful bit of code, a digital network that emerged around that and gave these companies huge, huge -- I mean, near infinite return on capital employed, unbelievable cash generation models and digital monopolies. But they are now turning on each other. All those companies that I mentioned, those digital monopolies, a vast majority of their value is now linked to their cloud computing and AI capabilities, and they're all in a direct arms race to hit artificial general intelligence. And this has resulted in the much noticed massive capital expenditure that these companies are coming up. And the favorite example of this, I mean, it doesn't actually -- even -- it's not even measured in the capital expenditure. And it's not even the biggest bit, but the announcement of this. Now some of you might recognize this. This is Three Mile Island. This is a nuclear power station in America. And it was actually -- it almost blew up in the 1970s. It was the closest thing that America had to a Chernobyl moment. Three Mile Island is run by a company called Constellation Energy and Microsoft recently signed a 15-year agreement to buy 100% of the power from this power station for the next 15 years to power one of their data centers. I mean it's hard to get your mind around the scale. This isn't even CapEx, but if you're buying power forward for 15 years on a plant of this age, you have essentially bought a nuclear power station. This is so far away from a CapEx-light Software-as-a-Service model. It's hard to get your head around. So going back to think about comparisons with the dot-com boom again. As best I can tell, the peak CapEx in 2000, as I was setting up in my office in New York was about $150 billion, not an insubstantial sum. Today, obviously, there's been plenty of inflation in the interim. But today, the AI CapEx or the CapEx of just 7 of these hyperscale companies is $400 billion. And that's before you take into account CapEx from the likes of CoreWeave, Oracle, all these other companies, the Stargate initiatives in the U.S., the Middle East, China, all of these other companies supporting the AI infrastructure. The scale of it is absolutely vast. In fact, the scale of it is truly macro in scale. I was reading an analysis earlier that in the first half of this year. U.S. GDP, which actually grew at quite a decent clip. I think an annualized rate of over 3% would have been about 0.1% were it not for expenditure coming out of the software and information technology area, largely this CapEx. So essentially, we have a market which is valuing very highly these digital businesses, precisely a time that I would have some concern that their models are becoming -- their business models are becoming less rather than more attractive, by which I would point out that the AI revenues of these hyperscalers are about -- forecast to be about $20 billion in 2025. And if all the anticipated CapEx, which will probably be around $2 trillion to $3 trillion between 2025 and 2030, if all of that is spent, these companies to generate a 15% return on capital employed will need to be generating revenues that continue essentially in perpetuity of $2 trillion by 2030, $2 trillion. They've got GBP $billion, they're 1% of the way there. Now of course, it's possible that they will get there. But my God, if they get there, then we should be -- we should buckle in for the ride because we are about to witness societal change on a level kind of -- it will be a disruptive event of itself. I guess what I'd be more concerned about is whether a $400 billion bet on a revenue stream that is going to need to grow hundredfold. It looks a bit like a kind of venture capital type bet rather than the kind of investment that you necessarily want in the S&P 500. So taking a step back, looking at some other markets and thinking -- moving on from a multiple of revenues. This chart is a little bit complicated. But on the top, in the pink line here, it shows what's called the cyclically adjusted PE yield. That's a PE yield that seeks to kind of smooth out the business cycle. And we can also see the yield on a 20-year inflation-linked bond, which, as you can see, is the highest yield today that it's been over the last 20 years. Obviously, it's not the same bond this is rolling over. And it's not theoretically right to think of this difference as an equity risk premium, but it's not a bad proxy. And what's so different from when Chris last had this chart up in 2021 is the narrative then was not AI, ChatGPT hadn't arrived. The narrative was something called Tina. There is no alternative, which essentially evolved around the fact that interest rates were negative in real terms, essentially zero in nominal terms. And the assumption was essentially valuations could be supported because interest rates were going to remain low forever. That lasted less than a year after our Investor Day as the inflation spike following the Ukrainian -- the Russian invasion of Ukraine cause interest rates to spike. And then, of course, the AI narrative came up and it's driven the cyclically adjusted price earnings ratio, the inverse of this Cape yield to levels that historically have suggested a 10-year real return of sub-1%. So there really is an alternative. There's quite a good alternative. The bond yields are at historically interesting and relevant levels at precisely the time that equity valuations look very stretched, which raises the question, of course, that why is that? And I'm predictively running a little short of time, so I'm going to try and speed up a bit. But many of you would have been following the budget over the last few days in the U.K. and there just is this incredibly difficult balance in modern politics between trying to deliver the kind of services that the elector expects of the state whilst keeping the bond markets on the side. And there just is no solution that any politician in developed markets has come up to this sustained fiscal deficits. And this has resulted in a massive run-up of government debt. And deficit spending remains at extraordinarily high levels, considering we're not in a time of recession and we're not in a time of war. This is the good time when we should be fixing the roof, but there just is no solution in modern politics for these issues. So as Jamie Dimon memorably said in May, the bond market is going to crack. Whether it will take 6 months or 6 years, I don't know. But this situation is completely unsustainable. And so at least as we look at it, you look at a lot of the major investment markets, the bond markets, at least the valuations have moved significantly to incorporate some of this risk. But it still remains very fragile given the elevated level of debt. The equity markets just seem extraordinarily stretched. And I doubt many people in this room would dispute too much of this. However, what we often and quite rightly are asked is fine. That's all fine, but we've heard it all before. And what is the catalyst? And that's a very valid question. It's very hard to identify what a catalyst is in advance. But again, it might be interesting to think a little bit about the comparison with 2000. As you'll remember, 2000, we had a high level of political consensus around figures like Tony Blair and Bill Clinton. Brit Pop was riding high in the charts, which was obviously a great feel good factor. The Berlin wall had fallen, the cold war was over. This was an ideal kind of societal backdrop out of which a technology story like the TMT boom to emerge out of. But today, the situation just seems so different in so many ways. I mean, politics, as far as I can tell, is in a terrible state across the developed world. I picked France here, but there's no particular reason why it need be France. Geopolitics, I'll leave that to Ben Wallace. But suffice to say, bombing of this year, we have bombing of nuclear facilities in the Middle East. The Middle East itself is a powder cake. We have a land war in Europe. We have high tensions in the South China Sea, in the Arctic, bombing of railway infrastructure in Poland anyhow. It's hard to feel particularly positive from that angle. We have a slowing economy. I've already pointed to the fact that the U.S., in our view, at least is propped up by a combination of AI spending and deficit spending, both of which could be somewhat fragile. Europe clearly is caught in sclerotic growth, and it's not clear how that can unfold. And the bond market itself, there has been significant bond sell-off. But as Jamie Dimon said, there's scope for things to be worse. So no one comes to CG Asset Management presentation to hear about the positive things. And there are lots of positive things. But why are we here and what role do we play in this ecosystem? We have a strange specialism in investment trusts. But ultimately, the core purpose that we play in our investors' portfolios or at least as we think about it, is the response to the what was I thinking question. We'll leave the upside to you guys. I know that plenty of you in the room will be making some extremely exciting investments. But we need to be the bit of the portfolio that when things look really tough, there's some real resilience. And I guess around Liberation Day in April, the equity markets actually sold off 20%. So we were delighted that our portfolios sold off only 2% against that what was a bear market, although it turned out only to be a dry run, but we hope for the big one, when it comes, if it comes, when it comes, that this portfolio can really show its metal. I'm not going to talk in detail because, frankly, I've run out of time. But suffice to say, 27% in risk assets or equities, that is pretty -- that is an extremely low weighting to equities because notwithstanding the fact we think that relative returns from investment trusts are going to be very good, we are concerned that against this backdrop of extremely elevated U.S. equity valuations that all equity markets could come under real pressure. We have a little over 40% in index-linked bonds in the U.K. and the U.S. We know that these assets held to maturity are going to deliver in the region of 2% in excess of inflation that is absolutely core to our mandate that we're delivering inflation beating returns without exposing our investors to undue risk. And we think in medium, not long duration, but medium duration, we think that these assets are very attractively priced given the risks in the world today. And then finally, we have 30% in managed liquidity, that is treasury bills and credit, the highest quality, lowest risk part of the portfolio. And that part of the portfolio is delivering between 4% and 5%. We hope the others more. And that exists to respond to some of these crises when the what was I thinking moment does occur. So as you'll be aware, this portfolio construct delivered about 6.5% over the last 12 months with very robust downside in what was a short-lived but quite sharp bear market. That's the kind of thing that we hope this portfolio can achieve with this kind of risk rating at the moment, this kind of risk aversion at the moment. So we will wait to see events unfold, but we hope that we are prepared for them when they come.