Examining the AI value chain with Horizon Global Partners

[Music]

Fawaz Chaudhry: It's a thematic, rich environment. I would say the opportunity set is wider today than it's ever been. We had a once in a hundred-year global pandemic. We had once in 80 year European land war. We have had now once in 100 year global trade war. And they all came in rapid succession, 2020, 2022, 2025 and it's kind of shaken things up.

Eloise Goulder: Hi, and welcome to JPMorgan's Making Sense. I'm Eloise Goulder, and today I'm delighted to be joined by Fawaz Chaudhry, who is CIO of Horizon Global Partners, which is the equity thematic team from Fulcrum Asset Management. So Fawaz, thank you so much for joining us here today.

Fawaz Chaudhry: Thank you so much. It's a pleasure.

Eloise Goulder: So Fawaz, could you start by introducing yourself and your background?

Fawaz Chaudhry: Well, originally I'm from Lahore, Pakistan. Born and raised there. And when I was 18, I got a full scholarship to go to MIT and that basically changed my life, as you would imagine. I remember my uncles, aunts, 20 cousins all at the airport seeing me off. My mom put me in a suit and tie. She saw movies, that's how gentlemen traveled. MIT was fascinating, I loved it. I did a triple major, computer science, finance, economics, also threw in a master's for good measure. Uh, graduated and joined Morgan Stanley's M&A department, did investment banking for four years (laughs) and moved to the buy side, came to London Business School to do my MBA and settled in London, which I love. And I started running this strategy 15 years ago at Hadron Capital in 2011, moved over to more capital. That was 2013 to '16, worked with Lewis. And I've been at Fulcrum now from 2017 onwards. I'm a partner and We've now rebranded my team as Horizon Global Partners, as you just said. And I'm very excited to bring the strategy to your group. Yeah.

Eloise Goulder: Fantastic. And I understand you really focus on global equity themes. Could you explain your investment process and your investment philosophy?

Fawaz Chaudhry: So the thought process is to identify themes and trends globally and then having a very rigorous structured access to that thematic view. And that, by that I mean in a relative value format, you hedge out all the risks you don't want to carry. This can be equity beta, market risk, but sector risk, style factor risk increasingly these days, or even positioning risk, so short interest or crowding. So we hedge out the risks we don't want to carry and are left with a trade in which we have access to the pure thematic that we want to have access to.

Eloise Goulder: And thematics as an investment style across the industry is growing, I think it's fair to say. Why do you believe it's so important and so valuable to be harnessing themes today?

Fawaz Chaudhry: So it's about breadth. You need to connect the dots across sector to sector to identify how one new idea or theme affects another sector. These days it's a thematic, rich environment. While I've been doing it for 15 years, I would say the opportunity set is wider today than it's ever been. A lot of top down macro shocks have playing out. We had a once in a hundred-year global pandemic. We had once in 80 year European land war. We have had now once in a hundred-year global trade war. And they all came in rapid succession, 2020, 2022, 2025. And it's kind of shaken things up and you have to have some top down view into any security, otherwise you will just be buffeted by those forces. So the opportunity set is rich and I built this strategy to take advantage of it.

Eloise Goulder: It's incredible the shocks we've witnessed over the last five or six years. The fact that we've seen the pandemic, we've seen the European land war, we've seen the global trade war, tariffs all relatively rare events in the context of history. What do you think the implications of all of these are for equity markets?

Fawaz Chaudhry: The big implication of this was these three kind of independent events all happened to be inflationary. If you think about the global pandemic that came in 2020, last one obviously was 1919, that create- shut down services in all the developed world and there was a demand for goods. We had a manufacturing shortage and called it the goods inflation cycle. You could say it's temporary. In fact, the central banks called it transitory, that it will roll over, except within two years, we got then hit with the European land war and an energy crisis from Russia. So that was an inflationary shock as well. And then we got hit with tariffs, which is again (laughs) inflat- so we've got three different independent shocks all pointing in an inflationary direction. We've come to the point where a lot of prices are up 40, 50, 60% from five, six years ago. It's caused the 40-year bond rally to come to an end because with inflation, it's the bond owners who are the losers and global yields in Japan, elsewhere, are starting to rise as bond owners have turned out to be the big (laughs) loser from all these inflationary shocks. But it's kind of jolted the system. I mean, (laughs) right now-

Eloise Goulder: Yeah.

Fawaz Chaudhry: ... uh, you can get four and a half locked in for 30 years in treasuries and investors need and want a higher return across all asset classes because the risk-free return has now risen from these shocks. And there's been a move back into equities, move back into risk. So I would say it's exciting times and that rise in long-term risk-free rate is actually what's shaking things up a lot.

Eloise Goulder: I mean, the pace of structural change over the last six years has been enormous and, you know, as most clearly represented in the rising bond deals and the ramifications from that. The other big structural change has obviously been AI-

Fawaz Chaudhry: Mm-hmm.

Eloise Goulder: ... the scale of technological change, possibly a deflationary force over the medium term. What's your view on the impact of AI and who are the relative winners and losers?

Fawaz Chaudhry: So AI is going to be a a seminal force, a general purpose technology that will basically permeate through all the goods and services we use. We will go beyond the AI apps to AI becoming like electricity or railroads or the internet, but it's actually bigger than the internet because it's, it's actually, uh, disrupting services itself and services is a much bigger part of the economy. So we are on the way to build the, a manufacturing footprint for AI, data centers, but ultimately everything you buy, whether you buy clothes, whether you go for a coffee, the, even the coffee machine will be t- fine-tuned through AI. The point being in every single thing we use, AI will be used, and that'll bring productivity. Now productivity, as you said, is going to be a disinflationary, unlikely to be deflationary-

Eloise Goulder: Mm.

Fawaz Chaudhry: ... most likely disinflationary force, but that means that real rates are start- is going to rise, that means real rates can be higher, back to the higher rates. This is good for the world, but it'll be disruptive and it'll be pretty quick. So generally speaking, in the past, where this kind of disruption has happened, there's been backlash as well. Let's see if that happens, if there's a lot of social upheaval on the back of it. There are winners and losers out of the whole AI thematic and, um, it's the, the shakeout is only just starting. People keep asking me is, are we nearing the end of the AI cycle or the bubble? I mean, we are just starting. We are in year four, year three, and this is going to be a multi-decade built seminal technology which permeates through society.

Eloise Goulder: Hmm. What a wonderful time to be picking themes.

Fawaz Chaudhry:  (laughs) Indeed.

Eloise Goulder: So in the context of everything you've just discussed, these three ultimately inflationary forces that we've seen over the last five or six years, the geopolitical pace of structural change, and then on top of that, AI and all of the ramifications and ripple effects of AI, what would you say, on a medium term view, are your highest conviction themes?

Fawaz Chaudhry: So AI is basically almost like a move back towards the hardware, putting the silicon back in Silicon Valley. So effectively, (laughs) we're back to the chip now. Whoever has the best chips can make the best AI-

Eloise Goulder: Mm.

Fawaz Chaudhry: ... large language models. And the value is shifting from software, application layer, to the hardware layer. And the highest market cap companies in the world, TSMC, it's high next, Samsung in Asia, ASML here, or Nvidia, Broadcom. These are all hardware chip companies.

Eloise Goulder: Mm.

Fawaz Chaudhry: So we're back to the chip, we're back to figuring out who has the best chips, who can make the best chips, and, uh, if I was to (laughs) give advice to new graduates, I would be saying, "Go do electrical engineering, (laughs) not computer science." So it's like, it's kind of, uh, interesting how things shift.

Eloise Goulder: But this is clearly the now, but if you look a few years ahead, will that remain the key place to be?

Fawaz Chaudhry: I do think that the value is sitting with the hardware for definitely three to four years. At some point, the application layer will kick on again, where the appl- new application will come, which will truly harness AI. And a lot of people use the analogy of the internet, so we're in like the mid '90s and I mean, and so, some of the seminal companies that emerged from the internet, Google was started in '99 and Facebook was started in 2004, or Amazon launched AWS in 2007, these came much later after the backbone had already been laid-

Eloise Goulder: Yes.

Fawaz Chaudhry: ... the fi- the fiber and et cetera. So right now we're laying down the backbone and the giants of AI, is it going to be OpenAI? Is it going to be Anthropic? Or is it going to be new companies which are, don't even exist right now? I would actually venture, there'll be a lot of new businesses that will still appear, will truly harness AI in our day to day lives, and which haven't even been started yet. But certainly near term, the free cash flows, the spend is going to these chip guys. So if you look at it in the '90s, Cisco became the largest company in the world at, by the end of '90s, because we didn't know who will win, which dot com company will win, but it'll be a lot of routers.

Eloise Goulder: And when you look at the market cap of global equity markets right now, how many of these companies would you wager are going to be the net beneficiaries versus the losers? And in terms of trades to make right now, is it more about, uh, identifying the winners or identifying the losers?

Fawaz Chaudhry: So a lot of our alpha comes on the short side, in fact, more than half. And thematic investing, it's almost synonymous with long investing. People even think of thematics as what are the long-term winners.

Eloise Goulder: Yes.

Fawaz Chaudhry: I think what are the long-term winners or long-term losers, so them- their thematic tailwinds to businesses, they're also thematic headwinds to businesses, and finding those structural losers, the melting ice cubes, that has been a much easier to harness than on the long side. On the long side, you get retail participation, valuation challenges, crazy rallies, volatility high. And on the short side, it's, um, much more hope generally available to businesses, they'll get the new management in. They never get too cheap. So it's just easier to sit on short, structural shorts for many, many years. Currently, we are looking at the services side. I mentioned software who are losing from some of these trends. And one of the main things that we are focused on is productivity, AI is bringing productivity, there's a hiring freeze at the moment, staffing companies are not seeing much businesses. So we are sh- shorting services, staffing, taking advantage of where are the AI losers as the shakeout is happening in the world, and those positions have been some of our biggest winners in terms of those trades.

Eloise Goulder: Yes. And when we look by country and by region, there are more AI losers in certain pockets of the globe. What are your views on that?

Fawaz Chaudhry: That is true. I mean, some of the biggest software giants are in America, actually.

Eloise Goulder: Mm.

Fawaz Chaudhry: But in the, which index has the biggest percentage of market cap coming from software, that's US. And guess what? S&P 500 underperformed the MSCI ACWI for the first time in a long time last year.

Eloise Goulder: Mm.

Fawaz Chaudhry: So the hardware supply chain is winning, so that means Asia's coming back to the fore, [inaudible 00:13:15] was up 70% last year.

Eloise Goulder: Yeah.

Fawaz Chaudhry: So the hardware's coming back in the, in the tech landscape to Asia and even Europe, rather than the last two decade kind of out-performance from US on the back of the software dominance in America. So big shakeout, Asia's back. (laughs)

Eloise Goulder: Totally. And quite counter to the consensus view of 6 to 12 months ago when I think US was definitely seen as a winner on tech without this differentiation necessarily between hardware and software.

Fawaz Chaudhry: Yes, there's a shakeout happening, but overall, this is a big positive to productivity, to the global economy, and hence to equities, et cetera, globally.

Eloise Goulder: Yeah.

Fawaz Chaudhry: A lot of the investment in AI, the data centers are being built in America, and hence American GDP is actually getting revised up. So ultimately, that must flow through into the entire economy, because while the Asian supply chain is building all this stuff, yes, but it's all being sent to America, all these data centers and et cetera. So that will permeate to the economy in terms of construction, in terms of power grid, et cetera. So (laughs) I would say overall it's just good for equities. And it is interesting that when I talk to people, the general consensus seems to be that AI is a bubble, equities are expensive. So the real contrarian view I'm taking is they're right tail risks, not left tail risks. What if AI is real? Are you under-invested? I mean, this is good for equities.

Eloise Goulder: Mm.

Fawaz Chaudhry: If AI actually plays out the way I'm describing it.

Eloise Goulder: Mm. And what about other asset classes? I mean, you've spoken about bonds and the fact that ultimately the inflationary environment we've been through has been bond negative. What about commodities and precious metals which have been incredibly strong this year and last year?

Fawaz Chaudhry: Yeah. Gold and silver pressure's doing great. We still like them. Two weeks ago, gold reserves at central banks overtook treasuries for the first time in 70 years. So the central banks are at the forefront of buying gold. That trend started on February 2022. Russia saw all its US treasuries, and the only reserves that really counted were (laughs) precious gold, and that really triggered a move by all central banks globally to move out of treasuries, bonds, and JGBs and into gold. And the pace of buying has been very strong and consistent and often volume targets being set, not even a dollar amount targets.

Eloise Goulder: Mm.

Fawaz Chaudhry: That they'll buy certain tons of gold, so this is price insensitive buyers. So you have to keep, find a marginal seller at what price is someone willing to sell, and it's not going to be a central bank. Who's hoovering it all up? So I think that it is fair that gold will keep rallying on the back of it.

Eloise Goulder: Yes. You've touched on the retail investor a few times. Obviously, the retail investor's been heavily involved in precious metals, especially silver this year. Uh, we track the number of posts on social media associated with equities and those asset classes. And in fact, we saw the number of posts in silver peak at 20 times the average number of posts over the last five years. What's your view, Fawaz, on the retail investor more generally? You mentioned that they are heavily invested in themes. Does that sometimes support your thesis because ultimately you've got another marginal buyer for those themes?

Fawaz Chaudhry: Flows do matter. Sentiment matters, obviously on the long side, mostly, not on the short side as much. So they are participating. We do follow the retail flow index where which securities are getting the retail flow. Clearly, money supply has grown significantly since 2020, whether it was QE, balance sheets are bigger. So there is a lot of capital out there, and when it moves, it moves prices, and we do keep an eye on it. And then you've mentioned silver, you can see when it comes out, retail flow, it can come out very quickly as well. And we've seen some massive pullbacks in silver. But a lot of the fundamentals of silver are very different from gold, and is linked to solar panels, and a lot of the thrifting in solar panels for amount of silver has already happened, and it's very difficult to reduce the amount of silver in solar panels. And we've now come to the point where the cost of a panel is so low that in big swathes of the world, unsubsidized, solar is the cheapest energy of the world. So in my home country, Pakistan, if you go there. Same with India, if you go to Africa, Asia, people are unsubsidized putting up solar panels on their roofs, and that's silver. And the silver is so little at, as a, it's, price of silver doubles or doubles again, it doesn't really move the cost of installation of a solar panel, the way for the sale, the module, the installation, the inverters, it's the minimist cost to it. And now, as EVs keep coming, in China, for example, 60% to 80% of new cars sold any month are now EVs, and China's putting up massive amounts of solar panels, solar farms, as its source of energy. We're replacing oil with silver. So (laughs) and I-

Eloise Goulder: Yeah.

Fawaz Chaudhry: ... I said like, silver's the new oil. You're now running cars on solar and that, the, the one metal that matters to that at the moment is silver. So that will continue.

Eloise Goulder: Yeah. Incredible.

Fawaz Chaudhry: Yeah.

Eloise Goulder: So you mentioned earlier these enormous geopolitical shifts with the European land war, with tariffs, that must be providing a lot of opportunities for you.

Fawaz Chaudhry: Well, US used to have, if you rewind back, an isolationist policy all the way till 1942, and they were stuck to that, uh, they were not gonna participate in these European wars, and that was a seminal moment, the Pearl Harbor, which changed that. So there was a new world order created after World War II, in which America was involved globally. And now, the cost of that, which is being borne by the American taxpayers and consumers, the benefits are not apparent to them anymore. We're just generations apart from those events which led to the decision to police-

Eloise Goulder: Mm.

Fawaz Chaudhry: ... the creation of United Nations and to set a world order. And there was some efficiency gains from it, keep in mind. Japan or Germany and others could spend 0.5% of GDP on defense, everyone else could spend a lot less. You have one country spending all the money, can have mobile assets going all the way in the world, we let them police it. And now they don't want to police it anymore. So Defense budgets have to rise, it's one of the thematic we've been very involved with, but not just in Europe. This is global. And it is going to be very different era, resource nationalism, so the rise of trade has kind of peaked and it's going to start falling. And China is building its own vertical stack, very different world from what we're used to, a lot less trade, like things used to be pre-World War I. It would [inaudible 00:21:29].

Eloise Goulder: Yeah.

Fawaz Chaudhry: Yeah.

Eloise Goulder: It seems like a complete shift in the global world order, and yet if you go back over long swathes of history, this is more the norm.

Fawaz Chaudhry: Yeah.

Eloise Goulder: And actually what we lived through in a post World War II era was quite counter to the norm, but it was also more efficient in the sense of global trade per Milton Friedman, efficiency of labor.

Fawaz Chaudhry: Yeah.

Eloise Goulder: You know, if we're going back to local supply chains, then it is in many ways much less efficient as well.

Fawaz Chaudhry: Of course. Yeah, it'll be much less efficient. I think Stephan Miran came out and said that we want to move away from this comparative advantage kind of order that we've moved to-

Eloise Goulder: Mm.

Fawaz Chaudhry: ... where everyone just specializes in what they're good at to back to everyone doing everything, which is a lot less efficient.

Eloise Goulder: Yes.

Fawaz Chaudhry: But it is more secure. (laughs)

Eloise Goulder: Yes. (laughs)

Fawaz Chaudhry: So-

Eloise Goulder: That's the trade-off. Yeah.

Fawaz Chaudhry: ... i- if security is what's important... But we'll definitely have some inefficiency. Now maybe AI can offset a, some or all of it (laughs) because the AI will bring some productivity. We are definitely moving to a more vertical stack in every s- space, yeah.

Eloise Goulder: Yes.

Fawaz Chaudhry: So I think the biggest, most important question I'm asked is, do I think AI is a bubble?

Eloise Goulder: Mm.

Fawaz Chaudhry: I would say no. And AI is real, the data centers investments will not turn out to be in waste. The people keep comparing it to the late '90s, the 2000 tech bubble, but it's very different from the '90s. In the '90s, we built out all this fiber optic cable and we just couldn't get it to the homes. The last mile couldn't be bridged, the fiber optic that they laid down was sitting dark. We just could not light it up. It's so different from that. Microsoft, Amazon, everyone is coming out and saying they're supply constraint or demand constraint. There's more and more demand coming. And the productivity gains from AI itself are so underappreciated. So January of last year, if you asked the best model at the time and you asked it some question and got an answer and thought it was a pretty decent answer, you can get the same answer today at less than 1% of that cost. This is the pace of productivity change. If you think the answer was good, imagine you spend 100 electricity, 100 Nvidia cards and that is the Jevons paradox playing out.

Eloise Goulder: Yeah.

Fawaz Chaudhry: The cost of the token is dropping by 100X, but the demand is growing by 1,000X or 10,000X. So overall demand is growing faster than the cost because more use cases are coming. The elasticity of demand is so high with the cost falling.

Eloise Goulder: Yeah.

Fawaz Chaudhry: So it's not a bubble. The demand is going to keep growing. The productivity gains are s- incredible. Like I'm sort of giving you two orders of magnitude per year or faster. Which industry you're seeing that kind of productivity? And we will keep finding new use cases for AI as the cost of token keeps dropping at such an incredible pace. And it's all about the societal impact that we'll have to start contending with. So it's going to cause disruption but I would say for equities, AI is not a bubble, and generally speaking, I would say it's going to become a general purpose technology and we're on our way. So super exciting times with... We will look back at, uh, November 2022 when ChatGPT-3 came out and took off as a seminal moment, similar to Netscape Navigator in '94, but even more so-

Eloise Goulder: Mm.

Fawaz Chaudhry: ... than the internet or the internet browser. Yeah.

Eloise Goulder: Mm.

Fawaz Chaudhry: Yeah.

Eloise Goulder: Well, I think on that very optimistic note for equities at least, we should close. Thank you, Fawaz. It's been such a fascinating conversation, a very far-reaching conversation, also really helping put into context these phenomenal structural forces we've witnessed over the last six years. And as you say, it's the beginning, not the end. There are so many ripple effects and dynamics that will happen as a result of the events we've gone through. So thank you very much, Fawaz, for taking the time to speak with us today.

Fawaz Chaudhry: Thank you. Thank you for having me. A pleasure.

Eloise Goulder: Thank you also to our listeners for tuning into this regular podcast series from our group. If you'd like to get in touch or ask questions, please do reach out to us at jpmorgan.com/market-data-intelligence. And with that, we'll close. Thank you.

Voiceover: Thanks for listening to JPMorgan's Making Sense. If you've enjoyed this conversation, share your feedback by leaving a comment or review wherever you listen to podcasts, and be sure to follow our channel so you don't miss an episode. The podcast views do not necessarily reflect those of JPMorganChase & Co or its affiliates, together JPMorgan, and are not from JPMorgan's research department. They do not constitute recommendations or offers to buy or sell securities. Intended for institutional and professional investors, not retail use, it is for informational purposes only. Products and services mentioned may not suit all investors or be available in all jurisdictions. JPMorgan may make markets and trade and discuss securities and asset classes. Visit www.jpmorgan.com/disclosures/salesandtradingdisclaimer for more disclaimers and regulatory disclosures. External speakers' opinions are personal and not JPMorgan's views. Copyright 2026 JPMorganChase & Co., all rights reserved.

[End of episode]