Go back

Stonepeak: ‘Power scarcity gave data centres their moat’

35m 29s

Stonepeak: ‘Power scarcity gave data centres their moat’

In this Infrastructure Investor Podcast episode, host Bruno Alves speaks with Michael Durrell, chairman, CEO, and co-founder of Stonepeak, the world's seventh-largest infrastructure manager. Durrell explains how Stonepeak evolved from a US-focused firm into a global platform, beginning with an Asia expansion driven by the opportunity to earn significantly higher returns in renewables than in Western markets. He emphasizes that geographic diversification is opportunistic rather than strategic, requiring the right teams and scalable opportunities. The conversation covers Stonepeak's Middle East expansion, including a Riyadh office and a billion-dollar partnership with the Arab Energy Fund, where asset creation and government involvement play larger roles. A central theme is moats. Durrell argues infrastructure assets either possess natural moats, like airports, utilities, and cell towers, or rely on long-term contracts. He notes that power scarcity has transformed hyperscaler data centers, giving providers genuine bargaining power and enabling 15-year contracts, a shift from the previous five-to-seven-year norm. He also discusses Ithco Group's reusable packaging business as an example of a logistics network with high barriers to entry. On US renewables, Durrell remains cautious, citing thin returns, early cash flow dynamics, and regulatory uncertainty, though recent legislation provides clarity. Finally, he observes that infrastructure is gaining share in LP portfolios relative to real estate, but views this partly as cyclical rather than purely secular.

Transcription

6177 Words, 34218 Characters

English
Speaker 1Hi, I'm Bruno Alves, and welcome to the Infrastructure Investor Podcast. My guest today is Michael Durrell, the chairman, CEO, and co-founder of Stonepeak. Stonepeak is one of infrastructure's premier managers, the seventh largest in the world, according to the latest edition of our Infrastructure Investor 100 ranking of the biggest global GPs. Michael's an industry veteran, and we spent a lot of our conversation talking about moats, how some infrastructure assets naturally have them, and what's the next best thing if they don't. He made some really interesting observations on how the current power scarcity has actually created moats and better contracts around hyperscaler data centers that would otherwise lack them. We also spent time talking about the U.S. renewables landscape, Stonepeak's geographical diversification, and how the asset class is evolving. Hi, Michael, welcome to the podcast.
Speaker 2Hey, Bruno, good morning from my time zone. Good afternoon to yours.
Speaker 1So, rightly or wrongly, for a long time, if I thought about Stonepeak, I would immediately think about North America, and that's obviously not you anymore. You guys have spread out, you're in more regions, but I'm interested in learning more about the internal processes that, you know. That prompted that regional diversification, and specifically, with thresholds had to be
Speaker 2cleared. Great question. So, I'm Aussie, and I've had most of my career over here in the U.S. just by way of background. So, we were very U.S.-focused for the initial half dozen years or so, and what first prompted me to look internationally was we were really struggling to find interesting deal flow in renewables in the U.S. There was lots of it, but the returns we were seeing in that sector were pretty thin, certainly lower returning than what we were targeting in our various platforms. And what I noticed is that there was a lot of deal flow being done by a group over in Asia, and that group worked for my old firm, Macquarie, who's had a lot of success themselves over in Asia. They're obviously in the region, and I was quite friendly with the folks who were running the group. And so, to cut a long story short, it became pretty clear to me, talking to them, that the returns they were seeing on wind projects, solar projects, and now battery projects in some of the Asian regions were, you know, a thousand basis points better than what we were seeing in some of the more westernized markets. I guess I'm just talking about the U.S. and Europe. And so, we ended up hiring that group, and it put us in. And then we went from there. So, we had really great success with that group, really doing what. Continuing to do what they were doing, developing wind and solar and battery projects in Asia. And it was pretty obvious there weren't many offerings of more broad infrastructure in Asia either. And look, it's obviously a very fast-growing region. It's a big population. You know, there's a lot of GDP coming out of that part of the world. And so, we. Then parlayed that group we had over in Asia doing renewables into a broader Asia platform. So, we'd now be, you know, amongst the largest infrastructure platforms over in Asia. Whilst we were doing that, we were seeing what I'd call a bunch of niche deal flow in Europe. And so, you know, slowly but surely, we started to build a team up over in Europe as well. And then it just gets a life of its own. So, that's how we. We're certainly not trying to be everywhere and do everything. We've done it, I think, in a very targeted way where we see a niche opportunity. We think we've got the right people to go and tackle that opportunity. And then we go for it. And you do that enough times. Over time, you start to build a, you know, a decently sized footprint globally. So, that's how it's come about.
Speaker 1Yeah. Just in terms of the Asia business, if I look at how you've done it, you know, you've done a lot of work. You set up, you know, your own dedicated team there. You have your own dedicated strategy just targeting those countries. Is that kind of the model for how you ideally expand regionally?
Speaker 2It's a little bit opportunistic, more than a grand strategy that will replicate. You know, it's always driven by you think there's a really good investable opportunity that you think is scalable and you've got the right. It's not much more complicated than that. You can only do so many things new at once. I would say going into Asia was a real learning experience for me in terms of what it takes to expand geographically. So, for example, it's one thing to go and start a new product in the U.S. with a team that I've worked with for, you know, a decade or so. And they're doing something quite similar to what they've done historically. That's a very easy thing for us to go and do. Because it's the same people, unfamiliar with the deal flow. It's nothing real new to that. Whereas when you go and expand into a different geography, you know, you typically. Certainly, we were talking about folks that I knew, but they hadn't been a part of the firm before. And I knew like a handful of the folks. I didn't know all of the mid-level and the junior folks. They didn't know the Stone Peak culture so well. I would say the challenges of going. To a new region are much, much higher versus just going and doing something that's closer to what you've done traditionally. So, I certainly got a lot of learning experiences, positive and negative, out of doing that, you know, foray into Asia. I'm super glad we did it. I think it's worked out fantastically well. But, you know, I think there was a lot of, you know, the duck legs under the water swimming very, very hard to make that work over time. And so, we take that lesson. And I think that's what we're going to do with this as we look at, you know, what else we might do in the future.
Speaker 1Just keeping on this geographical diversification for a bit longer. Earlier this year, you opened an office in Riyadh in Saudi, which was your second in the region, if I'm not mistaken. And then you also signed a billion-dollar partnership with the Arab Energy Fund. And so, I'm just curious, what are Stone Peak's plans for that particular region? And both from the point of view of new LP relationships, but also as an investment destination.
Speaker 2So, look, there's a lot going on in that region at the moment. You've got several very high-population, big GDP countries over there, more or less all of whom are really focused on growing their infrastructure base. I think in many ways, it's got a lot of parallels with our Asia experience in that what we've really focused on in Asia is asset creation. You know, I contrast that maybe with the U.S. and with Europe, where there's a huge, big existing base of companies and assets to buy in the U.S. and in Europe. And so, you can make a very good business out of buying just existing operating businesses in those big markets. I think that's more difficult in the more faster-growing, maybe a little less mature economies in Asia. And I'd say the same in the Middle East region. And so, when you're doing asset creation, it requires a whole different skill set. You need construction experts. You need development experts. You just need folks who have been in the rhythm of developing and creating assets over time. And so, that's a skill set that we have developed, I think, really well through our Asia experience. It just so happens that the person who runs our Asia business ran the Middle East business. For another large firm. And I would say was arguably the pioneer in the space of PPP investing in the Middle East, particularly in the UAE region. So, again, we happen to have a team that is, including a leader, that is very well experienced in the GCC region. You've got a very accommodative set of governments over there who want to drive this investment and growth. In their economies and, you know, certainly infrastructure is a big component of that. I think the difference perhaps with Asia is that the transactions in the GCC will have more government involvement. The government, I think, is a more important factor to consider in those countries versus Asia where it's maybe more on the private, just purely private enterprise side. And maybe all I mean by that is that I think it'll be a lot of PPP type activity in those regions that involves the government as a part of it. Or certainly a very involved stakeholder. And so I do think there's a somewhat more of a government element in those markets than maybe what we have in the U.S. or Europe or Asia.
Speaker 1You know, it's hard, I feel, not to have a discussion these days and not to discuss digital infrastructure. And I think this conversation is not going to be an exception to that. But I wanted to ask, I wanted to touch on something you actually said in a presentation to the Washington State Investment Board. earlier this year, I think it was in February, and you basically told them that you were positive about digital infrastructure, but you said that the level of exuberance is high at the moment. And so you have to be careful essentially. And I was hoping to hear more about what you had in mind specifically when you said this at the time. And also, you know, if you still feel this way, it's only been six months, but curious about that too. Yeah.
Speaker 2Look, I think that's a really good succinct summary of my thoughts on the space. Look, on the one hand, you've got the most incredible macro tailwind behind that sector. And it's not much more complicated than our consumption of data is just growing at exponential rates and has been growing at exponential rates for a long, long time. And it looks like it will continue to grow at exponential rates for a long, long time. And you throw AI, you know, fuel on that fire, and it's really a pretty potent macro tailwind there. So that's, brilliant. Because the more data that we consume, whether it's on our, you know, phones or through our streaming or, you know, it's commercial data or it's AI or whatever it is, you just need more infrastructure to support that. You need more cell phone towers, you need more antennas, you need more small cells, you need more spectrum, you need more data centers, you need more fiber cable, et cetera, et cetera, et cetera. So that's all tremendous. And you look at the amount of capex that's going into digital infrastructure, the number of, you know, the number of, you know, the number of, you know, the numbers are absolutely astronomical. You know, every quarter or maybe even more frequently, you get new announcements from the big, the big tech companies as to how much they're spending on their cloud businesses and their cloud businesses, you know, they incorporate a lot of the AI stuff because it involves the data centers that the AI sits upon. And, you know, the amount of spend has gone up four or five, six, seven times where it was before AI, so it's an incredibly positive trend. But where it's more treacherous, I would say, is that there are a lot of very large, very sophisticated players in that space who are either your counterparties or your competitors. So I'd think of Google or Microsoft or AWS, you know, that's more likely to be a counterparty than a competitor. But then you've also got, you know, a lot of sophisticated corporates who are looking to compete. And you've got frontier technology companies that are inventing things that you can't foresee today, may provide a competitor or may undermine a technology that the infrastructure is resting upon today. So you've got that going on. And it's also a space where it is, it's infrastructure. We're always trying to, to use a simple analogy, we're always trying to look for a toll road. People find it hard to bypass where you charge a toll. That's essentially what you're looking to do in all these different sectors. And so, you know, an electric utility is essentially a toll road for your power. Like you've got to use your local electricity to get power. And it is quite difficult to find toll roads, if you like, in the digital space, again, because the ground shifts technologically so quickly, but also because Microsoft and Google and Amazon and Apple and so on and so forth, they go out of their way to make sure that there are not infrastructure toll roads because, you know, they want to be the person with the bargaining power. They don't want you to be the person with the bargaining power. So that's the kind of jungle in which we're all doing this digital infrastructure investing. And so you can get very specific on certain asset classes, and ultimately you just have to get very micro in the particular investment opportunity you're considering. But that's the backdrop in which you're doing all of this. So it's trickier, in my opinion, than going and buying your local utility or your local airport or your local toll road. So that's maybe just in a macro broad sense, what I was getting at.
Speaker 1Yeah, no, that makes perfect sense. And it's interesting because when you hear a lot of the tagline, you know, behind this tremendous tailwind, certainly from our community, you hear that you're investing in the picks and shovels, so to speak, and so supposedly you're more insulated against some of the stuff that you just mentioned, such as tech risk, et cetera, which I always felt is true only to a certain extent, obviously, also for the reasons you mentioned. So I guess my next question would be, you know, within reason and what are your do's and don'ts? How do you protect about, you know, against some of the stuff you mentioned against getting caught up in, you know, quicksand, so to speak, or, you know, how do you try and do
Speaker 2that? I think there's really two ways to think of any infrastructure asset as a business proposition. You've either got a big moat, around that business, it's just a natural toll road. And so a very good example of that is like an airport has a natural moat. And again, an electric utility has a natural moat. A cell phone tower, I think, has a natural moat. It's not very easy to go and site new cell towers. So I think, I think in the digital space, the cell phone towers are a really good example of a business with a great moat. I think by and large, though, most digital, digital businesses, from an infrastructure standpoint, don't have a natural moat. And therefore, the other way to come at it is, do I have a long term contract? You know, so a good example is a power plant doesn't have a great natural moat. And so you go and get a 20 year power purchase agreement on that asset. And that's where you get your comfort from your cash flows, you've got a contractual basis to it. And so that's the other way to come at it. So, you know, you know, asset cost is your at the moment, which is data centers, of course. So I think that I'll talk about, you know, hyperscaler data centers, because they're that's where most of the money is going. And by hyperscaler data centers, you know, they're supporting the cloud. And it's both traditional cloud, but it's also AI is utilizing the cloud. So I would have said, traditionally, there's really no moat around a hyperscaler data center. It's just a warehouse that's got access to power to drive cooling. And everyone's hyperscaler data center kind of looks like the other person's hyperscaler data center. And so I do think that traditionally, you're relying more on contracts than you were on some great moat around that business. And you almost, you know, that space was unusual in that there's also a, you know, the other bet you can make, you don't have a moat, or you don't have a great contract. The other bet you can make is that just supply is not going to be able to keep up with demand. Like there were some moments in time, and in certain industries where demand is just so strong and growing so quickly that supply can't keep up. And so that's the other bet that I think you could make in the hyperscaler data center space is that demand was growing so quickly that supply just couldn't keep up. Now, I think that before the AI age, there was a little bit of an element of that, but really, I think supply and demand were pretty evenly matched. And you weren't seeing, you know, there's probably a good metric to look at. What sort of returns on capital am I getting? If I go and look on the balance sheet of a, or balance sheet in P&L, I guess, of a company, what sort of returns on capital am I getting? When I just do the simple accounting analysis, and you weren't getting any sort of amazing returns on capital in the hyperscaler data center space historically. And so I think it was pretty evenly matched. And so kind of long story short, it was a little bit of a contract bet you were making and a renewal of that contract. And then AI came along, and AI has had a profound effect on the hyperscaler data center space. And it's done it in a couple of ways. It's certainly sentiment-wise, it's really been great for sentiment. But more importantly, what it did is it put such demand for new data center space, the demand did start to outrun supply. So it's a big positive. But what fundamentally changed that space is that we ran out in many of the cities, access to power for new data centers, where it certainly became hard to get power. And so all of a sudden, I'm not just a warehouse wood cooling that looks like, you know, everyone else's warehouse anymore. I've got power, and you can't get power elsewhere. Like you've got to wait, depends on the city, but you've got to wait two, three, four, five years in a queue to get access to power. So all of a sudden, you know, whereas historically, Microsoft, Google, Amazon, they had all of the bargaining power when it came to negotiating, your rental or contractual arrangement with the data center provider. It's not the case anymore. Now the data center provider actually has a good degree of bargaining power because they've got power. And that's a golden ticket at the moment. And so if I just go back to that simple statement earlier, which is you've either got to have a moat around your business or a good contract, you know, I would have argued that maybe historically, you didn't have a good moat around your hyperscale data center, and you practically didn't have a great contract either. Contracts were five or seven years long. That's totally changed today. Now you've actually got a good moat around your business because you can't get access to power is now a scarce resource. So whilst that scarcity remains, and I think it will remain very city specific, but I think it will remain for quite some time. You've got a good moat. And not only that, because you've now got a good moat, the contractual structure has changed. So whereas historically, you know, Microsoft or Google or Amazon, you've got a on would enter into a five or seven-year career. contract with you. Now, the most common contractual structure is 15 years.
Speaker 1I think that's really interesting because, to be honest, you sometimes hear, you know, if you hear the word speculative development or you're used to and you're thinking about an infrastructure strategy, then you think about it as a negative, right? Because what you want is the long-term contracts and a good counterparty, etc. I think what you introduced there with the scarcity is what flips that on its head. And if you do, I'm guessing if you do have access to power and you have that, then you can do a little bit of, you know, I'll build it and they'll come. And then you can also set some contractual terms. So I can see it remaining in place while the scarcity is there, as you've highlighted, which is quite interesting.
Speaker 2I want to go a little bit further. I don't think many folks are building data centers and hoping they'll come. I mean, I think I will come, okay? I think it's a good bet, but I don't think you even need to do that. I think. I think what's happening, in fact, more than think, I know what's happening is that you'll go out and get land and you'll go and get in the queue for power. And eventually you'll get access to that power. And at that point, you haven't spent much money. You know, you may be optionally, you've spent a bit on lawyers and filing fees to get your access to power. But once you've got land with access to power, you can go to one of these big tech companies and you go get your contract at that point. So by the time you're putting real dollars into actually, you know, putting shovels in the ground and all the rest of it, you've got a 15-year contract. Now, you may take the view that I'll contract out my whole, all my capacity. And the beauty of that is, well, your cash flows for 15 years are fixed. You know, you've got great credit counterparties and you might take that bet. Or alternatively, you may say, you know, the tech companies drive the hardest bargain. I'll rent out half my facility to Microsoft. And that will underpin, you know, a decent. I agree with my cash flow. But the other half, I am going to take a little bit of a bet that I can get other customers into the data center. And so folks do that as well. Now, I think that's a very easy bet to make because, you know, even in the pre-AI era, new data centers in particular, they all filled up. So I don't think that's a crazy bet at all. And you'll get better rentals from the second 50% that turn up that aren't Microsoft because they just won't drive the same bargain that the big hyper-competitors have. So I think that's a very easy bet to make. So that's how folks go about it. So I don't think there's a whole lot of speculative dollars going to work in that space. I think it's pretty well controlled.
Speaker 1Yeah. Let me stay with the idea of moats and what is traditional infrastructure just for a while longer, because I'm a bit curious about hearing more about this recent transaction, this recent acquisition, actually, of Ithco Group, which you guys have been involved with. Ithco focuses on reusable packaging solutions for fresh food. So again, I'm thinking, if I'm thinking about traditional asset class characteristics and long-term contracts, moats, et cetera, how does that fit in for the benefit of our audience?
Speaker 2Yeah, sure. So I think that logistics network businesses are some of the best places to be in infrastructure. And I'll use Ithco as an example, but a lot of what I'm saying applies to logistics network businesses more generally. Not all of them, but certainly a lot of them. And so just for the uninitiated, Ithco essentially provides plastic pallets, plastic pallets that you put fruit and vegetables on, and those pallets transport the fruit and vegetables from the farms to the stores, to Walmart and Kroger and et cetera, et cetera. So on the face of it, plastic pallets don't sound real interesting. Like, how the hell is that infrastructure? Well, what happens is once you become the largest network in supplying those plastic pallets, it is very, very difficult for someone else to come in and either compete with you or take business away from you. So no one else can compete with us from a price perspective on going and picking up the next customer or the next customer or the next customer. And on the other hand, when a big food retailer chain decides they're going to use plastic pallets instead of cardboard boxes or what they've traditionally used, once again, like we're already going collecting fruit and vegetables from all the folks who supply that. And so it's very difficult. I mean, it's impossible for anyone else to be able to provide the same price that we can provide to that big retailer of food and vegetables. And so it's in terms of winning new business, once you've got the incumbent network, it's very difficult for anyone else to be able to provide the same price that we can provide to that big retailer. And so it's very, very difficult for someone else to come and win new business. Then on the other hand, once you've got the business, what happens is that both the suppliers of the food and the folks who sell the food, all the logistics that sort of are between those two and incorporate those two, they all get set up to deal with your plastic pallets. And so for someone else to come in and try and disrupt that, again, it's just the switch out costs are very, very high. And so we've just found that the network is going to be very, very difficult for someone else to come in and win new business. Natural barriers to entry for big logistic network businesses are very high. But the other way to come at it is look at the numbers. Look at the, again, I'm a big go and look at just the accounting return on capital, because what that tells you is if you've got a very low return on capital, that says to you that other folks have come in and competed away, profits. Whereas you've got an attractive return on capital, that says to you that, in fact, there is a money around this business. And because if you can get a nice return on capital over a long period, it tells you others have not been able to go and chisel away at that.
Speaker 1Michael, we started talking about, well, we started talking about North America, specifically the U.S. in this conversation and actually renewables. And I want to kind of circle back to that because, you know, as a predominantly North American investor, certainly in terms of your history and also with a strong energy investing background, I noted that when you spoke to the Washington Post, you said, you know, I'm a big go and look at just the state investment board in February. You said you were now being more cautious about renewable investments in the U.S. and particularly new projects that required some kind of federal support, be that subsidies or permits. Am I right in thinking that events since you made those comments in Feb have actually probably reinforced that view?
Speaker 2So I have been, I wouldn't even say cautious. I've been a bit negative on the U.S. renewable investing landscape for a long, long time. And the reason for that is that when I look at the, look, what you're trying to do with any single investment ultimately is you try to predict the future cash flows for as far out as they go and discount those back. It's that simple. And when I do that for wind and solar and batteries in the U.S. and I compare that to the, I look at the future cash flows and I compare that to the construction costs of those projects, I don't think the returns are going to be that high. I think the returns are going to be that high. Particularly in the early years. It's interestingly, it's an asset class, particularly solar, that the cash flows, the revenues diminish over time because the efficiency or the production of a solar panel reduces over time. Anyway, so you get these big cash flows early and we've seen this in certain MLP investments when the oil and gas industry, I think you had an element of this in data centers and we see it in renewables that assets that cash flow a lot, particularly early, have a tendency to get overpriced. And I think that's what happened in the wind and solar space. And so I've been negative on it for a long, long time because of that in the U.S. And I'd say a similar thing about Europe. And there was this thing where developers would develop these projects and flip them very early on, often before there even, there was a shovel in the ground. And it just became a space that was priced more on what the next person was willing to pay for it, maybe more than what I would have paid for it. And so I think that's what happened. And I think that's more than I have to hold this thing for 20 years and I'm happy with the return. So that's the tradition of it. And then obviously there's been a lot of negative regulatory stuff around it more recently. And there's a lot more certainty around it now because of the, you know, big, beautiful bill has now been enacted. We now know when the wind and solar subsidies will be rolled off. What we don't know is, and not to get into too much of the detail, but the key trigger as to whether you're going to qualify for subsidies or not is whether your project has started construction. And, you know, historically all that meant is, you know, someone's gone out with a tractor and dug a hole and that's starting construction. So you didn't have to do a whole lot, but it was pretty clear. They'd given a lot of discretion to the regulators to define quite strictly what starting construction means. And I don't think it's going to be so loosey-goosey. Going forward. So all I'm saying there is there's still some uncertainty around which projects will qualify for subsidies going forward because it's uncertain what starting construction means. But anyway, so there is at least a certainty around it now. So I think the space is investable because there's certainty, whereas previously, if you get back three months, I didn't even know what the rules were going to be.
Speaker 1Michael, just one last question, slightly philosophical if you want it, but I just wanted to get your take on how the market is broadly evolving. And for context, we did a recent cover story focused on the evolving role of infrastructure in LP portfolios. And to give kind of a broad summary, some people felt infrastructure was on its way to becoming sort of the defining real asset class within portfolios and so kind of stealing a bit more market share from real estate. Other people felt some of the platform building and value-add opportunities you could do in the mid-market were actually capable of generating some private equity-like returns, thus sort of drawing some money that could have gone to private equity. What are you hearing from your, you know, clients? What are they telling you when they engage with the asset class and what do they want out of it these days?
Speaker 2So, you know, if you look at the countries that have been in the asset class the longest, which Australia is the longest, but then Canada and Europe is not too far behind, you know, if you go and look at the average Australian superannuation fund or pension fund, their allocation of infrastructure is typically 10% to 15%. Europe is probably not. Not quite that high, but Europe is nearing, I'd say somewhere here, it's probably between 5% and 10% and Canada is probably around 10%, roughly speaking. US is later to the game. US is probably around, across the whole pension fund market, is probably around 3% to 4% to 5%. So I think that the US is going to continue to increase their allocations over time. And historically, it hasn't come from any particular asset class or not. It's tended just to come, you know, prorata from public equities and credit and some private equity allocations and some real estate allocations. No one's been eroded away, I don't think, particularly more than others. I do think that it's pretty common to hear from pension funds that they're pretty negative on real estate. And that's simply because real estate has been in the doldrums since interest rates went up. And, you know, unlike infrastructure, which has a lot of operating levers often to pull, spare real estate often doesn't. You've got rent and you've got cap rates. And rent, you know, tends to go up with inflation or thereabouts in a steady state. But real estate's gone up by, sorry, interest rates have gone up so much that there's no way rental increases can outweigh the cap rate changes. And then on top of that, you've got what's gone on in our office for the last couple of years. So there's been a lot of bad news coming out of real estate for a couple of years now. And so human nature is when that happens, well, you know, I'm sick of this sector that's causing me a lot of grief. I'm going to go elsewhere. And so I do think at the moment, there is a bias or a trend to lean more into infrastructure and less into real estate. I would not go so far as to say that's a secular or longer term trend. It may be, but I wouldn't, I'm not willing to go that far today. I think there's a strong cyclical element to what's going on at the moment. And, you know, real estate's not going to be in the doldrums, forever. And at some point, supply and demand will match up in all these barriers, you know, in office and logistics will continue, I think, to be a good underlying trend. And at some point, cap rates will move in the right direction for real estate. And all of a sudden, you'll get a bunch of wins in real estate and folks will say, oh, hang on, real estate looks more interesting. And they'll pivot back to that. So I wouldn't be so bold as to think that real estate is going to suffer forever as infrastructure picks up, but I do think at the moment, infrastructure has shown itself to perform a lot better than real estate through this recent cycle. And so because real estate and infrastructure are often thought of as, you know, part of the real asset sector, I do think there's some truth to your observation there that infrastructure is winning out versus real estate at the moment.
Speaker 1I think we'll leave it there. Michael, thanks very much for your time again. It's very generous of you and it's great to see you.
Speaker 2Thank you. Hey, Bruno, great to see you. I appreciate the time and we'll do it again sometime. Yeah, we sure will.
Speaker 1Thank you. value today at infrastructureinvestor.com. Infrastructure Investor is your central source for navigating the global infrastructure market. Gain exclusive access to deep industry insights, real-time market intelligence, and expert analysis. Our platform is designed to help you identify opportunities, expand your portfolio, and connect with key players in the infrastructure sector. Registration
Speaker 3is now open for the Infrastructure Investor Global Summit 2027, the definitive gathering for infrastructure capital, taking place February 22nd to 25th at Station Berlin. Join more than 1,200 limited partners, 400 speakers, and 3,500 industry leaders from more than 50 countries, all converging in one location for four days of access, insight, and dealmaking. Where capital, opportunity, and strategy converge, secure your place today. Early bird registration is live now at peievents.com. That's peievents.com.

Podcast Summary

Key Points:

  1. Michael Durrell, chairman and CEO of Stonepeak, discussed the firm's evolution from a US-focused investor to a globally diversified infrastructure manager.
  2. Stonepeak expanded into Asia by hiring a Macquarie team that was achieving returns roughly 1,000 basis points higher than Western markets in wind, solar, and battery projects.
  3. Geographic expansion is driven opportunistically rather than by a grand strategy, requiring the right people and a scalable investable opportunity.
  4. The firm opened an office in Riyadh and signed a billion-dollar partnership with the Arab Energy Fund, focusing on asset creation and PPP-style transactions in the GCC.
  5. Durrell remains positive on digital infrastructure's macro tailwinds but warns of high exuberance, noting that hyperscaler data centers traditionally lacked moats and relied on contracts.
  6. Power scarcity has created a genuine moat for data center providers, shifting bargaining power away from hyperscalers and extending typical contracts from five to seven years to around 15 years.
  7. Durrell has been negative on US renewables for a long time due to thin returns, early cash flow dynamics, and developer flipping, though new legislation provides more certainty.
  8. Infrastructure is gaining allocation share versus real estate in LP portfolios, especially in the US, though Durrell sees a strong cyclical element rather than a purely secular shift.

Summary:

In this Infrastructure Investor Podcast episode, host Bruno Alves speaks with Michael Durrell, chairman, CEO, and co-founder of Stonepeak, the world's seventh-largest infrastructure manager. Durrell explains how Stonepeak evolved from a US-focused firm into a global platform, beginning with an Asia expansion driven by the opportunity to earn significantly higher returns in renewables than in Western markets. He emphasizes that geographic diversification is opportunistic rather than strategic, requiring the right teams and scalable opportunities. The conversation covers Stonepeak's Middle East expansion, including a Riyadh office and a billion-dollar partnership with the Arab Energy Fund, where asset creation and government involvement play larger roles.

A central theme is moats. Durrell argues infrastructure assets either possess natural moats, like airports, utilities, and cell towers, or rely on long-term contracts. He notes that power scarcity has transformed hyperscaler data centers, giving providers genuine bargaining power and enabling 15-year contracts, a shift from the previous five-to-seven-year norm. He also discusses Ithco Group's reusable packaging business as an example of a logistics network with high barriers to entry.

On US renewables, Durrell remains cautious, citing thin returns, early cash flow dynamics, and regulatory uncertainty, though recent legislation provides clarity. Finally, he observes that infrastructure is gaining share in LP portfolios relative to real estate, but views this partly as cyclical rather than purely secular.

FAQs

Stonepeak initially struggled to find attractive renewable deal flow in the U.S. with target returns, so it hired a Macquarie team in Asia that was seeing returns about 1,000 basis points higher than in Western markets.

After developing wind, solar, and battery projects in Asia, Stonepeak saw few broad infrastructure offerings there and used that renewables team to build a larger Asia platform, making it one of the largest infrastructure platforms in the region.

In a new region, the team is often unfamiliar with Stonepeak's culture, and the firm may know only a few senior people rather than the full mid-level and junior staff, making execution much harder.

The region has high-population, large-GDP countries focused on growing infrastructure, and Stonepeak sees parallels with Asia in asset creation, with more government involvement and PPP-style activity.

He remains positive on the macro tailwind from data growth and AI, but warns that the space has high exuberance, sophisticated counterparties and competitors, fast technological change, and few natural toll roads.

Power scarcity has created a moat for data centers that have access to power, shifting bargaining power from hyperscalers to data center providers and leading to longer contracts, commonly 15 years instead of five to seven.

Chat with AI

Loading...

Pro features

Go deeper with this episode

Unlock creator-grade tools that turn any transcript into show notes and subtitle files.