Infrastructure investing: what the label doesn't tell you

This is the Investment Researcher, going beyond the benchmark, brought to you by Zenith Investment Partners, bringing you insights and analysis for smarter investment decisions. Hello and welcome to the Investment Researcher podcast. I'm Dan Cave, Deputy Head of Income Research at Zenith Investment Partners. Today's episode is dedicated to infrastructure investing, which Australia has been a pioneer of. Whether it's been investment managers and super funds in private assets or wealth investors being early adopters of global listed infrastructure, which brings us to today with recent product innovation, advises now of greater access to private infrastructure for their clients. And so to discuss infrastructure's key drivers, investment themes and where the opportunity lies, we're lucky to be joined today by Steve Kempler, co-founder and portfolio manager at Maple Brown Abbot global listed infrastructure. And from Macquarie asset management, we have Kirin Zabrinich, Head of Transaction Strategy, Asia Pacific and Lee Portfolio Manager of the Macquarie Private Infrastructure Fund. Thank you for both being here. Liza. Thanks Dan. So Steve, starting with you, let's set the scene with today's macro and fundamental environment and really what this means for infrastructure. Yeah, thanks Dan, a really good place to start. So where we are today, I think there's multiple forces pulling in different directions. So it's a really interesting time for infrastructure investors. On one hand, you've got real rates having increased from lows over recent years, and that is obviously putting a bit of pressure on the multiple that people are willing to pay. But on the flip side, you have a really interesting tailwind coming from inflation, from the record capex cycle that we're saying, and all of this is driving momentum in earnings for these infrastructure assets and pushing positive accretion to valuations through time. And if you break that up into the individual components, most of your investors will be familiar that interest rates have continued to rise around the world, more at what the long end of the curve, but it certainly rise generally. But if you break up the interest rate yield into two components, you've got the real rate, as I mentioned at the start, and then you have the inflation component of the interest rate. And it's that inflation component that is doing a lot of heavy lifting for valuations. The types of assets that find their way into infrastructure portfolios, whether it's listed or unlisted are, in many cases, assets that have an explicit inflation escalator through a tariff or through a toll of some kind or indeed through their asset base. And it's that inflation tailwind that we've been living with for quite a number of years that in providing nice valuation support for a lot of the assets we invest in. I think the other big macro trend today and that we've really seen since the tail end of the pandemic has been an acceleration in capex by infrastructure companies around the world. That capex has come from a number of areas. It's initially been policy-driven energy transition, which has now become much more demand-driven. And more recently, it's been the increase in electricity, low growth coming from, I guess, all faster society, including data centers, which I'm sure we'll get into in a minute. So, as I said, the macro environment is a really interesting point right now. There's probably something there for the bulls and the bears. For the asset class, certainly has reduced correlations with equities, bonds, property, things are really an interesting time for investors to be considering and as part of their overall portfolio allocation. And Kiran, anything else to add to the macro discussion, sort of, thematic tailwinds? Let's talk about inflation, and that's obviously infrastructure is an asset class where you have high inflation. You see earnings rising as a result of that high inflation. He's sort of touched on that. There's sort of the four main structural trends we see today, which have been going on for a number of years now. One is digitalisation. So, you've got any infrastructure that underpins data creation, storage transmission is seeing a lot of growth and expansion in that space, de-globalisation is another one where people are realigning the supply chains, trade flows are changing and those geopolitical shifts are causing a revisit of movement of goods and other things that businesses have around the world. So, the infrastructure that sort of enables that is changing, demographic change. So, that's not just population growth, but it's also increasing in middle class in certain economies around the world and then lastly decarbonisation/electricisation, which Stephen also touched on. We're coming out of a period of 20 years of almost flat energy or electricity consumption and it was sort of going into a new period where we're seeing growth and energy in electricity consumption specifically and that's driven by two main themes. One is electrification of everything and that is pockets of the world where we're seeing people roll out that electrification of everything into their local communities, you know cars, hot water systems, obviously PV panels on roof batteries, that kind of thing, but also just the cap expand on the digitalisation. So, in particular, the data centre growth is driving a lot of increasing both consumption and new build for electrification. Thank you. Infrastructure is a very sort of technical asset class and very bottom up and for listed infrastructure definition really matters and recently our growth team covered this again in a recent GLI global listed infrastructure sector report. So, Steve, just so we're all on the same page, can you outline the infrastructure characteristics you're looking for and perhaps those that you're trying to avoid? It's a good question Dan and I think when you look across the infrastructure universe, they're clearly asked that everyone agrees. Our infrastructure and then there are assets that have some people say the infrastructure and their others that are saying, you know, they're a property asset or a transportation asset or an industrial company or some kind. And there are a couple that expose technical differences around the commercial frameworks between both buckets of assets and so when we think about infrastructure we're mainly investing in what we call core infrastructure defined by commercial frameworks as being either long-term contracted nature, operating in a regulatory environment or operating under a long-term concession agreement that is granted by a government or public agency of some kind. The commercial framework in our mind that really underpins whether an asset is what we call core infrastructure or not philosophically and sort of expose back to what we think infrastructure can bring to the end investor is a long-term alignment with some of those trends but specifically it's the inflation tailwind and the low volatility or reduced correlations with other asset classes and so specifically for us we're looking for infrastructure assets that provide a combination of low cash liability to equity and inflation protection through time. So I think probably just as important to think what doesn't meet that definition or doesn't sit within those guardrails and for us things like ports, rail businesses, integrated utilities, satellites and probably the more vex topic right now is data centers they I guess the commonality of most of those types of sectors that I've just mentioned are the elements of competition that creeps in and when you've got competition that you compete on price and you compete on volume and those characteristics themselves aren't naturally aligned with them and optically asset. I think within some of those sub-sectors and Kirin will certainly have some views there are definitely assets available in private markets that would suit core infrastructure investors saying a port but what we see is available in listed markets in the port sector for example are the operators that sit on top of the docks there are three or four in any given port area and they compete against each other for the volumes coming from a small number of global shipping like in a little bit different to what private investors might get for example in a port sector. Generally speaking risks that we're deliberately avoiding when investing in in core listed infrastructure are commodity price exposures merchant volume risk material construction and development risks and anything where price is set by user demand as opposed to a long-term contract or a regulatory agreement. Stephen maybe just in terms of lease length just confirm what you would typically see as lease what you define as long-term. So there's probably two components to that this first the actual contract in terms of time and then there's also what the alternative is for that customer so you know what you typically find is there'll be anywhere between five and 20 year contracted agreements for certain assets and these aren't regulated assets these are contracted assets but even in the cases of those operations where we'll be investing are where there aren't really any material alternatives so in our company operating a droply environment that has a 20 year contract is going to seize significantly reduced volatility in its long-term contract so it's a combination of time and reasonable probability of any alternative infrastructure actually being available. Kiran at a high level what Steve described sounds exactly like what core infrastructure on the private side is maybe can you comment on that and also what is core plus and I guess value out investing in private infrastructure terms. The core infrastructure is perhaps what people are most familiar with from the early days of infrastructures and asset class it's sort of toll roads and capital city airports and regulated electricity grids those kinds of things and then obviously as the definition of the asset class is expanded it does expanding to you know these new terms like core plus value add probably the simplest way to think about it is once you start increasing additional elements of risk into the business then you're moving from core into core plus and that could be things that like Stephen talked about might be you've got a bit more competition so you might be a landlord port where you're just renting out space and providing port services to a number of operators on your land and maybe that's sort of sitting or in the core bucket or you might be a steved-ouring operation where you've got a long-term lease over a site where you unload containers, but you've got maybe one at a or two other competitors within that port precinct that you're competing with. And so that might sort of shift into the core plus bucket. Data centers is a good example where you're probably more in the core plus bucket or your outside, the infrastructure definition, it really depends on the risk profile as you unpick the business case. But the other things that sort of move you into the core plus bracket would be things like, if you've got more operating risk, you know, what are your EBITDA margins in the business? How much complexity is there in the operations? It could be that you've got an element of construction risk where you've got a renewable energy business that has some assets in operation but some assets in construction so there's a bit more risk in the portfolio. So that, again, could be to delineate it, but it's really around that risk profile. But I think what it needs to look like, typically you've got hard assets, you've got low obsolescence of those assets, you've got relatively inelastic demand, an ability to pass on cost increases to maintain margins, things like that. Ultimately, in the definition of infrastructure, whether it's core or core plus, you're getting your capital back with some returns, what the risk profile should look like. And if it starts looking a bit more, this binary risk sitting inside your business could be technological obsolescence or other things, then that starts to move you out of the definition entirely. Picking up on Stephen's point, what it also should look like is it gives you an equity-like return but with much lower volatility. More closer to bond-like volatility, but equity-like return is what this asset class should deliver. And in a period of inflation, which we touched on the earlier question, it should do better in a high inflation environment, which is exactly the opposite of what you see listed in bond market. So again, it sort of sits nicely in a portfolio for that reason. Thanks for that, Karen. And maybe just, I guess, if you are taking more incremental risk than core, having control of that company does allow you to control those risks or at least sort of manage those risks, that sort of active asset management is that how McCory sees things? Yeah, that's our approach in the unlisted space. We'll typically, as we make these investments, we'll have a defined value creation plan that we work up right to making the investment in order to deliver that. You need to have control, so you can actually get done what you want to get done. And that includes operational changes, which could be driving additional revenues, making bolt-on acquisitions, it could be a cost-out program, transforming the business, rolling out new technology, upskilling management capabilities, all those sorts of things. So to do that, you need control. And then also, as you move up the risk curve from so core to core plus, you also want to have a think about your capital structure. So the leverage you put in a core business might be different to the leverage you might put into a core plus business taking to account that different risk profile as well. And again, you need to have control out of that cap structure. Steve, we've already talked about data centers not being within your investable universe. Can you explain why? And are there any other types of digital assets which you favor or avoid? I think data centers are one of the more vexed discussions we tend to have with investors in the infrastructure market. Largely because they are just so relevant to all of our daily lives today. When we talk about what is infrastructure, why are you investing infrastructure? Infrastructure is about providing an essential service to society. And I don't think we could have done this podcast itself without having data centers running the engine behind them. So certainly providing essential service to most of our daily lives today. But just because something provides an essential service doesn't mean it's infrastructure, office towers. We use every day, but they're not infrastructure. Houses are not infrastructure. And so similarly, in the case of data centers, we see them as sharing a lot of the characteristics or attributes of property assets rather than infrastructure. The barriers to entry are a lot lower, notwithstanding current short term concerns around electricity access amongst other things. The contract links that you typically find for data centers are also typically shorter end of the spectrum in anywhere between one and three years for typical retail co-location contracts. And ultimately, it comes down to a pricing question more than anything. The pricing is set by the market, not set by a regulator or a long term contract. It's not inflation linked. And I think the proof is in the pudding. If you look over the last 10 years, 2017, it's through 2021 on average, US asking rents for data center capacity, the client year on year every single year. And then we had the advent of AI and you've got this demand and supply and balance for data center compute capacity. And it's basically gone up 10, 15% per annum over that period. So it looks much more like a demand supply-driven pricing environment than I guess what you would typically see in an infrastructure environment. I guess on the barriers to entry point, probably worth noting that even in the largest data center operator in the world, Equinix, which is a US-lister company, if you have a look at their most recent 10K filing, they cite, or one of the largest risks has been competition coming from a highly fragmented industry of more than two and a half thousand companies operating similar services globally. So whilst these are amazing assets that do provide an incredible service to societies around the world, they resemble much more like a property asset, an infrastructure asset. And when you look at the indices as well, the FTSE Global Coin Infrastructure Index, the S&P Index, the Dow Jones Brookfield Global Infrastructure Index, none of them have data centers. Included as index constituents instead, data centers sit within all the REIT indices. So I think even the index providers are effectively classifying these most property assets. But they are great assets. And I guess, I'm here inside the world in the quarry. The Australian market made a pretty significant investment in business called AirTrunk, which they sold about 18 to 24 months ago at a very impressive price and multiple and generate a lot of money for their investors. You know, that in my mind sits more in the core plus of infrastructure as opposed to the core infrastructure. So here in credit where credit's due, our quarry has done very well on a number of data center investments. And this hasn't been a recent thing for you at Macquarie. So maybe you can sort of talk about these data center platforms in road terms and they've alluded to the risk return profile. But really, why are you investing them? What are the sort of return drivers? Maybe just to put some context around it, it's a very big capital expenditure program going on globally at the moment. And it's mainly from the hyperscalers and to a similar extent, the chip makers and the model makers as well, but it's all linked. The amount of growth that's going on in that CapEx program is also very high. So it's not just the CapEx numbers are big. Like this year's going to be estimate is 725 billion. Next year, 1.3 trillion. But the year before was like 14. So you can see like big numbers, but they're growing at sort of 75% year-on-year. And if you try and put that in a historical context, it's percentage GDP is bigger than Pollo Space Program Manhattan Project, but there's not many times in history we've seen something similar. It's bigger than the fiber rollout in the '90s by the Telcos. The railroads was bigger in percentage turns of GDP, but it's big, right? They're spending a lot of money. That's leading to a lot of demand, which as Stephen was alluding to pretty good returns for some of these data center businesses and developers at the moment. It will be seen what that looks like over the next 15, 20 years to things sort of normalize. Certainly unusual times at the moment. I would agree that there are many data center investments you'd look at that you'd say that's more of a property play. That's not an infrastructure asset. One to three-year contracts, co-location data centers, enterprise customers that have an ability to switch. There are churn rates that you can look at historically where people have switched location, switched data centers. So that starts more and more like a real estate business where you have tenancy managers making sure you're customer-stay or if they move, you've got a replacement tenant and you're constantly negotiating renewals and things like that. That I would agree with. In some instances, particularly what we look for is much longer-term contracts, 10, 15, 20-year contracts, you're still providing some infrastructure services to that client. You will own and operate high-voltage substations. You're providing water, fiber optic cable connection and other services around it. So in that sense, it's slightly different to a real estate business in that. It's not a triple net lease. You're actually providing infrastructure services to the client. But typically what we're looking for, if we're investing in an infrastructure fund into data center business, it would be long-term contract, investment-grade credit counterparty. So you have de-risked that business as much as possible. There's certainly some examples where you would regard them as a risk profile that sits outside infrastructure. Stephen touched on air trunk. We had also one in the US called Aligned, which we just recently sold this year, which we don't since 2018. So we've had many invested in the space for quite a number of years. And it looks like the amount of CapEx going into this space is accelerating. Steve, on the energy transition, there's been some ups and downs and data center build-out really seems to only be accelerating everything in energy markets. So where are you seeing the opportunities? Probably good segue from data centers themselves at the energy transition story. If you want to back the clock a couple of years, I think I mentioned this at the start, but very much policy-driven as opposed to demand-driven. Top-down by governments, I mean, targeting net zero across a lot of markets. And that drove how regulators thought about CapEx plans or integrator resource plans depending what market you're in, which obviously drove future generation mix planning. That story has shifted, I think, in my mind, far better for the end investor, certainly for the end consumer as well, in that it's become a little bit more balanced. And whilst decarbonisation as a driver has slowed, because of probably more than anything concerns around cost of living and bill pressure, what has replaced it has been an increase in energy demand generally and Kiran mentioned this earlier, how the last 20 years it's probably longer. And we've seen flat and negative energy demand and that's really pivoted in the last couple of years driven by combination of factors. We talked about data centers, but it's also the electrification digitalisation of our societies. It's the decarbonisation that is taking place as well, which is seeing an insuring of manufacturing in places like Mexico and Canada for the US and probably now going forward within the US itself only. but this is. certainly driving a big pickup in CapEx. And when we look over the last few years, where we have seen a lot of the investment opportunities around the energy transition, it has been in the transmission and distribution infrastructure that has been so under-invested for the last year, really, generation, as electricity, low growth has asymptote toward zero. So that is where we have seen a lot of the interesting opportunities, less so on the generation itself, because that has been competed away in terms of returns, particularly on the renewable side, but much more on the significant build-out of transmission networks and at a local level, distribution networks to handle increasing low growth and volatility in power demand coming through. As I said, we find this in the regulated space mostly, which is attractive for end investors because the investment and the CapEx that is taking place is completely disconnected from economic cycles. Every dollar of approved CapEx goes into a rate base earning an allowed return, but decades to come, most that asset remains under-appreciated. Probably the most topical one in our universe for some time has been what we've seen in Louisiana with meta-building its Hyperion development in Richland Parish. So I think that's close to $30 billion US now in development, which could scale quite significantly, but they've got roughly nine gigawatts also approved under construction, with plan of 13 or 14 there. So that is a single data center complex, that energy, which is a large US-regulated utility, will begin serving the data center through significant investment in the network, and that'll be only in that. But we're seeing that all across the globe at this point, a little bit of pushback is certainly creeping in, partly from a rate-past perspective, but the energy transition has become certainly a bit more balanced. That's not just the carbonization anymore, but it's serving in this increased demand for electricity and tired of that is coming from a need or a view by governments that need to balance that decarbonisation with energy security as well. - Kieran, what are you seeing on the private side as opportunities? - So there's probably four main drivers, and Stephen's touched on most of these, the data center build out to the obvious one. In a lot of parts of the world now, governments are saying, if you want to build a new data center, you need to also arrange your own power supply because the system doesn't have surplus power to meet your demand. That's an obvious one. The next one is electrification of everything, which is unrelated to what's happening with data centers. It's just people moving to electric vehicles, moving away from the usage of gas in their homes. And so that's part two. Part three is just the population growth and the growth in the middle class, and the third one is energy security where you've got geopolitical shocks, supply disruptions that in certain markets, they're sort of focusing more on what can we generate locally rather than rely on externally. So all those things are driving that increased a build out, but just the US example, and I think it's an interesting one that will apply that application across the world, but maybe it's more at the beginning of what's happening is that data center build out, you need power, you need it quickly. The fastest quickest way, relatively, to get that is actually wind and solar. I agree with Stephen that it's not really being driven by renewable energy credits, government policies, subsidies, grants, carbon taxes, anything like that. It's just now the fastest quickest way to build generation in most of these markets. What that then is translating into though is, obviously opportunities as a developer to build new generation, but it's also providing opportunities for grid owners like utility grids, whether it's transmission, high voltage transmission, or the low voltage distribution grids. There's a very big spend needed in addition to generation in the grid, and this is probably not a surprise to most of the listeners. We're seeing that in Australia, we're seeing it in the US and other markets, and so that's another area where we're invested in infrastructure class, where you've got these regulated utility grids, where your CAPEX programs are increasing because of this need for expansion and upgrade of the grid. And part of it is just more capacity, but part of it is dealing with two-way flows of the electrons as well. Steve, take privates of global estate infrastructure companies, especially in Australia, has been a theme we've been seeing for a long time, and has this been positive for investors? And I guess what are you seeing currently in the market? It's an interesting discussion, Dan, and I think with all of our Australian hats on, we've probably seen it a little bit more in this country than elsewhere in the world, and maybe that's the starting point where the Australian market was very heavily penetrated with listed infrastructure companies, relative to, I suppose, the global capital universe. Yes, since the pandemic, we have seen a lot more. Take private activity here than elsewhere around the world. And if you take a step back, probably two things going on, the combination of the tailwind of significant amount of money, going into private infrastructure up until a few years ago was in a drive-in demand for assets where they may find them, but there's also a message that I think in those cases, at least you've got private capital telling the market that that point in time to sit infrastructure was always looking cheap. So it's probably just as much about size of the checkbook or the availability of capital of these valuations. You know, when you look at the valuation gap, it really depends where you are in the cycle, probably depends on the sub sector, it depends on the geography, and it's not always going to be the case that there are going to be ample take private opportunities and perhaps, you know, at different points in the cycle, you might find listed infrastructure trading at a premium to private infrastructure as well. Probably the most recent example in the Australian market that listeners would be familiar with being the IFM tilt at Atlas Arteria, which they managed to close about 67% ownership of the end of June. So not quite a full take over of the business, but that's certainly in a trying to take advantage of what many believe was their view of a significant disconnect between where the shares for trading and list what the fair value of the company is. And there are many reasons for that disconnect to exist, but probably one of the more recent examples. When you look globally, take private phenomenon has been a little less obvious. Part of that is because in the US and Europe, you tend to find listed infrastructure and regulate utility businesses, market caps being pretty significant and really reducing the buyer universe who could, you know, ride a check of the size that we're talking and, you know, look at the FTSE global core infrastructure universe, I think an average market cap is close to $50 billion in US dollars now. So on average, I'd say a lot of these businesses are in the two big basket to be taking private opportunists to claim. Kieran, we've mentioned take private. So how do you think about the different ways private infrastructure managers can source in investments? And where are you seeing the opportunities? Stephen talked about Australian experience where we had listed airports, regulated gas and electricity, assets listed, fiber optic cable networks listed and other things as well. And there had been a lot of take privates in these train contexts over the last 15, 20 years. And perhaps more so, but certainly it's also been occurring in overseas markets. When we look at these, we've made some of these take privates ourselves. Most recently, we closed on cube and before that, we closed on focus from the ASX. You need to be seeing some value that isn't able to be achieved in a listed environment. And so it's not so much that the listed market has misprice those assets. It's just you can do something in an unlisted environment to drive additional value. So that's, I guess, the comment I'd make in terms of what's driving it. But in terms of our approach to sourcing investments, which was your question, the moment we're tenting to find most of their opportunities coming through exclusive or bilateral opportunities. And that's not necessarily always a case, but if you just look at the last three years, almost 70% maybe a little bit over of our investments we've made have actually been sourced by laterally. So it is very much large global team, local teams in each market we operate in discussions with potential sellers, potential partners. One of the things we've seen more recently is partnerships with corporates. So whether it's joint ventures or them selling some of their assets to us, and we provide services back to them. And just some example of that, we have a joint venture in the US with Dow Chemical, where they've put some of their infrastructure assets into a joint venture with us. And we run that as a separate company and provide services back to Dow for the large chemical parks in Louisiana. We've just announced an adventure with Anthropic to build data centers together. And we've acquired some base stations from the Japanese mobile network operator called Rackerton, where we ought to base stations off them and we lease them back to them for their use as part of their core business. So there's some of these opportunities we're finding globally that are not just tight privates, but also just partnerships in some form with strategic to try and meet an objective that they might have. We've covered a lot of really interesting topics today, but in terms of takeaways, three things that really stuck out to me were. One, the all-weather nature of infrastructure, particularly in the current investment environment. Two, if we can put aside discussions on listed V unlisted, we can see that there's some really strong common characteristics as well as some complementary opportunities. Could be sector, sub-sector exposures, could be core, core plus risk, or those sort of value creation, asset level sort of management of private. But most importantly, for both listed and unlisted key takeaways really having an appreciation of the definition of infrastructure and a sort of risk profile of a given investment, given fund, or the allocation overall allocation. Kieran, and Steve, thank you for joining me on the Investment Researcher podcast. - Listen, thanks Dan. - Thanks Dan, thanks for hosting. - You've been listening to the Investment Researcher, going beyond the benchmark. For more research and insights, visit zenithepartners.com.au.