Digital Realty Trust, Inc. (DLR) Earnings Call Transcript & Summary

Welcome everbody to our afternoon session. I'm John Atkin with RBC. We're going to spend the next 30 minutes doing some Q&A with Digital Realty Trust. From the company, we have Matt Mercier, the Chief Financial Officer and Jordan Sadler, the Senior Vice President of Investor Relations. Welcome, gentlemen.

Thanks for having us.

And I think maybe just to kind of level set, give us the latest sort of highlights around scale of your platform and trends that you've seen in 2023 to date in your various business segments.

Sure. I'll start off with -- I think 2023 started off for us on a pretty positive outlook. So -- and this was before the AI, call it, wave craze that's been more recently a buzz and discuss. So even before that, we talked about coming into the year, we saw pricing power. We're starting to come back into, we'll call it, the landlord, the owner's favor. And we demonstrated that in terms of our guidance where we guided towards positive re-leasing spreads across both our segments being the 0-1 megawatt as well as the greater than 1 megawatt. This also led us to positive same-store growth guidance, which we had not had for several years. And that was all -- that was brought on by sort of, we'll call it, overall positive fundamentals that have only, I think, strengthened, meaning we've seen tremendous amount of demand, again, even before AI, we saw supply that was starting to become constrained in a number of markets, which I think has also become more challenging from both the power and the general equipments, broader supply chain dynamic. And then we added on top of that now more recently, which has been -- in terms of the demand equation, which has been a multiplier effect, if you will, as a result of the type of AI demand that we're starting to see from -- largely from our larger cloud-related customers. So this has been, I think, a great year to be in the data center business from positive pricing power, solid growing demand, and we think we have the global platform to be able to compete effectively and continue to see, I think, what we're expecting is improved long-term growth going forward.

So the occupancy in broad terms, you expect that to kind of increase given the elevated demand? Or are you able to keep up in terms of your development pipeline? Or are you strategically keeping some inventory available for productized colo? How do we sort of think about the occupancy metric going forward?

Yes. So we -- I kind of address it in sort of 2 parts. I think if we look at our in-place operating portfolio today, we're near the mid-80s in terms of occupancy, I think we would like to continue to press and improve that from -- to lease that vacancy, especially considering the dynamics of the markets that we just talked about. Part of where that becomes slightly challenging in today's market is the demand that we're seeing is typically for larger contiguous blocks of space. So finding those pockets of inventory -- available inventory within our portfolio that meet the customer demand requirements that are in place. But we're working towards finding creative ways to free up some of that available inventory and continue to press from an occupancy standpoint in our operating portfolio. And then from the development side, I mean, we're developing 375 megawatts currently. It's largely -- it's a little -- it's around 50% preleased. And typically, by the time we deliver, we're sort of in the 80% to 90%, if not 100% pre-leased. So I would expect that sort of dynamic to continue around where our development portfolio is.

I want to maybe ask about kind of the piece parts around core FFO per share and how to think about future periods, because there's a lot to unpack around top line, around EBITDA, things below the line as well. A lot of it, you may not even have perfect visibility into, but what are kind of the puts and takes around the trajectory for core FFO per share?

Yes. So I mean, we're doing -- we have done, we're doing a number of things this year that are, I'll say, limiting our bottom line growth. And I think we've been pretty transparent about we're -- we've made a lot of progress this year in terms of deleveraging. We started the year at 7x. We've now brought that down closer to 6.3x on a pro forma basis. We've done that through a mix of executing, as we said, as we started out the year on development -- stabilized joint ventures of core assets where we've sold the north of $2 billion to date in that area, which is about close to 3x what we set out and stated to when we provided guidance. So we've done the investment team, in particular at Digital have done an excellent job in executing on that front. We raised additional equity capital earlier in this year. And the kind of the third leg of the stool that we are in progress on is looking to put in place development joint ventures given the growing opportunities that we see in the market, given AI and just a broader set of demand that we see. We think it's prudent from, call it, a risk management perspective to partner with additional capital providers and sort of, call it, contain the amount of capital spend and development that we do on balance sheet going forward. So all those things are put in the mixing bowl are offsetting what is -- what has been as we started the conversation, positive fundamentals in terms of pricing and same-store stabilized growth. And we expect those positive fundamentals to continue into next year, given we still see a favorable pricing environment, but at the same time, we're still not where we want to be long term in terms of our leverage. So we've talked about getting down to our long-term target of 5.5x. As we said, we've made considerable progress this year, but there will be more to do next year. But we think overall, we've laid the foundation towards getting back to what should be long-term stabilized growth and improving growth next year and hopefully even more so years following.

So anything around things you didn't mention. So a lot of that was sort of balance sheet related and financings, and that obviously has an impact on FFO per share growth. You talked about pricing, anything else to kind of throw into the mix? I don't know whether it could be things like property taxes or renewal spreads for next year. But some of the key things to maybe focus on or look out for?

Yes. I mean, usually, it's a lot of things that are, generally speaking, or, I would say, outside of our control. So things like property taxes over the last couple of years, 2 other items have been outside of our control that have had an impact our exchange rates, right? We're a global company. We're -- over 40% of our business comes internationally, so we're subject to exchange rate. There's interest rate risk. I think the theme of hired for longer is here to stay. I think -- I don't think we're expecting it to have the same level of increases we've seen over the last year, but that's something we have to consider as we look towards '24 and beyond. And then maybe last, but not necessarily least is power prices as well, although those, again, have been a lot more -- a lot less volatile than they were in, call it, at the end of -- most of '22, especially towards the end, but those are some things that we're still looking at, again, given the global portfolio that we have and some of the volatility we see in different markets that we operate.

I want to touch on development JVs, but before doing that, kind of the things that you did accomplish earlier this year around noncore asset sales and stabilized JVs, should we look forward to seeing similar targets for years to come as you continue to kind of recycle assets? Or are you going to focus primarily on other sources of financing and deleverging?

Yes. I mean -- so I think that's going to continue to be a part of the, we'll say, the financing playbook for Digital. Look, we mentioned the one that we're in motion on that we haven't executed is the development joint venture. So that's going to lay some foundation for not only capital that will come into the door this year, but also extend into years beyond depending ultimately on what's executed, but that should serve to reduce the overall capital needs and capital spend for Digital directly. And then I think on top of that, we're going to continue to see growth in EBITDA from the organic growth in not only our portfolio, but the developments that are coming online and related yields that have also been continue to ratchet higher as we focused on making sure that we're deploying capital in the areas in the markets where we see the best return. But also as a result of the, again, what I think we see is a long-term favorable pricing dynamic within most of our core markets that those 2 things will help continue to provide some level of funding as we're able to then get back to our leverage levels and become more of a mix of debt as well as joint venture development to fund our business going forward.

Yes. I would sort of layer on that, John, that in the beginning of the year, we -- 1 of the 3 priorities, Andy laid out for folks was the bolstering and diversifying of our balance sheet and capital sources. And this is part and parcel with that priority for Digital Realty. So we're going to continue to recycle capital. We're going to continue to avail ourselves of multiple avenues of capital to raise capital, to sort of capitalize on the opportunity that we're seeing, without having to be overly reliant on any individual source of capital.

For development JVs, what flavor or flavors does that take around disparate partners depending on geography or metro or somebody, who you're going to be partnering with kind of across geographies? Anything that we should be focused on there?

I mean, we obviously haven't announced anything. But I think we're -- in general, we're looking -- we should expect to have -- not to be, call it, a single asset, right, development. Ideally, we have -- there'd be a initial deployment and project oriented towards the development JV, but has the option or ability to expand and have other assets, part of that again is part of creating and partnering with a capital provider that can offload some of the capital spend given the demand that we're seeing, again, that's growing across our global markets. So I expect ultimately it should be not a single project ideally in multiple markets with some ability to grow and expand over time.

John, just to frame it. You recall and just for the benefit of the audience, maybe, we framed it originally as a hyperscale development JV, so it will be a hyperscale portfolio generally of assets and similar to the hyperscale, stabilized JV that we did. So -- and the sizing of that initially was $750 million. So that's still our expectation.

For this current year?

Yes, from the development change.

Got it. And then with a stronger currency and you did use the ATM earlier this year, but thinking of your currency as a delevering tool or perhaps as a tool to aid and inorganic growth, what are some thoughts around that?

I mean, yes, so I mean we did raise some equity capital earlier this year. But looking at -- and then we've executed as we noted on stabilized JVs brought in $2 billion at the -- as we talked about on the last quarter's call, we -- at that time, we're -- obviously, the business is continuing to run and we're continuing to develop assets, but we had close to $4 billion of liquidity at that time. So in terms of where we stand today, we feel very good about our liquidity position. You then add to that the fact that we're still executing on the development joint venture, which we'll expect to bring in additional capital as well as reduce our long-term capital needs. We feel today, we're in a good place from being able to fund our business going forward with some more of the transactional activity that we've accomplished as well as what we expect to accomplish at the end of this year and then going into '24.

What are you seeing in terms of larger requirements, whether it's AI related or not? Any kind of a regional contrast that you could draw in terms of what inning we're in, in North America versus or Americas versus APAC versus EMEA for the industry and then maybe for yourselves as well?

From an AI perspective?

Yes. Yes.

So I would say the majority of the demand we're seeing is North America oriented. That said, we are seeing demand within other regions as well, although we expect that North America will get -- you'll see the majority of the demand initially in North America and then, I think, somewhat quickly thereafter, you'll see it in EMEA and APAC. Similar -- I mean, we see the along the similar lines of how the cloud ended up rolling up to where North America was -- largely had the initial wave of cloud demand that's now been a big driver of our recent signings in the last year in particular, where we had record signings going into EMEA was heavy cloud-oriented as well as APAC. So considering that you've got a lot of the same players on the cloud side that are driving a lot of the AI demand. We're somewhat seeing us, I'll say, generally similar playbook. So call it, 50% to 60% of what at least what we've been in discussions about are North America oriented. And then I think as you -- and this is largely on the training side of the equation in terms of AI demand. But we are, as a result of either we'll call it, language or other data sovereignty restrictions, we do expect to see that within a fairly short start to deploy within some -- in EMEA and other more country-oriented markets, especially as it starts to get into more unique or specific use cases around enterprise and AI-oriented training that needs to be within more data sovereign type structure.

I want to open it up to questions. So if there are any from the audience, feel free to raise your hands at any time. I want to maybe hit on just a couple of things. So in India, you brought in a third partner. Talk a little bit about what's going on there given the size of the market and opportunity?

Yes. So we -- I mean, we see India as a -- obviously, it's a high-growth market for us. Originally, we had a partnership with Brookfield to similar to our joint venture that we have with them in South America had a long operating history of being in India. And in addition to that, we then saw an opportunity to bring in a third partner that really, I think, can help bring in additional -- help us not only from a local perspective, they're are obviously Jio's, a major player within the India market. So we see them being able not only to be a capital provider, but also really help us from generating incremental demand within the local market back to kind of our overall strategy, which is not only Digital, but can be applied to our India venture, which is being able to be to satisfy some of the larger workloads, cloud, hyperscale requirements, but also establishing a really strong enterprise connectivity oriented business, which we think Jio can have -- has capabilities in experience and knowledge and can really bring that to the mix and help us accelerate that side as well.

So you quantified your current development pipeline. So it's a large number. The velocity around which that gets built commissions, starts generating revenues. So a lot of unknowables, I suppose, around supply chain, given the larger size of the requirement that the whole industry has seen is supply chain, something that could become a constraint energy, if you want to throw it and just maybe related to that energy in more geographies, whether it's transmission or energy generation, permitting, but that seems to be a more regular issue that we're seeing globally. So any reactions to that?

I just want to make sure, I'm clear on the question. For Matt to answer, which is regarding the existing development pipeline in terms of supply chain and equipment. Rest assured recovered. So are you asking beyond?

Covered with longer lead -- are long lead time items getting longer or...?

We've procured that equipment.

Yes. I'll kind of answer it in -- again in 2 ways. So to Jordan's point, our -- we have, call it, roughly 375 megawatts under development across our global portfolio today. We have the power. We have all the major mechanical electrical equipment, everything we need to be able to deliver that capacity for our development schedule. Going forward the next, call it, the next wave of megawatts, that ultimately depends on the market. So from -- what's happening now is, which I'm sure most of this group is aware of what used to be, I don't want to call it easy, but what used to be a lot easier in terms of power and supply chain on equipment has become a material challenge. You've got to really plan ahead for both of those. You have to be in the markets participating with power providers. You have to be partnering with your major equipment vendors well in advance, be part of the queue that they've established in order to bring that equipment to market on time. And that's in markets on the power side, where there isn't an essence of power constraint. So there's a number of markets now in the U.S. and really globally where power is constrained. I think most people are aware, Northern Virginias, in essence, tapped out for the next couple of years. Santa Clara, it's going to be not until 2030, if not beyond. Even the market here in Chicago is becoming more supply constrained. There's a number of markets in EMEA, Singapore's harder to bring power on. So there's a number of markets that even you can't bring on power even if you wanted to build. So obviously, dealing with that. The good part is we've got one of the largest portfolios in some of these power-constrained markets that we have an operating portfolio that we think will give us an advantage within that context plus in some of these markets, we still do have developable capacity at least over the next couple of years. So we have in Northern Virginia, in particular, we had 75 megawatts underway, and we have line of sight to bring on another 100 over the next 2 years. So we think that gives us a good runway in terms of our ability to, call it, weather some of this power-constraint storms that are hitting us in several markets. And obviously, we're not the only ones that are being constrained by the power equation.

Question from the audience? So the question is pricing power in markets where there are power constraints versus rest of your portfolio?

I mean, I would say, broadly speaking, because of the strong demand we're seeing across the majority of our major global markets, even those that don't have necessarily the [indiscernible] level of power constraint. we're seeing pricing -- and improved pricing dynamics, especially over the last, call it, year to 18 months. The one -- the market that most people can appreciate because one, it's in essence, the largest market in the world, Northern Virginia, where we've -- I think we've seen in the last 12 to 18 months, we've seen pricing go from, call it, the 70s and 80s of KW to close to [ 140 ]. So that gives you kind of a sense of the type of pricing dynamic and swing that's happened within one of the major global markets that we operate.

Would there be consideration to do more stabilized asset JVs to bring down leverage? I think you answered that earlier, yes.

Yes. Yes. We are -- we're not -- we've done -- this year was not the first year we've done joint ventures, right, from a stabilized perspective, we've done. We've had partnerships with a number of different partners in different markets. So I think we see that as a tool, we'll call it in the financing toolkit that we expect to be available not just this year, but going forward. And we do -- I mean there's appetite, obviously, for this product, this industry, this space, and we don't expect that to change.

[indiscernible].

I mean, I want to get down to 5.5x. We've talked about in doing that -- getting there by sometime in 2024 is our ideal setup.

Can you just talk a little about in an environment that's competitive, just second-gen space. Also, obviously, in spaces are [ acquiring ], [indiscernible] but just kind of wondering what you're seeing on renewal, is there just as much demand for used space that [indiscernible]?

So second -- question on second generation space, expectations around renewal spreads given higher power density requirements.

Sure. So demand is strong across the board. So looking at breaking down the 2 segments separately, perhaps, so our 0-1 versus greater than 1 megawatt, right? So while I think there's been a lot of excitement within the market around the larger capacity blocks and around AI and some larger leasing that's been happening and were rumored in the market, bringing it back down to sort of our 2 core products. We've seen steady demand on the enterprise side in the 0-1 segment. We saw a little bit of a pullback, particularly in timing about a year ago right now when we saw the capital markets volatility we saw. That -- the time line or lead times in between sort of initial dalliance with the customer and execution of closing, those lead times have -- they widened out a bit, and they have kind of remained somewhat consistent. They have not widened further. But they remain reasonably wide. And we think that financial scrutiny around deals remains something that's significant. But the demand of the pipeline in the 0-1 segment remains very strong. And we think a lot of that will lead to and will drive some of the lease-up of some of the vacancy we have. The greater than 1 megawatt segment, I think, as a result of some of the leasing we're seeing in the market, but also less power availability, which is a larger theme, there certainly is an expectation that we'll continue to be able to drive up the occupancy or utilization across that aspect of the portfolio over time as well.

Yes. I mean the only thing I'd add is we've seen churn declining over the last, call it, year to several quarters, and part of that is related to this dynamic of the industry that we're sitting in where there are supply constraints and there's demand that's growing. So the ability to move, it was already hard to move, that has become even harder. So there's some level of like -- there's some level where people are -- as a result, churn is lower, and we kind of expect that to continue as well.

So the question was does it make sense to redevelop older assets as densities increase and/or push rate higher?

Well, I mean, we're seeing -- I would say we're seeing some level of that now. I don't necessarily like to turn -- I think our assets like have -- in my view, have stood the test of time. So we developed -- we have developed our assets in a flexible way that allows us to be able to deploy different types of densities within a shell. And that kind of goes back to somewhat of our strategy of we're looking to be able to satisfy a broader spectrum of product, meaning within the 4 walls, we can satisfy 0-1 megawatts like we typically have different type of density levels as well as up to cloud and hyperscale, which has, generally speaking, has a higher density level. And we're seeing now requirements that are in the 40, 50 and even higher in terms of the kW [ rack ] that we're able to satisfy within our existing footprint with little to no incremental redevelopment. That's not to say that, that can happen in every single one of our facilities today. But I think we're -- to answer your question in maybe a shorter way, if -- if we see an opportunity to redevelop and we think we will be able to achieve a risk-adjusted return that's adequate. That is definitely something that we'll look into.

Before you jump ahead, [ Chris ], I just want to layer on the -- come back to Alex's question earlier on retrofitting a little bit, which is some of the older assets, 2 things about them. One, the scale or the megawatts of capacity in a 10- or 15-year-old asset, right, what was built that many years ago is going to be significantly smaller than a current generation asset, right? So maybe a sort of first-generation hyperscale asset was built at 36 meg probably about as big as they were built. Today, we build 96-megawatt building. So that's one thing in terms of -- and those 36-megawatt buildings may not be right for redevelopment yet or necessary. So you might be looking at old -- smaller first-generation type assets that are 9 or 18 megawatts. So how much redevelopment do you necessarily need to do is a question. And then speaking to what actually has to be done in those facilities. I don't know if you were going in this direction, but some of those older facilities were designed with chillers and cooling towers, so plumbing in place and plumbing to the floor already, which makes them, in some ways, more conducive to bringing water to the floor today than sort of waterless cooled facilities like [ DSEs ]. So they can be more conducive to redevelopment, those smaller assets, but they don't necessarily have the scale to accommodate larger workloads, if you will.

So 2 questions on this side further back. Yes, go for it.

[indiscernible]

So I think there is some -- like performance clauses around delivery and how do you deal with that? And do you mind rephrasing that little?

[indiscernible]

So back to -- the demand in the market is significant, right? And I think to your point, the contracts that we have, I mean, they -- most of our -- especially the larger customers, they're looking for space in the proverbial way yesterday, right? So -- and look, I think we've -- the way we manage that is we have a long history of development. We've been developing 300 megawatts or more a year over the last 5-plus years, right? So -- and we have the supply chain, which we kind of talked about earlier in terms of making sure that we're -- we have all the equipment lined up, ordered and ready to deliver according to those, the timetables that were set within the contract. We have great relationships with our GCs. In most markets, right, we're sort of perpetually going, not in every market, but in some of our largest markets, we continue to develop that keeps people on site there. Now there's always room for potential delays. But I think based on just our history, experience dealing with our vendors and having relationships with them as well as our GCs really helps us minimize those running into any sort of penalties, but also try to maximize our delivery and making our -- ultimately our customers happy, which is the end game.

We are out of time, but I appreciate the comments, and we'll make that to the next panel.

Thank you.

Thanks for having us.