American Tower Corporation (AMT) Earnings Call Transcript & Summary

All right. Good morning, everyone. Hope you're doing well. And first, I hope, actually, you and your families are doing okay with the COVID situation. Welcome to the 42nd Annual Raymond James Institutional Investor Conference. This is actually my 25th conference, 100th earnings season, I think. So not my first rodeo. This is the first virtual Raymond James Institutional Investor Conference. We missed not getting together in person. Certainly, we miss not being in Orlando, Florida, for those of you in the colder areas of the country. So hopefully, in 2022, we're back in person in Orlando and that we can get back to normal. We will be taking questions via the chat and raise your hand, question mark, you can also shoot me an e-mail. We're really -- Igor Khislavsky is with us today from American Tower to talk about American Tower. Igor, welcome.

Thanks, Ric.

Let's kick it off with -- obviously, you guys reported earnings last week. It was a very detailed earnings call. That was a very good earnings call to lay out a lot of the uncertainty of what's happening there. But I think probably the top questions certainly around the market are -- medium and long-term growth rates. What gets us back into that mid-single digit? Why can't we get back into the high single digit? And just kind of the puts and takes on the U.S. side first?

Yes, sure. So I think, like you said, what we tried to do last week was to really lay out the long-term trajectory of the business, given the predictable nature of the business, given our long contracts, and so those characteristics enable us to have an extremely long-term view into our baseline growth trajectory. And that's what we laid out. And essentially, that trajectory is split out into 2 main pieces. The first is the next couple of years in the U.S., so 2021 and 2022. And then the second piece of that is 2023 through 2027. And so 2021, and particularly 2022, is going to be a period where we have the most heavy weighting of Sprint churn, which we've talked about at length for a while now. And as part of our T-Mobile MLA that we signed last year. And so during that period, on a reported basis, we think we're going to grow at an average of 2% per year in the U.S. organically. In ex spread churn, we're going to grow at about 5%. And so on a normalized basis, very solid growth, that growth is going to be backed by a lot of the same things that you've seen over the course of a number of years driving our business, network densification, the deployment of new spectrum. We obviously had the C-band auction just wrap up here recently. The deployment of 5G is well underway, and I think will accelerate over the next couple of years. And so on a gross demand basis, next couple of years look really solid, right? And we think the gross activity levels will be ramping through 2021 and particularly into 2022. The offsets to that, of course, is going to be some near-term churn from Sprint. And a lot of that comes from the fact that unlike our peers, we've never really had any historical heightened churn of significance. And so some of that is going to flow through that. Once you get through 2022, that growth trajectory actually steps off quite a bit, right? Starting in '23 and then ramping through into '24, '25, '26 and '27. And on a reported basis, we think we can hit at least 5% growth organically during that period. On a normalized basis, around 6%. And so that, by the way, includes roughly 2/3 of the total being contractually committed today. And so that's where kind of that visibility that I referenced earlier really comes in. We know today, that 2/3 of that growth path is contractual, right? It's going to happen no matter what. Now we have to go out and find the other 1/3 of that piece. And then to your point, I think one of the main thrusts of putting out this guidance on a multiyear basis is it's a way for people to understand the baseline trajectory that we're expecting today. It doesn't incorporate material contributions necessarily from things like edge compute or other platform expansion initiatives. It doesn't contemplate anything on the new entrant side beyond DISH, right? I'm not saying that you're going to see new entrants. But if you do, I think that could be some additional offside. And so I think the goal internally for us is this is a realistic number, a set of numbers, rather, for us to hit what we're working today, frankly, on initiatives internally that hopefully will enable us to exceed those numbers over time.

Right. And when you think of DISH, how much DISH is in your estimates and over what time frame?

Sure. So for 2021, DISH really is not in our guidance at all. I think going forward, starting in 2022, there's some DISH in there. It's -- I think DISH publicly has been quite clear about their intention is to have a progressive build-out, it will take time. It won't be done on day 1 on a nationwide basis. So as folks of impactive DISH or American Tower specifically, I think what you're likely to see is activity build on itself over the course of a number of years. So out of the gate, it's unlikely to be material. But a couple of years out, I think it can certainly be noticeable, and we would expect it to be noticeable. We have a 40,000 plus site portfolio that for a nationwide build-out will need to be utilized.

Yes. I think Tom was pretty emphatic and clear about do you think you'll get your fair share, because there's been a debate in the marketplace given the staging of MLAs with Crown, Vertical Bridge, numerous small guys within that spec, but no MLA yet with you in DISH. So you do feel you'll get your fair share over the couple of year period?

Absolutely. I think we're in a position where we can afford to be patient and we're certainly open to a comprehensive MLA agreement with DISH as we've signed with a number of other tenants. At the same time, we're more than open to just going pay by the drink as well. Both of those structures have worked for us historically, and I think both can work in the future.

Yes. I'm not going to put you in an awkward position talking about your customers because clearly DISH has given some insight, but not a ton of insight. But is it maybe fair to say that if DISH were to get an anchor tenant capital partner or some kind of -- they talk about [indiscernible] to sign a cloud provider, that it could be a bigger build than you maybe have assumed, depending on what happens with DISH?

Yes. Certainly. I think there's upside to the projections that we have internally, potentially depending on how that build evolves over time. So I think that's fair on a broad level.

Okay. With the C-band, how should we think about how you've reflected that into that multiyear guidance of '21, '22 versus 23 or 27,

Sure. I think it's always tough to explicitly carve out spectrum contributions from a specific spectrum asset into a long-term view of growth. It's certainly embedded in the numbers. I think on the one hand, we have some comprehensive MLAs that are long-term in nature or multi-year last -- in some cases, throughout that entire period. And so some of that C-band activity can be encapsulated within the lease agreements and the terms that are in them. And then in other cases, you're going to have the opportunity to either go pay by drink, we have C-band or take a look at putting into place new master lease agreements that incorporate C-band deployments explicitly. And so the way that I would look at it is, I think, given we just issued the guide last week, and given the C-band was just wrapping up at that time, we've layered in some assumptions for the impacts of those deployments. There may be upside to that eventually. It's tough to tell upfront. But it's certainly not going to be all incremental by any means to what we've laid out. I think we've sort of taken our best guess at what that baseline level of activity will be incorporating our assumptions for C-band.

Makes sense. And the carriers themselves have to be very careful with the quiet period. Our understanding is they haven't even been communicating much recently with the tower guys just because there is nervousness, although there had probably been communications prior to the quiet period.

Right. Yes. I mean, I wouldn't want to get into any specific conversations. But I think as a general rule, it does take time for Spectrum to eventually be deployed. I think there's certain sense of urgency that's been discussed publicly with the C-band specifically in terms of its importance for 5G. I think that's a good thing for us. We do continue to think that the majority of C-band spectrum will be deployed on macro towers. If you would have said that 3 or 4 years ago, you've gotten some weird looks in the room. I think today, it's a fairly well accepted conclusion. And so we do think C-band is clearly a great catalyst for the industry. No question.

And more spectrum auctions to come as the FCC continues the pipeline?

Right.

You mentioned that there's no material edge in the growth rate in the guidance. First, can you help people understand what the heck is the edge and what do you think it means to American Tower? Help us kind of size the opportunity?

Sure. Yes. So I think the general concept is, as you progress with the 5G ecosystem developing, and which at this point is still in it's very early stages, I would say. You're going to need lower latency applications, you're going to need tremendous processing power throughout the network. And a lot of that sort of application power and that application strategy, I think, is going to evolve towards the edge to complement what has historically been at a large data center but maybe a couple of hundred miles away, right, because of that latency need. And so whereas with 3G and 4G, if you could get away with latency that wasn't that small. With 5G for applications like AR, VR, if there's autonomous driving involved, you really can't have any latency within that network. And the way to deliver that is to have some of that processing power that historically has been centralized moving towards the edge of the network. And so what we're trying to do at this point, and we have some trials out there, we've had a lot of conversations with numerous stakeholders in the value chain, is figure out how to make our tower sites a preferred location or what we think will be these micro edge data center facilities. And I think the -- at a high level, you're going to have a number of different potential locations here. You're going to have some of these edge facilities at central offices for the carriers. I think there have been a couple of those, trials already ongoing. You're going to have some other venues that may work. And I think at select tower sites, all the precursors are there to house this type of a solution. You've got ready access to power, you've got a fiber connection, you've got the communications operators that, chances are, they're already at the site. And so what we're trying to figure out how to do is to optimize a solution that brings together multi-cloud, multi operator into one location. And much like the tower industry has evolved over time, where it became clear that a neutral host was the most efficient way to deploy a terrestrial wireless network. Our thesis is that the neutral host that brings together multiple cloud players and multiple carriers in 1 location is going to be the most efficient way to deploy the edge over the long term. And so what we've been doing in our platform expansion initiatives internally is trying to figure out how to create that solution, right? And I think to this point, some of the challenge has been 5G is just very early. And so the demand case for these edge deployments is at the very, very beginning stage. So in terms of TAM, it's tough to explicitly quantify right now. I think given that we are continuing to talk about it, it's safe to assume that our view internally is that this can be something material over time. Is it going to be on all 40,000 plus of our towers in the U.S.? I don't think so. It's going to be a subset of those, probably sites that are more closer to the population, things like that. I think the other very interesting aspect that perhaps sets us apart a little bit in this regard from other folks in the industry is our global presence. And so the edge is not going to only be a U.S. phenomenon. You're going to see the edge deployed across the world eventually. I think in developed markets first, just like any other technology. And so you look at a place like Germany, as, an example. There's going to be an edge solution in Germany, right? Through our pending Telxius transaction, we're going to have a significantly expanded presence there. And so that's another market where you kind of take a look and see the long-term opportunity. So to circle back to your original question, we're excited about it. I think in many ways, it's still tough to explicitly quantify right now. It's not really baked in to any significant extent in our long-term numbers. And so there, again, I think as you kind of look at the upside case to that baseline growth trajectory, something like the edge probably plays a meaningful role in that upside case.

Okay. We get a lot of questions on the technology side because, obviously, we have real estate infrastructure generalists that are also in the tower investment, not just technology and telecom investors. Can you talk a little bit about all the acronyms that we've seen, O-RAN, vRAN, DISH talking about a cloud-native network. But also, we get carrier aggregation and massive MIMO, so we get a lot of alphabet soup out there. Help us unpack that for some of the generalists in the audience, what it means for towers?

Sure. Yes. So I could probably spend a good half hour on this question, but I'll try to...

You get 2 to 3 minutes, max.

No, it's a good question. I think over the years, there have been a lot of acronyms, as you know, that have filtered through this industry and a lot of interesting products and things that have been introduced or folks have had different takes. I'm thinking back to the -- like radio days back in the early 2010s, as an example. But at the end of the day, when you look at something like O-RAN, what the carriers are trying to do, and in this case, DISH specifically Rakuten has done this in Japan, is migrate to a software-heavy, hardware-light model on the network side. And I think the other distinction that is really important to make within that equation is there are 2 pieces to the network. There's a piece of a network that's sort of the digital side, the processing power. Historically, those have been in base stations that are located on the ground at the tower site. And then the other piece is the actual equipment that propagates signal, which ends up on our tower sites, and that's where we derive the economics for our model. And so when you talk about O-RAN and vRAN and those things, what you're really talking about is digitalizing and putting the processing power events on the base station in the cloud. And so for us, the impact really is pretty minimal. It doesn't change the equipment that ends up on the site, at least not in a negative sense. We've actually seen as part of some of the C-RAN deployments in the past, we've ended up with more remote radio head equipment on the site itself as a result of these deployments. And so in some ways, it's actually been a net positive. But there's going to be, I think, less in the way of base stations, physical base stations in future networks. The way that our leases work, of course, is we don't really charge for that space per se at the ground. It's included as part of a lease, you're not going to pay any less if you don't use the space. And so that's a very critical distinction that I think folks need to understand. But at the end of the day, what drives our business is the physics of signal propagation. And so things like O-RAN and vRAN and all of these other acronyms really don't change that side of the equation. When you talk about carrier aggregation, massive MIMO, these are other methodologies where the carriers are trying to optimize their sort of network design, right? And so there, again, there are wrinkles and permutations. But what we've seen as a result of something like massive MIMO, for example, is antennas get bigger. And as antennas get bigger and heavier that end up on our tower sites, we actually get paid more. And so there again, I would look at that and characterize that as essentially neutral to net positive to our model. I think network design will continue to evolve with a lot of these elements. But net-net, we look at that evolution and unless somebody figures out how to actually change the physics signal propagation, which, to this point, nobody has figured it out. We feel really good about our position as sort of the base of these networks. As they evolve, as they look a little bit different, depending on which element you look at. But at the end of the day, when we've tracked this over time, the trend has always been, and we expect it to continue to be that there is more and more equipment that ends up on our sites as mobile data usage growth throughout these networks continues to accelerate. And at this point, there's nothing to suggest that that's going to change. We, as you would expect, we have kind of quarterly sessions where we bring in an external tech consultant, we bring in our internal tech folks, and we literally just pepper them with questions about, okay, what's this new technology? What does it mean? What can it do? What can't it do? What are the impacts? And so we're continuing to look at it. But I think at the end of the day, when you look at the overall equation, what ends up on our sites is still very similar to what has ended up on our sites over time and continues to grow in size and weight overall with network usage.

So we always say it's the Moore's law makes things smaller, but the laws of physics are the laws of physics. And that's really where the tower and it's like -- it's physics radio propagation, radios, antennas, et cetera. And smaller base station equipment or moving base station equipment uplifts more ground, which opens up the edge opportunity. Speaking of edge opportunity, we have a question coming in. With the Congress talking about an infrastructure bill, how does that effect into American Tower and other tower companies and does that have any aspect to the edge as well?

It may. I think it's too early to tell. On the infrastructure side, look, clearly, there is a need for -- and an opportunity for incremental development of telecommunications infrastructure in this country. I think, particularly in rural areas, but there is still a digital divide. And so we have participated in trying to help bridge that. We have a number of rural WISP customers on our sites throughout the country. And so I think that is clearly going to be an opportunity over the long term. I think our business certainly doesn't rely on incremental infrastructure proposals to be successful. But to the extent that there is funding available over time, to the extent that there are infrastructure programs that are implemented that involve telecom infrastructure specifically, which I think is a critical piece of the economic equation, particularly in today's world, we -- the industry may very well benefit from that. Then, again, I think you sort of have to look at the fact that infrastructure bills have been proposed for quite some time. And to this point, we haven't quite seen make it to the finish line. But look, I think to the extent that we can participate in that, to the extent that the industry can help build out incremental infrastructure on the digital side, we're all for it. We're ready to go. We have the relationships in place, and we're more than willing to work with everyone to make better reality.

Right. And clearly, COVID-19 has emphasized the digital divide, homework divide. So the ability for telecom to play a role in the infrastructure development, it's as important, if not more of the roads, if we're not traveling as much. You guys have gotten -- you're about to get even bigger in Europe. We get a lot of questions, because the Europe market really is different than the U.S. market. Can you talk a little bit about revenue sharing aspects in Europe, network sharing aspects in Europe and where we're at, in particular, in Germany with Drillisch.

Sure. It's a good question. And from our perspective, Europe, in the context of telecom infrastructure, is almost a misnomer. You have to look at each and every market individually because throughout the continent, you can have market X that has completely different characteristics from market Y. And that candidly is part of the reason why before the Telxius deal that is pending. We've passed on a lot of portfolios throughout Europe. We've participated in other processes and simply not gotten anywhere close to the valuation where those assets eventually traded. A lot of the reason for that is not that we don't think Europe as a whole is attractive, or was not attractive, rather. But because the specific market characteristics as well as a specific assets in question weren't sufficiently attractive from the perspectives of some of the things that you just mentioned, Ric. In some European markets, there's a lot of network sharing that goes on, and there's been a long history of active network sharing. That obviously makes a third-party tower model less attractive. And so you have to incorporate them. In other markets, you had a situation where you didn't have too little infrastructure, you had too much infrastructure. And so the network was oversaturated, and the opportunity was less an organic growth opportunity and more of a decommissioning type opportunity, which for us was inherently a little bit less attractive. And so if you think about the Telxius deal in and of itself, I think there are a couple of really key characteristics that you have to hone in on as to why we went ahead and did the deal or in a process of doing the deal. Number one, Germany, we think, is the best market in Europe from a telecom infrastructure perspective. You've got a very strong carrier base. You have, as you alluded to, and what looks like a new entrant coming into the market with Drillisch, you have a very solid asset base that we're picking up with an excellent anchor tenant and a counterparty. And so -- and you're getting 5G built out in the country now. So from a demand perspective, Germany looks really good. We've operated in Germany for, I guess, 8 years now with a much smaller portfolio. And so we've seen the growth in Germany. It has been tempered as of late by some consolidation churn in the marketplace. So as a whole, our German plus French business this year, as an example, is going to grow at 3% organically. That 3% includes about 2.5% of churn. And so if you look at the Telxius assets, on the other hand, our expectations is that churn will be well under 1%. And so what we've talked about from that perspective is we expect our Telxius assets to grow at 5.5%, 6% long term. You basically get there just through the churn differential between Telxius and our existing business. And so to the extent that you have someone like Drillisch come in and really get aggressive on the build side, which I think just recently, they've referenced their intention to build out 12,000 sites in the near term. And so we're optimistic on that. You can potentially grow at a better rate than that. And so the organic growth opportunity associated with these Telxius assets is not one that we felt was replicated in really any other portfolio that we looked at. So that is really the kind of foundational key point as to why that was so attractive for us. And we've known Telefónica for 20-plus years, right? So we know the counterparty well. We know an anchor tenant well. There's a long-term noncancelable lease in place on these sites. But we're really excited about being able to lease up this asset in Europe, where you have great stability, you have a very solid backdrop overall. So we expect it to be quite additive to our portfolio in a way that, frankly, for other assets that we've looked at in the region, we've really never felt.

Rooftops will also be a large part of this transaction. Talk to us about getting into a larger stake in rooftops when you really haven't in the past?

Yes. It's a good question. Rooftops, I think you have to really make the distinction between a rooftop model in a place like the U.S., for example, where, by and large, we're not owners of rooftops. We don't have owner economics. We simply take what is essentially kind of a management fee or leasing those rooftops up to wireless tenants. In Europe, the structure is more of a owner economic structure, right? So we're paying the building owner a rooftop lease, and we are collecting effectively owner economic type of returns from leasing activity. Now there are select rooftops, even within this portfolio, where you're going to have some more restrictive terms, either from a capacity perspective or from a revenue share perspective, where the landlord takes an outsized cut of incremental lease-up. That is -- are from ubiquitous, of course, in these assets. And that was, as you would imagine, one of the key things that we looked at when looking at this portfolio is, okay, you've got a significant contingent of these sites about rooftop assets. We know that in some European markets, rooftop lease terms are not the greatest. And so let's make sure that we get very, very comfortable with where we're at from a rooftop leasing terms perspective with this portfolio. And we got comfortable. And so I think in that regard, the flip side to all of that is when you have a new entrant, when you're starting to build out 5G in a market like Germany where so much of a population, unlike in the U.S., is concentrated in urban and dense urban areas. Most of those network builds for the foreseeable future are going to be really concentrated in those urban and dense urban areas. That's where these rooftops are located. And so that was a component of our German portfolio that previously was missing. Today, pre-Telxius, essentially our entire German portfolio is rural and suburban. We don't have any urban presence to speak of. And so this deal fills in that part of the network for us. And I think it's going to be especially relevant over the next couple of years when you have not only Drillisch presumably entering the market and obviously looking to get their biggest bang for the buck in urban and dense urban areas, but also the other carriers really looking to deploy 5G in those same locations. And so we feel really good about those rooftops as actually positive differentiators and I would stress that the rooftop model in a place like Germany is inherently quite different from what you see in the U.S.

And Drillisch, for those on calls, is already an operator, but they're a virtual operator. They're moving into the marketplace as a network operator as well. So they're in the commercial side, but this would get them into the tower side of the business and the rooftop side of the business. I want to jump over to India. Clearly, India has not been fine. It's going through carrier consolidation. I think you guys knew it would come, but it happened fast especially in this line. It's been disruptive. You've had the AGR tax issue. Help people understand your positive thoughts on India and over what time frame does it get back to like earning the return on capital that we need?

Yes. It's a good question.

That's 3 in a row, good questions.

You're full of them, as usual.

I'm just full of it.

No comment on that one. But look, I think India has been a market that, to your point, has been challenging for us over the last couple of years. We've actually been in India since 2007. And so up until about 2017, 2018, India was probably our fastest growth market on an organic basis. Even with a very inefficient market structure and some other challenges there, we were growing organically at regularly, 8%, 9%, 10%, 11% per year. That has obviously not happened the last couple of years. And the catalyst for that, as you alluded to, was a wholesale carrier consolidation process, where you saw the wireless industry in India go from 14 or 15 carriers, which we knew was far too many, down to basically 3 large carriers plus a smaller government-funded entity in the course of about 12 to 18 months, which is an astronomical pace for carrier consolidation. And so we've taken a pretty decent hit on the churn side as a result of that, we were compensated for much of that churn through onetime payments. But today, the situation in India is really that you have 3 large carriers remaining, as I alluded to before, plus a government entity. You have -- one of those carriers that is in a position where they need to really raise some funds and they're working on kind of reestablishing their strategic path forward at this point from a network perspective. So they're working through that. That carrier is a large tenant of ours in India. And so a lot of the churn that we continue to see is from them as they figure out their path forward. So having said all that, why are we optimistic on the long term? I think a lot of it comes down to the fact that India is a market with 1.3 billion plus people, significant portions of that market are still legacy 2G, 3G rather than 4G. And to fundamentally deploy 4G technology across the country, there has to be a tremendous amount of incremental network investment on the part of all the carriers. There have been some hiccups in that process clearly over the last couple of years. But insets from a pricing perspective have come down materially. You can get a cheap 4G handset for $25 in India right now. The demand for mobile data usage is there. Indian consumers who have smartphones use them at a greater rate than U.S. consumers do, believe it or not. And so the missing link is really getting through these industry issues once and for all and kind of looking at the path forward. And so structurally, the market actually looks much better than it ever has, right? You have a reasonable wireless market structure, you have a reasonable tower industry structure at this point, you have a large population. That population is more data hungry. There's no fixed line alternatives to speak of. And so all of those things are positive, long term, in our view. The real thing that you have to see happen is for the carriers to get healthy, right? And there's certainly optimism there. Timing-wise, it's very difficult to lay out an explicit time line. I think we're certainly optimistic that the churn trend that you're seeing coming down will continue and will accelerate on the way down, frankly. This year, we still have double-digit churn in India at the midpoint of our guide. We're optimistic that, that trajectory from 14% last year to 11% this year kind of steepens to the downside. And sooner rather than later, you're going to have a much lower churn number. And accompanying that, we're certainly optimistic that you have a lot of gross new business activity, which has actually been fairly strong over the last couple of years even in the face of all these headwinds. So long term, I think we're still very optimistic. It's going to take some time. But from a structural perspective, there's no fundamental reason why you can't get back to high single-digit growth rates in India over time, right? I don't think that's going to -- certainly not going to happen this year. It's not going to happen in 2022, I don't think. But longer term, there's just so much activity that has to occur that we think you're going to have a pretty good shot at monetizing a lot of that activity on the tower side.

And is that on the net basis, that high single digit you hope to be able to get back to the...

Yes. Yes, it's our antenna billings metric. Exactly.

Well, perfect timing. We've reached our allotted time. Igor, I appreciate your time today. Everybody on the phone and the Zoom and the e-mails I've been getting, appreciate your time today. Everyone stay well. Look forward to next year, 2022, Orlando, back together in person. And of course, we're hopeful that our -- highly confident that our Park City Summer Summit we'll be back live in person. And I know American Tower has been there many times in that nice event setting. So August for the Park City Summit looks good and March in Orlando. Igor, thanks for your time. Stay well.

Thanks, Ric. Thanks, everyone.