American Tower Corporation (AMT) Earnings Call Transcript & Summary

Well, good afternoon, everybody. Tim Horan, our cloud and communications analyst here at Oppenheimer. It's my pleasure to host American Tower here for 40 minutes. We have Rodney Smith who was with us here last year. And thanks for returning, Rodney. Rodney now this year though is CFO. I think it's been a big promotion between last year and this year, although probably some of us do -- or some of you knew the writing was on the wall. But...

Yes. Thanks, Tim. It's great to be here, and thanks for inviting me back.

Absolutely. And if anyone wants to ask questions, you probably all know the drill now, but you can do so online. And I will actually be asking those questions as I'm going through asking my own questions at the same time.

So Rodney, maybe just to kind of start off, can you just kind of summarize the quarter for us and what you thought the highlights were?

Yes, I'd love to. So again, thanks for having me, Tim. I enjoy being here with you and the audience. And let me start off by wishing you well. Hopefully, you're doing okay and the audience is doing well. I'm sure there are people in the audience that I would recognize and familiar faces. I just regret that we can't be together. So hopefully, that happens at some point soon. So we had a solid second quarter. We had really good revenue, EBITDA, AFFO growth on an FX-neutral basis north of 7%, kind of right across the board right now. AFFO and AFFO per share both coming in with growth rates in the 7.5% range on an FX-neutral basis. We posted solid organic tenant billings growth of about 5% consolidated around the globe. That was led in our Africa market at about 10%. Latin America came in at about 7%. So really solid organic tenant billings growth. We did have some FX headwinds, as you all probably saw at the end of Q1, really as a result of the COVID-19 pandemic. When that started off, we did see kind of a sharp weakening of some of the currencies in the emerging markets that we operate in. So there is an FX headwind that we're facing this year when you look at year-over-year comps. In addition to that, we built about 500 towers mostly outside the U.S., 250 towers in Africa, 200 in India and about 75 in Latin America. So really happy with the pace and the rate of tower construction around the globe. The pandemic did slow us down a little bit when you think about tower development, tower construction, primarily in India and in Africa, where the lockdowns prevented some tower activity. We do think that, that is a timing issue and we will end up building those towers for those carriers in the coming months, quarters or even into next year if the pandemic and the lockdowns persist in any individual markets. And then the final update I'll give you for the quarter is we ended the quarter in a really strong liquidity position. We went into the capital markets during Q2. We raised another $2 billion across 3 tenors. We raised $500 million of 5-year, really attractive economics, at 1.3% coupon; $750 million 10-year at about 2.1% coupon; and another $750 million for our second 30-year bond at a coupon of about 3.1%. So we were really pleased with the way the capital markets recovered after the COVID-19 pandemic initially shocked the markets. So we took advantage of jumping in and raising some capital there. We now sit with our average tenants of over 6 years, and our average interest rate's down below 3%. So we're in a really good position there. We ended the quarter with $2 billion in cash, $4.5 billion of available revolvers, which gave us total liquidity of over $6 billion, almost $6.5 million (sic) [ $6.5 billion ]. And we now have no senior notes due until 2022 as we've been able to redeem our 2020 notes and all of our 2021 notes. So we couldn't be in a stronger position from a balance sheet perspective or from a liquidity perspective. We saw good demand for our assets around the globe. And I think the one thing that we saw from the pandemic is just how resilient our tower infrastructure portfolio is in the United States and around the globe. And it really highlighted for us that the wireless services that get provided around the globe is really an essential service. It's a critical service for the global community, and our assets are a critical part of that service globally. And we see that. We saw that as this pandemic kind of rolled across the globe. And we continue to see strong demand for our tower assets. And in a lot of cases, in the international markets in particular, our customer -- our carrier customers are reporting that they've seen sharp increases in mobile data consumption as a result of the pandemic. That's the case in India and Latin America and in Africa. And in places where they have a shortage of fixed line infrastructure, people are really relying on these wireless networks, and that's certainly good for us. To the extent that persists, over time, it should drive more demand for our tower assets. So we couldn't have been happier with our results for Q2 and how our portfolio and our company performed through this pandemic.

Do you have any idea, in those markets you mentioned, how much the wireless volumes kind of picked up sequentially? And were the networks under a lot of stress?

Yes. I can give you a couple of snippets. I wouldn't say that they were overly under stress. But one example is in India, our customer, Vodafone Idea, did report that they saw a jump in data consumption that was equivalent to a full year's growth rate, and they saw that in the first week of the lockdown. So that's a pretty dramatic kind of shift. In the U.S., carriers have been reporting that they've seen a shift in patterns of usage, so going away from the highways, out of the urban centers and more in the suburban areas. And again, I think over time, that would end up putting pressure on the networks. And we saw similar things in Latin America and in Africa.

And any other ways that COVID has impacted your business? I guess did you really have any impact here in the U.S. so far?

I think the only impact that we really saw -- there's 2 things I will point to. One is the way we work is different, right? Just like all of you and people around the globe, our employees are not in the offices or haven't been for a number of months. But we've proven we can effectively manage this business and manage our portfolio and take care of our customers remotely. So we haven't missed a beat in terms of the way we work, but we are working differently. But the systems that we have and things like Zoom that we're using here today really have made that work very well for us. So that's been great. We did see the FX headwind that I talked about a few minutes ago. That was the most significant effect that we saw from COVID. Right when the COVID-19 pandemic began and began to spread, we did see a sharp weakening of foreign currencies in a lot of the markets that we operate. We did see that bounce back a little bit in the second quarter, which was good to see. But the foreign currency is probably the bigger one. And then the final one that I would say is we did see a couple of other small impacts here and there. We reduced our build plan in India like I mentioned a few minutes ago, where some of the crews that go out and build towers were locked down and didn't have the ability to go out and build. In the U.S., we actually are designated -- our employees are designated as essential employees so our employees were able to get out and service the wireless site, service our tower, service our customers throughout the pandemic. So -- and that's the case internationally as well. So particularly for existing sites, maintaining those sites and getting out and doing the work, our employees were essential really around the globe and were able to continue to work. But where we were building new sites, if they weren't essential, if they weren't going to be done in a short amount of time, those projects -- some of those projects were put on hold, which we think is really just a timing issue.

And obviously, you kind of pointed out the fact that there's been a little bit of a shift away from dense urban areas and dense venues, airports, stadiums and definitely urban areas, more to the suburban areas and even rural areas, depending where we are. Do you think that's kind of a permanent shift out of the urban areas? And I know you don't have a huge amount of towers in the urban areas. So I would think -- has that put pressure geographically on the wireless networks where they need to add more capacity in these suburban areas?

It's difficult to say. I mean I -- there definitely has been a shift away from the urban areas and away from the busy congested highways during rush-hour traffic. That phenomena has dissipated quite a bit. And the traffic has risen in the suburban areas and back where people live and vacation and those sorts of things. Whether or not that's permanent or not, I couldn't really say, but the -- but for us, our towers -- the vast majority of our towers are in the suburban and rural areas, and our towers are specifically not located in the dense urban areas. We certainly have towers along travel corridors and major highways and things like that. But we do think that if there is a shift where more people stay in the suburbs more often and are using and consuming mobile data that, that could put demands on the network and increase spending on our tower assets over time. But at this point, it's really hard to predict what will happen in the future, how quickly the world kind of gets back to normal, if and when the majority of people will be back in the urban centers and back in office buildings and those sorts of things. But the good news is if there is a reduction in capacity required in the urban centers, that's a location where we don't have towers. Our towers are in the suburban areas and the rural areas.

Yes. That's a great, great point. I guess -- so just on that, just talking out loud, Rodney. I'm at home obviously a lot more, but I'm on Wi-Fi an awful lot now, right? Where I used to -- my cellular usage has kind of plummeted versus what it was. I used to be kind of a nomad. Is that offsetting some of the pressure on the wireless networks? Have you had much conversations with the carriers on that?

Yes. I think in the U.S., there's a chance for that. So I'm -- like you, I'm in my home and I can operate on my Wi-Fi network. But out of convenience, a lot of times, my phone stays on the wireless network as well in my iPad. So you really have to make the conscious choice sometimes to switch over to the Wi-Fi. But in the U.S., I think you're right. There is pretty good infrastructure here in terms of Wi-Fi networks in building and in people's homes. That can provide some pressure relief for the carrier networks here in the U.S. But at the same time, I do think that you've seen a shift of demand onto the suburban and rural networks that very well could not persist over a longer period of time. When you get into the international markets where we have the majority of our tower infrastructure, a lot of those markets don't have the luxury of fixed line broadband and Wi-Fi and things like that. So the full burden of that shift in demographic and the increase in data consumption falls on the wireless networks. And to the extent that, that persists, that should drive additional demand for our towers in those markets over the long term.

Yes. I've heard from people that design and maintain towers -- sorry, cellular networks, not towers, in Africa, Latin America and India that the governments have kind of come to the realization and the companies have come to the realization that wireless is their key broadband infrastructure, and they need to actually accelerate the build-outs, is what I'm hearing out there. Just curious if you're hearing the same thing that they realize this is going to be a 10-year secular upgrade that they really need to do in these wireless networks.

Yes. I think in general, I definitely agree with you. I mean we have seen a shift in the way governments and I think people around the globe, including the U.S., view wireless networks. And they really are taking on a much more essential, critical nature. Certainly, that's the case around the globe and in emerging and developing markets, where there is not enough fixed line infrastructure, as we said. These wireless networks is the way people communicate. It's the way they entertain themselves on tablets and phones. It's the way that they work certainly when they're remote, which a lot more people are working remotely nowadays. So that is very clear that the governments have recognized the importance of wireless services and wireless infrastructure. And I would highlight India has recognized that. So in India, they have a India -- a Digital India initiative. And the whole premise there is that their economy will become digitized and on a digital platform. In order for that to work appropriately, they need to have access to telecommunications networks. And throughout India, primarily the wireless networks fill that need. So that's a place where the government there, the Modi government, has identified that Digital India is a key priority for the country. And therefore, wireless services is essential for that. So we do think that they will work to protect the industry there and make sure that there are the right amount of players and they have the right capital investments and those sorts of things. But even in the U.S., we've also seen the government recognize the importance of wireless networks and telecommunications networks, particularly when it comes to 5G. And you've heard, I'm sure, a lot about the competitive nature now of 5G deployments from country to country. And the major world powers are looking to be the first to deploy a full 5G network. And you may have seen just as early as yesterday, the administration in the U.S. is -- announced that they're going to release another 100 megabits of -- or megahertz of spectrum, mid-band spectrum, for full power use across the U.S. So that's really another indication of a government, in this case the U.S. government, taking proactive steps to give the industry what it needs to get full 5G deployed and to make sure that they have healthy wireless network across the country. So that's really constructive. So it's not just outside the U.S. realization of how critical these networks is. It's the U.S. as well.

And does that -- more spectrum, does that help your business, the incremental 100 megahertz? Is that generally positive on the towers?

Yes. I think it's generally positive for 5G. I think it's definitely positive for the wireless service providers to have access to more mid-band spectrum so they can deploy 5G. And as they deploy that spectrum, they put up additional equipment to handle it, and that's good for tower companies, us included, of course. So yes, we think the more 5G spectrum that becomes available and gets in the hands of the carriers, the carriers will deploy that, and it will become essential part of the 5G networks in the U.S. And it -- we expect 5 -- we expect mid-band spectrum to run on macro towers, propagate from macro towers and probably be paired with massive MIMO antennas, and that requires amendments on our contracts and drives revenue for us. So we do think as additional spectrum gets deployed, it will drive revenue opportunities for us. Certainly, if the spectrum that gets in the hand of the carriers is mid-band spectrum or lower-band spectrum, that will directly drive revenues for the tower companies.

And you kind of touched on MIMO. And sorry to jump around a little bit, but could you talk about the 5G builds, what type of equipment are going on? Are they -- are these MIMO antennas? Are they larger? Are they heavier? Do they cost more to kind of put on monthly rent on the towers?

Yes. So it's an interesting question. I would say that in the U.S., the 5G deployments are really early on. So we haven't seen a lot of equipment, let's say, being deployed. But I'll give you a sense of what we -- of what's happening and what we expect to happen. So each carrier is deploying 5G maybe a little differently. I think we've seen Verizon and AT&T initial 5G launches beyond millimeter wave spectrum, the high wide-band spectrum that's being deployed more on small cells and low transmission points in the urban centers. We've seen T-Mobile take a different path, where they've deployed nationwide 5G on the low-band spectrum, on the 600 megahertz spectrum. We've also seen AT&T do something similar with their 700 megahertz spectrum. So 5G could be deployed in a lot of different ways. And the way we think about it is the higher the band, the more robust the 5G service. The lower the band, the longer propagation, but it will be a lower grade of 5G service. It won't quite have the same speeds as you get with millimeter wave, on one hand. And the capacity increases, the latency reductions, the spectrum efficiency won't be there in the same way that it will be with a millimeter wave, high wide-band spectrum. With the high wide-band spectrum, it propagates very -- not very well at all. So it has trouble penetrating buildings, and it only covers very small geographic area. So you need lots more transmission points, and it becomes a very costly way to roll out 5G. And it's really only required when you have significantly high bandwidth requirements in dense urban areas. We believe that the main part of the 5G networks in the United States in the suburban and rural areas will be on that mid-band spectrum. It will be the CBRS spectrum, the C-band spectrum, this new spectrum that the government has just announced that they'll be releasing from the DOD over the next year or 2. That mid-band spectrum will be paired with the massive MIMO antennas most likely and will be transmitted off of a macro tower similar to the existing 4G networks. That's where you get the best blend of extended coverage, good propagation, true 5G services, where you get the benefits of the higher speeds, the higher capacity, the lower latency, the spectrum efficiency, the cost benefits to the carrier. You get all that, and the customer gets a good experience. So it seems like the industry is kind of quickly moving towards the realization. And most of the carriers have said this, and certainly, the government has echoed it, that mid-band spectrum is essential for 5G deployments in the suburban and rural areas in the United States. That will be paired with macro towers. And it probably will be propagated from massive MIMO antennas, which are basically -- the way to think about it is the higher the spectrum band, the lower or the smaller the equipment can be. When you get to mid-band spectrum and massive MIMO antennas, that massive MIMO antenna has many, many antennas embedded within it, which are the smaller, lighter antennas that propagate the higher-band frequencies like the mid-band, but they're all combined in one antenna, which makes the antenna very large. That's one of the benefits of mid-band spectrum and massive MIMO, is you can have very good spectrum efficiencies and you can broadcast lots of spectrum off of one antenna. What that does is it requires a big antenna. And that's the way we end up monetizing the 5G on the mid-band deployments, is when they're paired with the massive MIMO antennas, it will probably be a pretty significant revenue event for us because they're very large antennas. But it also gives the carriers that spectrum efficiency. It lets them use lots of spectrum and increase the bandwidth and propagate it in a pretty good distance. So that's really constructive for them and for their cost base and the quality of services they provide. The other thing I would say is the mid-band spectrum doesn't -- it propagates much further than the millimeter wave spectrum, but it doesn't propagate as far as the lower-band spectrum, the 600, 700, 800, 900 megahertz spectrum. So in that scenario, there will be densification. So not only will they be upgrading existing sites with mid-band, but they'll have to get on additional sites to densify the network because the propagation of the mid-band isn't as good at 600, 700 or 800, 900 megahertz spectrum, which is now the kind of the -- doing the heavy lifting on the LTE networks.

Very good, very good. And so the mid-band MIMO antennas, are they twice as large as legacy 4G antennas? Or just in order, how much bigger are they?

No, I would say -- I mean I'll put it in the ballpark of it may be in the range of a 600 megahertz spectrum antenna. So you start thinking about these antennas, the 600 megahertz spectrum deployments, those are pretty big antennas. I think you can think about it as kind of a refrigerator door sort of a size. I mean they're pretty good-sized antennas. And these massive MIMO antennas will be roughly in that range. We've seen a few of them get deployed, but really not a lot yet. So we still are in the early stages of this mid-band spectrum deployment for 5G with these -- the larger massive MIMO antennas.

So as a total aside, Rodney, we're having thunderstorms here in New Jersey after getting a hurricane, and my power keeps going in and out.

Okay.

If you lose me, just proceed to talk for the next 15 minutes.

Okay. Yes, I can do that. I'll do that.

I know you can. So I had a bunch of other questions. But on this point, one of the main questions we get asked about 5G and the carriers in general is what can they grow revenue and free cash flow at in the United States over the next 5 to 10 years to support all the CapEx and all the money they need to spend on the towers and the same thing outside the United States. And I know it's not your expertise, but if their revenues are declining 3%, 4% per year, ultimately, they're not going to be able to kind of spend more with you. But if they're growing 3%, 4% per year, we have really, really good visibility on runway. I don't know if you guys have any insight on new kind of services or what the industry can kind of grow revenues and free cash flow at.

It's an interesting question. I'm probably not the best person to address that. The carriers really would be much better to talk about that. But we do think that in a 5G world, that there will be additional use cases for the carriers. Certainly, when you think about the Internet of Things, automated cars, machine-to-machine communications, literally millions of sensors kind of talking to each other over a network, those are the kind of things that can happen in the 5G network that really would have more difficulty happening over a 3G or even 4G network. So not only are there automated cars, but -- so drones, a drone network that will be in place at some point. The whole idea of Amazon delivering packages and sandwiches via drones all day long, that's potentially tens of thousands, if not hundreds of thousands of drones kind of flying around and needing to be guided. That's all -- that communication network can be the 5G network, where a 3G and a 4G would have trouble, I think, managing through that. So we do think that there are additional use cases. And there's also the potential that a 5G network can really compete with fixed broadband and bring kind of a convergence to the cable companies and the in-home broadband and the wireless networks with 5G and the speeds that are possible and the increase in capacity and those sorts of things. It very well could be that it could be a substitute or a competing service for fixed broadband. So that's pretty interesting. And then the other point is there are cost benefits. So deploying 5G networks and with the spectrum efficiency that you get in the way that the networks work is you can reduce -- the idea is that the carriers will be able to reduce their cost per megabit of service, and that would be a good thing. So it's not just about potential revenue use cases, but it's also controlling costs and providing more at the same or lower cost base.

So you've kind of started to deploy capital in some new areas -- well, I guess specifically data centers. I know it's really early, but can you talk about the thinking behind that process as opposed to maybe buying back stock and entering kind of a new business here?

Yes. So we have a pretty consistent approach, very consistent approach to capital allocation. First, we fund the dividend, and our dividend this year is growing at 20% over last year. So we're certainly committed to returning capital to shareholders through that dividend. Next is we build assets. And like I said, we've built over 500 towers in the second quarter around the globe. We'll continue to do that. So we'll spend $1 billion to $1.1 billion on our capital programs around the globe. That's important to us. We get really good returns when we do that. And we build towers and expand our footprint in existing markets where we can leverage the SG&A and other costs that we have already in the cost base. So that's the second priority that we have. And then we get into the innovation. We do have an innovation program which is kind of coupled with our internal CapEx programs. So in that scenario, we really are looking at ensuring that we understand what the future network needs are within the markets we operate in, whether that's for our existing customers or potentially new customers, whether it's expanding existing services or new services to our existing customers. We want to understand what this 5G network and beyond will look like and what natural abilities and skills do we bring to the table and where can we support these networks in broader ways than just on the towers. So that's where we get into looking at edge computing, data centers, maybe some fiber optic cable outside the U.S. We do have some fiber in the U.S. where we talk about our outdoor DAS networks. We own fiber in buildings, of course. So we do have experience with that. But we do kind of see a convergence of the networks as the towers and the computing power, the edge caching of content and the latency requirements continues to squeeze and reduce to the point we're getting that compute power and the content closer to the end user is going to be important. So that means that whole network chain kind of gets scrunched down where you want that computing power and the caching of content closer to the end user. And I think what that means is maybe as close as being in the handset, but more likely, it's going to be in servers and in storage -- electronic storage that's close to where the base radios are, where the C-RANs or the O-RANs are. So to the extent that we understand how that architecture works, maybe we can own some facilities close to the edge of the network, which could be on tower sites. And we could own many data centers where we're caching content, doing some compute power, connecting the radios and the towers back to the C-RANs and that whole environment and participate there. So we're not straying too far away from towers. We like the tower business. There's still a lot of growth in the tower business. But we also want to make sure we're exploring other ways we can service our customers of today as well as potential new customers with their network needs. And we do think that edge caching, data center is a place that is -- a place where we should really explore. It's early. So no predictions on where it'll head or what will happen with that, but we're interested in learning more. We've made a few small investments. We have a few trials, and we're learning a lot about the potential for edge computing, the potential for O-RAN and C-RAN being stored somewhere outside of a traditional data center or back of a telecom switch, but more distributed and out in the field, more closer to the tower, those sorts of things. So we're exploring that from a lot of different angles. But we are focused on having these things be an extension of the tower, having long-term contracts, trying not to take technology risk, staying within our competency as a company in terms of the things that we're good at managing, which are contracts and real estate and all those sorts of things. We also like the idea of low capital intensity, expanding returns as you get revenue growth. We like assets that can monetize growth in data consumption, which certainly towers do, and we're looking for other things that could do that same thing.

And on these edge data centers, I know it's very, very early, but do you think like we'll need one for every 10 wireless towers, every 100 or every tower, just a range? And how big do you kind of envision these buildings being?

Yes. It's really hard to say. It is early. I don't think that you'll see one per tower like you see shelters nowadays. I think everything is moving back a little bit. So you'll see aggregation points. And depending on the location of the towers and the density of the certain network, you could see larger facilities being connected and managing lots of towers, let's say dozens of towers or even 100 towers, where in other areas, you might see one facility supporting 10 or 20 towers. So the network will be different, but we do think, just on a ballpark here, that you're talking about thousands of these things across the U.S., not dozens or hundreds but thousands across the U.S. And I do think it's important to point out that they would be varying size and they would connect towers -- some may connect lots of towers, some may connect just 25 towers, those sorts of things. But with all that said, it is early. It's hard to predict exactly what -- how it's going to roll out or what the important characteristics will be. But that's why we have the innovation program. That's why we're investing in facilities like the Colo ATL in Atlanta that we have and the other few -- half a dozen sites that we've kind of popped up and that we're trialing with support of the carriers.

So I did -- before I lost my computers, I did get a question on the public media group contract you announced yesterday. I believe it's for broadcast TV, and I think it was 40,000 towers. I don't know if you guys are getting all of them or not, but -- and I apologize, I didn't have much chance to look into it yet. But could you talk about that contract win at all and how material it could be?

Yes. I wouldn't -- I mean I wouldn't want to overdo it in terms of materiality. It is part of an innovation program. It is a contract where, if you can kind of think of these -- a new way to distribute the digital broadcast frequencies. And it's single-frequency antennas where the broadcaster is now -- can put single-frequency antennas out in -- let's say on macro towers, and they can basically shrink their broadcast footprint. So the idea is that a broadcaster can, let's say, show a sporting event in a major metropolitan area. And typically, when they do that, when they do it on over-the-air broadcasting or even through the cable companies, by and large, the commercials that accompany that sporting event, everyone sees the same commercials because it's broadcast over the air or through the cable companies, and everyone sees it. The idea here is that with the single-frequency antennas, these broadcasters can show the sporting event, but they can change the commercials for every one of these single-frequency antennas. So you can have targeted advertising at a much greater extent, which maybe that increases your -- the cost for the ads. Maybe it increases demand for the ads and helps with pricing and things like that. So it's a different way to do broadcast TV, and it gives you certain benefits at more of a micro level. So we're trialing some things out. We've been working on these sorts of things for a few years. This is not the first time that we've had a trial. So it really is in early stages. And I wouldn't want to put any kind of parameters or numbers around in terms of the total addressable market. It's just an interesting thing to do, and it could lead to something more interesting. But at a minimum, it leads to a few tower leases. So that's always a good thing.

And this would tie in with the high-definition antennas, which have become more and more popular, I guess, the digital antennas?

Yes, that's correct. Yes, that's exactly right.

And it would improve the quality, I'm assuming, of the digital antennas or the reception?

Yes. I think that's the other part of it, is that with the new digital antennas and the high-def antennas and things like that, over-the-air broadcasting has the ability to make a comeback. And I think we've seen that happening where there was a time when the number of subscribers just kept dropping, dropping, dropping. Now it's actually going the other way, and people are going back to over-the-air broadcasting because the quality of service is really good. And the other thing that you can do with these antennas in the digital nature of the broadcast networks now is you can multiplex the frequencies and you can broadcast more than just one thing. So it makes it more efficient. So it could be that you see more of the broadcast companies trying to get back into over-the-air broadcasting on digital platforms through these frequencies, and then that could also use or incorporate some of these single-frequency antennas that get distributed out of our macro towers. So it really is pretty neat and pretty interesting. But at this point, it's small, it's early stage, and we'll continue to work on it.

And I know you're a little loath, and I don't blame you, to talk about specific customers. So I'll bundle them together. The combination of T-Mobile, Sprint and DISH, do you think this is a net positive to 5-year revenue growth or net negative? Any kind of color you can provide around the whole combination transactions here?

Yes. I do -- we do view it as a net positive when you think about the combination of T-Mobile and Sprint. So we were pleased when the merger got approved. I think that's a good thing for them. Certainly, it's what T-Mobile and Sprint both wanted, and I think it's good for the industry. So now having 3 large, kind of equally matched competitors in the wireless landscape, all with good funding, good balance sheets, that's a positive. So we do think that the combination of the new T-Mobile with Sprint will invest more in their network over time than Sprint and T-Mobile had in the past when they were separate. So we think that's constructive. T-Mobile -- the new T-Mobile is going to have a pretty -- they have a pretty ambitious build plan. They expect to add 10,000 to 15,000 additional macros to the network. They expect to spend $40 billion over the next few years, next 3 years. That's more than what T-Mobile and Sprint together invested, let's say, in the last 3 years. So that's where it's an uptick in spending overall, which is good. So we do think that, that's a positive. We think T-Mobile is very ambitious. They've stated that they want to have the best 4G network. They want to have the first and best 5G network. So that takes capital investment, and I think that's constructive to the tower industry and certainly for us as well. And that may also kind of heighten the level of competition amongst AT&T, Verizon and T-Mobile. So competition is good. I think it's good for the consumers. It's good for the level of service, and it's probably good for the tower companies. So that certainly is a factor in there. And then you have the addition of DISH. So DISH is -- they bought the Boost network, and they've committed to building out a 5G broadband network nationwide. So that's very interesting and important, and that's going to take an awful lot of investment over the next 5 years and beyond, quite frankly. I mean we think 5G deployments, it really is a 10-year CapEx cycle at a minimum. And I think that will be for any kind of a nationwide broadband-only 5G network. So we think that DISH being a fourth player -- what we've seen from DISH to date, not only do they have real obligations to build the network, but they're showing the signs that they're beginning to build a network. They're doing their planning. They're negotiating with tower companies. We've certainly been talking to them. We want to be as helpful and supportive to them as we can and help them with their network build plans. But having that fourth player, again, is a very constructive, positive thing for the industry. On the downside, there will be some churn from the Sprint side, right? So we've talked a lot about that level of churn there. But with all these other constructive, positive things, I think if you look over a 5-year period, we do expect that it will be net neutral to a net positive.

And should we expect like new master lease agreements maybe for both of them at some point? Or do you think they can kind of work under the existing lease structure?

No, I think you're going to -- I think you would need to see new leases for both T-Mobile and DISH to kind of move forward with the 2.5 deployments on our sites as well as DISH building out their network. It would require either contract amendments or new contracts. So one way or another, there will be some signatures on a page, let's say, for both of them to move forward.

And is an MLA the most efficient way to do it?

Yes. We're really agnostic about the structure of the contract. And MLAs is certainly -- we've got a lot of experience with that and we like those. That's basically where you have unified terms and conditions in a master lease. You have a pricing table attached to the back, and then you have individual site licenses that get executed as and when the carrier wants to jump on an individual site. The other contract structure that we have, which you and the audience, I'm sure, has heard us talk about, is more of a holistic arrangement, which basically, you still have the master terms and conditions but you kind of predefine what the use rights are across some segment of the portfolio. And then you price that up, and you build it into a fixed fee over the time period that you have the holistic rights. And then that -- it doesn't have a change or any difference in terms of value -- NPV value to us or for our shareholders, but what it does is it just makes the administration a little bit easier, gives faster turnaround time for the carriers when they want to get on. It gives them more predictability in terms of what their fees will be over an extended, a couple of year period, let's say, as well as what the use rights are. So that's a structure that's acceptable to us as well. And in some cases, we've used that structure to not only provide benefits and have fixed fees where the carriers know what they get and what they pay and have it be firm and fixed over a time period, but we can also throw churn into that as well and kind of manage that and spread that out over time and use it to offset some of the upsides and those sorts of things. So it's particularly where the carrier doesn't know exactly when they will be able to churn off site, having that flexibility could be a benefit.

Yes. And you've deployed them very successfully historically and obviously managed the business incredibly successfully. So it sounds like a little slow -- well, not slowdown, a little delay in T-Mobile kind of stepping up the spending. It's just a timing issue. It's not a secular issue. It's just a little bit timing. And it sounds like DISH is going to come to market, obviously, also here pretty soon. Any idea when DISH will really start deploying equipment, do you think?

It's hard to say. I can tell you we can tell they're doing a lot of work. They're working hard. Their teams are busy. So they're planning. They're looking at contract terms. They're trying to figure out the way they're going to move forward. Exactly when they start, I don't know. I could tell you we don't have anything from DISH in our 2020 outlook. So if there was any kind of an early start here in the back half, that would be upside to our outlook. And we wouldn't put anything into 2021 unless we had clear visibility as to what the activity level was going to be. So that's the way we'll approach DISH kind of going into the back half of this year and planning for next year. We'll only put it into our plans and into our guidance if we have visibility and a line of sight into exactly what they're going to be doing with their assets.

Rodney, I really appreciate the time. Hope to see you next year. We are out of time, but have a great rest of the week. And thank you so much for attending.

Yes, you too. Thanks a lot, Tim. I enjoyed it. Thank you.

Okay. Bye.