CCI - Crown Castle Inc. (CCI) 2023 RBC Capital Markets Global Communications Infrastructure Conference Transcript
2023-09-27 11:08:04 ET
Crown Castle Inc. (CCI)
2023 RBC Capital Markets Global Communications Infrastructure Conference
September 27, 2023 09:30 A.M. ET
Company Participants
Daniel K. Schlanger - EVP and CFO
Conference Call Participants
Jonathan Atkin - RBC Capital Markets
Unidentified Analyst -
Presentation
Jonathan Atkin
Let’s get Started. I'm Jon Atkin, I'm with RBC and I'm pleased to have Dan Schlanger for our kickoff session at our Annual Global Communications Infrastructure Conference. Dan is the CFO of Crown Castle and maybe just to kind of warm things up, give us a little bit of an overview of the company for people that maybe need a little bit of a refresh, and then we'll dive into just some Q&A for the next 30 minutes.
Daniel K. Schlanger
Great. Thanks for having me, Jon. I appreciate being here. So Crown Castle, we are the largest shared infrastructure provider in the U.S. where we focus primarily on shared infrastructure for wireless communications, where we own around 40,000 towers and then 60,000 small cells on air and 60,000 in backlog. Those are -- tower business is about 70% of our business. And then we've recently or somewhat recently started investing more heavily in the small cells which we can get into, I'm sure, in some Q&A around the decision-making and what we're doing and why. And what we believe this provides us is an opportunity to benefit from what seems like an insatiable growing demand from all of us of wireless data in the U.S. And as that wireless data demand continues to increase, the demand for our assets continues to increase because our assets basically provide space for the national carriers and the other company that wants it to put antennas at height in order to deliver wireless data to any device. And we believe that the combination of towers and small cells gives us the opportunity to take advantage of whatever that market may bring and are excited about the fact that in most industry analyst estimates, data demand will continue to grow at 20% to 30% per year, which provides a solid amount of support for the growth in our businesses going forward.
Question-and-Answer Session
Q - Jonathan Atkin
So at a very broad level, give us an idea of what's driving activity on your portfolio, what's different in 2023 versus 2022?
Daniel K. Schlanger
Yes. So the driver of activity ultimately is what I just said is the data demand traffic in the U.S. as more demand as we use our phones more and you want more data to come to you, the carriers basically have three ways of supplying that demand. They can utilize spectrum they already own. So the spectrum is what carries the data from an antenna to your phone, the wireless part of the wireless network. They can use the spectrum that they already own more and more. They can utilize new spectrum and they can use better spectrum efficiency. What drives demand for us are the first two, where they utilize spectrum more and more, the same spectrum more and more or they utilize new spectrum. That means adding antennas to assets they already have antennas on or adding antennas to assets they don't have antennas on, either way that drives growth for us.
In our tower business that has over the course of the last several years, been a very robust amount of growth, our carrier customers have been extremely active. And it drove what would be -- what we would call outsized growth. We generally are trying to shoot for in our tower business, 5% to 6% revenue growth per year. And over the course of the last couple of years, not including 2023, we were over 6% because the level of activity was very high as the customers were deploying new spectrum. They got access to new spectrum, wanted to put out new antennas that had that spectrum on our towers and small cells, mostly on towers though. As the generational upgrade to 5G has occurred, the initial move, which is typically the case in our business is to push that generational upgrade onto sites they already know, so towers. Tower is just exactly it sounds like at height or put an antenna on it. They already have those for 4G and they're adding 5G to the tower. That adds revenue to us, which got us to outsized growth.
As we started looking into 2023, when we gave guidance in October of last year in 2022 for the 2023 fiscal year, we said to the market that we believe activity levels were coming down from 2022 to 2023 because we saw the initial surge into 5G, like you had the initial surge into 4G happen, we saw that abating a bit. And then as typical happens in these cycles kind of leveling out and we have seen that come to bear. The activity levels in our tower business in 2023 are below that of 2022, we believe we will grow about 5% in 2023 which is lower than what we did in 2022, but still within the range that we think is a very good range for a business our size and with the stability and predictability of the growth that we have. The reason for that slowdown is kind of what I said, it is a natural evolution in a generational upgrade.
So the wireless network gets generationally upgraded about every decade or decade and a half. So 1G to 2G, 2 to 3, 3 to 4 and now 4 to 5. If we look back at the 4G cycle, we saw our customers start to act and there's a surge of activity and then they kind of come down and it plateaus at activity plateaus. We are seeing that currently. We believe the initial surge of activity to deploy 5G over our towers and the industry in general is over. And therefore, we believe we're going to reach the plateau part of the curve of a generational upgrade. And will allow us to continue to grow our tower business looking forward through 2027 at about 5% a year. That's our expectation.
And the reason that we believe that there is growth in that business is that the demand for data continues to grow, and therefore, the demand for antennas on towers continues to grow. And if we look at what we believe will happen in 5G, that plateau I'm speaking of, we believe we will be higher than the plateau was in 4G. 4G plateau the general spending level across the industry was around $30 billion. We believe the general spending level across the industry for 5G will be higher than that. And so we have good visibility into our tower growth. And despite the slowdown that we saw coming into 2023, we're excited about where the industry is headed overall and see if we continue down the growth path that has happened over the last few decades.
Jonathan Atkin
So there's activity level that you spoke about, and then there's revenue growth, and they would normally be correlated, but there's MLAs that you have with various of your carrier partners. So how does that kind of affect the equation, how much visibility do you have into rest of this year, next year, following year as you think about your forecasts?
Daniel K. Schlanger
Yes. So what you're referring to is we have agreements with each of our customers at all times. That's true. The term MLA is a master lease agreement that is just -- it just sets up the obligations and requirements of each entity. How that structure can be, various ways, all the way from here just the legal structure, and we have to do everything else outside of it to okay -- all the way to, here's a pricing structure for every site all the way to, there's a different way of pricing, which is -- what I believe you're referring to, Jon, is MLA that includes some sort of incremental escalator or incremental value in order to get access to our towers or a specific amount of space on our towers. We do everything. Our MLAs would be structured, however, we believe, is the best deal for us that works for our customers. Within the MLA structures that we have today with all of our customers, of the 5% growth we have in towers looking forward through 2027, three quarters of it is already contracted. So not only the base of the business is contract but 3.75% per year growth is contracted as we sit today through 2027.
I want to make sure that excludes -- so everybody understands that excludes Sprint churn. T-Mobile bought Sprint, there's been a consolidation of that network and then some synergies between those two companies. We have $200 million of churn, revenue that is going to decrease in our business in 2025 in our tower business. Excluding that, $200 million decrease. We believe we'll grow 5% through 2027, of which three quarters of which is already contracted. And the MLA structure gives us that visibility into 3.75%, which we think is a great floor. And as we look back over what happened in 4G, what we tried to do when negotiating with our customers was bring up the floor without really cutting off a lot of the upside. And we believe we've done that. The 3.75% growth is reasonable growth. It's not what we think will happen, but it's not zero even if activity goes down. And we feel good about that. But we didn't cut off the top end because as I mentioned, we were growing 6%, 6.5% on towers in 2021 and 2022, which we think are -- is a really substantial level of growth given the size of our business. So we do have structures in place that we think protect us and generate good visibility into growth into -- for a long period of time.
And to answer the other part of your question, that was the longer term -- the other part of your question on 2023, we have very good visibility about our business generally. By the time we get to this time in the year, September, we will be working on whatever it is that’s going to be done by the end of the year. So when we update our guidance in three weeks, we will have very good visibility for the rest of the year and whatever we were going to say then.
Jonathan Atkin
Great. And maybe pivot to small cells, Chicago is the venue for our event each year for towers and you have quite a lot of density of small cells going back quite a number of years. But the corporate journey started, I think, in the early part of this century initially and then it kind of grew through acquisition, a lot of your discretionary CAPEX is related to small cells and fiber. But how does that kind of fit into the mix differently going forward, say, the next two years the way you see demand and capital allocation?
Daniel K. Schlanger
So when we first looked at going into small cells, which was 10, 15 years ago, our premise was that as we modeled out the amount of demand that was happening in the U.S. and the growth in that demand that at some point, there weren't enough towers and they weren't close enough together to be able to service all of what we would need as consumers of wireless data. And therefore, there was going to have to be a second leg of network architecture beyond towers that was going to fill in the gaps. If you think about downtown Chicago, there's a tower it covers a radius of call it a 0.5 mile, but there are so many people that there's not enough capacity in the spectrum that the tower covers that the tower can then serve all of the people. So we build small cell systems for that purpose to offload some of that traffic onto a small cell, which is just the utilization of that same spectrum again. So spectrum has a finite amount of capacity. So if you try to cover a half mile, that's the amount of capacity for the half mile. But if you try to cover quarter mile, you've doubled the capacity.
Small cells allow our customers to reduce the radius and therefore, increase the amount of data traffic they can carry with the same spectrum. And small cells because they fill in really work in dense areas, but they also work in dense suburban areas. But downtown Chicago is a good example. There's a small cell right there, just looking at it. And it is -- and my guess is lots of people walked around yesterday, and my guess is nobody saw it. So just when you think about small cells, they are not obtrusive, but they make your service better. So if you live in downtown they're saying, no, just go talk to them and tell them you want them. But the reason for small cells was because we thought that towers were not going to be sufficient for the amount of data demand that was coming. And that's proven to be true, where we wouldn't have small cells because it is always cheaper to build a tower. Towers are a more efficient, more effective way of deploying spectrum. But when they can't get closer together or can't get built, which is in most places, most big cities in the U.S. at this point, small cells have to fill in. That's exactly what has happened over the last decade.
And as we're looking at a 5G cycle that on the towers we said was a surge that came down in a plateau, it's actually a very different cycle for small cells. In 4G, there were no small cells to speak of until kind of halfway in and then small cells came in, in the back half of 4G because we didn't reach the point at which demand was really outstripping the ability for towers to supply. There was a little bit of small cells, just not a lot. But in the back half of 4G, we started getting a lot of small cells. We are now starting 5G, our customers are starting 5G, planning that small cells are necessary because the data density is already in the system. And what we saw that even during a time when our customers were really focused on towers over the last -- from 2020 through 2022, really focused on getting as much spectrum as towers as they possibly could, driving great growth for us and our competitors, they also had to focus on getting small cells on air because they knew they weren't good enough to just focus on towers.
So we got two orders, one from Verizon and one from T-Mobile that equated 50,000 small cell nodes. As I mentioned in my preamble, there's 60,000 that we have on air. To get 50,000 orders is a huge amount, basically almost doubling the size of our business. At a time now where towers are or plateauing, coming down off the surge, we're actually seeing an acceleration of small cell activity because we're converting that 60,000 node backlog into revenue in 2023, 2024, 2025 and beyond. So in 2023, we expect to put 10,000 small cells on air. That is a doubling of the number of small cells as we put it in 2022 because you can see the shift from towers towards small cells or from, if you want to think about it this way, coverage to capacity. So towers give you coverage that everybody has the coverage. You get the bars on your phone, the way coverage and no capacity looks to you as a user is you get five bars and you go to your Facebook page and you get a little spinny thing. That means that the phone sees that the tower has this spectrum, but there isn't enough capacity in the spectrum actually to the deliver you anything. That's actually where I think most people get the most frustrated. I have five bars, why do I have the little spinny thing. That small cell solves your problem. And that's where we think the world is going, and we're seeing that in the activity of our customers right now.
And we believe that on the 10,000 that we're putting on air this year, we're going to do at least 10,000 in 2024 and grow our business at double digits. So whereas tower growth has come down off of a very strong level, and we believe will plateau at some point and grow 5%, like I said, grow 5% through 2027, we think small cell growth is accelerating. And the whole premise is playing out. Not only is the architecture playing out like we thought we would but it also allows us as Crown Castle to have access to a growth avenue that no other company has like we do. And we think that, that positions us to take advantage of growing data demand in the U.S., no matter how our customers want to address that data demand.
Jonathan Atkin
How capital-intensive is the business as you convert that small cell backlog because there's co-location on existing polygons and minimal route extension and then there's essentially new builds, so what does that mix look like?
Daniel K. Schlanger
Yes, it's a good point, and I appreciate you bringing it up. The -- what Jon is talking about is when we build a new system, what we would call an anchor build, where there just isn't a system that exists at all. We got to put capital in place. Most of that capital goes into building a fiber network because every small cell needs to be connected back to the network with a pair of dark fiber strands. So a strand going from the network, the hub to the small cell and going from the small cell back to the hub. Therefore, we have to build net fiber through all of areas like this in order to service our small cell business. That capital is compared to the tower business extremely high because the tower business spent all that capital 20 years ago to build and buy towers 20 years ago and now we're reaping the benefit.
We are in the investment phase of the small cell business, much like we were in the investment phase of the tower business, actually 10 to 20 years ago. And you see what's happening in the tower business, we're not putting any more capital in to speak of, and we're getting more revenue, which generates really great incremental returns and driving up our returns over time. Right now, we're in a position where we're building anchor builds on small cells, which take capital. When we do so with that first customer on a small cell build, we get about a 6% to 7% return. That is not enough to clear our cost of capital like any infrastructure, shared infrastructure business, we can't charge more than our cost of capital for the first tenant. That would be like charging a single tenant in office building for the entire building, but only allowing them to have one floor. It doesn't work that way. We have to charge less than what it would cost our customers to do it themselves or we're not providing any economic value.
We believe we charge less than our customers would be able to build for themselves. They are recognizing that value, which is why they're giving us as many small cell nodes as they have. And we then get co-location where another customer and goes on to that same small cell system. When we get the second customer on that small cell system, we get to returns that are in the low double digits, call it, 10% to 12%, which we believe does clear our cost to cap. And when we get to a third customer, we get to mid to high teens, which clearly exceeds our cost of capital and potentially almost doubles it. That's a huge amount of return for the amount of investment we're making, and we believe will drive significant growth over a long period of time.
Right now, in our backlog building out this year we're about 50-50 between co-location and anchor build. That number historically has been about 30% co-location and 70% anchor build over the last seven years. So we've made a move more towards co-location meaning the capital intensity comes down because once we've built the fiber, we don't have to build it anymore. It's there. We have put lots of strands in place and therefore, if we want to put another customer on that small cell, we just pull a strand or two pair of strands into the small cell. It's already underneath the ground. It doesn't cost very much, and we put more revenue on the same asset. That's why we get increasing returns.
So if you go from 6% to 7% on the first customer to low double digits, the incremental return has to be significant, 20% to 30% or more. And that's exactly what we're seeing in our business. It is just that we continue to invest in the anchor side because we don't want to give up the upside of the future like we do -- like we have in towers now. If we had made the same decision in towers, I just hate to stop, we would have stopped at 5,000 or 10,000 towers. We own 40,000, it would have not allowed us to generate a tremendous amount of value for our shareholders. So we believe we are investing well in the anchor builds that will lead to co-location in the future, and we are seeing that co-location today in our backlog.
Jonathan Atkin
I have got three more questions. One relating to kind of the M&A side. Strategic options are being explored for one of the few remaining U.S. carrier portfolios upside, is this of interest, how do you think about accretion to your portfolio's value if you were to embark on that?
Daniel K. Schlanger
Yes. We love towers, and we would love to own more. It is a great business. And if we can own more in the U.S., especially, we would love to own more but there's a price for everything. And what we have seen recently is that the amount of growth that we expect out of the towers that have been sold either in the U.S. or in Western Europe, which is the only place we really look internationally, developed countries, but think of it as much. The amount of growth that we see does not justify the valuations that are being paid. And therefore, the returns aren't good enough to compete for our cost to cap. If it's different for the tower business, if it comes out of U.S. Cellular, I don't know if it will. If it does, we would be very interested. But that return would have to be in excess of our cost of capital or we would be destroying value by owning those towers. And what we look -- what we've seen is that the growth has not been sufficient to justify the pricing. And we'll see.
Jonathan Atkin
So you bought the AT&T towers, you bought the T-Mobile towers, looking backwards has there been pent-up demand that you saw in those portfolios in the first couple of years post ownership or was the growth rate that you saw more or less on par with legacy CCI and I know that some of this is before your time?
Daniel K. Schlanger
Unfortunately, I don't know the answer to that question because it was before my time. So I don't know. What I can tell you though is on the U.S. Cellular, I think there might be demand for the U.S. Cellular towers. I think one of the characteristics of those towers is that U.S. Cellular is the major customer. And so that provides a different dynamic when buying a tower.
Jonathan Atkin
Next question, AI, Vapor IO, modular data centers and towers, what are some of the thoughts that come to mind around use cases demand profile, capital allocation?
Daniel K. Schlanger
So we have an investment in a company called Vapor IO that does a very small data center, what you would call mobile edge compute or an edge data center at tower sites, we did that in order to learn more about the business. They also have a good form factor we think will be accessible for the market. But generally, what we're trying to do is figure out what use cases could be for infrastructure. And our premise, again, just generally speaking, is much like the wireless network through small cells has to distribute and densify, we think computing power will have to ultimately distribute and densify. We can't keep it in Northern Virginia forever. And that's already playing out because you're seeing data centers being built out in more and more cities. We think that, that general trend will continue and get to a point where their data center is much more densely populated in large metropolitan areas. And we believe that our system of fiber and small cells and towers, including the distributed real estate we have under our towers gives us a competitive advantage to house those data centers. The used case has not been defined yet. I think the used case is as more data gets generated in the network, there has to be a place to offload that data closer to the user, so you reduce latency. And that the computing power is happening closer to the consumer, so you can get rid of that data that doesn't matter and not send it back to Northern Virginia because that would congest the network too much.
So a good way to think about it, although I don't think this is the first use case, but an easy way for me to think about that is if you get to an autonomous car, you can't have that autonomous car offload its data all day onto the network, send it to Virginia, realize that most of the things hadn't changed because a lot of that data is just mapping around it. Nothing's changed and get rid of it in Northern Virginia because too much data is traveling and it will increase latency. And then the car isn't getting feedback fast enough to drive. But if you have an edge data center that's a quarter mile away, you get rid of that problem. And that's an extreme example, but I think it's an example that is illustrative of what could happen with the computation and storage of data closer to consumers because you just basically reduce latency, you take latency out of the system, which I think is what is going to be a key factor in developing use cases that are different in 5G than it was [Technical Difficulty].
Jonathan Atkin
I had two quick questions, maybe just an update on the execution of the restructuring that you announced shortly after earnings?
Daniel K. Schlanger
Because of the reduction in activity levels that I spoke of earlier, we had a reduction in our services business. Our services business had two sides. One is what we would call preconstruction, one is what we would call construction. Preconstruction does everything to get a tower site ready to put more antennas on it. So structural capacity, is there enough capacity on the tower to put something else on it won't fall down. Permitting, zoning, utilities, land, all of that is preconstruction. And then construction is the actual management of the process of getting somebody to climb the tower and put the antenna on. We got out of the construction business because we don't see a huge competitive advantage for us, and it was a low-margin business. It was not generating a lot of profit for us over time. We maintain the preconstruction business. But at lower levels of activity than we had seen in the first half of the year because we did come off a surge and believe activity levels are lower now. Those lower activity levels will result in less preconstruction work because there's less antenna is going on towers. Therefore, we brought our guidance down in the second quarter for the remainder of 2023. And we executed a restructuring around that activity level coming down. We needed to resize, right size the workforce and some of our physical footprint, and we took charges to do so. That restructuring is I don't know how to characterize it, I think the words are going well, but I hate that because I didn't -- it's not fun and it's bad for people.
Jonathan Atkin
Maybe thinking about AFFO per share and dividend growth over the next few years, when do you think you might be able to get back to the targeted long-term dividend growth rate of 7% to 8%?
Daniel K. Schlanger
We believe that when we exit 2025 and no longer have the Sprint churn in front of us, we will be back to growing our business like we expect. As of right now, the revenue growth in our business has been very consistent with what we would have wanted at this point. Like I said, 5% in towers, double digits in our small cell business, and then the fiber solutions business, enterprise fiber business, we're going to grow at 3% in our expectation. That level of growth will lead in a normal business cycle to 7% to 8% dividend growth. And we think we will be back there once we get through Sprint churn, the $200 million Sprint churn I talked about earlier in 2025 because the underlying data demand is there. So it's a relatively tight correlation for us over a long enough period of time that as data demand goes up, demand for our assets goes up, and that's what's driving our confidence in the ability to drive 7% to 8% growth over a long period of time.
Jonathan Atkin
We have time for one quick question if there is one, raise your hand. Yes, Richard.
Unidentified Analyst
[Question Inaudible].
Jonathan Atkin
So the question is, have you contemplated the use of third-party capital to fund some of your expansion?
Daniel K. Schlanger
We would contemplate anything that we thought generated better returns for our shareholders. The premise though would ultimately be that, that third-party capital would be less expensive than the capital we could get somewhere else, and we have not found that to be the case. If that were to change, we would get the less expensive capital.
Jonathan Atkin
We are out of time. Thanks so much.
Daniel K. Schlanger
Thanks, Jon. Appreciate it.
For further details see:
Crown Castle Inc. (CCI) 2023 RBC Capital Markets Global Communications Infrastructure Conference Transcript