Crown Castle Inc. (CCI) Earnings Call Transcript & Summary

Welcome, everyone, to the next session at our Media Internet and Telecom Conference. I'm Matt Niknam, comm infrastructure analyst at Deutsche Bank. And for our next session, we're very pleased to be joined by Crown Castle's CFO, Dan Schlanger. Dan, welcome to the conference.

Thanks, Matt. It's great to be here. I really appreciate you having us.

Great. Great. Well, maybe just to start, for the purpose of everyone on the webcast as well. So obviously, I'm going to -- we're going to do a fireside chat. But if you do have any questions, feel free to type them in on the web portal. I'll see them on my end, and I'll obviously filter them and make sure they are asked during the Q&A session. So maybe just to start then, can you give us a high-level overview of your top priorities for Crown Castle in 2021?

Sure, Matt. Like I said, thanks for having us, and sorry, we're not at the breakers this year. Yes, 2021, like most years, as we're moving into it, we have similar priorities that carry across all years. And basically, that is to increase the number of customers we have utilizing the assets that we already own, build additional assets that we'll ultimately use to drive co-location and ensure that we're providing the appropriate value to our customers to make that happen. So our business model, I think, most people are pretty well aware of it, is build long-term, relatively high-cost capital-intensive assets upfront with an anchor customer on it, and then try to add additional customers over time. And as we do that, we increase the return that we can generate for our shareholders. But because we charge less to the customer than they could do in any other way or certainly by building it themselves because we share economics, we're able to provide a value proposition to the customers who are generally the large wireless carriers in the U.S. And by doing so, we think that we create a very good incentive for them to use us as an outsourced provider because it's economically more attractive and, in many cases, faster time-to-market. So our goal, as it is every year, is to ensure that we live into that value proposition by making our offerings on that shared infrastructure as competitive as possible from a value standpoint, an economic standpoint, but also as competitive as possible from a time-to-market. And as we do that, we'll generate really good returns for our shareholders that over time, we anticipate will allow us to meet our stated goal of growing AFFO per share and dividends per share at 7% to 8% per year. And that's how we think about the business in the short-term and the long-term. And it comes through very well because it's -- like I said, it's a great value proposition for our customers and generates really good returns for our shareholders.

And so maybe let's start with the tower business. Obviously, sort of still a core within Crown. So the '21 outlook you've laid out, call it, for 6% net organic growth. And that's also on a consolidated level, but you're also expecting that for the tower business and it's an acceleration relative to the 5% growth we saw last year. So maybe to start, can you talk about what's driving the improvement? And maybe give us an update in terms of the latest you're seeing across your top customers in the U.S.

Sure. I appreciate you kind of taking my pitch for me, though. It's good that it's gotten out there enough that we do see an acceleration from 5% tower growth in 2020 to 6% in 2021. We're really excited about that because we think that, that incremental growth is -- before we see it -- before, we believe we'll see a significant amount of 5G spending, we think this is just a continuation of building out networks as they sit today. And the driver of that is a slight increase in new leasing activity that we see on the tower business and a slight decrease in churn. And when you add those 2 things together, we get incremental year-over-year growth. And we think that's really important, not only because of the point I made earlier about being really early on into 5G spending, so this is kind of the precipice of what we think could be a decade-long-or-so investment cycle from our customers, but it also looks good when you look at it relative to the peers we have in the market that are seeing decelerating growth. And we're just excited that we've positioned our company purposefully through the agreements we've entered into and the relationships we have and the assets we own to drive some outsized growth in 2021. And it's something that can't really speak specifically on '22 and beyond, but we don't see anything that would be a major impediment for us to be able to continue to grow and have a relatively positive growth rate compared to our peers over some period of time.

And so I don't mean to jump the gun, but I do have a couple of questions in terms of some newer builds, some of the churn headwinds the industry is facing. But I just want to go back to your comments around '22, noting that we're still very early days in '21. If we think about the tailwinds, right? You've got 5G, which obviously is very, very early innings, you'll probably start to see some C-band deployments and you're not really prone to much churn until '23 as it relates to Sprint. I mean, is the implication that 2021 growth levels could be sustainable into next year? Is that -- am I reading this fairly?

I wouldn't say that. What I was trying to say is that over a several-year basis as we look out, there's nothing that we look at -- as you mentioned, we have a potential churn event in 2023, but there's nothing that we look at to say that there's a significant impediment to growth on the tower business, potentially other than that one churn of that 23. I wouldn't say it's going to be 6%. That is not at all what I'm saying. I just think that the tower business in specific, but also the overall industry of providing shared infrastructure or communications is poised to do well because we are at the beginning of what we think is an investment cycle. And we're excited about how we have positioned ourselves to take advantage of that, both on the tower side because we have industry-leading growth, at least compared to our peers, our large peers in 2021, but also because we have small cells where you think we have differentiated access to the growth going forward, as we think there's a tremendous amount of capital that will be spent to build out networks, both on towers and on small cells. And I'm sure we'll get into that a little bit later. But we just think that we positioned the company really well to take advantage of the growth drivers that we see, which is basically the demand for wireless data in the U.S. is growing at 30% to 40% per year. And that's the underlying growth that drives our business. And because we have access to both towers and small cells, we think we will have an outsized portion of that growth going forward.

Got it. And I appreciate you clarifying that comment around '22. So around the C-band, we got the results about 10 days ago. And so obviously, we saw record bids, total outlay, including some of the relocation costs of close to $95 billion from the carrier. So it obviously speaks volumes about the importance of the band, but it also does raise the risk of stretched balance sheets and potentially slower near-term investment. And so I'm wondering how you're thinking about these deployments as a growth driver both in terms of timing, duration and order of magnitude?

Sure. A couple of thoughts there. One, I don't believe that C-band specifically will have a tremendous impact on 2021, just given when it will clear and when we think it will be deployed. Will be, at best, late in the year and probably not a huge impact on 2021 for Crown Castle. But we're not really thinking about band by band, what happens. We're relatively spectrum-agnostic. All we want is for our customers to have plenty of spectrum to deploy, which we believe they do, both in terms of current spectrum they own that has not yet been deployed, but also, as you pointed out, the C-band spectrum that will be released later this year, and a reason to deploy it. And that reason to deploy it, we would think about in 2 ways. The first is there's competitive pressure to do so because all of the major carriers in the U.S. are competing with each other on network quality. And as that happens, there is a requirement or at least [indiscernible] that's continuing to spend on [indiscernible] in order to [indiscernible]. And when you have a combination of spectrum available and some competitive dynamic to push that to be deployed, that makes for good power and small cell dynamics. And the third leg of that, that I would say is that they've also created -- we as an industry have also created an expectation that 5G is coming and will benefit consumers, and consumers are not buying handsets that they believe will lead to better performance and better usability for them. And that end consumer, looking at this dynamic and saying, okay, when am I going to see this 5G performance increase? How is that going to impact? And I just got a new phone. Now what? It creates another point of impetus to push increased spending and investment in the network. And like I said, when we have this confluence of events where you have lots available, lots of competition on network quality and a consumer base that is looking for an incremental benefit, that spells a very good system for a shared infrastructure provider like us, both on the tower and small cell side. And C-band is a part of that, but it is not the only part of it. And we believe that C-band will be deployed. It will be deployed quickly, because it seems illogical to me to -- for any company to spend somewhere, like you're talking about, in an aggregate $90 billion or so, and not want to make a return out of that $90 billion. So if they buy the spectrum and let it sit, they're not making a return. The only return they can make is if they deploy it. So we don't think it would be logical for them not to deploy C-band even though they spent a lot on it. I think precisely because they spent a lot on it, they need to deploy it. So we think the confluence of events, in general, are very good. And having the C-band out there in specific is very good. And so we're excited about what that can lead to over a period of time.

And do you sense this will be primarily deployed using macro sites? Or are there small cell solutions you think carriers are going to seek as they begin deploying C-band as well?

Yes. I would -- again, well, I think it's really up to our customers to talk about how they're going to deploy their spectrum. But I do believe just that C-band follows a pattern for most spectrum is that the most efficient way to deploy spectrum over large geographic and population areas is to do so using macro towers because they are the most economically efficient way of deploying spectrum. But because the C-band mid-band has a good combination of propagation characteristics and capacity, we believe that ultimately [indiscernible] on tower [indiscernible] the demand will force the density of the network into utilizing cells. So we think over time, that C-band will be deployed both on tower and on small cells, although my guess would be that they would probably -- the carriers will probably default to towers upfront. But again, it's hard for us to do this on half of them. I'm just talking general outlook, how I would think through the deployment. But we don't know. It's really [indiscernible] the carriers have designed. What is great for us, though, is that we're -- we benefit whichever way that decision comes out because we have 40,000 towers in the 70% of the top 100 markets in the U.S. and we have 80,000 small cells on there under development, which is the largest outsourced small cell deployment in the [indiscernible].

Dan, I'm just going to ask if you -- or if it's possible to move somewhere with a little bit of a stronger signal. I'm having trouble with the audio. If it's possible. Can you hear me okay?

Yes.

Is that better? This is better there. You there?

Okay. All right.

Yes. So good. Good. Okay, great. Maybe just on the tower discussion in terms of '21. One other follow-up question is I know you guys typically don't like getting into carrier-specific activity. So I'll ask it maybe from a higher level. We know one carrier, obviously, has been very active post closing a bigger acquisition in 2020. I'm just curious as you think about the expectations you've laid out for '21, is it pretty evenly distributed? Are you seeing activity broadly across the 3 nationals? Or is there one that may stand out relative to the others?

Yes, I'm not going to get into the specifics around our customers one by one. I will say that in our 2021 guidance, we anticipate activity accelerating across the board. And we anticipate some modest benefit from our agreement with DISH towards later in the year, which is more based on the fact that DISH has been very open about equipment not being available until the back half of the year, which means this is hopefully a lot of time to have much of an impact on our business. But it's going to be across all of our carriers because of the situation I was talking about earlier of having spectrum, having competition on network quality and having a consumer base that's looking for a network improvement. And I believe and we believe that, that will lead to investment across the board, and that's what's baked into our outlook for 2021.

Got it. Okay. Yes. I want to dig into DISH a little bit. I mean, you signed a long-term agreement with them last November. They've got access to up to 20,000 Crown sites for their 5G build. So I'm curious if you can just talk about the contribution to your business. Obviously, they've made commitments by mid-2023. And so I'm wondering how investors should think about DISH and the contribution in terms of levels of activity and in terms of contribution to growth exiting '21 into '22? Because I would presume, without getting into specifics, that, obviously, you're their largest tower partner today, and they do have targets to hit. And given where your portfolio is in terms of over-indexing to larger markets, Crown should be a little bit of an outsized beneficiary. So maybe if you can help us think through how big of a driver of that could be in the next several years.

Again, I appreciate you taking on some of our pitch. I think we are their largest provider. We are skewed to the right places. And we are going to help them meet their requirements of covering a substantial portion of the U.S. population over the next few years. And it's precisely for those reasons that we believe that we were able to enter into an agreement that will allow us to drive more than our fair share [ of tower business with DISH ]. We are excited about that because it's been a long time since we've had a, from-a-scratch new network build in the U.S. and a long time since we had a new entrant into the wireless business that needs to spend money to build out a network. And we were very focused on utilizing our strengths to try to make sure that we would get more than our fair share, and we believe we have done so. And those strengths, as you pointed out, were the location of our towers, which is due to, as you said, to the larger population centers in the U.S. The number of our towers, just the sheer number, it's hard to sign a deal for 20,000 towers if you only have 20,000 towers because it's not going to be a one-for-one type of deal. So they have access to up to 20,000 towers of ours, but that's still only half of how many we own. So just the scale and size of our business is important. And our ability to provide not only towers, but also fiber, because they included that in the deal as well, for some fiber transportation as part of this agreement. So when you put all of that together, we think that we leaned on our strength, utilized them to put ourselves in this position of getting an agreement in place that is very meaningful. And the last thing we did as part of that agreement is we entered into it with a significant contractual payment that is due to us from DISH over the 15-year life of the contract. And that payment, like I said, is contractual, whether they go on 2 towers or 20,000 towers. And the reason we believe that's important is it gives them an economic incentive to go on more and more of our towers because that payment will be amortized over however many towers they are. So on a per tower basis, it's better to be on more. So not only do we think we have the right set of capabilities and assets, but we also have a great agreement in place that we think will drive them towards our business as their anchor provider, which is what they have called us -- their anchor provider of infrastructure is what they've called us. And we intend to utilize that as exactly what it is, which is to be their anchor and to be a partner with them as they meet the kind of stringent coverage requirements that they have to maintain their spectrum.

Got it. And so we've talked a little bit about C-band and DISH, the ongoing 5G growing 5G builds from the nationals. And so I want to sort of pivot a little bit and think about Sprint and the decommissioning headwind. That's obviously a bigger headwind for the tower group over the next several years. And so I'm wondering if you can maybe refresh us in terms of how investors should handicap that risk for Crown Castle? And then maybe segueing somewhat, obviously, there's bigger risks on the macro side, but are there meaningful enough risks on the fiber side to consider from Sprint churn as well?

Sure. Again, you stated it well. I think that there is more risk of decommissioning due to the T-Mobile and Sprint merger than there has been historically over the last several years, at least, in the tower business. And T-Mobile has been very clear about what they want to do. They want to do decommission some of the Sprint network and put that network back on the T-mobile network, which means that there is some risk to the legacy Sprint sites that we have. And we provide a disclosure in our supplemental package each quarter about how much of our contractual agreements are coming -- are terminating or expiring in any given year by customer. And when Sprint and T-Mobile were broken out before the merger, you could see that around $200 million of spread comes due in 2023. So that's a way of sizing what we have as any significant risk over the medium to longer term because the next sizable churn that we could have from Sprint sites would be in 2028. So anything within a reasonable time frame, that $200 million in 2023 is the biggest risk we have. We don't yet know how that all is going to play out because we need to have conversations with T-Mobile about what they want and what we want in order to come up with some sort of agreement. And if we can't come up with some sort of agreement, then we'll live with the ones we have, including that churn risk. But we believe that we've set ourselves up with the best possible outcome that we could, which is to give time to both us and to T-mobile, after their merger, to figure out what's good for them, what's good for us; how many sites they want to move on to, because in order to move off of Sprint sites, they got to go on to T-Mobile sites; how much they want their network coverage to grow given that they are investing heavily in both 4G coverage and in 5G cases going forward; and how that will play out versus the potential risk for decommissioning. And we're in the middle of those conversations, and we hope that we can come up with some solution that addresses whatever the key requirements are for T-Mobile while addressing the key requirements for Crown Castle. And as we do that, we hope that we can come to a new agreement. But if not, we're very comfortable with where we are because even though that risk is out there, that churn risk of $200 million is out there, that would just be a reduction in our growth over some period of time in 2023, not an elimination of our growth. And we feel really good about how we positioned ourselves. And that was all kind of purposeful when we saw this merger coming. We renegotiated our agreements with Sprint and T-Mobile to give ourselves more time until those churn events were happening so that we could really focus on all the things I just said, which is how do we come up with a win-win agreement that we'll both feel good about and be able to move forward with. And we think we've kind of positioned ourselves uniquely in the industry because of it.

Let's pivot over to small cells. Maybe if you can -- maybe just to start, if you can give us high-level expectations for that business in '21 in terms of both the growth outlook. And then maybe what you're seeing on the demand backdrop from your carrier customers.

We think revenues on our small cell business are growing in the mid-teens range in 2021 based on our ability to put on in the neighborhood of 10,000 additional small cell nodes in 2021. And that's consistent with what we've done in 2020, and it's consistent with where the demand is for small cells as of right now. There's no magic to 10,000 small cell nodes per year. There's no cap to it. It's just that's how much demand we have, and given the length of time it does take to put a small cell on air, which is somewhere between 18 and 36 months, this is -- we think 10,000 is the appropriate number for 2021. We did sign a deal with Verizon on small cells where they committed to entering into agreements with us to cover 15,000 additional small cells, which is the biggest agreement that we have entered into in terms of number of small cells to be deployed. So just on the sheer magnitude of it, we're really excited that we were able to have that conversation with Verizon and come out with an agreement of that size because we think it's meaningful to their network. It's certainly meaningful to our network and our capital spend to be able to support them in that way. And we think that it is a -- just beyond the sheer magnitude of it, we think that it's a helpful signal as well because I think it helps us and investors understand that even Verizon, who had been very vocal about building a lot of their own fiber and small cells, sees value in having an outsourced provider. And we want to provide -- we want to be that provider, and we want to make sure that they continue to see that value. So we think we'll work very closely with them to determine the exact sites of these 15,000 nodes to deliver them on time for them so that they can improve their network. And we think that, that could lead to more business over time if we continue to do a really good job and improve our value proposition. And hope that it's the start of what could be an increase in activity on small cells, although that's not something that we can point to yet, but we think that there's a lot of underlying demand growth, like I was talking about earlier, that should lead to the necessity of more and more small cells over time, which we think will get us above 10,000 small cell nodes per year, which is -- it's just hard to tell when that's going to come, and it's really dependent on when our customers see the value for small cells and the necessity for them. And we're hopeful to see that soon, and we'll wait and see.

And you referenced that 18- to 36-month window. And I believe in the past, I think it was 18 to 24, and it's kind of lengthened. And obviously, COVID hasn't helped either. I'm wondering, has that changed at all? Or is this still -- are we still sort of firmly in that 18- to 36-month book-to-bill kind of window?

We are still in that 18- to 36-month book-to-bill window. It takes a while because there are a lot of discussions and, ultimately, negotiations that need to happen with municipalities and local governments and utilities on zoning and permitting and, ultimately, pulling the utilities, and that just takes a long time. And if you think about it in relation to even the tower business, we've been in the business of putting new equipment onto towers for 1/4 century, and it still takes 9 to 12 months for that to happen. And that's a well-established already existing asset. So I don't see a tremendous amount of change coming to shorten the time between when we get a contract and when we're able to build the system itself, which is -- we would like it to go faster, but it's not required to go faster. It just means that it takes a little bit of time to go from -- start to generating revenues. And if we can -- one of the things we think it might do, though, is it might push some of our customers to more closely examine where we already have small cell deployments because in the case where we can co-locate on existing assets, we can cut that time frame in half from 18 to 36 months to about 9 to 18 months and that's a huge difference when talking about when a network is going to see the benefit of spending money and getting new spectrum deployed on small cells. So we think that those long periods of time may actually benefit our co-location abilities or prospects going forward as our customers get more and more focused on speed to market, which we think will be a differentiator going forward.

Got it. I've got a couple more on small cells that are coming in from the audience. So maybe first, in terms of the competitive backdrop, can you give us any updates -- in terms of the competitive backdrop you're facing on the small cell front, is carrier DIY build still the biggest risk? Or are you seeing more from newer comm infrastructure players? And I've got one follow-up that I'll ask after.

Sure. And we have not seen a significant change in the competitive dynamics around small cells. We still get a significant market share of outsourced small cells or even of the small cells that we know in general. And the pricing has not changed very much. We still target 6% to 7% yields on our anchor builds. And we haven't seen that move very much. So our win rate stays about the same. Our pricing stays about the same, which, to me, is a result of the competitive dynamic staying about the same. We do see the biggest competitive dynamic coming from carrier self-build. But I wouldn't necessarily call it, as you pointed out, a risk, I would just say that it is a structure of the market where there are going to be some areas where our customers want to build on their own and for their own account. And there are going to be some areas where they see lots of value in our ability to provide a lower cost and potentially faster time-to-market solution, which would be through the outsourced model. And as long as we see a significant ability to add co-location, add additional customers to the same deployment over a reasonable amount of time, we're going to be willing to put that capital to work which provides another benefit for our customers, which is, as you pointed out earlier, they spent a lot of money on spectrum, and it might be good to have an extra source of capital through outsourcing like we could provide, to allow for them to deploy that capital onto new systems, both towers and small cells. And that just provides another value proposition that we can articulate to our customers, is not only could we be faster and less expensive, but there's a separate source of capital that is available to them that we would take the brunt of as opposed to them. And all of that speaks to why we think that the competitive dynamics are positive for Crown Castle. We have the most experience, the most scale, the best relationships with our customers, the best relationships with municipalities and the best relationship with utilities in addition to the largest balance sheet that can be put to -- brought to bear to build small cells as an outsourced provider. We believe we just had the right -- we put in place the right dynamics in order to make us very successful. And the way you see that is we have the most small cell-ready fiber in the top markets around the U.S. of any third-party provider. And that's where we think we've just kind of gapped out on the competitive advantages. We have the right assets, right capabilities, right relationships to provide a significant value proposition.

And so last question on small cells. So of the 10,000 nodes you're bringing online this year, what's the mix between co-location and sort of brand-new greenfield nodes? And then maybe as a secondary to that, any color you can give in terms of the co-location mix among the 30,000 nodes sitting in your backlog?

The amount that we have for 2021 is in the neighborhood of about 30% co-location, which is on the higher end of our normal. Typically, we see somewhere between 20% and 30% co-location, and conversely, 70% to 80% anchor build. And we think we'll be about 30% co-location, 70% anchor build in our 2021 nodes. Going forward, in the remainder of our backlog, it's closer to the lower end. It's somewhere in the 20% to 25% co-location and 75% to 80% anchor build. But that could change because if -- like we talked about earlier, it takes us 18 to 36 months to put an anchor build in place. If we get co-location nodes signed in the next, call it, 9 to 18 months, and we can put those on air in 9 to 18 months, we could actually just add, within the same time frame, more co-location without adding any more anchor builds, so it would increase that percentage. And we think there's a good possibility of that, given all of the things we just talked about around how it's faster to market and co-location can happen to where it can alleviate the capital call on our customers' account themselves to build these small cells. We think there's a chance that the market is looking good for co-location. But again, the amount of co-location, the amount of nodes we book are kind of just a truism, they're based on how much demand there is from our customers in the market. And we're hoping to see that demand go up both for co-location and anchor build over time.

Got it. I want to pivot because there's an interesting segue, I think, to capital allocation, and specifically around the CapEx discussion. As we've seen discretionary CapEx now start to stabilize around $1.5 billion, and that's a gross number both last year, and I believe that's in your '21 outlook as well. So if you can give us a little bit of color in terms of what's driving stabilization? And then also help us think about what you would need to see to justify that spend ticking back up from here?

The stabilization is a result of how much we are putting into our small cell business. So we spend most of that money on our small cell and fiber business or our fiber segment, and that's -- that discretionary capital is to put in place those 10,000 nodes we've been talking about. So the reason it's stabilizing is we have about 10,000 nodes in '20 and about 10,000 nodes in '21. So we have about the same amount of discretionary capital to be able to deliver those nodes in those calendar years. And going forward, what it would take for us to feel comfortable is to see demand pick up. If we could enter into a 50,000-node agreement with a customer and they would want those as fast as possible, it would push us above that $1 billion because we would be trying to deliver that as quickly as we can. And we would love for that to be the case as long as -- like I mentioned before, as long as that initial capital was going into assets and markets that we thought would lead to a proximate co-location opportunity, then we'd be very willing to spend that money because we think that there's a tremendous return over time to putting that capital in place, like I said, in the right markets with the right assets. Because that 6% to 7% anchor yield that is what we see when we first put a greenfield project in place, goes to low double-digits when we add the second customer. And when we add the third customer, we get the mid-to-high teens. Well, that's significantly above our cost of capital and would be a very good investment and drive significant value for our shareholders over time. So as long as the demand was there and it was in the right places, we would very much like for there to be some additional capital to spend. What would offset that, though, is the -- is what I was talking about a bit earlier as well, of if more co-location are to come, that, as I mentioned, lower capital intensity, and therefore, we could increase the number of nodes without commensurately increasing the amount of capital. So it depends on how all of those things play out and play with -- interplay with each other because we can have increasing node counts, increasing co-location and having somewhat stable capital intensity, or we could have increasing node comps there are lots of anchor build or would increase our capital intensity, or we could have a lot of co-location that stayed in this range of 10,000 a year, and we can have decreasing capital intensity, all of which would be good signs for us. We just -- we're going to be in reaction to what the market is demanding at the time that we see that demand coming.

Perfect. And maybe to shift now to inorganic. Can you maybe just update us on what you're seeing on the M&A front in terms of opportunities and valuations as you look at both macro tower and fiber assets?

Sure. I'll start with the fiber assets. We do not see a tremendous opportunity to expand our footprint with some sort of acquisition. Because the fiber assets that are available in the market these days typically don't fit the criteria that we have articulated of high capacity, meaning lots of strands of fiber; dense, meaning under a lot of the streets; metro, meaning in the top 30 markets. So high-capacity dense metro fiber is what we're looking for. That is what we bought in our past, but we stopped once -- and we said we were going to stop once we had basically bought the ones that we thought were the most applicable for -- or met most of those criteria. And the reason those are our criteria is because that is what is required to put small cells on air. You need lots of strands of fiber because every node of small cell requires 2 strands of dark fiber. So any significant deployment requires a tremendous amount of fiber. And you need a lot of density of that fiber. And we think that the most likely place that small cells will be deployed the fastest is in the top 30 markets. So those criteria were developed in order to support our small cell aspiration. We're not getting away from that. That is the reason that we want to own that fiber and build that fiber is to put small cells on air and drive really good returns over time. But we just don't see those available right now in the U.S. because the assets that are coming available have less strands of fiber or may not be in the top markets or may not be as dense as we want. So there could be some tuck-in acquisitions, but it is not going to be through acquisition that most of our fiber growth will occur will be through organic build. On the tower side, we would love to own more towers in the U.S. because we think it's a great business. It's the best business ever, and it will be great to have more exposure. But we need to make sure that we are getting a good return for whatever we spend money on. And because we are in a bit of a different position than most companies where we have this great organic growth opportunity in front of us through small cells and building fiber to support small cells, we are -- we always balance that acquisition against, could we spend that money and make a better return with this organic growth opportunity we have? Or would it be the best use of our capital would be to buy back shares? And we look at that all the time, and what we have determined is that given valuations of the acquisition -- in the acquisition markets, that our organic growth is the best allocation of our capital and what we believe will drive the most growth in our dividend per share over the long term.

Got it. Maybe just one follow-up there. I think we've seen some of your peers, I guess, maybe one in particular, acquiring related or what we'll call complementary infrastructure assets like electric utility towers. And so I'm just wondering, how does Crown think about greater ownership of these types of maybe tower-like assets?

I would call what SBA did with PG&E kind of buying towers. They weren't buying electric distribution poles. They were buying structures to put antennas on. And I think it's a perfectly reasonable thing to do and believe that there is a market for that. You just have to, as the acquirer, bake into what your assumptions are going forward, the potential for lease-up on those towers, which I'm sure SBA has done. I think they've done a good job of acquisitions in their best. I couldn't imagine they didn't do a good job this time. So I would not actually say that, that was an ancillary type of business. I believe that is their core business. But there are ancillary type of infrastructure investments available, and we are very much interested in them, including things like edge data centers, where we think that there is going to be a requirement in order for 5G to deliver on the performance promises of very low latency and very high speeds, we will have to have edge compute and edge storage that would fit at edge data centers. And the way we would define the edge is where the wired network becomes a wireless network, which is at a tower site or a small cell site. And because we own 40,000 towers with parcels of land that sit underneath those 40,000 towers, which would be -- which are already connected to power, already secured and already connected to fiber for the most part, that we think that those tower sites are going to be great places for edge data centers, and we would -- we have already invested in edge data centers as a thought for what could happen in the future through an investment that we have in a company called Vapor IO that has developed both hardware or form factors, hardware and software solutions to help with edge data centers, and we think that, that will be a great market going forward. And we're trying to figure out, as Crown Castle, how would we play in that as kind of the next big thing. But as we sit now, when we think about the business, the next big thing and then the most ancillary -- or the biggest upside ancillary investment we can make right now is small cells because they are so much like towers. They're the same customer, same asset, same utilization, same use cases and same economics, just better anchor yields than what we saw with towers. So we feel we've already gotten to the next thing, which is small cells. And we're looking into next, next thing like data centers.

Got it. One on leverage, in the balance sheet. Maybe if you can update us on how you're thinking about target leverage for the business and whether the recent rise in interest rates changes your approach to sort of optimal leverage for the business?

The target we have is somewhere around 5x debt-to-EBITDA. And we have targeted that because we believe that is what is necessary to maintain an investment-grade rating, which we very much intend to do. We think it is hugely beneficial for our cost of capital for both the equity and debt side. And although we think that the business could withstand more than 5x debt-to-EBITDA, that's kind of where the discussions are with rating agencies at this point. So we'll continue to try to push on that. And if we can make it a little higher, we would like to do so. We think it is reasonable to carry more debt on a business that is as stable and has lots of underlying growth demand drivers as ours does. But until that point, we'll remain in that kind of 5x debt-to-EBITDA range as our target leverage. The interest rate increase has not impacted that. We think the limit on our debt, like I was just talking through, the amount of debt we have, our interest coverage has never been in question. So rising interest rates wouldn't have an impact even if they impacted us. But they're not going to impact us that much because we have made conscious decisions over the last several years to extend the term of our debt and reduce the amount of fixed floating rate that we have in our balance sheet because we know that as we make very long-term investments in these infrastructure assets, that cost of capital is one of the most important aspects, whether we can make a return. And we want to lock in that cost count, so we've moved our average tenor of our debt from a little over 6 years to almost 9.5 years over the last 3 years, so that we didn't have any near-term maturities and we could more closely match our assets. And we've moved the amount of floating rate debt from around 20% to less than 10%. We're now greater than 90% fixed rate debt, meaning that rising interest rates are not going to have a significant impact on our interest expense. And we did that on purpose because we wanted to ensure that we could meet our longer-term target of 7% to 8% dividend per share growth rate on an annual basis without having an issue with financing costs and interest costs. And we could have gone short term in floating rate over the last several years, and it would have driven a little bit of incremental return to our equity holders. But we just didn't think it was the right decision given how long-term our assets are and how long-term our investments are, and we think we are positioned really well for a rising interest rate environment.

Got it. And then just last question to sort of tie this all altogether. We talked about a lot of puts and takes for the business. How comfortable are you with the long-term sustainability of your 7% to 8% AFFO per share and dividend per share growth targets?

We are comfortable with that. It has been our goal since 2017 to grow at 7% to 8% per year. We have either met or exceeded that in each of those years. And our goal is to maintain that 7% to 8% in every year. Obviously, we can't guarantee that, but we're hopeful that the way we've set up the business, both in terms of the revenue side that we talked about earlier in the conversation and in terms of the cost side, which is predominantly lease expense under our towers, which is not inflationary, other than the fixed escalators they have in place. And our interest cost, which is the second biggest cost we have, are very fixed in nature. So we believe that we have controlled the cost side with lots of revenue upside. And believe that the combination of towers and small cells provides us with a lot of visibility into how that revenue is going to grow over time. And our goal is to deliver on that 7% to 8% and believe that we've been able to do that for the last several years, and we'll continue to be able to do that.

Perfect. I think we're just about out of time. So Dan, thank you very much for joining us. And I'll reiterate what was mentioned before. Hopefully, we can do this next year at the breakers in Palm Beach.

Absolutely. Thanks, Matt. We'll talk to you soon. And hopefully, we'll see you at Palm Beach.

Sounds great. Thanks.

Bye.

Bye.