Good day, and welcome to the Crown Castle Q4 2021 Earnings Call. Today's conference is being recorded. At this time, I would like to turn the conference over to Senior Vice President of Corporate Finance, Ben Lowe. Please go ahead, sir.

Great. Thank you, Paula, and good morning, everyone. Thank you for joining us today as we discuss our fourth quarter 2021 results. With me on the call this morning are Jay Brown, Crown Castle's Chief Executive Officer; and Dan Schlanger, Crown Castle's Chief Financial Officer. To aid the discussion, we have posted supplemental materials in the Investors section of our website at crowncastle.com that will be referenced throughout the call this morning. This conference call will contain forward-looking statements, which are subject to certain risks, uncertainties and assumptions and the actual results may vary materially from those expected. Information about potential factors which could affect our results is available in the press release and the Risk Factors sections of the company's SEC filings. Our statements are made as of today, January 27, 2022, and we assume no obligation to update any forward-looking statements. In addition, today's call includes discussions of certain non-GAAP financial measures. Tables reconciling these non-GAAP financial measures are available in the supplemental information package in the Investors section of the company's website at crowncastle.com. So with that, let me turn the call over to Jay.

Thanks, Dan, and thank you, everyone, for joining us on the call this morning. As you saw from our results last night, 2021 was a tremendous year for Crown Castle. We delivered 14% AFFO per share growth. We grew our dividend by 11%. We benefited from the highest level of tower application volume in our history, resulting in a 35% increase in core tower leasing activity and a 6% -- and 6% organic growth, leading the industry by a wide margin. We saw an inflection in the demand for small cells, securing commitments for more than 50,000 new small cells during the last 12 months, which is equal to 70% of the total small cells booked in our history prior to 2021. We entered into a new 12-year agreement with T-Mobile that provides committed long-term tower revenue growth, and we made significant progress towards our goal to achieve carbon neutrality for Scope 1 and 2 emissions by 2025 as we successfully sourced 60% of the total electricity we expect to consume this year from renewable sources. These results reflect the positive fundamentals underpinning our U.S.-centric business, as our customers are busy upping and deploying their nationwide 5G network, resulting in robust activity across our towers and small cells from new installations and amendments. As we start 2022, we expect elevated levels of tower leasing to continue this year and anticipate leading the industry once again with the highest U.S. tower revenue growth, supporting our announced 11% dividend increase, which is well above our 7% to 8% target. Our customers are also committing to the next phase of their 5G build-out that will require the deployment of small cells at scale to increase the capacity and density of their networks as more spectrum deployed across existing macro towers is not sufficient…

Thanks, Jay, and good morning, everyone. As Jay discussed, 2021 was a great year for Crown Castle, and we expect the momentum to continue in 2022 as 5G deployments continue at scale. With our comprehensive offering of towers, small cells and fiber solutions, we're able to support our customers' expanding infrastructure needs as network architecture evolves. Turning to Slide 4 of the presentation. Full year 2021 results exceeded our prior expectations, with site rental revenues increasing 8%, adjusted EBITDA increasing 12% and AFFO per share increasing 14% when compared to full year 2020 results, excluding nontypical items. Some of the outperformance in 2021 was due to approximately $10 million of additional site rental revenues, $10 million of additional expense reductions and lower-than-expected sustaining capital expenditures, the majority of which we do not believe will recur in 2022. The 8% year-over-year growth in site rental revenues included approximately 6% growth in organic contribution to site rental revenues, consisting of approximately 6% growth from towers, 10% growth from small cells and 3.5% growth from fiber solutions. Turning to Page 5. We have increased our full year 2022 outlook to reflect the additional $250 million of straight-line revenues associated with the long-term agreement with T-Mobile that we announced earlier this month. Other than these additional straight-line revenues, our 2022 outlook is unchanged. These additional straight-line revenues reflect the significant additional contracted tower revenue growth that comes with the new agreement, but they do not contribute to 2022 AFFO. In addition to the contracted tower revenue growth, the agreement with T-Mobile includes a contractual commitment for 35,000 new small cell nodes over the next 5 years. The agreement with T-Mobile also results in several events related to the decommissioning of the Sprint network, including tower nonrenewals that are expected to reduce site rental revenues…

[Operator Instructions]. Our first question will come from Michael Rollins with Citi.

A couple of questions, if I could. First, on the tower side. I'm curious, given the comments you made about leasing activity, if you could frame the backlog on the tower side that Crown is carrying into 2022, and maybe give us a sense as you look at what the carriers are doing, how much of their footprints are covered by the bookings and the billings that you've recognized to date? And what might be on the come? And then just separately on small cells, if I could just ask one other question. Given the comments about the ramping small cell demand that you've highlighted over the last few months, does it make you want to take a more expansive strategy to add fiber outside of maybe that top 25, 30 markets that you've been historically focused on and be prepared for a more expansive small cell deployment cycle from your customers?

You bet. On your first question around the backlog in towers, obviously, as we spoke to, we had the highest level of activity in the company's history during 2021, and we're expecting that level of activity to continue into '22. And if you frame it in terms of historical context, what's really unique about this cycle is that we've got 4 carriers deploying -- you've got AT&T, Verizon, T-Mobile and DISH, all deploying network. They've got significant amount of spectrum to be deployed, and they have the capital to be able to deploy that. I can't think of another time in history of our business where we've had 4 well-capitalized carriers with spectrum and the desire to deploy network. So we're certainly riding the wave of that. In terms of what they're touching similar to past upgrade cycles, the focus for the 3 legacy carriers is to touch the sites where they are already existing on the assets, on the macro assets. And we would expect the next phase of 5G build-out will be to densify their network. And we expect that some portion of that densification is going to happen through new installations on towers that they're not located on. But a big part of that, we see that happening in terms of small cells and the 50,000 nodes that we booked over the last 12 months. The commitments from them, I think really just speak to that is that second phase of network deployment as they start to try to densify the network and the need there is going to be both macro sites as well as significantly needing a lot of small cells in order to do it. So in terms of the footprint being touched, they're going to touch virtually all of their existing sites as…

Moving on, we'll go to Simon Flannery with Morgan Stanley.

I want to talk about M&A, if I could, for a little while. You've been fairly quiet in terms of your activity over the last couple of years here. I know you've talked in the past about interest in developed markets. There's a lot of activity in Europe. So perhaps you could just update us on that. And then the other would be on ground leases. It looked like it was a fairly quiet year in terms of extending and purchasing ground leases. So any color there on perhaps being able to continue to own more and push the maturities at?

Simon, on your first question around M&A and how we think about this, core to the business around towers and small cells is certainly our focus and really don't see anything outside of our core business of towers and small cells that would be of any interest to us. Around what markets to be in, I mentioned this a couple of times in my comments, we look at the U.S. market as the most attractive market in the world for investment in the kind of infrastructure that we want to own. We think the growth profile is most attractive here. And we also think the risk is the lowest. And so as we've assessed both developing markets as well as developed countries, we just -- we don't see those 2 characteristics exist in the markets. And so we've stayed solely focused on the U.S. We continue to watch what's developing in the world and to see if maybe our calculus in our view would change over time with some characteristics. But based on the work that we've seen today, we just don't -- we don't -- we haven't seen anything that's attractive to us -- and frankly, we're really excited about the opportunity to put capital work and continue to invest in the U.S. market because it does have that low-risk and high growth -- and we see that growth continuing unabated for a long period of time. So really excited about the opportunity in the U.S. I think we have plenty of places to put capital to work here in the U.S. On the second question around ground leases. Now this has been an initiative we've been out for about 15 years. Our average ground leases are well over 30 years remaining on them on average. And so as we look at the ground underneath our assets, we're focused on both extending the leases to the extent that they start to get inside of the kind of a 10- to 20-year period of time. And then where it makes sense to bring those leases on balance sheet in the form of an acquisition, we're certainly open to doing that. And at different points in time, as the markets move around, we get opportunities that arise. And we'll continue to do that. But given the maturity of the portfolio, there's not really any pressure on us to feel like we need to put capital to work to buy ground leases. So we'll do it. It's opportunistic when the financial returns make sense.

Yes. And just to add to the maturity of the portfolio point there, Simon. Over 80% of our leases are either owned or greater than 20 years and the average life of the leases that are not owned are over 35 years. So as Jay mentioned, there's just no pressure for us to do anything unless the financial returns make sense at this point.

And next, we'll go to David Barden with Bank of America.

I guess just two. The first would be, just over the last couple of days, we've heard new news to a degree from Verizon pulling forward, accelerating their C-band build with extra capital, getting in the hands of [indiscernible]. And then T at the same time kind of pushing back their build-out with respect to waiting for 110 to do a 1 tower climb. I just wanted to kind of see how that lined up with what you were expecting for the year and how you set your guidance, et cetera? And then I guess the second question for the T-Mobile deal, in terms of the nonrenewals, how do those nonrenewals manifest themselves? Does -- is it a hot cut on January 1? Does it happen ratably over the course of the year? And specifically with respect to the $10 million we're losing in fiber in '22?

On your first question, I would just -- without being really specific about individual carriers, I think the activity, as I spoke to a minute ago, that we saw in '21 and expect to carry into '22, we haven't seen anything that would cause us to look at our guidance and have to shift our expected growth in revenues. The -- we have pretty good visibility, as you know -- pretty good visibility into what their deployment plans are. And so to the extent that to adjust those plans, we're probably 6 to 9 months away from seeing any impact in what we do. So it starts to -- if there is a movement upward, it would push it pretty late into this year and wouldn't have a real meaningful result impact on our results in calendar year '22. It probably more portends what's going to impact as we start to think about site rental revenue growth going into 2023. So we'll continue to watch it and be responsive as they work on their 5G network.

Yes. And David, to speak to the nonrenewal point, as we talked about when we announced the agreement, on the tower side, we think it's going to be a $200 million impact to 2025, which you can assume happens the first of the year and its $200 million through the course of the year. So it's all of that. On the small cell side, it's a similar portion. It's about $45 million, the majority of which is in 2023. Again, you can just assume that happens at the beginning of the year and impacts the entire year. And the $10 million is a fiber solutions churn in 2022 is an impact of, again, $10 million to the total revenue in 2022. So I think you can -- for modeling purposes and how you think about it, just assume that happens on January 1. And it just -- it impacts revenues negatively by $10 million.

Next, we'll go to Nick Del Deo with MoffettNathanson.

First one for Dan. It looks like your cash tower cost of service is up about 9% year-over-year in the quarter. It was up about 7% last quarter. In the past, it's been pretty muted. And looking at unallocated G&A in the quarter, it was kind of higher than it's been before, too. So maybe talk a little bit about what's behind that cost growth and how we should think about the trajectory from here? And then I have a follow-up.

Yes. If you look year-over-year, it's relatively stable. So it's just when costs come in and what happens on a quarter-to-quarter basis. I would think that when you think about the cost structure, especially in our tower business, it's pretty stable over a long period of time. So I would just ask that when you look into it, Nick, look at the year-over-year growth as opposed to any quarter-over-quarter moves, and that probably is more indicative of what's going to happen through the course of long periods of time, including in our 2022 guidance.

Okay. Okay. So nothing unusual going on there to call out. I guess the second one on small cells. You kind of thinking about the T-Mobile and Verizon deals, do they presumably structure those deals to cover their anticipated needs for the foreseeable future? Kind of thinking about the cadence of signings going forward from customers with long-term deals like those, is it appropriate to think that it might be several years before they come back to do more with you and kind of a feast or famine outlook? Or are there reasons to think that they might come back to the well more -- on a more regular basis before those deals conclude?

Yes, Nick. I would say framing that, I would think about 2 things. The first thing I would think about is the 50,000 node commitment that we received is a hard contractual commitment. So there's a balance whenever the carrier -- we're working with the carriers around contracted future leasing activity, whether it's on towers, for amendments or new installations or in the case of small cells, we're not going to get 100% of what they think they're going to build contracted and committed that early. So I wouldn't -- we don't view this and don't think this is 100% of everything they're going to need or contract for because they're just not going to make that level of commitment. The second thing that I think is -- when I think about your question is the amount of and scale of the commitment relative to historical activity is really, frankly, astonishing. We've said this a couple of different times, but we've been at this for 10 years. We've got 50,000 nodes on air. And in the last 12 months, we've gotten 50,000 node commitment. So it feels to us like this is a very significant turning point. And I think that turning point and the commitment from the carriers while we think about it in terms of -- in my comments, I talked about the return elements, the revenue growth elements and why it's important there. But the other aspect of this is how difficult these are to build, I think, is very well understood by our carrier customers. And so there's an element to this of the carriers know they're going to need the site. And unlike towers where we can get them on air in 6 months, there has to be significant planning that goes into this. And they recognized the difficulty of it, which I think is really the value proposition that we bring to the table of and ability to deliver these nodes and implement. And so it's really a statement of confidence also in our operating ability in addition to the macro elements of what's needed in order to make their networks work. And I think the combination of those 2 things is just really compelling.

And moving on, we'll go to Jon Atkin with RBC Capital Markets.

So I noticed that the cash yield on invested capital showed a nice sequential increase during 4Q for fiber. And can you speak a little bit as to what drove that?

Yes, John. A couple of things drove it. One is just performance in the business as we continue to add more revenue to existing fiber, either through -- both through fiber solutions and colocating small cells on the assets we already own. So I think part of that is just the continuation of proof that this business does generate incremental returns that add to the overall yields. But I think there was a little bit of it that the -- when we called out some expense reductions in the fourth quarter that we didn't expect. Those hit on the fiber side and therefore, increased the gross margin and ultimately, the yield because we -- that calculation is in last quarter annualized. So we got the benefit of that kind of onetime $10 million reduction in expenses. So some of it is from that too, but the majority of the increase in the yield is because the business is just performing well and we're increasing the revenue and returns on the assets that we already own.

So on the tower side, it was in the kind of the low 11% range. And do you expect fiber to -- what pace does fiber kind of converge towards that 11%? How do we think about that?

Yes. So first of all, on the tower side, there is substantial growth there, and that's all just in the performance of the towers. We are continuing because this business is a great business to add revenue on existing towers, which increases the yield, which we would anticipate to happen in 2022 as well. The timing of when we're going to get the fiber business to an 11% yield, we've always had a hard time answering that question specifically because it really matters not only the amount of small cells we get, but how many are co-located. As Jay pointed out, though, we're going to make significant progress towards that, given the 50,000 that we just booked. Because not only do we get a lot, a significant portion of them are going to be co-located on the existing assets, which will drive that yield up. And what's so important to understand about that is at the same type of maturity time frame of towers, we were much below the 8% that we're on at the fiber business right now. So we think that the proving out of the small cell business is happening faster than what happened with towers and that we will get to those 11%. And as we've talked about, as we get to 3 tenants for an overall small cell deployment, we'll get into the -- that drives the returns up to the mid- to high teens. So we feel like there's a lot of upside that is going to be realized in our fiber business, and those yields will creep up and approach in where we are with towers and towers will continue to increase also. So hopefully, you'll continue to ask the question of when are you going to get to the 12% on towers or the 13% on towers. And we'll keep saying we're going to chase it with small cell.

On the 50,000 small cells, you talked a lot of that co-location. Any kind of rough numbers you can share on the capital intensity of that? And then more broadly for the business, given all the MLAs you now have in place, how do we think about the 2022 growth outlook? And how much of that is now contractually locked in under MLA terms?

The capital intensity on the colocation is substantially lower than capital intensity on anchor builds. And as we start getting clarity around what the timing is of building those nodes, both anchor builds and co-locations within the 50,000, we'll continue to update you on the capital. But right now, what we believe is that the capital in 2022 will remain relatively consistent with what we saw in 2021. But it's clear that, over time, we would anticipate that to go up because we're going to increase the number of nodes being put on air from what we expect this year to be 5,000 to what we expect next year to be over 10,000. So we expect an increase in capital but not commensurate with the increased number of nodes because the capital intensity is coming down. On the long-term MLAs and the percentage under those, we think of our MLAs as the way we structure the growth. So whether we put it in an MLA or take it a la carte, that growth is going to occur because our assets are so important to the functioning of the network. And we've signed a lot of MLAs, but we always think about what is the best present value we can get within them. And whether we go a la carte or more holistic pricing is really dictated by the NPV we get plus the benefits that both we and our customers get from having holistic pricing, just making it easier to operationalize. But we also always leave room for upside through that. So we're not pricing all of our tower capacity and saying that's what we get paid within our MLAs. We're pricing it with respect to how much activity we think is going to happen. So we believe that the contracts provide all of that, both the underlying contractual obligations plus the upside and the operational ease that comes with having more pre-described pricing, which is really all it is because -- like I said, however, we want to monetize the growth, we think there's going to be growth and we are monetizing it.

And Jon, one of the things maybe just to think about that, that I would add to Dan's comments around the MLAs is the more nearer period that you think about, the more certainty we have of that revenue stream. And as we get towards outer years, the more optionality we have towards the upside. And we would think about it that way. And obviously, our counter-parties think about it that way as well because the nearer the term, the more certainty they have as to what exactly what they're going to need. And as we get into outer periods, then we end up with more optionality towards future growth. So this is just the nature of the way those contracts end up coming together usually.

And next, we'll go to Rick Prentiss with Raymond James.

A couple of questions, if I could. One, definitely appreciate the focus on core, we call it cash leasing. Also like you guys reporting by segment, the core leasing activity. Hopefully, we get that by quarters as we go forward. A couple of questions on the Sprint T-Mobile items that you discussed earlier this year and touched on this call. Are we right that -- I think you said maybe 5,000 nodes would get decommissioned by -- small cell nodes get decommissioned by T-Mobile in the '23 time frame. Should we assume that the 10,000 is a gross add of nodes, not a net add of nodes?

Yes. Yes.

The net add of nodes.

No. No, the 10,000 of gross out of nodes. The 5,000 will be decommissioned in 2023. So yes, the 10,000 is the gross or more than 10,000, and then the 5,000 will be decommissioned.

Makes sense. Because the gross would obviously lead to new lease activity and the decom will go to churn?

Yes.

Correct.

Okay. Second question on the T-Mobile small cell. I think you also mentioned there was a 5-year specific member commitment, but is the 35,000 node commitment spot over the 5 years? Is it spread over a longer period of time and be getting several questions on that one.

Yes, the 35,000 node commitment is spread over 5 years.

Great. And then it looks like you guys simplified the Sprint churn, the $200 million hitting in 2025 and then churn kind of hangs in the 1% to 2% range other years. Should we assume that it goes more towards the 2% range than the 1% range though as we think about it? Just trying to understand -- because I think you previously acknowledged that you had $679 million ballpark of Sprint leases. So just trying to think of how we should think about the $200 million versus that $679 million and over what period of time might they be coming off?

Yes, Rick, as we get into those outer periods, obviously, there's probably a little bit of movement as you would expect from year-to-year. So we're trying to give you a range that in each of the years we would be within. So we would expect it to kind of be 1% to 2%. And I don't think, at this point, we're ready to be more precise than that range.

Okay. Can you give us an idea of how much total Sprint churn you expect to occur if it was 679 was the most exposure and 200 is kind of lump sum? Can you give us an idea of how much Sprint churn you expect to occur over a multiyear period then?

Yes. I think when we get out into the outer periods of time and history has been such great place to go back and look at how these networks get deployed and what ends up happening when there's consolidation churn. Today, putting a precise answer on that question, I don't think we have -- we don't have great visibility towards that. We can range-bound it. We know kind of the outer limit of it around the 2%, the lower end of kind of 1%. And as they deploy the network and upgrade it to 5G as T-Mobile goes through that process, there were going to be some sites for sure that they're going to not need. There will probably also be some sites that today they look at and may be on the list of potentially losing that ultimately will have need as data growth occurs. So I think as we get into those outer years, we'll be able to give you a better estimation. But in terms of modeling, when we look at the total business, we think as we look at the contractual committed revenues on our assets today, we're pretty confident we'll be inside that -- in between that 1% and 2% churn in any years other than the specific years we were calling out.

Makes sense. One other quick housekeeping one. In the supplement, I guess, the settlement has not been updated for the new straight line adjustment for T-Mobile. But should we assume, again, ballpark ZIP code, that the straight-line adjustment that went up by $250 million in the '22 guidance, should that straight-line adjustment kind of dropped by 50 year change by $50 million a year, just kind of theoretically?

Yes. I won't speak specifically to the $50 million a year, but you're right that, that $250 million will drop over time. And it will ultimately, obviously reverse towards the back half of the 12-year agreement. So yes, it's going to be somewhere generally in that range.

That's good. Yes. I expect in 1Q, you'll give us kind of the more detailed schedule as you have time to kind of finish it up?

Yes. Yes. Well, it's just -- it didn't happen in the period that we are reporting on in the supplement. So we wanted to put it in the period that it actually occurs. So we'll have that all laid out really well in our supplement after the first quarter earnings.

And next, we'll go to Phil Cusick with JPMorgan.

This is Amir for Phil. Two, if I may. You guys have talked over the years about a second and third tenant on small cells being accretive to return. How should we think about these co-locations and pricing that first tenant and later driving up the return?

Sure. So the pricing environment that we've talked about for years is unchanged. We've seen the economics, whether it's the 2 large deals that we spent a lot of time talking about this morning or as we do one-off markets with the carriers, we've seen that pricing stay in line. In terms of our underwriting, we typically see a 6% to 7% initial yield on invested capital as we add a second tenant to those same assets or on that same network, we get into the double digits return, so above 10% yield. And then as Dan mentioned earlier, by the time we get to the third, then we're into the mid- to high teens in terms of our yields on invested capital as we add that third tenant. We've seen that pricing basically be unchanged for the last 5 or 6 years. And so those are the economics you should expect as we add additional tenants. Typically, when we're underwriting these investments, we will be -- we don't do anything speculatively. So we'll start off with some sort of 6 to 7 -- in the neighborhood of 6% to 7% initial yield on that investment. We're building it for a carrier purpose built. And in order for us to sign up for that, we're going to have a view on future tenancy and what will go there. And the things that we underwrite are going to have at least one additional tenant beyond that anchor tenant. So we're underwriting things with yields that are into the double digits and then over time, getting and achieving that lease up to drive our returns above our -- well above our cost of capital.

And one of the things that we've talked about historically and remains to be true is that when we talk about a second tenant, it doesn't have to be a second specific customer that the first tenant coming in and densifying on their own network can be their own second tenant. And we will see the economics that look almost exactly the same, whether it's the same company or a different company as the second tenant. And that's true in these agreements as well as any time that we're seeing the densification on the existing network, that those will drive returns that are very consistent with everything Jay just talked about.

That's helpful. And then my second question, in the material, you guys note that the carriers are planning the next phase of the 5G build that requires small cell that scale. Can you expand on what those conversations are looking like?

I'm sorry, what was the last part of the question?

Can you expand on what those conversations are looking, like talking to the carriers about like the next stage of the 5G build that requires the small cell?

Yes, Amir, I'll only speak to it at a really high level, and then I would let you inquire of each of the carriers as to how they're thinking about it. The conversations that we have brought to fruition the 50,000 nodes that they committed to. And so we have an understanding of how they're thinking about their markets, where they're wanting to densify, how they're thinking about putting it in places where we have existing fiber or not. So we have a good view of that. But specifically how they're thinking about spectrum management and densification, that's really a question that I think they should speak to.

Moving on, we'll go to Colby Synesael with Cowen.

It's Greg Williams sitting in for Colby. I have 2 questions, if I may. One, I just wanted to revisit the idea of AT&T's messaging yesterday about 1-tower climb. I don't want to talk about that customer specifically, but what is the 1-tower climb or a 1-touch program mean in terms of impacts on your leasing expectations? Specifically, if the radios aren't ready for late spring, early summer, if it's one piece of equipment or 2 pieces of equipment, how does that impact your lease-ups and the way your contracts are constructed? Do you price on space equipment? I know some of your peers comprise on a frequency-specific level. The second question is just on the services business. How are we going to see services volume on the tower side compared to 2022 versus 2021? And what are the gross margins looking like for 2022 on service margins?

Thanks, Greg, for the questions. On the first question, I think it's a thoughtful approach to try to limit the number of truck rolls that you have in the network. And so obviously, AT&T is being thoughtful about how they're thinking about truck rolls and a number of times to touch a site because it just increases the cost of that activity. It doesn't matter that much to us in terms of our ultimate financial results on site rental revenue, whether it's 1 or 2 touches in order to get to the steady state or the final state of the network. We're obviously happy to work with our customers as they do things like try to eliminate the number of truck rolls that they have in order to help them accomplish those lower cost deployments. In terms of the pricing in the second part of your question related to that, we price based on space used up on the top of the tower, space used at the base of the tower, the ground space that the equipment takes up as well as our agreements are oftentimes spectrum-specific as to what they're deploying or using on those sites. So all 3 of those elements can impact the ultimate price on the site. On the services question, we expect services, as we talked about in terms of tower activity, to be similar in '22 as it is in '21. We have a similar expectation around the services business in '22, relatively similar to what we saw in 2021. The timing in the quarters may change a little bit. But as you think about it in the totality of the year, our outlook assumes that services is about the same in '22 as it was in '21.

Operator, may we have time for two more questions?

And next, we'll go to Sami Badri with Credit Suisse.

There were some references made earlier about the difficulty of the small cell business, and this is kind of the rationale for the competitive advantage that you guys have built over many, many years and even trenched yourselves and have been able to watch where the puck is going. And another dynamic that really kind of entered in 2021 was the increases of costs and decreases of labor firepower for a lot of organizations involved in the telecommunications industry. Is this environment conducive of a dynamic where telcos will begin outsourcing more of their telecom infrastructure builds over time rather than in-sourcing?

Yes. Thanks for the question. We believe that to be the case. Certainly, the economic proposition that we have to the wireless operators on the tower side has proven out over 25 years that the shared infrastructure model reduces the cost of the network for the operators, and it has played out over a long period of time very successfully. That exact same dynamic is at play with small cells, where we're willing to put up the capital and then deploy the capital initially and then share it across multiple operators such that each operator has a much lower cost of operating than what they would have if they own their own and each of them built their own network. So we absolutely see those dynamics at play. And what you're highlighting around an environment where labor costs increase, inflationary pressures may drive up interest rates, I think it just underpins our value proposition to our carriers that we're happy to provide the capital at a much lower cost than what the markets can provide them capital because of the opportunity that we have to see returns from multiple operators across that same asset. So that's the business model and feel like we're really well positioned for that. And in a period of time where we're trying -- where everyone is trying to figure out ways to reduce overall costs, we're a very good alternative as they think about network deployment.

Got it. And then one comment or question actually on just some of the FAA and airline type things that we've been seeing. Has there been any change in payments or negotiations or on an account contract that you have with your major customers regarding some of the noise that non-telco constituents are making?

There has not been any change in the behavior with our customers. And we don't expect there to be any impact to our 2022 outlook. Operator, maybe we have time for one more.

And that final question will come from David Guarino with Green Street.

Two questions I'll ask them up front for you. The first one is on valuations for fiber assets have moved significantly higher over the past year. Do you think it might make sense to recycle a portion of your fiber assets that maybe aren't as well suited for small cell co-location? And then the second question is on data centers. You guys have made some small investments in data centers, but what are your thoughts on a larger scale data center acquisition? And do you see enough evidence today that data centers and towers under the same roof makes strategic sense?

You bet. On your first question around valuations for fiber, I think the world is starting to see the real value of fiber and the necessity of it for small cells. And a desire for folks to try to figure out a way to invest in that space has certainly driven up the valuations. We've talked for years around our strategic focus and what the interest in fiber is for us. We wanted to be in the top U.S. markets and we wanted to acquire fiber that was dense in those markets and had a lot of capacity. And as we said, I think it was coming on now about 4 years ago, we really didn't see any more meaningful acquisitions in the U.S. or opportunities to acquire fiber that kind of met those criteria of dense urban markets, high-capacity fiber. And so the majority of what we have been investing in since then has been building fiber in markets for carriers to deploy small cells. So when I look at our capital base and the assets, there's not a portion of those assets that are extraneous to the strategy of any meaningful amount. The assets that we acquired and bill were really in the locations that we chose, and we didn't buy large nationwide portfolios of assets that we look at and say, okay, these are underperforming or unlikely to perform long term. So like the asset base that we have, I think we're going to be able to drive really nice returns across them over a long period of time. And I think the actions of the last 12 months really speak to the opportunity that lies ahead. So feel good about where the assets are. On your second question around data centers, large-scale data centers, we don't…

Yes, just to clarify. When I said large scale, I was referring more to a larger platform, not necessarily a hyperscale data center, but it sounds like based on your response, that same would hold true. The timing is not there yet from your view. But in the future, you could be more open-minded to a smaller edge data center strategy? Is that correctly agreeing on what you said?

Well, I think I'm trying to separate the difference between the large box data centers and the edge data centers, we see potential on the horizon for edge data centers to come to fruition in the wireless networks. In the use cases, in order to get to a place where edge data centers are a meaningful opportunity for us, we're talking about cases that are upside cases. As we think about small cells, we'd have to get to that kind of level of data demand in the U.S. So if we get to kind of the upside cases on small cells and fiber and edge data centers are an opportunity, we feel like we're really well positioned with the assets that we have to capture that, and we would absolutely be willing to build out sites. We've made some investments in Vapor IO. We have a good position there. It's an edge data center company. They've done well. We've built a number of their assets on our sites. And so we've got a good partnership there, and we're watching the development of that space. But for it to become meaningful in our operating results is a pretty big run from where we are today in terms of total data traffic that would be needed. And so that's where I would see our opportunity. But don't see or frankly, I don't see an opportunity in the big box data centers today or in the future as being really a part of our core strategy.

Well, thank you, everyone, for joining this morning. And let me just conclude by thanking our employees for a job well done in 2021. The level of activity is remarkable. The way you delivered for our customers was outstanding. And so thank you for all of your work. I know you're already busy working on 2022, but thank you for what you accomplished in '21, and I look forward to talking to all our investors next quarter. Thanks for joining.

And this concludes today's call. Thank you for your participation. You may now disconnect.