AMT - NYSE

Ladies and gentlemen, thank you for standing by. Welcome to the American Tower Fourth Quarter and Full Year 2025 Earnings Conference Call. As a reminder, today's conference call is being recorded. [Operator Instructions] I would now like to turn the call over to your host, Spencer Kurn, SVP of Investor Relations. Please go ahead.

Thank you, and good morning. Welcome to our fourth quarter 2025 earnings call. I'm Spencer Kurn, Head of Investor Relations for American Tower. Joining me on the call today are Steven Vondran, our President and CEO; and Rod Smith, our Executive Vice President, CFO and Treasurer. Following our prepared remarks, we will open the call for your questions. Before we begin, I need to call your attention to our safe harbor statement. It says that some of our comments today may be forward-looking. As such, they are subject to risks and uncertainties described in American Tower SEC filings and results may differ materially. Additional information is available on our Investor Relations website. With that, I'll turn the call over to Steve.

Thanks, Spencer. Good morning, everyone. Thanks for joining today's call. As you can see from our published results, we had a great year and an excellent fourth quarter. For the full year, we delivered attributable AFFO per share as adjusted growth of 8%, including over 13% growth in the fourth quarter. These results were underpinned by robust leasing demand across our tower and data center businesses and strong execution against our strategy. Over the past year, we've taken meaningful steps to improve our earnings quality and durability. We've steered capital toward developed markets, globalized and simplified our operations and brought leverage back down to our target range. These actions put us on strong footing to capitalize on future growth opportunities and deliver on our goal of industry-leading AFFO per share growth. Before turning the call over to Rod to review our detailed financial results and 2026 outlook, I'd like to spend a few minutes discussing our key priorities for 2026, as outlined on Slide 5 of our earnings presentation. First, driving durable revenue growth. The backbone of our revenue growth is mobile data consumption, which continues to grow rapidly alongside growth in mobile customers, 5G adoption and fixed wireless access. This secular demand growth is expected to require a doubling in wireless network capacity between now and 2030. On top of this, with trillions of dollars being deployed into AI, it's likely that new AI applications will propel mobile data consumption even higher and require greater bandwidth, lower latency and more uplink capacity than today's typical usage. In our largest tower market, the U.S., carriers are in the middle stages of the 5G cycle, where they broadly completed their initial 5G coverage-oriented activity and are shifting toward capacity-oriented activity. We anticipate carriers will densify their networks not only to meet the capacity demands of 5G, but also to plan ahead for the 6G cycle. We're excited about the 800 megahertz of higher frequency spectrum that's been earmarked for 6G and believe its deployment will drive significant activity on towers. As carriers invest in this capacity, we expect our U.S. portfolio to deliver durable long-term mid-single-digit organic growth. As you saw in our 8-K form from January, DISH has defaulted on its payment obligations. We continue to pursue legal action to recover the value of its remaining lease obligations. And while DISH's default negatively impacts our 2026 outlook, in the long run, we expect our business to benefit from a healthier, well-capitalized customer base that can invest more heavily in their mobile networks. Internationally, we see parallel trends of rising data consumption driving durable network investment. In our European market, 5G progress lagged slightly behind the U.S. and strong demand for new sites is prompting exciting levels of newbuild activity with top-tier carriers. In our emerging markets, 4G-related activity continues to dominate, but we see increasing levels of 5G rollouts in key metros with significant runway for growth. We continue to expect our international tower portfolio to deliver faster organic growth in the U.S. as our less mature portfolios lease up over time. In our data center business, strong demand for hybrid and multi-cloud deployments and positive pricing actions continue to yield impressive double-digit growth. Demand for AI-related use cases like inferencing and machine learning is driving an increasing portion of new leasing, and CoreSite's AI-ready platform is equipped to accommodate these higher-density, interconnection-heavy workloads within its existing cost structure. CoreSite is also benefiting from sustained migration of enterprise IT infrastructure from on-premises to interconnection-rich colocation facility. These powerful demand trends, combined with our unique interconnection-oriented infrastructure, continue to support CoreSite's achievement of mid-teens or higher stabilized yields on new data center deployments. Our second priority is operational efficiency. This has long been a key operating principle at American Tower. Over the past 3 years, we worked diligently to improve our cost structure by centrally aligning our regional groups, divesting noncore business units and automating leasing transactions. These initiatives have helped deliver over 300 basis points of cash EBITDA margin expansion across our global tower portfolio since 2022. And today, we have the highest like-for-like tower cash EBITDA margins amongst our peer group. The bulk of our recent cost efficiency efforts have focused on reducing SG&A, which for our tower business is best-in-class at approximately 4.5% of revenue. With the creation of our global COO position last year, we've undergone an extensive review of the direct costs within our tower business in an effort to bend our cost curve and grow direct expenses at a slower rate than revenue. We identified four key areas of expense savings across our global tower portfolio. First, managing land expense, which is our most significant direct cost, by expanding our highly successful U.S.-based land optimization program to other markets; second, implementing a global unified sourcing and supply chain to enable economies of scale, gain pricing advantages and improve inventory management; third, accelerating the adoption of our well-developed standard of care for U.S. assets across our global portfolio to improve repair and maintenance costs; and fourth, simplifying and standardizing internal technology platforms to optimize customer service and accelerate automation. We expect these new initiatives in conjunction with continued strong conversion rates to drive 200 to 300 basis points of tower cash EBITDA margin expansion over the next 5 years. On top of this, we're investing in AI to accelerate efficiency gains even further. While we're still in the early stages of AI adoption, we expect AI use cases to target process automation, predictive maintenance, power and utility management and workflow optimization. We look forward to updating you on our AI endeavors and accelerated efficiency targets in the future. Moving to our last priority for the year, capital allocation. We remain disciplined stewards of capital and strive to generate durable cash flow growth with high returns on invested capital. Now that we're back within our target leverage range, we have significant flexibility. After funding our dividend, we will opportunistically assess the best uses of our capital among internal CapEx, M&A, share repurchases and further delevering. This year, we plan to deploy the vast majority of growth CapEx to our developed tower markets and CoreSite, and we'll continue to manage our global portfolio in ways that accelerate growth and reduce volatility. Before turning the call over to Rod to discuss our 2025 results and 2026 outlook, I'd like to thank our incredible employees for delivering another excellent year. We've established a best-in-class platform for capitalizing on strong industry demand drivers, and I'm confident that we're well positioned to execute our 2026 priorities and drive accelerating durable growth in 2027 and beyond. Rod, over to you.

Thanks, Steve, and thank you all for joining the call. I'll start by walking you through our 2025 highlights and then share our 2026 outlook. Slide 7 shows a snapshot of our full year highlights. Consolidated property revenue grew approximately 4% year-over-year and approximately 5% when excluding noncash straight line and FX impacts. Our growth was primarily driven by organic tenant billings growth of approximately 5% and complemented by data center revenue growth of approximately 14%. Adjusted EBITDA grew approximately 5% year-over-year and approximately 7% excluding noncash net straight line and FX impacts. Property revenue growth was magnified by record services contribution and disciplined cost management, resulting in 20 basis points of consolidated margin expansion. Attributable AFFO per share as adjusted grew approximately 8% year-over-year, firmly within our long-term range of mid- to high single digits. This growth was supported by strong conversion of adjusted EBITDA growth and management of below-the-line costs. Excluding refinancing headwinds of approximately 1% and normalized for FX impacts, AFFO per share as adjusted grew approximately 9% year-over-year, demonstrating the underlying strength of our business model. Finally, on the capital allocation front, we brought leverage back down into our target range of 3 to 5x, and we ended the year at 4.9x. Also in the fourth quarter, we repurchased approximately $365 million of American Tower common stock, our largest quarterly and annual buyback since 2017. We've continued to repurchase stock in 2026, buying back approximately $53 million year-to-date. Now let's turn to our full year 2026 outlook, starting with organic tenant billings growth on Slide 8. As Steve mentioned, DISH failed to meet its payment obligation and is in default. This did not impact our 2025 financials, and for the full year 2025, DISH represented approximately 2% of consolidated property revenue and approximately 4% of U.S. and Canada property revenue. In order to reset true run rate expectations for the U.S. and Canada, 100% of DISH's revenue was removed from organic growth beginning on January 1 and reflected in churn. Any payments collected from DISH subsequent to year-end 2025 will be reflected in other non-run rate revenue. For 2026, we expect consolidated organic tenant billings growth of approximately 1% or approximately 4% excluding DISH churn. In the U.S. and Canada, organic tenant billings growth is expected to be approximately 0.5% or approximately 4.5% when excluding DISH churn. This is comprised of colocation and amendment growth of approximately 2.5%, escalations of approximately 3%, DISH-related churn of approximately 4% and normal churn of approximately 1%. We remain constructive on growth for towers in the U.S., supported by a healthier, well-capitalized customer base. In Africa and APAC, organic tenant billings growth is expected to be approximately 8.5%. This is comprised of colocation and amendment growth of approximately 7%, representing a modest acceleration off of 2025 levels, CPI-linked escalations of approximately 4% and churn of approximately 2.5%. Churn is expected to be back half weighted, resulting in approximately 10% organic growth in the first half of the year and approximately 7% in the second half of the year. In Europe, organic tenant billings growth is expected to be approximately 4%. This is comprised of colocation and amendment growth of approximately 3%, consistent with 2025 levels, CPI-linked escalations of approximately 2% and churn of approximately 1%. In LatAm, organic tenant billings is expected to decline by approximately 3%. This includes steady colocation and amendment contributions of approximately 2%, CPI-linked escalations of approximately 4%, churn of approximately 8% and other run rate revenue headwinds of approximately 1%. As communicated over the last couple of years, we have expected low single-digit organic growth in LatAm through the end of 2027 due to elevated consolidation-related churn in Brazil and for organic growth to accelerate in 2028 once the churn passed. On average, our multiyear expectations remain consistent, though we now expect more acute churn in 2026 and the acceleration in organic growth to commence in 2027, 1 year earlier than previously expected. The higher churn in 2026 is driven by a combination of delayed churn initially expected in 2025 and accelerated churn initially expected in 2027. Overall, we are encouraged by the prospects of an earlier-than-expected market repair in Brazil and from the forthcoming acceleration of organic growth in 2027. As a reminder, we still have an ongoing arbitration with AT&T Mexico. We remain confident in our legal position and note that the outcome of the arbitration may impact organic growth. Turning to property revenue on Slide 9. We expect our outlook for approximately 1% organic tenant billings growth to be complemented by the selective construction of approximately 2,000 new tower sites at the midpoint of our outlook and approximately 13% growth in our U.S. data center business. Excluding noncash straight-line revenue and FX impacts, property revenue is expected to grow approximately 3%. Normalized for the impact of onetime DISH-related churn, our outlook for property revenue implies approximately 5% growth on a cash FX-neutral basis. The FX assumptions contemplated in our 2026 outlook, which reflect our standard methodology and are conservative relative to current spot rates, contribute approximately 1% of incremental growth. And noncash straight-line revenue represents an approximately 2% headwind to our GAAP outlook for property revenue. Moving to Slide 10. Adjusted EBITDA is expected to grow approximately 2% when excluding net straight line and FX impacts as growth in towers and data centers is partially offset by a decline in services. Normalized for the onetime impact of DISH-related churn, our outlook for cash adjusted EBITDA implies approximately 5% growth. Cash adjusted EBITDA margins are expected to be 66.8%, down a modest 20 basis points versus last year as steady margins in towers are offset by contributions from lower-margin data centers and services. In towers, due to a continuation of high conversion rates and cost savings initiatives, we expect cash margins to be flat year-over-year even while absorbing approximately 60 basis points of onetime pressure from DISH-related churn. In data centers, we expect cash margins to decline approximately 270 basis points year-over-year as onetime benefits from property tax adjustments and legal settlements in 2025 are not expected to reoccur in 2026. Normalized for these onetime items, we expect cash margins to hold steady as strong lease-up of existing facilities is offset by putting new capacity into service. In services, we expect healthy levels of carrier activity to drive our third highest services contribution in the history of the company. While this level of services contribution is robust relative to historical standards, following our record 2025 and taking into account an increasing contribution of lower-margin construction services, it weighs on consolidated growth and margins in 2026. Turning to AFFO on Slide 11. Our 2026 outlook assumes attributable AFFO per share growth of approximately 1% year-over-year. Normalized for the impact of onetime DISH-related churn and excluding the impact of FX and refinancing costs, our outlook for attributable AFFO per share growth implies approximately 5% growth. Bridging from our 2026 outlook for cash adjusted EBITDA, tailwinds from lower maintenance capital and share repurchases executed in the fourth quarter of 2025 and year-to-date in 2026 are partially offset by higher interest expense as debt is refinanced at higher rates, higher cash taxes and higher minority interest and distributions, consistent with our expectations. While our outlook for 2026 growth is negatively impacted by churn events in the U.S. and Latin America, we believe that we are well positioned to deliver our goal of industry-leading attributable AFFO per share growth and compelling total shareholder returns in subsequent years. On Slide 12, I'll review our capital allocation plans for 2026. We expect to grow our dividend approximately 5%, resulting in approximately $3.3 billion in distributions to our shareholders, subject to Board approval. Next, we're planning for $1.9 billion in capital deployments, of which $1.8 billion is discretionary in nature and includes the construction of 2,000 sites at the midpoint. Approximately 85% of our discretionary spend is directed towards our developed market platforms, including over $700 million in success-based investments in our data center portfolio to replenish elevated levels of capacity sold over the past several years, increased spend in the U.S. primarily toward land buyouts under our tower sites and continued acceleration in European new builds with over 700 new sites planned. Our plan also includes approximately $180 million in maintenance capital, a reduction of roughly $15 million due to an acceleration of maintenance capital projects into 2025, reducing 2026 anticipated spending. Moving to the right side of the slide, we remain disciplined as we utilize our balance sheet, which is well positioned for a variety of macroeconomic scenarios. And we are focused on allocating capital to optimize long-term shareholder value creation. As I mentioned, we repurchased approximately $365 million of American Tower stock in 2025, plus another approximately $53 million so far in 2026. We will continue to be opportunistic in utilizing the remaining approximately $1.6 billion that the Board has authorized for share repurchases. Turning to Slide 13 and, in summary, we are pleased with our 2025 results, which demonstrate the fundamental durability of our business model. Robust mobile data consumption growth and demand for our interconnection-rich data centers underpin a long runway of growth opportunities for American Tower. With our best-in-class portfolio of towers and data centers and strong balance sheet, we are well positioned to capture these growth opportunities and deliver on our goal of industry-leading attributable AFFO per share growth. And with that, operator, we can open the line for questions.

[Operator Instructions] Our first question comes from the line of Batya Levi of UBS.

Great. On the domestic side, can you provide a bit more color on the pacing of activity that you're seeing from the carriers as we enter a lower contracted revenue cycle that you had under the holistic deals in the prior terms? And are you seeing a change in the amendment versus densification activity today? And maybe just to compare that 2.5% leasing growth guidance for '26, how does that compare to '25 if we exclude DISH?

I'll start out with leasing trends, and then I'll let Rod talk about the numbers on it. So what we're seeing, Batya, is we're seeing the customers providing a steady level of activity, kind of broad-based across the entire ecosystem there. And we are seeing a higher incidence of new colocations coming in, but we still have a pretty healthy amendment pipeline as well. And this is what we would expect to see at this point in the cycle. Some of the carriers are broadly done with their initial 5G overlays. So there'll be some fill-in sites that happen there, but they're broadly done with their initial targets. One is still a little bit further behind on that. And so we are still seeing some amendment growth there. And when it comes to the densification, we're seeing both some amendments because they're adding more equipment to existing sites that they've already overlaid, but they're also adding new sites. So we are seeing a little bit of a shift in that. But we still would expect the majority of our new leasing to come from amendments this year, as we have historically.

Batya, this is Rod. I'll take the other piece of your question relative to the colo and amendment contribution to organic tenant billings and how it relates to prior year. So if you look at our 2026 guide for organic tenant billings growth, it's about 0.5%. Within that, there is about 2.4% contribution coming from colocation and amendment revenue. Now that doesn't have any contribution from DISH at all in it. If you go back to the prior year, the 2025 numbers, we were at about 3.1%, 3.2%, which included some activity from DISH in terms of the contribution from colocation and amendment revenue. When you remove that contribution in the prior year number from DISH, you come right in at that 2.5% level. So we are seeing, as Steve outlined, very consistent activity levels in the U.S. marketplace ex DISH. And we're seeing about 2.5% contribution from colocation and amendment revenue in each of those years from the carriers in the U.S. ex DISH. The only other thing I would add to the pacing of the new business, as Steve said, it's pretty consistent. You will see a little bit higher number in the first half of the year and it drops down just slightly in the second half of the year.

Yes. Thanks, Rod. Yes, I think you meant to say 2.5% contribution from new leases and amendments this year.

Our next question comes from the line of Rick Prentiss of Raymond James & Associates.

Hope you're doing okay with the snowstorm. Can you hear me okay?

Yes. We can, Rick.

Hope you're doing okay with the snowstorm. Obviously, crazy up there in the Northeast. I want to start on the DISH. Appreciate it's out of the guidance. We had taken it out of our numbers as well. Can you provide us the amount owed? Like Crown Castle mentioned that they're owed $3.5 billion when they terminated the agreement with DISH. Are you able to tell us how much is owned and that you're looking at trying to work out a payment from them?

Yes. Thanks, Rick. Yes, I think the key takeaway that we want everybody to have about DISH from today's call is that we have derisked our business going forward by taking it out of the numbers. And we fully plan to fight in the litigation. We think our contract is enforceable. We're going to do everything we can to collect that. But that would all be incremental upside to the current guidance that we're giving out there. When it comes to the exposure on DISH, we've given you guys the numbers. We can kind of back into it, where it represents about 4% of our U.S. revenue. So that's approximately $200 million a year and we've disclosed that it goes through 2035 into 2036. So that gives you guys kind of the ballpark on that. We haven't put a specific number out there and don't plan to put a precise number on it, but that gives you guys kind of the

Excellent. Along those lines, obviously, settlement or payments would be upside to the capital allocation. You mentioned opportunistic stock buybacks, also pursue M&A. How should we think about M&A out there, what you're seeing across the global landscape? And maybe address also kind of the disparity between public and private multiples.

Yes. Thanks, Rick. We continue to evaluate everything that's out there. As you probably know, there's a lot of portfolios that are talked about right now. There's not a lot of active deals that we're seeing. But we are still seeing a disconnect between private and public multiples. And we think that, that reflects the attractiveness and the durability of revenue in the tower business. And so that's kept us on the sidelines for the past few years because there has been that delta there that's made it hard for us to participate. But just to reiterate to everyone, our capital allocation strategy is to focus on developed markets. And so you should not expect to see us participating in M&A in emerging markets. We'll continue to invest a small amount of capital there, opportunistically doing redevelopment to support our organic tenant billings growth there. And then we do have some build-to-suits that we're doing as part of multiyear commitments we entered into previously. But as we think about capital allocation going forward, it's really focused on developed markets, predominantly the U.S. And then if there's an opportunity in Europe or elsewhere that's developed, we'll certainly evaluate that. But we're not seeing a lot of deal flow out there that we find attractive today. And we hope that changes. We hope that there is an opportunity for us to scale in some of those markets. And if there is, we'll keep you guys apprised when it happens.

Rick, I would just add a couple of quick things here. Once you had mentioned any possible future settlement from DISH could be a balance sheet item. I'll just highlight the fact that DISH is in default at the moment. They're not paying us. There is the potential for future collections that may come in. And if they do, it could be accounted for in other non-run rate revenue. So there could be some P&L impacts to the extent that there are future collections from DISH as we go forward. And the only other thing I'd highlight on capital allocation is we are now down below our 5%, within our 3 to 5x leverage target. As you've heard Steve and I talk about over the last several quarters, that brings us into financial flexibility. Just to remind you the bits and pieces here, Steve talked about this. We're a REIT. We provide the dividend. We think that's essential to our long-term TSR, total shareholder return. Then we have been consistently investing between $1.5 billion and $2 billion in CapEx. And we've been able to rotate that, as Steve said, into the areas where we see the best returns. Today, that's going into developed markets and it's increasing capital investments in CoreSite. And then we look at M&A and buyback, and we will make the decisions between those two pieces in terms of which one provides the best outlook for long-term total shareholder return. And of course, if paying down debt and building capacity for future deployments make sense, then we'll do that. So we have a lot of options available to us. We're willing to use them all. And we are now in a place where the balance sheet is very strong and we've regained full financial flexibility.

Our next question comes from the line of Michael Rollins of Citi.

So the margin guidance for cash margins to go up by 200 to 300 basis points by 2030, how much of that is just organic from the natural operating leverage in the business? And then how much is represented by the acceleration of the activities that you outlined earlier?

Michael, this is Rod. Let me take that one. So as you highlight in our 2026 guide for cash EBITDA margins, we're guiding to about 66.8%. That is slightly down from the prior year, down about 20 basis points. Of course, within there, there is organic revenue growth as the benefits of cost management that I would remind you and other listeners that we, as a company, have been focused on cost management and efficiencies over the long term historically and certainly over the last several years, you've seen us talk about absolute reductions in SG&A year-over-year over the last few years. So this is not new to us in terms of focusing on cost. A couple of things that are offsetting those expansionary pieces that are driving the margin is it's offset by contributions -- higher contributions from our data center business which is lower margin as well as contributions from our services business that also has lower margin. And I would highlight that it's absorbing about 50 basis points of contraction because of the DISH churn. And within there, it has a step back in the CoreSite margins by about 270 basis points. A lot of that is a onetime nonrecurring benefit we got in property tax as we reversed a prior property tax accrual in 2025. That's not expected to be recurring again, of course, in 2026. With all of that said, I'm not going to go through and try to break out the bits and pieces of that margin expansion that we are expecting. We have increased margins about 300 basis points over the last several years. We expect to do that again going forward in the next several years going out to about 2030. And I'm not going to break it out in terms of which pieces are the organic growth pieces and which ones are the cost savings. And it really is a continuation of what we've been doing. We've been focused a lot on reducing and managing SG&A. We'll now pivot towards global operations and look to reduce and manage our direct expenses down. That will help contribute to that continued expansion.

And just to confirm. Over the last several months, I think you and Steve have been talking about the incremental effort to drive efficiency, and we were going to get an update at some point. So does today's target for 2030 fully encapsulate the activities that you've been describing over the last several months just to continue to push those efficiencies forward?

It encapsulates the things that I talked about in my script, where we talked about the four initiatives that we're taking on today. We do think that AI could offer some incremental upside to that, but it's too early to predict exactly what that's going to be. So when you think about what we're doing here, the direct costs typically rise with inflation. And so we thought the best way to explain a target to you guys was to do it in terms of margin. We could put out a number that's sort of a voided cost number that wouldn't mean anything to anybody. But we didn't think that was the right way to explain it. We thought it was really to focus on what's it going to be in the bottom line and what's something you could actually model out in terms of our expectation. And so when we looked at it, and we looked at what the growth would have been in terms of margins just from the operating leverage and where we were in terms of direct, we set a stretch target for ourselves. And we do think that, that 200 to 300 points of margin expansion represents some nice improvement over what it would otherwise be if we weren't able to recognize these cost savings. So that's the guidance you're going to get from us, is that margin expansion piece. If theres's a chance to do something else with AI, and we think there is, once we've been able to sort of figure out exactly what those numbers look like, we'll share it. But until then, focused on margin expansion. As Rod said, look at it on the tower side, not on the data center and services side. And we give you guys enough information in the supplemental to do that. And we'll continue to expand those margins and update you on that quarterly like we always have.

And Michael, I would just add that, that margin expansion is off of a base that is already industry-leading.

Our next question comes from the line of Nick Del Deo of MoffettNathanson.

First, I was hoping you could expand a bit on two of the tower revenue growth drivers you highlighted, fixed wireless and AI. So with fixed wireless, are you seeing the carriers invest behind it as the primary motivating factor for work on a site versus it piggybacking mobile-led deployments both in the U.S. and overseas? And what AI use cases do you see as most promising for driving wireless traffic growth?

Sure. I'll take that one. When it comes to fixed wireless, the carriers are still using their existing installations to support that. So you wouldn't necessarily see a stand-alone deployment for fixed wireless. The way we think about it is overall mobile network traffic and mobile data demand. And when you look at the percentage of mobile data usage that's coming from fixed wireless, it's accelerating. We also look at our carrier customers and what they're saying publicly, and they're all raising their targets for fixed wireless subscribers. So what that tells us is that's driving demand on the network and that's underpinning growth in our sales, even though we can't necessarily pinpoint this amendment or this colocation to fixed wireless, we know it's kind of an overall driver. And when it comes to AI, we're in the early days of this. And most of the AI that we're all doing on our telephones is text or maybe a still photograph. That doesn't put a ton of stress on the networks. It's really video that puts the stress on the networks, and it's both video upstreaming and downstreaming. And that's what we think is going to drive a lot more activity over time. Some of the projections we've seen are showing that the upstreaming effects of AI could require a change in network architecture. Where most networks today have about 10% dedicated to upload and 90% to download, varies by customers, so some of them could be different, we think that in the future, AI could change that trajectory a bit so that you're seeing north of 20% in terms of uploading capacity. So again, it's early days. Too hard to predict exactly when it's going to happen or exactly what app is going to drive it. But it's really that video upstreaming, video manipulation as well as things like the Meta glasses that are live streaming kind of everything around you, those types of applications that we think are going to really result in some network traffic over time.

Okay, Steve. And can I ask one on CoreSite as well? I thought there were some local news reports that indicated that you recently bought land in the Bay Area and might be pursuing a new campus in the Dallas-Fort Worth area. Assuming those reports are accurate, kind of what's the time frame for the Bay Area land? And how many megawatts do you think it will be able to support? And maybe talk a little bit about the vision and rationale for de novo market entry in Dallas.

Sure. So we're not ready to announce any new markets yet. We are selectively looking at opportunities in other key metros that would be complementary to our existing portfolio, and we have purchased the land in various areas as sort of an exploratory foray into those. And when we make a decision to break ground there, we'll tell you guys that we're doing it. The rationale is we're seeing incredible demand on our campuses. And this is our fourth consecutive year of record sales growth and this is the first year we've seen AI really manifest itself as a huge use case. So when you look at what's driving the success of CoreSite right now, we still have our kind of bread-and-butter customer, and that's the enterprises that are -- need to be in an interconnection-rich data center that's directly connected to multiple cloud on-ramps. That's our core customer. There's still a huge tail -- a long runway of demand from that customer. But we're also seeing AI workloads like inferencing and machine learning, things like that, and that's our fastest-growing new use case. So from our perspective, to maximize the value of CoreSite, which is what we should be doing, it's both investing in our current campuses and expanding those. And it's looking at what other markets our customers would like for us to be in to drive even more accelerant sales over time. From a timeline perspective, from the time we break ground until the time we open the facility, it depends on a couple of factors. Power availability is one of them, but also just the size of the facility zoning, et cetera. But you can expect it to be approximately 2 to 3 years from the time we break ground till the time that we can bring that capacity online and start realizing revenue from it.

Our next question comes from the line of Jim Schneider of Goldman Sachs.

I was wondering if you could maybe just elaborate on the cost reduction program. Maybe specifically talk about what -- it sounds like many of the actions you mentioned, Rod, would be things you would have done in normal course already. So I'm trying to understand, are you basically saying or is the message that you'll be able to sort of achieve this 50 basis points on average per year in spite of some of the cost headwinds and margin headwinds that you mentioned earlier? Or is there something sort of above and beyond that? And would you expect any nonlinearity in achievement of those?

Yes. Thanks for the question, Jim. And I would highlight the fact that in the last several years, you've seen us really manage our SG&A and manage that down. Of course, we don't announce when we're reducing staff and those sorts of things, but some of that activity certainly happened over the last 3 years as we rationalized SG&A across the board. When I mentioned a continuation, it really is a continuation of the mindset around cost management and cost controls. The thing that is different going forward is that we have a different global structure. We have the addition of a Chief Operating Officer that is going to be bringing best practices around the globe in terms of the way we manage land expenses, the way we execute on supply chain and sourcing. We're going to be looking to expand the standard of care and the way we manage tower operations in the U.S. globally, which we expect really will drive efficiency and bend that curve down. And that will be a contributing factor to the margin expansion that we expect going out, out to 2030.

And then as a follow-up, can you maybe comment on the Europe property growth expectations? 9% new site seems like a lot. I'm just kind of curious where that's coming from. And maybe talk about any kind of the flavor or underlying color on a country-by-country basis.

Sure. I'll take that one. So we're seeing a lot of good opportunities in Europe right now, and we have a strong portfolio anchored by Telefonica that's largely insulated from some of the potential consolidation that's out there. So as we look at Europe, we're continuing to see a long runway of mid-single-digit organic growth that we expect to realize there. And as part of our acquisition but also as part of some of our other agreements out there, we have the opportunity to build new sites. So we're expecting to bring onboard a record number of new builds in Europe next year. And so that's underpinning a lot of that growth. And it's largely in the countries you'd expect it to be. Our two main markets are Germany and Spain there. So you're going to see a lot of towers there. We will bring some new towers online in France as well. And we'd like to bring more online in [ Italy ]. We like that country, and we just don't have as much presence there as we'd like to have. But what you're seeing there is a reflection of some really solid performance by our teams and earning the trust of our customers and then giving us more opportunities to build sites for them. And that's what's underpinning the growth there, along with good leasing expectations over time. I would note that Europe in general is behind in deploying 5G compared to what the U.S. is. And so I think that from that perspective, there's a lot more runway to continue to deploy 5G there as well. So we feel good about the market. We feel good about the investments there. And what you're seeing in that 9% is really us continuing to build new sites as well as realizing organic growth there as well.

Our next question comes from the line of David Barden of New Street Research.

I guess regarding capital allocation, we talked a lot about returning capital and making new investments. But something we haven't talked about for a while, Steve, was kind of the pivot away from emerging markets. And I think the term of art is called capital recycling. And if you go to Slide 11 and you kind of look at that left-hand side, it looks like there's a lot of markets that are small enough to be distractions and that money could be put to a higher and better use. So if you could kind of talk to us a little bit about what the strategy there is at this stage, whether currencies and market valuations have perhaps stopped you from doing things or if things are on the burner. And then I guess my next question is a little offbeat. But for the last 5 years, if anyone asked, hey, how's satellite going to affect the terrestrial wireless business, you'd roll your eyes and you'd say, it's never going to have an impact. But now that you're starting to talk about 2030 margin expansion, 2030 6G is a driver and the reality that these LEO constellations are going to evolve over the next 5 years in material ways, how do you kind of get comfort right now looking into 2030 that this kind of evolution of connectivity, towers in the sky, so to speak, isn't an equivalently disruptive -- an equivalent to, say, the AI evolution, which you also expect to happen in the next 5 years?

Yes. Well, let me take the emerging market question on that. And our goal is always to establish a real estate portfolio that's giving us industry-leading AFFO per share. And we made a pivot a couple of years ago to focus more of our development CapEx in the developed markets because we thought that was going to give us the best durable growth over time and because we had gotten a little ahead of ourselves in terms of emerging market exposure given some of the challenges that we saw there with India and other things. So we've already made a number of changes to our portfolio mix there. You referenced some of the smaller countries. And we'll always evaluate those countries to see what the best use of capital is on that, and you could see us do something there, but only if we think we're realizing the value of those that's accretive to our shareholders. So we're not going to do any fire sales. We're not going to eliminate something because of the distraction. What we've done to fix that is we've changed our operating model so that they're operating from regional hubs or through these kind of global organizations. So it's really not a distraction for us. So it's all about what value can we realize and does that make sense? And if it does, you might see us take action. Otherwise, we're going to harvest that cash flow and redeploy it into our capital priorities that we outlined there. So again, just when there's some -- when there's news there, we'll let you know. But until then, we're going to continue with those. When it comes to satellites, the reason that we made an investment in AST SpaceMobile was to get a Board seat so that we would have a front row seat to this technology as it unfolds. And we certainly continue to monitor that. We talk to the engineers in the space. We talk to the dreamers in the space and what they're trying to do. And that gives us a lot of confidence that satellites will be complementary to the terrestrial networks. 6G is likely to be designed with satellites being an integral part of an integrated network. But the simple physics of spectrum, the simple economics of having a constellation that has to consistently be replenished over time means that towers will always be the cheapest and best way of deploying the level of content, the volume of mobile data that consumers demand. So while we think the satellite business is a great business, it's going to be a good complement to the network and we certainly are excited about the developments we're seeing in that area, we see no risk at all to the tower business over the long term because, again, towers will always be a cheaper form of delivering a mass amount of data to the consumer. And the satellites just can't compete with that.

David, I want to add a couple of just data points for you to think about to support Steve's discussion around us being an active portfolio manager. The evidence here is that we sold India and we took the proceeds from that sale. And we delevered and helped us get down below our 5x. You also saw us exit [Technical Difficulty] in different markets around the world. And again, we used those proceeds from those sales to delever the balance sheet and to help us regain that financial flexibility. And most recently, and we announced it in the prepared remarks and in the press release, we sold half of our stake in AST Mobile. Remember, it was just a modest investment. We really invested in to stay close to the satellite business and to learn about that business going forward. There's no secret, the stock has done well. The company has done well. And we looked at it as a good opportunity to recycle some of that capital. So we sold half the stake. We used that to actually buy back shares in the last quarter. And we maintained a Board seat on AST. So we still have the ability to continue to learn.

Our next question comes from the line of Brandon Nispel of KeyBanc Capital Markets.

Can you guys hear me?

Yes, we can hear you.

Great. So two questions. Obviously, the U.S. ex DISH is pretty steady. I guess if we could remove DISH from like the last 3 years, it still seems like colo and amendment activity is down. Is that right? And just to nitpick a little bit, why is the second half of the year lower than the first half? And how should we be thinking about that in terms of the exit rate? How should that inform our view in terms of 2027? And then separately, in Africa, one of your largest customers just announced their intent to acquire some towers for their own. Sort of how you're thinking about that in terms of your view when one of your largest customers now wants to own a captive tower portfolio? How does that impact your growth expectations for that market?

Thanks. I'll take the Africa question first. We don't expect it to have any effect on our business there at all. We view that transaction is unrelated to the business we have there. If you look at the sales success that we had in 2025, we're doing very well in Africa in terms of our new business there. And our projections for 2026 are to have another good year of that. We don't think that the acquisition of the other tower company really has any bearing on that at all. As we've said previously, our goal is to reduce the incremental capital that we're putting into Africa over time. And that means that we're not going to be building as many new sites for the carrier customers there. And so I think that what you're seeing play out in the various changes that are happening in Africa are reflective of our customers there looking for other alternatives versus American Tower in terms of how they're going to build some of those sites in their network. So we feel good about that business. We think that we're going to continue to have a nice long runway of organic growth there.

Brandon, I guess keeping with the theme here and working backwards, I'll answer your second question, which is the timing and the pacing of new business. So I referred to the fact that there will be a slightly higher number or contribution in the first half of the year and it ticks down really ever so slightly. It's simply just a function of the way that our holistic agreements work as well as the timing of activity that we expect to see. And there's nothing more to it than that. Your first question again is related to kind of going back over several years and the contribution and activity level that we've seen in the U.S. and how it relates to our organic growth. I'll point out a couple of things. A few years ago, you did see us achieve record levels of new business. And there were a couple of drivers to that. One is there was an initial phase, initial wave of 5G networks, carriers deploying and upgrading their network with C-band spectrum. That came with an initial spike in CapEx, where we saw the carrier CapEx in the U.S. come up over $40 billion. So there was a significant push to begin that 5G launch. And that's typically what we see in the industry when a new technology is deployed. After that initial wave, we do see a moderation of the CapEx in a more steadying, albeit a step-up in terms of CapEx. So we may not be seeing a repeat of the all-time high that we saw in the initial wave of 5G. But the steady state now, more in the mid-$30 billion range, is higher than the steady state that we saw in CapEx under the 4G cycle. So it moderates but there is a step-up that's consistent over time. And the other thing I would highlight is over the last several years, you saw contributions from DISH to new business and organic growth for the tower companies over the last several years. Going forward, that's no longer in the numbers. So that said, when you think about the state of the industry today and the 3 wireless carriers, they're well capitalized, healthy. And they are contributing a consistent level of activity in '26 that we saw in prior years, and we expect that to continue going forward.

Our next question comes from the line of Brendan Lynch of Barclays.

Rod, maybe just to follow up on that commentary there. That was very helpful in kind of framing out the longer-term outlook. You previously guided to 5% organic growth through 2027. Obviously, with DISH not in the picture now, that is coming down a little bit. How should we think about that long-term growth going forward, we get back into about 4.5%, and that's what you're suggesting for this year? Should we anticipate that, that continues out into the future as well?

I'll actually take that one. Back in early 2021 when we set out that expectation for U.S. and Canada organic growth of 5% or better between '23 and '27, that was based on the growth drivers that we saw at the time. And what we didn't have in our view shared then was DISH trying to sell its spectrum and exit the market effectively and also T-Mobile completing the transaction of U.S. Cellular. So those are things that have taken us off of that guide, as you know, this year in particular. However, everything else is kind of playing out the way that we expected it to. And if you look at the past several years, we achieved 5% up until last year when those transactions were announced. And so as we think about going forward, and we're not going to give '27 guidance here, but as we think about going forward, our long-term growth algorithm that we've laid out for you guys, we believe still holds true. And that is organic growth in our developed markets in the mid-single digits, a little bit higher in our emerging markets, higher contributions from CoreSite because we see double-digit growth in revenue in CoreSite. And we're going to have expanding margins because of our cost control. So as we think about that long-term growth algorithm that we laid out, we're still confident in our ability to deliver that even in the 3-carrier market.

Great. That's helpful. And maybe just one on the data center front. Can you describe to what extent you're seeing actual inferencing demand and specifically low latency inferencing demand at CoreSite?

I can speak to inferencing demand. It's hard to say if it's low latency because the whole campus is low latency based on the way we've organized it. But what we've seen is an uptick in inferencing. It is one of our leading new use cases that's coming in. And quite frankly, we have more demand for it than we can meet with our supply. So we're able to curate our mix of inferencing partners there, and that's helping us keep the risk, the business model risk down because we're only choosing the best names in the space in terms of who can go in our facilities. We could do more if we had more space, quite frankly. There's a lot of demand out there for it. But because we're still curating our customer mix, we're still trying to make sure that we have that right balance of cloud, networks and enterprises, and I would throw inferencing in, is kind of the new fourth characteristic of it, but because we're still curating that mix, we're just not taking everything that comes in the door.

Our final question comes from the line of Richard Choe of JPMorgan.

I wanted to ask a follow-up on the data centers. What kind of renewal pricing are you seeing and overall pricing for new business? And then back to the tower business, if you can give us a sense of what kind of pipeline of applications you are seeing? And has that shifted at all? And at some point, do you see it kind of inflecting higher?

Sure. So I'll start with the data center question -- I'm sorry, pricing. So we're seeing generally higher pricing. Our mark-to-market continues to exceed what it's been over kind of historical norms on that. So that's a really good indication of that. And then the market level pricing does continue to rise because there is this imbalance between supply and demand. And so I don't have any specifics for you in terms of percentages there. We're not putting that kind of level of information out there. But overall, it's going up. And that's enabling us to continue to underwrite mid-teens or better returns on the new incremental capital that we're putting in. Because as we're creating that new space, even though you do have a little bit of cost pressure from inflation, tariffs, things like that, we're able to pass that through in the form of higher pricing to keep those stabilized returns kind of in that mid-teens range. So we feel very good about that over time. In terms of the application pipeline on towers, we are expecting slightly fewer number of total applications this year. But that's not reflective of anything other than a couple of the carriers are largely through their initial 5G overlays. And so a lot of that was kind of amendment business that was likely covered in a lot of our holistic agreements anyway. So there's no real readthrough on that in terms of customer demand or property revenue. And in terms of an inflection, what we're seeing as an inflection is a higher number of new colocations coming in which is very positive because those come in at higher revenue per transaction than the amendments do. But I would say, again, it's an overall consistent level of activity year-over-year, and it's consistent with what we expected to see at this point. And we expect it to be at that level or better in the future as they switch to densifying their networks.