Crown Castle Inc (CCI) — Q4 2019 Earnings Call Transcript

Thank you, Katie, and good morning, everyone. Thank you for joining us today as we review our fourth quarter 2019 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, which we will refer to throughout the call this morning. This call will contain forward-looking statements, which are subject to certain risks, uncertainties and assumptions, and 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, February 27, 2020, and we assume no obligation to update any forward-looking statements. As you saw from our press release yesterday, we've restated our historical financials. Of the financial information we discuss in this call includes the expected effect of the restatement. 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, Ben, and thank you, everyone, for joining us on the call this morning. As you saw from our results, we closed out another year of solid growth in 2019, which included generating the highest level of tower leasing activity in more than a decade. I believe our strategy and unmatched portfolio of more than 40,000 towers and approximately 80,000 route miles of fiber concentrated in the top U.S. markets has positioned Crown Castle to generate growth in cash flows and dividends per share, both in the near term and for years to come. Dan will discuss the results for full year 2019 and the full year 2020 outlook in a bit more detail. So I want to focus my comments this morning on two key points. First, we expect 2020 to be another year of significant growth in cash flows and dividends per share. And secondly, I'm excited about the long runway of growth for Crown Castle as we are sitting on the doorstep of another investment cycle by our customers as they deploy 5G. On the first point, we expect to grow AFFO per share in 2020 by approximately 8% supported by similar levels of growth in our tower and fiber segments when compared to 2019. We expect the elevated level of growth that we experienced in 2019 to continue with similar levels of tower leasing this year as our customers respond to the ongoing growth in mobile data demand. Uncertainty around the outcome of the pending merger between T-Mobile and Sprint caused a decrease in activity during late 2019 and early 2020. However, we believe this slowdown will ultimately prove temporary and short-lived as we anticipate a significant increase in industry activity in second half of this year as clarity around the merger drives a ramp in 5G investments. Within our small cell and fiber businesses, 2019 was a terrific year as we've successfully deployed approximately 10,000 small cell nodes, making it the highest year of production in our company's history. We expect to deploy another 10,000 small cell nodes this year as we continue to respond to the significant increase in demand from our customers while, at the same time, navigating ongoing hurdles that remain challenging with many municipalities and utilities. We finished 2019 with more than 40,000 small cells on air and another approximately 30,000 in our construction pipeline as we remain the leading U.S. small cell provider in terms of scale and capabilities. Adding to the returns we are generating from attaching small cells to approximately 80,000 route miles of fiber, we generated 3% revenue growth from our fiber solutions business in 2019, and we anticipate similar levels of growth this year. We see a path to further improve our returns over time by sharing the same fiber asset across this larger addressable market of fiber solutions customers that require high bandwidth connectivity, including large enterprises, healthcare institutions and government agencies. Simply put, 2019 was a great year of growth. And 2020 is shaping up to be similar, albeit potentially more back-end loaded than we previously expected. And as excited as I am about 2019 and 2020, I'm even more excited about the bigger picture. We have positioned Crown Castle with the right assets in the right market with market-leading capabilities to deliver value to our customers and generate shareholder returns for decades to come. As is often the case, the natural tendency is to overestimate what is possible in any given 12-month stretch while underestimating the dramatic change that can occur over a 10-year period. Looking back over the last decade, we have significantly expanded our tower business from approximately 22,000 towers in the U.S., generating approximately $1.5 billion in annual site rental revenue in 2009 to where we are today with 40,000 towers generating nearly $3.5 billion in annual site rental revenue. We also established a common stock dividend during that time that provides a consistent return of capital to our shareholders, currently totaling $2 billion on an annual basis or nearly 35% of our total revenues. Further, we have built a market-leading position in the small cells industry and have invested approximately $15 billion of capital to establish fiber footprints in prime locations across the top U.S. markets where we see the greatest long-term demand. While making those significant investments in assets and capabilities that we believe will expand our future growth opportunity as 5G is deployed, our equity market capitalization has increased from less than $10 billion to over $60 billion, generating a compound annual total return of greater than 18% for our shareholders during the last 10 years. And the combination of the market dynamics and our unique portfolio of assets sets us up for a long runway of continued growth as the wireless industry embarks on an investment cycle to deploy 5G. This has the potential to make the next 10 years look a lot like the last 10. The current demand environment that is generating the highest levels of tower leasing activity in more than a decade is largely tied to our customers investing heavily in their 4G networks to keep pace with the 30% to 40% annual data demand growth. On top of that continued investment, we anticipate significant long-term demand for our infrastructure as 5G becomes a reality and wireless networks expand from connecting everyone to connecting everything. Adding to my optimism, I believe recent industry developments will help to accelerate the deployment of 5G in the U.S. We believe the new T-Mobile, along with AT&T and Verizon, are in a great position to leverage their scale and valuable spectrum assets, ultimately promoting more investment across the industry. Adding to the opportunity, this is the first time in more than a decade that we have had visibility into a potential new customer entering the wireless market at scale with DISH networks looking to deploy nearly 100 megahertz of spectrum over the next several years in order to compete with the established operators and meet significant build-out requirement. And finally, there are several large spectrum auctions on the horizon that we believe will bode well for the future tower and small cell demand. With our unmatched asset base and expertise operating in the best market in the world for communications infrastructure ownership, I believe Crown Castle is in a great position to capture these substantial long-term opportunities and consistently deliver a return of capital to our shareholders through a high-quality dividend that we expect to grow 7% to 8% annually. And with that, I'll turn the call over to Dan to go through some of the more specifics of the quarter and the last year.

Thanks, Jay, and good morning, everyone. We delivered another great year of financial performance in 2019 with several highlights. We grew dividends per share by approximately 7%, reflecting the underlying growth in our business and our commitment to returning capital to our shareholders. We generated the highest level of tower leasing in more than a decade. We accelerated the deployment of small cell nodes by delivering approximately 10,000 small cells last year, the highest annual production in our history, and we continued to improve our financial flexibility by increasing commitments under our revolving credit facility to $5 billion while also lowering our weighted average borrowing cost and increasing the average maturity on our debt by refinancing $1.9 billion of debt at attractive long-term rates. As I walk through our full year 2019 results and our updated outlook for 2020, please note that, where applicable, all financial figures reflect the impact of the restatement we disclosed in our earnings release yesterday, which I will discuss shortly. Turning to our full year 2019 results on Slide 3 of the presentation. Relative to the midpoint of our prior outlook, the outperformance in site rental revenues was primarily offset at the adjusted EBITDA and AFFO lines by lower contribution from services tied to a slowdown in activity during the fourth quarter. As Jay mentioned, uncertainty around the outcome of the pending merger between T-Mobile and Sprint led to lower activity levels in late 2019 that we believe will continue through early 2020 before rebounding later this year. As a result, we expect our financial performance in 2020 to be more back-end loaded than we previously anticipated, particularly in our services business. Turning to Slide 4 and looking at full year 2019 in more detail. Site rental revenues increased by $298 million, inclusive of $290 million in organic contribution to site rental revenues. That equates to just over 6% growth. That 6% growth is comprised of approximately 6% growth in towers, 17% in small cells and 3% in fiber solutions. Moving on to investment activities during the year. We deployed approximately $2.1 billion in capital expenditures, including $1.9 billion of revenue-generating capital expenditures comprised of $1.4 billion in fiber and approximately $450 million in towers. Additionally, during 2019, we returned significant capital to our shareholders through our quarterly common stock dividend totaling $1.9 billion in the aggregate or $4.58 per share, representing growth of approximately 7% compared to full year 2018. From a balance sheet perspective, we ended 2019 at approximately 5.5x debt to EBITDA. We remain committed to our investment-grade credit rating and anticipate a glide path back to our target leverage of approximately 5x by the end of 2020 based on the expected EBITDA growth throughout the year. Turning to our full year 2020 outlook. Starting on Slide 5 of the presentation. You can see our outlook remains unchanged, apart from the effect of the restatement. To wrap up, 2019 was another very successful year for Crown Castle. We are excited about the growth opportunity going forward as 5G deployments are just beginning and are expected to drive significant demand for our tower and fiber infrastructure. Currently, we are seeing the benefits from the investments our customers are making in wireless networks to keep pace with increasing data demand, which allows us to provide near-term returns through a high-quality dividend that we expect to grow 7% to 8% annually. At the same time, we're making significant investments in our small cell and fiber business that we believe will position Crown Castle to take advantage of the long-term growth trends Jay discussed earlier and generate shareholder returns for decades to come. Before opening the call up to questions, I'd like to spend a minute addressing the restatement of our previously reported financial statements as described and detailed in our press release yesterday. In connection with our year-end procedures and after receiving the previously disclosed subpoena from the SEC, we engaged in a review internally and with our independent auditors of our accounting policies for our tower installation services. Following that review, we decided with our auditors to seek additional input from the Office of the Chief Accountant of the SEC, also referred to as the OCA, regarding whether a portion of our tower installation services revenues should be recognized over the term of the lease of the installation work associated with. After consulting with the OCA, we determined that our historical practice of recognizing the full transaction price as service revenue upon completion of the installation was not acceptable under GAAP. Instead, a portion of the transaction price for our installation services, specifically the amounts associated with permanent improvements recorded as fixed assets, represents a modification to the lease to which the service work is related and, therefore, should be recognized on a ratable basis as site rental revenues over the associated remaining lease term. To be clear, this restatement only impacts the results in our tower segment and has no effect on our fiber segment. It is important to note two key facts as it relates to the restatement. First, over the term of customer lease contracts, we will recognize the same cumulative amount of total revenue and total gross margin as our historical practice. And second, the new accounting treatment will have no impact on our net cash flows, our business operations or our expected dividend per share growth going forward. As noted in our release, our consultation with the OCA was not part of the previously disclosed SEC investigation or subpoena. Based on our internal review, we continue to believe that our capitalization and expense policies, which were the subject of the subpoena, are appropriate. We will, of course, cooperate fully with the SEC, including in connection with the review of those policies. With that, Katie, I'd like to open the call up to questions.

A couple, first on the restatement. Can you give us some examples of the types of projects where the accounting on revenue has changed and why you're confident that the cost should be capitalized versus expensed in a ratio where they are today? And then second, can you give us any update on progress in the small cell business on applications and permitting timing?

Sure, Phil. I'll address the first part of your question and leave the second part to Jay. The projects we are discussing involve service work where we install new equipment on our towers. To do this, we need to make some permanent improvements to the towers, such as adding brackets or mounts to secure the antennas. By making these improvements, we create a permanent asset that we believe enhances the tower's value and revenue potential in the future. Consequently, we believe these improvements should be capitalized since they are permanent. This capitalization results in deferring revenue related to the services work associated with the capital improvements. So when we add that equipment to the tower, which increases its value, we defer the related revenue and capitalize the associated costs.

And does this now more closely match the expensing versus capitalization on the cost side?

Yes. It does. It defers the revenue and capitalizes and then it depreciates that over time.

How are those roughly related in terms of size?

You can see that the capital for 2019 was around $210 million. That's the revenue we are removing from the services business. Additionally, we have about $110 million from the amortization of prior and 2019 work related to what we had done historically, which we then amortize through 2019.

Sorry, I meant more on the cost side, the relative ratio of expense versus capitalization of those costs. How are those similar or different to the revenue side?

I'm sorry, Phil, I'm having a hard time following the question. The relative ratio of the capitalization to the expense?

On the cost side, how much of the costs you're generating are you capitalizing compared to expensing immediately alongside the revenue?

The amount of cost we are capitalizing is slightly greater, on a percentage basis, than the revenue we are deferring, mainly because some of that carries a margin. It's quite close, but the percentage of cost capitalization is probably a bit higher compared to deferred revenue as a percentage of total revenue. Overall, they are pretty similar.

Okay. And then on the applications and permit timing?

Yes, I'm glad to address that. I believe we made significant strides in 2019 by overcoming various challenges related to municipalities and utilities, and we've improved our approach to project execution and the preparatory work needed to launch new projects in different markets. The feedback and support we received from the FCC this past year was beneficial, especially in markets where we faced significant delays. They clarified the timelines and costs associated with the process, which was helpful. As I mentioned in previous calls, this support was crucial in allowing us to deploy 10,000 nodes during 2019, which is the highest production level in our company's history. Our team worked incredibly hard to reach this scale, and we expect 2020 to be similar, with another 10,000 nodes planned for this year. We are also continuing to address broader challenges related to working with municipalities and utilities. It's important to note that many of these hurdles may never fully disappear. I don’t foresee a time when we can complete these small cell nodes in less than 18 to 36 months, on average. The timelines tend to be lengthy because we need to collaborate with local governments and utilities to navigate existing infrastructure and ensure compliance with community aesthetics. The barriers to entry are quite high, much like in the tower business. Long term, we will continue to face challenges in deploying these networks quickly, as this is largely about building relationships with the municipalities and communities where we are placing this infrastructure and ensuring a balance between the need for the infrastructure and the community's aesthetic preferences.

Great. Continuing on the small cell, could you share some insights into same-store sales, what's happening with nodes per mile, and your ability to lease them? Is there any change in permitting if you have the fiber already deployed and someone wants to add another node? Additionally, regarding the enterprise fiber business, are there any modifications in your go-to-market strategy or sales force to help boost bookings?

We are observing consistent levels of lease-up for our existing assets, similar to what we've discussed over the past few months. There remains strong interest from wireless operators in utilizing the fiber we initially built for a primary tenant, particularly in the top 10 markets in the U.S., where most of our investment has occurred. Activity and lease-up on that legacy fiber continue to be significant. The density of new nodes, when a carrier requests additional fiber construction, has remained relatively stable, averaging around 2 to 2.5 nodes per mile. When carriers return to utilize existing fiber, the build-out patterns across the country are comparable. Looking ahead, particularly considering the growth in 4G and the anticipated developments in 5G, we are optimistic about the potential of our assets over the long term. Reflecting on the early days of tower constructions 20 years ago, we anticipated some lease-up, but the reality vastly exceeded our expectations. We are approaching our small cell investments with current data traffic levels in mind, and given the ongoing data growth of 30% to 40% per year, combined with the upcoming 5G technology, we believe the actual asset usage will likely surpass our underwriting criteria. This creates substantial opportunities for returns that exceed our initial assumptions. Everything we are observing in today’s market, including the activity we noted in 2019 and ongoing discussions in 2020, along with early signs of real 5G implementation, reassures us that lease-up is on the way, and the density we anticipated for these assets will meet or exceed our expectations.

And just building on that a little bit, Simon, we're still building about on activity levels, about 70% anchor builds and 30% lease-up. So we're continuing to add miles of fiber to accommodate the type of lease-up Jay is talking about. And what we think we're doing is getting those as kind of the first-mover advantage. So when we go into a market, we build that fiber, we're there. It does provide us a competitive advantage for getting that lease-up. And if we're right that the lease-up could be significantly more than what our underwriting has been, then we're just building more and more of that upside into our asset base as we go along. And therefore, one of the questions you asked is the node per mile, that really hasn't changed much. It's been in that neighborhood of 1 node per mile, on average, if you look at just the fact that we have 80,000 miles and about 70,000 nodes on air under construction. It hadn't changed much because we keep building miles of fiber that we think is just embedding more and more of that upside that Jay was talking about.

Simon, did you have another question on that topic before I go on to fiber solutions?

No. No. Go on, on the fiber, yes.

On the fiber solutions side, we continue to improve our operations. As previously discussed, we are projecting a similar level of net growth in 2020 as we experienced in 2019, which is approximately 3% growth for the business. We believe that these assets represent an appealing source of revenue as they enhance the overall yield. We are becoming more adept at identifying and developing future tenancy, and our strategy is unfolding as planned. Our primary focus is on the small cell opportunity, which we consider the key driver of long-term returns from the fiber business. In pursuing this major opportunity, we also aim to install enterprise clients on the same fiber network, and this is progressing in line with our expectations from last year into 2020.

It's Josh on for Dave. Verizon said they expect to deploy about 5x more small cells in 2020 versus last year. Maybe if you could provide any sort of insight as to what role you're playing in that, and maybe you can also speculate on how you think that's possible kind of given the permitting issues and the steady deployment volumes you guys are having. And then secondly, is there any relationship between the SEC subpoena and the restatement you did yesterday?

Regarding your first question, Josh, while I won't provide specific comments on Verizon or their plans, I encourage you to reach out to them for more detailed insights on their expectations for this year. Broadly speaking, their comments align with what we've heard from all wireless carriers. The current pace of small cell installations falls far short of the long-term demand and necessity for them. Today's network realities, shaped by data traffic and available spectrum, highlight the essential role of small cells in achieving the required network density. While macro sites are the most cost-effective means to deploy spectrum and manage data demand, they can't provide the density needed to meet total market demand, which is why small cells are critical. You'll likely find similar sentiments from Verizon and other carriers regarding the large-scale need for small cells. Many industry observers suggest that over 1 million small cells will need to be built in the next decade, and we certainly share that view. However, I want to clarify that we don't assume we'll maintain our historical market share. We've captured about 50% of the small cell market activity. We're not going to lay fiber in every U.S. location to keep up. Based on carrier feedback, there's an expectation that they will install small cells beyond the top 100 U.S. markets in terms of total node count, although we may not follow them to every market. Most of our investments have been concentrated in the top 25 markets, especially the top 10, and moving forward, our revenue growth and investment will likely be focused more on these top markets than on broader U.S. locations. As the total addressable market expands, we'll be selective in our capital investments, which will influence how many nodes we ultimately place on our fiber. The trend aligns with my earlier remarks about the increasing demand driven by 4G and 5G technologies, necessitating a substantial rise in small cell deployments compared to our current levels. When I hear our customers discuss this growth potential, it reinforces my optimism not just for 2020 but for the next decade, as we are well-positioned to capture a significant share of that growth in both towers and fiber as investments in small cells increase.

Yes. Josh, regarding the second part of your question about the relationship between the subpoena and the restatement, they are not related. The subpoena process is ongoing, and we will provide updates when available. It concerns our capitalization policies related to our installation services business. After receiving the subpoena and during our year-end review, we examined all policies regarding our installation services. We are still comfortable with our capitalization policies. However, we discovered that a specific aspect of the installation services revenue recognition was complex enough that we sought input from the Office of the Chief Accountant of the SEC. This led us to realize that a historical approach we had taken was not acceptable, resulting in the need for a restatement. So, while the subpoena and the restatement are not connected, the subpoena did guide our decision to consult with the OCA.

I wondered if you could maybe give a little bit of an update on edge compute and Vapor and kind of thoughts around those opportunities. And then as it pertains to your tower business, and I think we're still kind of working through the math as it pertains to the restatement, but you appear to be guiding towards a healthier leasing year this year compared to last year despite it being back-half weighted. So am I correct in that assertion? And if so, what's driving that given the back-end nature of this year?

Sure. Regarding the first question about edge and Vapor, we invested in Vapor IO a couple of years ago. They specialize in edge computing by enabling small-scale data centers at the very edge of the network, specifically at tower sites. We saw significant potential in managing data traffic and lowering costs as traffic increases and more people connect at the edge. With the transition to 5G, we expect further opportunities for both computing power and connectivity at the network's edge. Our tower sites are well-suited to capitalize on this potential. They provide necessary space for equipment and colocation facilities at the network edge. A substantial share of wireless traffic will be handled through these towers, making it logical to place these facilities at those sites. Another part of our business that will benefit is our fiber infrastructure. Fiber is essential to wireless networks as we advance into 5G, just as we’ve seen with 4G. As the industry moves toward C-RAN and O-RAN, connecting sites with fiber, including both macro sites and small cells, will be crucial. We believe our assets will appreciate in value as these trends unfold, and the synergies between our tower and fiber assets create unique opportunities such as with Vapor, which is why we invested. A word of caution, though; it’s still early days. We're generating revenue from Vapor and have set up various colocation facilities, but this is not yet a significant part of our overall revenue and isn't impacting our guidance at the moment. However, we believe the potential returns justify our investment, and there could be meaningful outcomes down the line if market conditions align. Now, regarding your second question about the tower update, Dan can provide more specifics if needed. The change in tower revenue relates to the restatement effects. We have kept our total leasing activity assumptions for 2020 consistent with our previous guidance. We are adjusting to be more back-end weighted, but we don't actually see a year-over-year slowdown compared to 2019, which is encouraging. As mentioned, 2019 saw the highest level of tower leasing activity in over a decade, and we believe 2020 is shaping up to have similar activity levels. The uplift you’re noticing seems to stem from the restatement and the amortization of deferred revenues that Dan discussed earlier.

And just to add a little more color to that, Jonathan, we're able to knock off some of the lower performers as we get through the restatement process too. So we think that gives us a better outlook for 2020 versus 2019. But we also, even with the back-end loaded performance, have enough visibility on the leases that we have. And so your general view on that being more optimistic certainly is what we're seeing.

The first question is about the economics of the network services business. For every $100 a customer pays for the complete range of network services, including any upgrades to your towers, how much of that remains with you after accounting for all investments and expenses? The second question refers to your glide path to return to a 5x net debt leverage. Have you considered using equity to speed up that process or to provide more flexibility for additional investments?

Yes. So on the first one, Mike, I would say some in the neighborhood of $40 is what we would make after that $100, if you just say that's the cash that we come out with regarding the services business.

And the glide path, have you contemplated using equity to try and accelerate that glide path or increase your flexibility to augment other investments?

Yes. What I would say is we're committed to investment-grade rating, and we will get back to our 5x debt to EBITDA. We believe, as we pointed out, that based on the year that we have and the EBITDA growth that we expect throughout 2020, that we will get back to that 5x by the end of the year. If something changes, then we might have to use equity, as we've always said, as a way to fill any gap that happens between the amount of capital expenditures and dividends we have and the amount of leverage capacity we generate by increased EBITDA and the cash flow we generate through our AFFO. So we always have that option open, but we believe, at this point, that we have leveraged capacity and that the glide path works. But we are committed to investment-grade rating, and we'll do what we need to do in order to maintain it.

Mike, you raised a valuable question about taking a step back, which I think is helpful. I'd like to expand on that beyond just the economics of our services and discuss how we view services and site rental aspects of our business. We fund our current dividend through the recurring cash receipts from site rentals. The services, including the deferred revenue amortization that Dan mentioned earlier or prepaid rent, are all activities that support our ability to increase the dividend over time. As has been noted several times, we believe we can grow the dividend by 7% to 8% annually in the long term, based on the positive trends in the industry that I’ve highlighted during the call. However, achieving this growth will require us to invest in CapEx to enhance our existing assets and develop new ones. We pass some of these CapEx expenses onto our customers through upfront payments and services, which effectively lowers our net required investment for growth. These upfront payments are not essential for funding our dividend. Therefore, we don't consider the cash margin from these payments as part of our dividend funding. In simpler terms, even if our business growth were to cease completely, I would anticipate that our dividend would remain stable. In that case, CapEx-related reimbursements and services would also decrease significantly from current levels. Nothing in this restatement alters that situation. Historically, we have determined our dividend and its payout based on the recurring cash aspects of our business, while more volatile elements, or net cash, serve as indicators of growth, helping to offset the necessary net capital investment.

Two quick ones, I think, on the restatement and then one more strategic question. On the revenue side of the restatement, it goes into amortization of prepaid rent, amortization of deferred rent. Am I right in thinking that, that level in 2019 was now like $460 million, up $50 million year-over-year? And how much do we expect amortization of prepaid rent then would grow from 2019 to 2020?

So first, the answer to your question is, yes, you're right. It goes through prepaid rent amortization. And to further, yes, you're right for $460 million in 2019, and that is $50 million of growth over 2018. We would anticipate that it grows in the neighborhood of $60 million to $65 million again into 2020. So the amount of prepaid rent amortization into 2020 is around $525 million.

Makes sense. And the other question, the restatement, is on the cost side. To Phil's question, you talked a little bit about how it is, capitalize those items you put on because it does help make the tower a better asset long term. But I assume that CapEx goes into growth CapEx and then gets depreciated, which would not be an AFFO, correct?

That is true.

Okay. And then the more strategic one for Jay or whoever, CBRS is another interesting topic. You've talked about auction bands coming out. A lot of people looking at the CBRS auction coming up in June as a potential to see more indoor systems developed. What is your thoughts on CBRS and indoor? And would that be someplace that you might put capital to work? Or is there just so much opportunity in outdoor, not that interested?

Yes. We're really excited about the long-term opportunity that CBRS brings. And I would say we think there's opportunity, both indoor and for outdoor applications. I think initially, you're right, and the bias will be towards indoor applications initially. It's going to be interesting to watch these private auctions for some of these licenses and what the opportunity is there. We also think that the open spectrum, the public spectrum that will be deployed, that there's opportunity for us to use the infrastructure that we own, both on the fiber, small cell and tower side, that there's going to be opportunities around that long term. The component of CBRS, that's the same. As you know, Ric, from everything that's happened in the past in terms of the deployment of spectrum is the broad deployment of spectrum, regardless of what name it gets, needs the infrastructure that we own. So we look at it and think maybe, initially, it may be more biased towards indoor, but we think there's outdoor applications. And over time, that spectrum will be used to meet the growing demand for mobile data, and that's likely to benefit well beyond venues or indoor applications.

First, as we look ahead, is there any reason to think that you won't be able to monetize the coming wave of integration-related amendment activity at the same level as your peers? I know you don't like to comment on specific customers. But to put a finer point on it, I'm trying to understand if the capacity rights you granted to T-Mobile back in 2012 as a part of that acquisition remain a relevant consideration today.

Sure, Nick. I don't see any reason why we wouldn't be able to monetize that. It's been a few years since this was a major topic during our earnings calls. For the most part, we've utilized the space rights that our anchored tenants had on those transactions. I believe it was around 2018 when we began discussing that, and nearly every new tenant we've had since then has generated additional revenue as they connect with the tower, and that's still true today. We haven't engaged in any transactions that would have altered that. As we continue to see incremental touches, whether through first-time installments or new amendment activities, that will drive revenue. Our results and expectations for 2019 and 2020 indicate that these higher levels of activity are due to both amendments and new first-time installs.

Okay. That's good to hear. Second one on small cells. Can you talk a bit about the mix of spectrum bands that are underpinning the small cell bookings that you're taking today versus what it was a few years ago? I guess I'm trying to understand the degree to which deployment, just given the pipeline today, are primarily in higher frequency bands like millimeter wave versus midband.

I think initially, when we first went into the business, many people thought that the only areas where spectrum bands would be used on small cells were the millimeter wave. The reality is the carriers are using the small cells across all of the bands that they owned. And we will see in a, given geography, a carrier will oftentimes start their initial deployment with one spectrum band, build out small cells for that band and then come back and add additional nodes across that same run of fiber, thereby providing lease-up for us, if you will, and adding additional spectrum bands beyond their initial deployment. But the vast majority, as we talked about the number of small cell nodes that we both have built, the 40,000 that we've built and the 30,000 that are in the pipeline, almost all of those would be midband spectrum. We're not at the point yet where we're deploying significant numbers of millimeter wave small cells. I think that would be unmodeled upside, if you will. When we talk about the fact that we underwrote 4G deployments and build-outs, that's under the assumption that small cells would only be used for kind of midband spectrum. So millimeter wave and things that will be used for 5G, those are unmodeled upside in the way that we underwrote the assets.

Yes. I think part of that, Nick, is because the carriers use small cells to offload some of the tower congestion. So whatever bands are on those towers can go onto the small cells because when a lot of people gather in one area, it takes up all the capacity the tower has, and putting small cells there then allows the tower to become useful again. And that does require a similar band and similar coverage that would happen with the tower. So we're seeing it across all spectrum bands, as Jay is talking about.

A couple of questions, first on the restatement. Can you just explain why the amortization of the tower installation work goes onto the site rental side as opposed to staying on the service side? And how much of the 5.9% organic growth for the quarter came from that amortization piece? And then secondly, you mentioned the services business being a little bit lumpier and soft right now. Can you talk about the profitability of it in the quarter, if there were any onetime expenses that really lowered the margin contribution? And lastly, on churn. I think churn picked up about 50 bps sequentially. Any items to call out there?

Sure, Batya. I'll begin with your first question regarding site rental revenue versus services. Since the services we are now deferring are considered a modification of the lease, they are classified as part of site rental revenue rather than services. This is due to the specifics of accounting practices. We discussed this with our auditors and the OCA because it involves a detailed accounting treatment. Ultimately, once we defer these revenues, we recognize the amortization under site rental revenues. As for the growth in site rental revenues for the fourth quarter, it is likely around 50 basis points related to this restatement.

You want to do services profitability?

On the services side, there were no one-time events in the quarter that positively or negatively influenced our reported outcomes. The profitability seen in our historical statements is what we anticipate moving forward. Regarding the churn question you asked, we discussed it last quarter. Most of the churn has been processed, and while we received notifications, it hasn't yet been reflected in our results. Therefore, the increase was expected. Similar to my earlier comments about revenue buckets, these would include leases that ended late in the year and will impact the full year in 2020. After 2020, we expect churn rates to decrease back to the lower end of our guidance of 1% to 2% annually. We are addressing the final stages of consolidation churn from several years ago and believe it will return to a more normalized level thereafter. And just to be clear on that, that is all tower churn that we're talking about there. So that increment came because the activity was on the towers, and everything that Jay said was around the tower side of business.

Can we focus on the fiber for a moment? Can you discuss how you plan to utilize the conduit and fiber on the ground, potentially alongside other services? What are your observations regarding the competitive landscape? When you mention other services, are you looking to expand in areas that connect to office buildings and apartment complexes? Are you seeing any instances of overbuilding in the regions where you are currently expanding? I also have a quick follow-up question.

Sure. When we consider fiber projects, we evaluate the overall potential return on that fiber. The main factor driving our fiber strategy is our assessment of what will be required for wireless networks, particularly for small cells. This understanding helps us determine which markets and areas in the country we are interested in for fiber assets, primarily based on the wireless potential. Following that, we aim to maximize revenue along that route from various entities such as universities, hospitals, large financial institutions, and other businesses that might need that fiber. However, our strategic decisions regarding location are influenced by our evaluation of wireless needs. From that point, we strive to optimize returns on our assets. Engaging as many customers as possible to utilize those assets increases yield and return, making it a worthwhile pursuit for us. We are not encountering significant overbuilding since the costs to construct these assets are quite high. Generally, if existing assets are available, companies tend to invest their capital in different locations where assets are lacking. It is becoming increasingly clear that the amount of fiber needed for small cell deployment cannot be met by current infrastructure. This necessitates substantial investment across the U.S. in areas where we intend to build, and we will not pursue dense, high-capacity fiber in areas lacking such resources. Thus, our growth strategy is significantly focused on opportunities to develop new fiber because we find limited chances to acquire existing fiber, as it simply isn’t available.

And just to clarify, does your 50% flow share on small builds pertain to areas where you have infrastructure, or do you believe it reflects the entire market?

We think it's the entire market as we've measured it over the last several years. And we have time for one more question, operator.

So you disclosed your total share of revenue from your tenants. You've got about 20% from each of Verizon, AT&T and T-Mobile, about 15% for Sprint. I was wondering if you could provide some color on how that breaks down for small cells specifically. And do you skew towards any of the carriers?

The general answer is that it's quite similar for small cells as it is for the tower side. We have observed activity from all the carriers regarding small cells.

Okay. And then one follow-up. It seems that T-Mobile and Sprint are likely to shift their focus towards macro towers over the next few years as they integrate their networks. Could you just provide some thoughts on how the merger impacts your view of your ability to ramp your small cells backlog over the next couple of years? Would it potentially stall growth? Or are you seeing enough demand outside of Sprint and T-Mo that you can grow through it?

Yes, I think there are two key factors to consider regarding the long-term outlook for Sprint and T-Mobile. They have indicated that they will streamline some sites where there is overlap, meaning that some macro sites may be rationalized over time. We have long-term leases in place, and some of those sites might be adjusted as they evaluate their network needs. However, we believe they will require many new macro sites to support the spectrum they will acquire from Sprint as part of the merger, which is not currently in use. They will also likely modify their existing spectrum broadcast. This will necessitate a mix of macro sites and small cells, and we are well-positioned to benefit from both. Our assets are aligned with the industry's demand for 4G and 5G expansion, and we are confident in our position with T-Mobile to assist in their 5G rollout. I appreciate everyone joining the call today. We anticipate 2020 to be another year of significant growth in cash flows and dividends, and we are excited about the longer growth trajectory as we prepare for a new investment cycle driven by 5G. Thank you, and I look forward to speaking with you soon.