Crown Castle Inc (CCI) 2016 Q4 法說會逐字稿

Good day and welcome to the Crown Castle International fourth-quarter 2016 earnings call. Today's conference is being recorded. At this time, I would like to turn the conference over to Son Nguyen. Please go ahead, sir.

Thank you, Alicia, and good morning, everyone. Thank you for joining us as we review our fourth-quarter 2016 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 investor section of our website at CrownCastle.com which we will refer to throughout the call this morning.

This conference 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 section of the Company's SEC filings. Our statements are made of as today, January 26, 2017, and we assume no obligations to update any forward-looking statements.

In addition, today's call includes discussions of certain non-GAAP financial measures. Tables reconciling these non-GAAP financial measures are available in the supplemental information package in the investors section of the Company's website at crowncastle.com. With that, I'll turn the call over to Jay.

Thanks, Son, and thank you, everyone for joining us on the call this morning. As you saw from our earnings release, we finished 2016 on a strong note, delivering another quarter of great financial results. During 2016, we generated AFFO per share growth of 10%, and increased our dividends per share by 8%. These results exceeded our long-term annual growth target of 6% to 7%, as well as our initial expectations for 2016. We generated this growth by focusing on serving our carrier customers' needs as they seek to enhance network quality and capacity to meet the increasing demands for wireless connectivity.

Additionally, during the year, we continued executing on our strategy of allocating capital to build and acquire assets that we believe will drive long-term growth in our AFFO and dividends per share. Towards this end, we strengthened our leadership position in wireless infrastructure, both by adding towers through the TDC acquisition, as well as by increasing our fiber footprint and small cell capability by completing the integration of Sunesys and the acquisition of FiberNet, which closed earlier this month.

I'd like to thank all of our employees, both FiberNet and Crown Capital, who are working hard to ensure that the integration process goes well. Our early view of the operating performance of FiberNet is consistent with our underwriting assumptions for the acquisition.

Turning back to our results, as Dan will talk about in more detail, we increased our 2017 outlook to reflect higher expectations for our base business, as well as the inclusion of FiberNet.

Looking out beyond 2017, we believe a steady and healthy leasing environment, driven by the continued significant growth in mobile data, positions us to deliver long-term growth. Mobile data traffic in 2017 is expected to be roughly twice what was consumed just two years ago in 2015. And by 2020, mobile data traffic is expected to be 6 times 2015's level, with the growth driven by a combination of factors, including the adoption of data-intensive applications, such as video streaming, and increasing number of connected devices and potential new applications, such as fixed wireless broadband.

In order to capitalize on this anticipated growth in mobile data, we believe that wireless carriers will continue to enhance and upgrade their network through cell site densification and deployments of additional spectrum for years to come. As the leader in shared wireless infrastructure in the US, we believe we are well-positioned to capitalize on the positive trends that we see in the industry, as the shared infrastructure model continues to represent the most efficient and cost-effective way for our carrier customers to deploy wireless infrastructure.

With our recently completed acquisition of FiberNet, our portfolio of wireless infrastructure across the top [tier] metro US market now consists of approximately 40,000 towers and 26,500 route miles of fiber supporting small cells. Towers remain the core element of wireless networks and the first option in network deployment, providing both coverage and capacity. Consistent with this, we anticipate a long runway of growth for our tower business driven by technology upgrades, deployment of additional spectrum, and cell site densification.

Importantly, there remains a substantial amount of spectrum yet to be deployed by our carrier customers across our wireless infrastructure, including AWS3, the build out of FirstNet, and spectrum from the current broadcast incentive auction. Additionally, our portfolio provides a substantial opportunity to drive organic growth and meaningful incremental returns from additional tenant leasing across those towers. Building on the foundation of a great tower business, we are investing in our small cell business, which we believe could represent an opportunity similar in size and returns as towers.

Today, our fiber footprint of 26,500 route miles provides us with significant growth potential in top metro markets such as Boston, Baltimore, DC, Chicago, Los Angeles, New York, Philadelphia and San Diego to name a few. In addition, FiberNet substantially strengthens our footprint in Miami and Houston, both markets where we are seeing significant small cell demand.

The route uniqueness, density, and capacity of our fiber footprint, combined with our extensive real estate, network engineering, and node construction capabilities provide us with meaningful opportunities to generate attractive long-term returns. Our returns on small cells increase like towers as a result of the shared economic model and our ability to leverage our initial fiber investment for multiple tenants.

Today we are building small cell systems with initial yields of 6% to 7% that increased to low double digits with the second tenant and higher yields with the third and fourth tenants. Given the growth in small cells, as our carrier customers densified in network and deployed spectrum closer to their customers, the lease up of our fiber assets is occurring faster than we experienced on our towers at a similar stage of maturity.

With faster lease up and higher initial yields in towers, we have good reason to be optimistic about the future of our small cell investments. As the only shared wireless infrastructure provider with expertise at scale across both towers and small cells, we believe we are uniquely equipped to help wireless carriers deploy their networks. This enables us to drive organic growth and significant incremental returns on our existing assets while deploying capital towards attractive opportunities that we believe will extend and enhance our long-term growth in AFFO and dividends per share.

Our strategy, focus, and disciplined approach related to execution and capital allocation have driven significant growth in value creation, delivering over a decade of consecutive AFFO-per-share growth.

Looking ahead, I'm excited by the opportunity from prospects for Crown Castle. I believe we are well-positioned to create shareholder value with a unique combination of meaningful growth and high-quality predictable cash flow. I also believe that our current dividend yield of approximately 4.3%, combined with our targeted annual growth of 6% to 7% in dividends per share, represents a very compelling total return.

As we start 2017, we're excited about the leasing activity we're seeing from customers, we're focused on increasing the returns on our portfolio, and we remain disciplined on looking for additional opportunity to drive long-term growth and dividends per share. With that, I'll turn the call over to Dan.

Thanks, Jay, and good morning, everyone. Capped off by solid fourth-quarter results, 2016 marked another terrific year for Crown Castle on several fronts. From a results perspective, we delivered 10% growth in AFFO per share, reflecting the attractive demand backdrop for our leading portfolio of wireless infrastructure and solid execution by our team.

With respect to our portfolio, we strengthened our access base with the acquisitions of TDC and FiberNet, the integration of Sunesys, and our continued investment in both towers in small cells. And we did all of this while increasing our financial flexibility, as we increased the average maturity of our debt, lowered our average interest rate, and achieved investment-grade credit ratings, which reflect the quality and stability of our business and cash flows.

Turning first to full-year 2016 results, as shown on slide 3, site rental revenues grew approximately 7%, or $215 million, as compared to full-year 2015, including 6% growth in organic contribution to site rental revenues. Our full-year 2016 results for site rental gross margin, adjusted EBITDA, and AFFO per share increased by 8%, 5%, and 10% respectively as compared to full-year 2015 results.

Moving on to investment activities for full-year 2016, we invested $557 million in acquisitions and $874 million in capital expenditures, including $90 million of sustaining capital expenditures. As Jay discussed, our discretionary investments are allocated toward opportunities in both our towers and small cell businesses that we believe will generate compelling returns and enhance the long-term growth in our dividends per share.

Additionally, during 2016, we returned significant capital to our shareholders through our quarterly common stock dividend, totaling $1.2 billion in the aggregate, or $3.65 per share, (sic - see press release "$3.605") representing growth of 8% as compared to full-year 2015. We continue to believe that providing a portion of shareholder returns in the form of a dividend aligns well with our business, which is characterized by high quality, long-term recurring cash flows.

Turning now to slide 4, we are increasing the full-year 2017 outlook at the midpoint for site rental revenues, site rental gross margin, adjusted EBITDA, and AFFO per share. Importantly, our improved outlook for 2017 reflects both an increase in our expected operating results exclusive of FiberNet, as well as the expected contribution from FiberNet, which closed earlier this month.

At the midpoint and compared to the previously provided full-year 2017 outlook, we have raised site rental revenues by $154 million, site rental gross margin by $115 million, and adjusted EBITDA by $95 million. These improvements result in AFFO per share being $0.02 higher than in our prior outlook.

Of the increases, FiberNet is expected to contribute approximately $150 million to site rental revenues and approximately $500 million (sic - see press release "$105 million") to site rental gross margin. Additionally, general and administrative expenses related to FiberNet are expected to be approximately $20 million.

Ignoring the contribution from FiberNet, on an apples-to-apples basis, we have increased our current outlook for site rental revenues, site rental gross margin, adjusted EBITDA, and AFFO as compared to our previously provided outlook for 2017. Our higher expectations for 2017 reflect the continued strength of our business and the healthy leasing environment we are seeing.

Turning to slide 5, I want to walk through the approximately $206 million of expected growth in AFFO from 2016 to 2017 at the midpoint of our updated outlook. Moving from left to right, our expectation for organic contribution to site rental revenue growth remains unchanged relative to the previously to provided 2017 outlook at approximately $155 million.

This expected growth in revenues is partially offset by an approximate $20 million increase in operating G&A expenses, which represents a less than 2% increase as compared to 2016. This is particularly noteworthy since we expect to continue to build additional assets and increase our capabilities in small cell throughout the year.

Moving to the right, our expectation for services gross margin contribution of $245 million during 2017 is unchanged from our previous outlook and is a reduction of $25 million as compared to 2016. Our assumption for network services is that we return to more historical levels of capture rate from the elevated levels we experienced in 2016.

Continuing to the right, the impact associated with the conversion of the mandatory preferred stock remains unchanged from our previously provided outlook at $44 million. This represents the amount of annual preferred dividends we paid in 2016 that we are no longer obligated to pay in 2017 as a result of the conversion that took place in November of 2016.

And finally, 2017 AFFO growth is expected to be positively impacted by approximately $55 million on a net basis from other items. Relative to our previously provided 2017 outlook, the $55 million impact incorporates the benefits of FiberNet, partially offset by higher interest expense from the debt acquisition financing of FiberNet and from higher expected LIBOR rates during 2017. With both an increase in the expected operating results in the base business, and the contribution from FiberNet, we are increasing full-year 2017 outlook for AFFO per share from $5.01 to $5.03, representing approximately 6% growth when compared to full-year 2016.

As Jay discussed, we remain excited about the long-term opportunity in front of Crown Castle as we have turned the page on another successful year for the Company in 2016. We believe our portfolio of towers and fiber-based small cell systems, combined with industry-leading expertise and delivering shared wired infrastructure at scale presents us with a tremendous opportunity to both meet our customer carriers -- our carrier customers' needs as they continue to build out their wireless networks and to create long-term value for our shareholders. With that, Alicia, I'd like to open the call to questions.

Thank you, sir.

(Operator Instructions)

Simon Flannery, Morgan Stanley.

Thanks very much. Good morning. Jay, I think you touched on the opportunities from FirstNet, AWS3, WCS; can you just give us a little bit more color about where we are in those processes? And if FirstNet is awarded say in March, how does that flow through 2017, 2018, 2019, et cetera -- your perspective on that?

I know some of the carriers have been waiting for some of these bands to get into the devices, into the iPhone, before they move too aggressively. So any more color on that, and how that impacts leasing first half versus second half would be great. Thanks.

Thanks for the question, Simon. We have not embedded any of that in our forward guidance for 2017. As we talk about things like FirstNet and the spectrum auction, those would affect and help our long-term forecasts. And one of the things that has been true about our business for a long period of time is, anytime there is new spectrum that is deployed across the network, it results in additional leasing for us. And so we look at it and believe that gives credence to our view that there is a long runway of growth in the business as that additional spectrum begins to be deployed.

On the timing first half/second half of leasing?

Yes, the year is back-end loaded as a typical year is in the tower business. Typically in most years, it's about 40% in the first half, 60% in the second half, and this year looks very similar to a typical year.

And are the carriers -- are they each going to continue at the same pace that they were at? Like some of them being pretty active but other ones being pretty quiet?

We're seeing activity across all of the carriers, and it is levels that are very similar to what we saw during 2016 as our estimation. So in total, as we talked about, we think our revenue growth will be in the neighborhood of about $90 million on the tower side and about $70 million from small cells, and we think the year is going to stack up in terms of revenue growth very similar to what we saw in 2016.

Great. Many thanks.

David Barden, Bank of America Merrill Lynch.

Hey, guys. Thanks for taking the questions. First one is just looking at the slight increase in the mid-point churn expectations for 2017. I was wondering if you could talk a little bit about the source of that. Is it just timing or is it -- some people would fear that maybe you are seeing some sort of abandonment from some of the higher-cost leases.

And second, if I could, Jay, I know you are trying to be -- you are kind of tentative to talk about things, but AT&T on their call talked very specifically about rolling out AWS3, maybe backfilling AWS1, working with FirstNet to layer on new rad centers and their 700 megahertz D block. Sprint has been talking about CapEx increases this year as their permitting process for small cells is starting to kind of flow through. T-Mobile's been talking about trying to get aggressive on deploying 600 megahertz as soon as they win the auctions, and the auction seems to be ending sooner than expected. So it does feel like there is a lot more moving parts this year than last year, and certainly because you have a services component, you've been hearing that AT&T is pre-positioning their turf vendors to get ready for this big build. Can you give us some sense of what things on the ground you are actually seeing that would validate getting maybe a little bit more excited about the back half of the year? Thanks.

Dave, I will take the first one on churn. Basically what we did is pulled up the bottom of the range. There is no real difference in what we see going forward from what we had seen before. It's just tightening up the range a bit from what we have seen.

And I don't think there is anything you can read through to what the carrier activity is specifically. Just more understanding of what is going on, and more conversations and thoughts around tightening the range a bit.

Yes, Dave, on your second question, it's obviously very early in the year. On the landscape of conversations that are going on among the carriers and what they said publicly, those are all positive milestones that you are speaking about. And as we go through the course of the year, we may need to update where we are today, but at this point we see the year stacking up pretty similarly to the way we saw in 2016.

And one of the things I would just draw everyone's attention to is the nature of our business. We have a lot of predictability around the cash flows. And so as we sit here, as we go into any given year, most of the leases that we will sign on, that will impact our financial results, we've got those basically circled and know where they are by this point in the year. And so the opportunity to outperform in any given year, or underperform, is relatively low.

And the activities that you spoke about, those are likely to be impactful and positive as we start to think about our 2018 results and beyond. We will certainly update everyone as we go through the course of the year. And if some of those start to materialize and start to impact leasing, I think those are more likely to impact our run-rate activity as we go into 2018 than they are necessarily to impact our financial results for 2017.

Great. Thanks, guys.

Brett Feldman, Goldman Sachs.

Thanks for taking the question, it's kind of a big-picture one. We look at the way the sector is changing before our eyes, just the US telecom cable sector, more consolidation obviously, continued growth in data traffic. I am wondering if it's caused you to have some big-picture conversations around what type of company you are supposed to be to be positioned for this? You've obviously made a decision that being in the small cell business is a nice complement to what you were already doing in towers, and I think we are starting to see some benefits there.

But as you spend more time operating fiber networks and engaging with customers, are you thinking that perhaps there is a broader infrastructure model that you guys can be pursuing? And if you do, what is the ultimate way you think about, whether those types of deals would make sense for you? Is at all about growing the dividend or are there other considerations that weigh heavily as you evaluate your M&A funnel and really the strategic direction of the Company?

Brett, that's a great -- that's a good question. I appreciate you asking it. From a broad perspective, our Business is a shared infrastructure model. And that shared model has been through a number of economic downturns, consolidations, technology shifts, and it has through all of those cycles and movements delivered really strong returns and growth because it's the most effective and cost efficient way to deploy wireless networks. And as we consider sort of our strategic focus, we believe we are best aligned by continuing to keep that our core focus as a business.

As we watch the landscape, and we do, we are trying to position ourselves as that shared infrastructure provider. And where we found the most benefit has been historically towers, and we believe small cells are very similar to that.

There may be opportunities for both us to use our towers and our real estate assets, as well as the fiber underlying the small cells. We may find opportunities for additional revenue growth, and we are happy to consider those. And to the extent they can increase the returns in yields, we will pursue them.

But the primary focus is today around wireless for both towers and small cells. And all of the conversations that you reference are encouraging to us because they indicate the comments that I was making around the expectation of continued growth in data, wireless data, and we think those trendlines are very supportive of our underlying focus on towers and small cell.

Great. Thanks for that color.

Phil Cusick, JPMorgan.

Yes, thanks. Following up on Brett's question, can you dig into what is going on in the Sunesys and the new business in terms of enterprise sales? Sunesys had a sales force when you bought it, and we've talked about them selling not only that but your legacy fiber. What is happening there? Is there any momentum, and how do you think about that changing in 2017?

Yes, we have found, as we talked about before, we have found some opportunity to use the fiber that we have acquired for some enterprise services in places where we have a dense urban fiber footprint. Our focus and analysis around what fiber is interesting to us is driven by the opportunities that we see on the small cell side. But to the extent that we can use the pipe in a shared infrastructure model in a way that drives additional returns and yields from fiber services in certain cases, we think that makes sense.

Obviously there was a tremendous amount of expertise that we received in the Sunesys acquisition up in the northeast in several markets. The same thing is true, as I mentioned, from FiberNet in the South Florida market, as well as in Houston. And so we believe there will be opportunities there, but the main strategic focus is around small cells and around the edges; that expertise enables us to increase the yield on the asset.

Has there been a lot of momentum in doing that, or any increased amount of investment there? Or is it moving along slowly?

It is very limited capital investment on that front. And the momentum, I would describe as being fairly consistent with what their historical practice has been around being able to use that fiber for those fiber services.

Okay. Thanks, guys.

You bet.

Nick Del Deo of MoffettNathanson.

Thanks for taking my question. I wanted to ask about your underwriting for the FiberNet deal. When you've done large tower transactions in the past, you've generally indicated they are predicated on securing something like one tower -- or sorry, one tenant over 10 years to make the math work, or something along those lines. With respect to FiberNet, what does the business pace require, both for the business you assumed, as well as for the small cell volumes you are going to layer on top?

Yes, Nick. Thanks for the question. Typically, we are assuming on the small cell side about one tenant over a 10-year period of time is roughly our underwriting assumption in these acquisitions of fiber. We obviously believe there could be a lot greater upside than that to these fiber assets, but that's what we are roughly underwriting in both the FiberNet acquisition, as well as the Sunesys acquisition.

And, Nick, just before you get on, we have been seeing faster than that lease-up, is what Jay mentioned earlier. So we think that those assumptions we could probably outdo, but for now the underwriting has been based on that small cell demand at that pace.

Okay. For the existing business you assume, I assume some sort of modest growth rate is baked in?

We basically assume that business is roughly flat. There are some growth assumptions in some cases depending on the location and the market, but the returns are being driven by small cells.

Okay. Great. And in a similar vein, these deals are partially about capital avoidance; in the past you've talked about investing something like $100,000 net per node that you roll out. How far does that fall in markets where you've acquired these big fiber inventories in advance?

It falls significantly because the fiber is very similar to a tower. So the analogy when we look at small cell business, fiber is the tower. And in the tower business we would typically describe a rad center, that's a level on the tower where the tenant would install their equipment. In the small cell, in small cells, a installation of a small cell node is like a rad center on a tower. So we are leveraging the asset, the fiber, and then adding small cell nodes across that fiber to drive the return.

And to the extent that the fiber is acquired through one of these acquisitions or it is previously built for another acquisition, then we are adding additional tenants to that fiber, and I talked in my comments about the incremental returns. If we're building it from scratch we are in the neighborhood of an initial investment of about 6% to 7%, which is about where we acquired Sunesys and FiberNet. And then we are adding to that, additional tenants, which moves the returns with one tenant into the low-double digits on a yield basis, and then the third and the fourth tenant take the returns well beyond that.

Okay. Great. Maybe one quick one and then I will hop off. Small cell network services revenue jumped a lot in the quarter. Any commentary behind that or if it's a leading indicator for the segment?

No. It is really, how we get there is we have equipment that we buy on behalf of our customers and we get a small margin on that, and that's the services you see in the small cell business. It's just a reflection of the activity we had in the quarter, which we had talked about before. There was a high level of activity in small cells in the fourth quarter, and that's where you saw that services coming from.

Nick, sometimes what we also see is the carriers come back and optimize their networks, their existing small cell networks. They will pay us to perform a service where we will analyze re-optimizing those networks. And ultimately, that looks like, to your question about co-location, as they go through that optimization process they end up coming back and adding additional nodes across the fiber or adjusting the network slightly which results in us getting some services fees from that.

Okay. Great, thanks so much, guys.

Ric Prentiss, Raymond James.

Hi, Ric.

Thanks, good morning, guys. A couple of questions, one on FirstNet. What would cause the FirstNet equipment to trigger a new rad center and be like a co-location versus being able maybe to slip into an existing rad center and come in as an amendment activity? In other words, I think back in the day when Sprint and LightSquared were thinking of doing stuff, it was seen as being more amendment activity. Just wondering how are you viewing FirstNet's opportunity when it does start materializing?

Ric, I think at this point it's probably too early to be able to answer that question. We really need to know ultimately who is going to be the provider and the distributor of the service, and then you need to understand the network that they want to build. It is probably too early to really be able to comment on that.

Both of them are good for us, though, obviously. There is a meaningful -- if they are deploying brand-new spectrum in that kind of scale nationwide and it's an amendment, that would be great for us. If it's a new installation, that would be good as well.

Right, but there's nothing contractually that says they have to go to a new rad center if it is spectrum that they don't own?

No. Whenever there is a transaction like this or a new deployment, we will sit down with them and understand what their needs are, and we will try to structure something that makes the most sense for both parties. There have been times, you brought up one with LightSquared years ago, where there was a co-planing arrangement where they wanted to go on the same level on a tower. There have been other occasions that I could point to where they wanted a separate installation and a separate rad center on the tower. We would have to just understand a lot more and then work with them on what makes the most sense.

Okay. And then just given all of the convergence and consolidation that others have mentioned on the call, can you update us just on what your Sprint/T-Mobile overlapping exposure is, and remaining life of leases? Maybe also just what percent of revenue AT&T represents at year-end 2016?

Okay, so, each of Sprint and T-Mobile are around 6% on overlapping sites, and we have about five or six years left on both of those contracts. So what we think is that if the two of them were to get together, the implication would be that they want to compete on network quality. And to do so they would have to increase their investment in their networks, not try to minimize the number of towers they would be on. As has historically been the case, we see that consolidation doesn't necessarily equate to them coming off every tower they possibly could, because they are really trying to invest in the network. AT&T I think is 28% of revenue for 2016, and you can see that in our supplement.

Okay. Great. One final quick one: Since you closed FiberNet earlier this month, a large portion of that was funded by the equity raise back in November, but I assume you used some debt. Can you let us know what kind of debt placement you did to handle the remaining portion of the cash for FiberNet?

Right now it's on our revolver, and we have not said how exactly we would fund that overall in the future. But it is in -- right now it's just borrowed on our revolver.

Thanks so much. It looks like it could be an exciting 2017 and 2018. Thanks, guys.

Thanks, Ric.

Jonathan Schildkraut, Guggenheim.

Great, thanks for taking the questions, fitting me in here. Two, if I may. The first is yet another follow-up on the fiber or small cell business. You guys have done a bunch of acquisitions here, and the Business has a bunch of different revenue streams. Is it fair to assume that all of the acquired revenue streams flow into the small cell business? And then sort of as a derivative question, is there anything we should know about that legacy business characteristic-wise, churn, contracts, things like that, which would distinguish it from sort of the business that you are organically building on top of that asset?

And then my second question, and I know you guys have addressed this in one form or another in the past, but considering the amount of work that is potentially in the pipeline from US carriers, whether it is FirstNet, AWS3, WCS, or other sort of spectrum rollouts, you guys have been very open, the MLA process in the past. And again this one seems like it would be fairly complex relative to some of the stuff we have seen historically. Is there any desire, conversations, thoughts about new MLAs? Thanks.

Sure. On the first question, I would not distinguish the revenues or cash flows on those businesses from our small cell business. Any of the revenues that you are referring to specifically, to your question, that is included in the small cell segment. So those revenues and cash flows flow through that. We are excited about that business, believe it's a good business, believe it is recurring, and do expect on some level we can grow it in some ways as I spoke to in my earlier comments.

On the MLA process, I'm assuming that you are referring to what was done in the past in some occasions where carriers have been willing to commit to certain levels of activity and then effectively paying for that ahead of their anticipated work. You know, Jonathan, those agreements worked out really well for us financially; we were pleased with the outcome. I believe our carrier customers were pleased with that because it gave them greater certainty and expedited the process of getting on towers. So we would certainly entertain those agreements again, and if we were to do them, then they would have to make financial sense for us and accomplish the same goal for the carriers. So we would be open to them.

I view those, frankly, in hindsight having done a couple of them with multiple carriers over a lot of different years, at the end of the day, I think financially they worked out well for us and they accomplished the purpose the carriers were looking for in terms of expediting activity. So if those two things aligned again, it would probably make sense. But we may just do it as we do on a normal basis of just taking leases as they come and obviously that works well for both parties as well. We have to see how it develops, but we leave it open as an opportunity.

Thanks, Jay. I appreciate it.

Matthew Niknam, Deutsche Bank.

Hi, guys. Thanks for taking the question. One on fiber and strategy: If you could maybe talk about your appetite for further expanding on the fiber and small cell front, just having closed FiberNet, and given where you are on leverage and a rising rate environment. And then just one housekeeping item on tower site rental expenses: They ticked up about $5 million sequentially, they have usually been down seasonally in 4Q, so if you could shed some more light on what went on there? Thanks.

Matthew, on your first question, we are, as we've talked about lots of times, when we look at investments, and whether that's capital investment, capital expenditures that we would do on a quarterly basis and report, or larger acquisitions, the filter through which we run those is whether or not they increase long-term dividends per share. And so if we have the opportunity to invest in additional fiber, which we're certainly looking for and seeking out opportunities to do that, they're based on our view that there's going to be significant small cells in dense urban markets around the US, and that our investment in those needs to be aligned with our overall cost of capital.

So you raised the point that in a rising rate environment, as we underwrote the most recent assets, we certainly didn't use current financing rates to look at what we thought the long-term returns were. We used a forward curve in terms of interest rates to account for that. And then we've got to get comfortable that over the long term we can get a return for the weighted average cost of capital.

Depending on the size, we may need to use equity at points, which we have done in the past. But we will consider that as a part of the overall return and make sure that it's compelling. And ultimately the measure of that is whether or not it increases dividends per share. And that dividends per share obviously accounts for the effect of however we finance the asset, we have got to drive incremental returns from what we would consider absent doing any of those actions.

As Jay pointed out, we have thought about and baked into our outlook an increasing interest rate environment. To the extent that we would have to look at an acquisition, we would do so again and just try to take a view on what that would look like.

The second thing I would point out on the equity, if you look at what we did for FiberNet is, we had assumed cost to capital in our modeling and, therefore, we went out and sold the equity where we could clear that cost to capital back in the fourth quarter, even before it closed. Because we want to be very true to that increasing dividend per share to make sure that we lock in that cost to capital and the returns that we're looking at.

With respect to tower site rental expenses going up, there are a few things that happened in the fourth quarter that -- I think the important part is that we did not include in our 2017 guidance and outlook. We have not made them continue through the 2017, so you can take away from that there are things that we don't expect to incur going forward. So there were -- it's just small things here and there that added up in the quarter that we don't include in the 2017 guidance.

Okay, and if I could just follow up, on the first question, are you seeing any demand from your carrier customers in markets beyond the top 10? I know I think in the past you've talked about supporting carrier needs for densification in some of the larger markets, but are you beginning to see that spread beyond these larger markets to date?

We are seeing it spread beyond the top 10 markets. There are a number of markets inside the top 50, well outside the top 10, where we're seeing small cell activity. If you looked at the majority of the activity, as well as the majority of the capital that we are spending, we're still in the top probably 10 to 15 markets of the US, but we do expect that will continue to grow well beyond just the top 10 markets [in the US].

And like we pointed out in our last quarter, in the case study around Chicago, it's not only expanding from the top 10 beyond, but also from the central business districts in the top 10 to more the suburban areas. We see the expansion both ways -- of the demand for small cells.

Got it. Thank you.

We'll go next to Amir Rozwadowski of Barclays.

Thanks very much. I just wanted to follow up on some of the prior questions in what seems to be a rapidly evolving end-market structure. A lot of the combinations that seem to be discussed in the market seem to drive a lot of discussions around the initial deployment of 5G and whether or not it actually can provide the type of services promised at a meaningful level. If you look at the opportunities over the next few years, are you factoring in a meaningful opportunity based on these initial deployments, or should we think about the deployment of current fallow spectrum and densification that's in the hands of the carriers at the moment as a primary opportunity for the foreseeable future?

Yes, it would be the latter and not the former. As we think about our Business, we tend to look at our forecast and the activity and we are really not assuming all of the potential upsides that could come, which has been the great thing about both the tower business and now we believe about the small cell business. You don't have to believe in upside usage case on mobile data in order to justify the investments that we are making or to deliver significant returns for the shareholder. That's a great place to be. We will let the upside take care of itself. We obviously think the assets have a tremendous amount of growth and upside.

It's interesting, and the nature of the question really drives home the point. If we were on a conference call 15 years ago talking about towers, the discussion was around the transition from 2G to 2.5G and the opportunity and the returns that would drive, and now we are talking about the opportunity of 5G. And throughout those periods and all of the successive technology shifts, we have seen additional revenues and additional tenancy across the assets. And we believe that's the beauty of the shared infrastructure model, both on towers, that we will continue to see activity on that front, and then as densification occurs and data traffic increases, we are going to see the benefit of that on small cells as well.

Thanks very much. And one follow-up, if I may. The debate on the sustainability of escalators seems to continue. Now this is something you folks have expressed your views on for some time, so I won't belabor the point, but I would like to ask: What is the potential middle ground that could take place between yourselves and the carriers? Is it re-engaging on some of the MLAs, as you had mentioned? I would love any color you can provide on that front.

Yes, Amir, we have not seen any change in the pricing environment. And as we've talked about, our business model and propositions of the carriers is based on a value proposition. The underlying cost of the real estate has escalated far faster than the escalators under their lease agreements, and far faster than the pricing increases that have occurred on the asset.

And we have been able to take upon ourselves that higher escalating cost and share that among multiple operators, which has limited the amount of growth in their costs that they would've otherwise occurred to them if there wasn't a shared infrastructure provider. So, I don't see any change in the pricing environment and, frankly, the value proposition, I think, continues to hold as we are able to sustain a higher level of costs and be able to share that across multiple operators in order to hold their costs down.

That's very helpful. Thank you very much for the incremental color.

Matthew Heinz, Stifel.

Thanks. Good morning. With respect to FiberNet and the footprint, I am just wondering what sort of visibility or funnel you had in those markets as you were underwriting the deal in terms of carrier demand for new small cell sites? And, I guess, whether that pipeline has changed at all since the deal was announced? Maybe the breakdown of anchor nodes versus co-locations, and when should we expect to see incremental site revenues showing up on that fiber?

Matthew, I think on the second part of those questions, I will probably beg off on that level of specificity. But on the first part of the conversation, we are having significant conversations with all of the carriers, and have a pretty good understanding, we believe, around how they're thinking about the deployment of small cells and macro sites across the country. And our FiberNet acquisition was based in part on the insight that we have from working with customers. So we absolutely had insight, and as I mentioned, there is significant opportunity and activity going on in the South Florida markets that we gained fiber footprint in, as well as Houston.

Okay. Thanks. And as a follow-up to that, as I think about your shared infrastructure focus, I guess that is inclusive of a fairly broad range of potential asset types. And as we consider the opportunity around commercial IOT enabling that with Edge computing nodes, is that something you would like to capture in the long term, beyond just the consumer mobility opportunity? What is your level of openness to alternative asset types beyond the tower and fiber focus you have today?

Yes. Our focus is in providing the infrastructure. As we look at the opportunity, and I think you raise a good one around commercial Internet of Things, those are likely to grow across wireless networks, for which we would expect to be a significant provider of space to deploy those wireless networks. But we don't [view-quarter] our strategy as ourselves owning the wireless networks themselves, or interfacing directly with the consumer, and whether that consumer comes in the form of an individual or a commercial Internet of Things.

I do think, as I mentioned in one of my earlier answers, I do think there are opportunities for our tower assets and real estate to be used in greater ways in the more upside cases around what will happen with the IOT over time. And we're certainly focused on what those opportunities might be, and could provide additional upside to our base-case assumptions.

Okay. Thanks, Jay. I appreciate it.

Yes.

Spencer Kurn, New Street Research.

This is [Robert Chatfield] for Spencer. Thanks for taking the question. I wanted to ask, given recent comments about potential tie-ups between US wireless carriers and cable companies, how do you see those potential outcomes or tie-ups affecting both the existing lease-up or the lease-up of existing small cell deployments, as well as organic demand for new small cell deployments, given consideration for the fiber plan that the cable companies might have?

Yes. I don't want to speculate specifically on any of the rumors that are in the market this morning. But I think I would take you back, Robert, to the shared economic model. Obviously for a long period of time, some portion of the carriers' wireless networks have been self-performed, whether that was towers or the ownership and build of their own fiber. But we found significant opportunities for a shared economic model.

And regardless of ultimately who owns the wireless networks and who is deploying wireless networks, we believe the shared economic model is the most cost effective and efficient way for carriers to deploy wireless networks. And we would expect, regardless of what consolidations occur and ultimately who the owners of those networks are, that a shared solution will be the most cost effective and will be a part of the solution that they would use.

Robert, just to add something Jay said earlier to an answer, in answer to another question, which is: The implication of all of this activity is mobile demand is going to grow significantly because people are trying to get at that mobile demand as efficiently as possible. It may not be the only reason for those types of consolidations, but it is one of the reasons. And that gives us -- you've got another data point to confirm our belief that serving as the infrastructure provider to meet that mobile demand is really a great place to be. We think that all of these things do have potentially a positive implication for our overall business model and something I think would play out well for us.

Thanks, that's helpful. Sorry, wasn't trying to catch you in comments about the rumors. Maybe I will ask the question a little bit differently. You look at the cable plan versus the fiber deployments that you might see in some of your small cell networks. What would be the differences between those existing in-ground plans versus yours? I think answering that question might get to what I was trying to get at.

Well, I would probably answer it, honestly, the same way. I think I understood that's where you were driving with the question. If you look at the wireless operators today, many of them have significant fiber footprints. And we are still seeing significant activity in small cells because the route uniqueness of our fiber, the density of that fiber, the capacity of the fiber, and the shared economic model that we offer is still the most cost-effective way for those networks to be deployed.

It's also true that today carriers self-perform some of those activities. So to the extent that there was a new owner of wireless networks, whether that was as an acquirer or as building out a new network, I think you would see a similar pattern where you have use of both the shared economic model as well as self-perform. And ultimately, I think this is sort of to the benefit of the tower industry.

But that shared economic model, because of its lower cost, the carriers are looking for opportunities to drive out costs in the network and we are the provider for that. And so I don't think any equation on that front would change as a result of a cable operator moving into the wireless space. I think, as Dan said, it is supportive of the long-term view that the world is going wireless and there is significant mobile data demand ahead.

Okay. Thanks.

Batya Levi, UBS.

Great, thank you. A couple follow-ups. First, going back to the churn question, if you look at churn aside from the acquisition-related churn coming from PCS, Clearwire, Leap, it looks like there's an uptick this year. Can you talk maybe what is driving that?

And going back to the small cell, I believe that you had guided a similar CapEx level for small cell deployment this year versus 2016, given all of the activity in the market, do you see potentially an opportunity to accelerate that spend in the near term? Thank you.

Yes. To speak on the churn, there really isn't an uptick. It's really that normal churn of the non-consolidation, non-renewals is still in the 1% range, which is relatively consistent with where it has been. To the extent that there may be a small uptick, it's really not in our big four. It wouldn't be of our big four customers, it would be in the other customers. But we don't think that's a material amount at all, nor do we think it's an indication of what's going on in the market. We see the non-consolidation churn to be relatively consistent with historical.

Our underwrite on CapEx for 2017 for small cells as you stated is similar to what we saw in 2016. Typically, the amount of timeline between when we sign up a customer and when they go on-air, that cycle is a 12 to 15 month cycle in order to get small cell nodes to be built, at a minimum. Sometimes it's longer than that. It can be an 18 to 24 month cycle. So the uptick as we would talk about activity, both that we're talking about, or talking with our customers about, or that they publicly start to talk about, the CapEx spend for that and then ultimately the resulting revenue would be beyond, in all likelihood, beyond the calendar year 2017, both in terms of material capital spend as well as then the ongoing revenue.

I think, absent other factors, I think we are biased to the belief that -- when I made the comment earlier in the call -- that ultimately we believe small cells could be of a similar scale as towers. We certainly anticipate that there's going to be significant investment and opportunity in the years ahead, and the scale of that would be much larger than what we are currently seeing. So our bias is that it is upward movement, but I think you are unlikely to see a meaningful increase to our activity in 2017.

Great, that's helpful. One more follow-up on the macro side: Would fiber, with public safety network about to be planned and deployed, and maybe some of the national carriers not having full coverage in the rural footprint, do you see an opportunity to build macro towers in more rural locations right now?

There may be some opportunity for that. Frankly, I would tell you that's not really where our focus is. We are focused right now on the major US metro markets, and believe most of the capital spend as well as the return opportunity is going to exist in those markets. There may be opportunities to do that and we may participate in that, but I think we are more likely to be focused on investment in small cells.

Okay. Great. Thank you.

Mike McCormack, Jefferies.

Hi, guys. Thanks. Maybe, Jay, just a quick comment on what you're seeing out there as far as the small cell deployments go, with respect to time to deploy and cost on your side to deploy as well, and where that's pacing and how that looks? And then secondly, maybe just a comment regarding the broadcast auction, and it clearly seems like there has been fairly tepid demand there. Just wondering, as you look out over your longer-term outlook, how that might change your view given what appears to be, again, low demand for that spectrum.

Yes, on the first question, time to deploy and cost, we have seen the cost stay relatively similar to what it has been over the last couple of years in terms of the costs to deploy these systems. Time to deploy is about the same as what it has been, maybe a little bit more extended than the last couple of years. I think it is supportive of our notion that there are high barriers to entry in that business and it takes a long time to build new systems, so there's a big advantage of owning fiber in the markets and being able to go out and co-locate tenants onto that existing fiber footprint. And the time and cost to deploy is really supportive of that view of the barriers to entry and the opportunity ahead of us for the assets that we own today.

On the broadcast auction side, I won't speak to how much the spectrum band goes for, or ultimately how it gets used, but what I would point to is, going back and looking at the business over a very long period of time, it's been terrific for the tower industry as spectrum is deployed and gets into the hands of an operator who has capital to deploy that and a use for it, and so I wouldn't at all take it as a negative. I think ultimately the spectrum that's in the market today under auction, as well as additional spectrum that the FCC has been pretty clear about, over the long period of time, looking to find additional spectrum to put in the hands of wireless operators, I think that is a positive trend for the tower industry and a real tailwind for future growth.

Great. Thanks, Jay.

Walter Piecyk, BTIG

Thanks. Can you talk about the cadence of new leasing activity at small cells? If I do the basic math, it looks like it yo-yos back and forth from quarter to quarter? Towers seem to be a little bit more regular, adding like $22 million, $23 million or so per quarter, where small cells I think it was less than $5 million, it was like $9 million last quarter. Is that going to smooth itself out at some point, or how do we think about small cells going forward?

Yes. I think as you described the yo-yo, it's probably more related to as these turn on as we are building systems, Walter, about 75% of the activity is related to us building new networks and about 25% is related to co-locating on existing systems. So you are looking at the timing of construction starts and finishes roughly in there.

Over time, I think the other thing that will happen is if we are right about the scaling of the opportunity, it will get so big that those movements are not as noticeable and you are comparing those two businesses -- one is about $400 million in revenue and the other one is about $3 billion. And so the tower business tends to look a little more smooth just given the relative size of it.

I would not point you to anything that you could, frankly, predict or indicative of the business. I would point to, as we've tried to do over the call today, really focus on a year over year. So, look at a 12-month period of time compared to the previous 12 months. I think that will give you the best indication of how we are doing.

Yes. And on that front, as you know in our outlook, that we think the small cell business is going to grow more in 2017 than it did in 2016 by about $15 million. So year over year (multiple speakers).

So you are sticking to the $15 million, because the $15 million is off, at least for me, a lower number because the fourth quarter as far as the new leasing activity shows that your new leasing activity for the year is like $28 million, so it's $15 million off of that number as far as that increase for 2017?

No. I would go back to year over year. We did about $55 million roughly of new leasing activities in the small cell segment in 2016, and we think that's up about $15 million in 2017 to about $70 million.

Okay. Just another question on the small cell boxes: I know I come back to this every quarter but I can't help myself. Again, if you look at those boxes where you can put two or three tenants in them, is your anticipation that one operator is going to take up that second slot in a lot of those locations? Or is it more likely that a second operator will take the second -- I don't know if slot is the right term. But I think you know what I am talking about, when you crack those things open, you got two tenants you can stick in those things that are up on the lampposts.

I think, Walter, you are going to have some of both. You are going to have some of those boxes where we have multiple tenants in them. I think we have showcased a number of systems including there in New York, walking around showing folks where in some cases we have three tenants in some of those boxes.

Yes. On the bottom part.

And in other places we will go next pole over and add the tenant.

Sure.

I think you are going to have a mix of both. Ultimately the way we price it, though, comes back to return, and the aesthetics of a community are going to drive, ultimately, do we put two or three on the same light pole? That's not really the economic reality.

The economic reality goes to -- the fiber is the really expensive asset in the mix, and it's across that fiber that we are driving additional return by adding tenants. Whether it goes on the same pole or the next pole over, we don't really care. We are focused on the return across the capital investment that we make.

I get it. And then just one last one on that same type of question: Take Verizon as an example. But this could be the case for any carrier. They have existing AWS spectrum that they are putting in those small cells today, those small radios.

When they are ready to do AWS3, is it contemplated that that AWS3 spectrum will also be in the same radio as the existing AWS? Or is that going to have to take a second slot? I'm not saying for -- obviously if they go back, they are not going to bother to replace, they will just the second slot. But let's say we're out into 2018, your small cell business is ramping. Is the expectation that they can pack all of that AWS spectrum, the existing stuff, and the AWS3 into one radio, so it's only taking one of those three slots that is in that box that is on a lamppost?

I don't want to speak specifically for Verizon, so let me answer the question more generically for all of the carriers. Small cells have amendment activity to them in the same way that towers do. And that amendment activity, as to the answer I just gave you, looks in a similar way.

So in some cases the amendment will be they will come back with an additional spectrum band or an additional technology and they will go next pole over because they've put so much equipment in the first box that they are going next pole over. Sometimes the answer is they are trying to optimize the network differently, so they will go next pole over. And then at other times, they will go in the existing box and take up an additional slot in that box, and that looks to us like any other new tenant that would go on there. So, and we would probably (multiple speakers).

Those are three examples. But that was not the example I asked about. I said in 2018, let's say it's a new operator into a new market, and they have all the spectrum, so they're not adding anything, this is brand new to them, so they go into there, can they stack all of their AWS spectrum into just one of those slots or are they going to need two slots for existing AWS, as well as AWS3? I understand everything you've said, I get that, and you're going to get more money from those, but since we are still early stages in the small cells, I'm just interested in how much spectrum they can pack into one of those slots that is on that lamppost.

The answer is that it is theoretically technically possible to stack it all on one. It is possible that they could do that in the same way that it's possible on towers. I think that's a less likely outcome. I think you are more likely to see them add additional equipment and either be an amendment or a new lease for us. But it's theoretically possible that they could combine the spectrum bands and do it with a single piece of equipment, but that's not been what we've seen so far.

Got it. Thank you very much.

I think we have time for maybe one more question.

Michael Bowen of Pacific Crest.

All right. Thank you very much for squeezing me in. A couple, if I may. Sorry if I missed this, but I wanted to make sure I understood, with regard to the reduction in the organic growth projection in 2017, if I'm correct I think it went from 4.9% down to about 4.8%. Can you help us with a little detail there of exactly what is baked into that?

And then another question is, we are hearing more and more about Project Loon over at Google. Is that more, as you see them, is that a friend or foe? Is that something you could potentially see benefit or is that something that is going to take business possibly away from you? How are you thinking about that? Thanks.

Michael, on the first one on organic growth, we kept organic growth consistent in our outlook from before until now, so it's not a reduction in growth. There's a higher base on which we provided it. So it looks like -- I think you did the math and it may be a little lower, from a percentage basis, but we did not change our assumptions on the activity levels or the growth that we expect in 2017.

On your second question, I think the carriers and operators will, over time, continue to look for opportunities to drive down the costs in their network, and so there may be a number of opportunities, whether it's Wi-Fi offloading or other projects that the carriers come up with to offload some portion of the capacity off of the networks. And that is good for them in terms of reducing the cost and good for us as consumers in terms of improving the coverage and capacity of the networks. We don't see anything on that front or otherwise at the moment or out on the horizon that is damaging to our overall business.

Okay. And one last one, if I could. You mentioned the yields on a per-customer basis for small cell. Can you compare and contrast those with tower? I think you said 6% to 7% for the first customer on a small cell and then increasing from there. Is it similar, or can you help us with that?

On the tower side where build projects are being done in the market or where we're seeing builds and for which we are not investing capital because we don't find the returns to be attractive, there are some folks out building towers that initial yields that are in the 2% to 3% range. We don't frankly find those that interesting given the other opportunity set that we have in front of us.

As we look at maybe the larger acquisitions that have been done more recently, those traded in the 4.5% to 5% initial yield on the tower side. So the small cell returns are a little higher at initial yields than what we've seen in towers from an acquisition standpoint, and meaningfully higher relative to new builds.

That probably has to do in part, though, with the complication and difficulty of constructing those. New tower builds are not being done in major metro markets in the US, so I think both the risk, the opportunity and the difficulty of completion is factored into those returns. It may not be fair, really, Michael, to compare those as apples to apples.

That's helpful. Do you have any historic yields off of towers from past years? I'm assuming much higher.

Yes. We were typically building towers in that 4% to 5% yield range, maybe a little bit higher than that at some point. So small cells, initial yields are higher than historical yields initially were for towers.

Okay. Great. Thank you.

Thanks, everyone, for joining us this morning. We will wrap up the call and look forward to talking to you next quarter. Thanks so much.

This does conclude our conference for today. We thank you for your participation.