Crown Castle Inc (CCI) 2016 Q3 法說會逐字稿

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

Thank you, Vicki, and good morning, everyone. Thank you for joining us today as we review our third-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 Investors 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 factor sections of the Company's SEC filings.

Our statements are made as of today, October 21, 2016 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 crowncCastle.com. With that I will turn the call over to Jay.

Thanks, Son, and good morning, everyone, appreciate you joining the call this morning. As you saw from our earnings release yesterday, we delivered another great quarter of financial results that exceeded our expectations. As the leasing environment continues at a healthy pace across the towers and small cells, allowing us to once again increase our outlook for full-year 2016.

We also increased our dividend by 7%, reflecting our expectations of continued growth in 2017, and consistent with our goal of increasing dividends 6% to 7% per year. As we discuss further on this call, we believe the healthy leasing environment will continue into 2017. Specifically our 2017 outlook assumes higher growth of approximately $15 million from small cells and similar growth from towers as compared to 2016.

The growth we see in our business is fundamentally driven by the strong growth in demand for mobile data, which we believe will continue given the continued adoption of data intensive applications such as video streaming; the increasing number of devices that are being connected with the growing importance of the Internet of Things; and the potential for new applications to be deployed in the years ahead that will rely heavily on mobile networks such as fixed wireless broadband, autonomous cars and augmented reality just to name a few.

With this expected growth in demand for mobile data we believe the wireless carriers will invest steadily and their networks, including leasing our infrastructure in the coming years, much like they have historically, to serve the growing and unmet demand for wireless connectivity.

As we have seen over the last two decades, our approximately 40,000 towers are the first option in network deployment given their efficiency and cost-effectiveness. More recently the carriers have looked to small cells to add to their capacity in their networks. Therefore over the last few years we have begun to invest more significantly in this area of our business.

This investment has been based on three fundamental characteristics that we believe make small cells very similar to that of towers. First, the shared economic model of small cells drives down our carriers' cost of deployment. Second, our margins and returns increase over time as the carriers lease more tenant nodes across our fiber. And third, the total addressable market is very sizable.

To illustrate these points I would like to walk you through our experience in Chicago using slide 4. In 2013 we had approximately 300 tenant nodes on air on approximately 100 miles of fiber, or approximately 3 tenant nodes per mile of fiber. Of these 300 nodes approximately 90% were focused in the downtown area.

Fast-forward to today; we have approximately 1,100 nodes on air and under construction on 250 miles of fiber with the tenant node density at approximately 5 tenant nodes per mile. This results on a yield on our investment in Chicago of about 10%.

Importantly, more than half of these nodes are situated outside of the central business district, which demonstrates both the expansion beyond the central business district and, at the same time, the densification in the central business district that has occurred over that time.

The expansion beyond the central business district into the surrounding Chicago suburbs is an example of why we believe the total addressable market for small cells will be so significant. Chicago is representative of what we are seeing throughout major US metro markets, which is why we remain so bullish on the opportunities we see in small cells.

We believe this growth opportunity from small cells, combined with the growth from our towers, uniquely positions us to capitalize on the evolution of wireless networks through the extension of our shared infrastructure model that continues to efficiently and cost effectively serve carrier needs.

The strength of our business model, the underlying strong fundamentals driving wireless investments and the uniqueness of our assets gives us confidence in our ability to continue to deliver 6% to 7% growth in dividends per share annually. We believe this expected growth combined with our current dividend yield of approximately 4% provides a very attractive total return opportunity.

And with that I will turn the call over to Dan to go through a few more details around our financial results.

Thanks, Jay. And good morning, everyone. As Jay mentioned, we have a lot of good things to discuss on the call today. We exceeded our third-quarter guidance range and raised our full-year 2016 outlook, reflecting the continuation of the strong operating performance our team has driven throughout the year and the healthy leasing environment we continue to see.

At the midpoint our updated full-year 2016 outlook for AFFO per share of $4.73 is up 10% compared to 2015. This growth includes approximately 200 basis points of benefit from acquisitions. We also provided our outlook for full-year 2017 results, which at the midpoint calls for AFFO per share of $5.01, up 6% over our updated 2016 outlook.

Additionally, we increased our dividend by 7% from an annualized $3.54 per share to an annualized $3.80 per share, demonstrating our strong belief in the growth in the cash flows will be able to generate from our high-quality asset portfolio.

Turning to slide 5, our third-quarter results for site rental revenues, site rental gross margin, adjusted EBITDA and AFFO each exceeded the high end of our previously provided third-quarter 2016 outlook, reflecting the continued strength of the overall operating environment.

For AFFO the quarter benefited from $7 million in lower-than-expected sustaining capital expenditures. To be clear, this benefit was due to timing and we expect that capital will now be spent during the fourth quarter. Adjusting for this timing benefit in lower sustaining capital expenditures, we still would have exceeded the high end of our third-quarter outlook for AFFO and AFFO per share.

Moving on to investment activities. During the quarter we invested approximately $220 million in capital expenditures consisting of $19 million in sustaining capital expenditures and approximately $200 million for discretionary investment opportunities. Included in the discretionary investments is approximately $17 million of land purchases we completed to further strengthen our control of the ground beneath our towers, which aligns with our central focus of securing long-term high-quality cash flows.

The remaining discretionary investments were focused on opportunities to expand our unique portfolio of shared wireless infrastructure assets and position us to capture more future growth opportunities as our carrier customers continue to invest in their wireless networks. We believe these net investments will generate compelling long-term returns and enhance our ability to deliver long-term growth in dividends per share.

As an example of the compelling returns we generate I wanted to highlight the investments we have made in our small cell business. To date we have invested $2.6 billion on a net basis in small cells. These investments have resulted in that business now generating over $400 million per year of site rental revenues and a recurring yield of 6% to 7%.

In broad terms this yield is comprised of three major investments. First, in 2012 we invested $1 billion from NextG which provided us with an industry-leading platform and significant first mover advantage that we see the benefits of on a daily basis. This investment was made at an initial yield of 3%.

Second, in 2015 we invested about $1 billion to purchase Sunesys, which provided us with an attractive, dense dark metro fiber footprint in top markets where significant small cell appointments are occurring. This investment was made at an initial yield of 6% to 7%.

Lastly, we have made $600 million of debt investments outside of these two acquisitions which give us an expanding asset base on which to deploy additional small cells. These investments are generating a yield of low-double-digits demonstrating the higher margins and returns we are driving through co-location.

As Jay mentioned, these results reinforce our view that small cells exhibit similar share economics as towers and provide us with attractive growth opportunities in the future.

Moving on to financing activities, during the quarter we once again accessed the investment-grade bond market to proactively manage our balance sheet, issuing $700 million in unsecured notes to refinance debt maturities coming due in 2017 from borrowings under our credit facilities. With this successful refinancing our balance sheet remains in great shape with no significant maturities between now and 2020, ample liquidity under our credit facilities and approximately 80% fixed rate debt.

Additionally during the quarter, we returned significant capital to our shareholders through our quarterly common stock dividend of $299 million in the aggregate or, $0.885 per share.

Turning to our 2017 outlook on slide 7, focusing on the midpoint and going from left to right, our expectations for growth in site rental revenues of approximately $100 million includes approximately $155 million from organic contribution to site rental revenues which is comprised of the following: Anticipated new leasing activity of approximately $160 million which assumes approximately $15 million of higher growth from small cells and a level of growth from towers consistent with the healthy levels we are seeing in 2016.

All of this is offset by approximately $25 million less growth from the amortization of deferred credits, which we commonly refer to as prepaid rent, as compared to the growth we saw in 2016. Importantly, we expect to collect a similar level of prepaid rent in 2017 as we were collecting in 2016.

In addition to new leasing activity our organic growth is also supported by annual contracted tenant escalators which we expect will contribute approximately $80 million to $85 million in growth during 2017. Growth from new leasing activity and tenant escalations is expected to be offset by tenant non-renewals of approximately $85 million to arrive at organic growth of approximately $155 million or 5% growth year over year.

The change in straight line, which is an adjustment from GAAP purposes related to our fixed-rate escalators, and contributions from acquisitions made during 2016, bridge the remaining amount to arrive at expected site rental revenue growth of around $100 million.

Turning to slide 8, I want to spend a moment walking you through the approximately $145 million expected growth in AFFO the midpoint of our outlook. Picking up where we left off on the prior slide with organic contribution of site rental revenues, the $155 million of growth the partially offset by a $35 million increase in normal ongoing operating and G&A expenses as well as a $10 million reduction in services gross margin from 2016 to 2017.

Additionally, with the conversion of our mandatory preferred stock, AFFO increases by $44 million representing the amount of annual preferred dividends we paid in 2016 that we're no longer obligated to pay in 2017. And finally, 2017 AFFO growth is expected to be negatively impacted by approximately $10 million on a net basis of other items including cash interest expense, sustaining capital expenditures and the contribution from acquisitions that have occurred in 2016.

The growth of $145 million at the midpoint and AFFO represents an approximate 9% increase over 2016. On a per-share basis AFFO growth is expected to be 6% including about a 100 basis point headwind from the preferreds that will convert into approximately 11.6 million additional common chairs on November 1 of this year.

Based on our expectations for 2017, and our outperformance in 2016 compared to our initial outlook, we have increased our quarterly dividend by 7% per share, further adding to our track record of growing our dividend.

As Jay mentioned, we remain confident in our portfolio of shared wireless infrastructure assets -- assets uniquely positions Crown Castle to meet our carrier customers' needs as they continue to build out and enhance their wireless networks to meet the rapid growth and demand for wireless connectivity.

These fundamentals give us confidence in our ability to increase our AFFO and dividends per share by 6% to 7% per year over the coming years, which, when combined with our current dividend yield of approximately 4%, we believe provides a very attractive total return profile.

With that, Vicki, I would like to open up the call to questions.

(Operator Instructions). Simon Flannery, Morgan Stanley.

Great, thanks very much, good morning. Thanks for the color on the small cells in Chicago. Staying on that theme, can you just talk about the latest thoughts on the zoning and the municipalities? There is a lot of articles about the opposition of some towns to certain carrier deployments and to the various providers including yourselves. So where is that coming out? What is Washington doing about it?

And then talk a little bit more about Chicago in terms of the lease up from the carriers. How are you doing in terms of getting additional tenants on obviously the node densities improving? Is that mostly from the initial tenant or are you also getting additional tenants on? Thanks.

Sure, Simon. On the first question, obviously one of the great benefits of our business, both on the tower side and the small cell side, is that there are very high barriers to entry in the business. And so, we are accustomed to working through the various local municipalities and what the zoning regulations in those municipalities are.

And over time we have worked with Washington to put in what we believe is some smart federal legislation that creates a path that gets to a timely answer from the local municipalities and communities. And we continue to see the impact of conversations with the local municipalities as we deploy small cells and I don't think that is going to come to an end anytime soon.

But I would also tell you that that really demonstrates some of the capabilities and expertise that we have in the business that we are used to dealing with that and can get our carriers deployed on time frames that are reasonable. And so it is not in any way inhibiting our ability to deploy these networks on a reasonable timeframe. But it certainly gives us an opportunity to set ourselves apart from some of the competition in the market.

On your second question, in Chicago, it is a combination of both lease up on the existing fiber, as well as us continuing to deploy additional fiber as we have expanded our fiber footprint around Chicago. As you can see from that slide and the picture, we have increased the number of nodes per mile on that first 100 miles of fiber that we had in the central business district of Chicago. So a lot of that would be lease up, if you will, or co-location on that base of fiber.

And then at the same time we've built out a number of miles, about 150 miles, beyond that central business district out into the suburbs. And while we were building that initially for one carrier we have seen additional co-location beyond that sort of initial build. And it is reflected -- the Chicago market is reflective of really what is happening across the whole country.

We are seeing co-location, which we have talked about, about -- in the neighborhood of about three-quarters of the activity that we are doing is related to expanding the fiber footprint and building new systems for carriers. And then about a quarter of our activity roughly is coming from the co-location of tenants onto existing fiber plan. And the Chicago example shows kind of that same thing as you look at one individual market and how that works.

Great, thanks for the color.

David Barden, Bank of America.

I guess two numbers questions if I could. Just the first one maybe Daniel or Jay. The escalators for the base case guidance for midpoint 2017 look I guess on my math to be about 2.6%, which is down actually from the escalators that were part of the equation as we looked into 2016 from 2015. So if you could kind of talk about why the escalators are moving around like that, that would be helpful.

And then the second part was just on the straight-line amortization piece of it. I was wondering if you could kind of maybe elaborate a little bit more on kind of as new prepaid rents are getting paid I think you said it would be flat in 2017 versus 2016. How you are amortizing those so we could kind of model those in.

And second, where they are coming from. Is it mostly the prepaid rents for the small cell builds or is it macro or is it some mix? If you could kind of give us some color there it would be helpful. Thanks.

Yes, thanks, David. On the escalators I think the best way to think about that one is we have about 85% of our contracts that have fixed rate escalators at 3% and the other 15% are in the CPI range. So with CPI being as low as it is right now the average is that 2.6% or 2.7% that we are seeing and what we have in our guidance for 2017.

On what you called straight-line amortization, the second question you asked, I think you were mixing concepts a little bit. What you had asked what was really on the prepaid rent which is different than straight-line amortization. So on the prepaid rent amortization, what it is is as we get reimbursed for capital upfront we do amortize that over the life of the contract.

So as you think about it, we have about six or seven years remaining on the contract, that is probably not a bad way to model it going forward is over that amortization [line]. And where that is coming from is both towers and small cells.

But the reason that it is reducing [and that] $25 million is lower growth in 2017 than it has been in 2016 is that as the tower business has a similar level of activity from year-over-year, the growth in prepaid rent that we are collecting is slowing down to be very similar to where it has been year-over-year, which is what we think is happening in 2017.

And therefore the growth and amortization on the towers is slowing down. Whereas on small cells, as we are building more and more systems, we are still growing the amount of prepaid rent we are collecting and therefore the prepaid amortization based on the activity that we are seeing in small cells.

Okay, yes, you are right. I confused straight-line and amortization. Amortization was what I was interested in, so thank you for that. And so, just to clarify on the escalators again. So if the 85% and 3% are fixed and the 15% is a CPI, how did the CPI change so dramatically from your expectation for this year versus last year?

Yes, I think if you -- Dave, if you look at it over time -- and this is where the supplement I think is really helpful, where we have laid out in the supplement over many years what we expect the escalation on those contracts to be.

You will recall a number of years ago where we went down the path with some of the carriers and they prepaid for rent and that was often in the form of increased steps in their contracts as they paid for the right to amend the sites rather than just doing it one at a time. And so those were embedded in the escalations on those contracts.

As you will recall, for virtually all of 2016 we have been back on a -- as they did amendments they would pay for each individual. And so we saw our last step during calendar year 2016. So that accounts for most of the movement downward.

But in terms with looking at what it will look like kind of over time, I think the best place to go is look at the supplement as we have modeled out your contractual steps into the future so that you can get your model right.

Thanks, guys.

Brett Feldman, Goldman Sachs.

Thanks for taking the question and I guess I will go back to Chicago. If I look at the slide, it certainly is supportive of the bullish view you had on the addressable market for small cells. But if you could flip it around and you look at it from someone who is purely interested in the tower space and say it confirms worst fears that small cells might be cannibalizing towers.

Because it would seem like from that map that some of those new areas would be traditionally tower areas. And so I was hoping you could just give us a little bit more color as to the capacity and coverage problems that these small cell deployments outside the urban core were solving and why was it that towers weren't being used in those circumstances?

Sure, Brett, appreciate that question. Small cells are a means by which the carriers can make the macro sites most efficient and effective in terms of utilizing the spectrum on that particular site. As I mentioned in my comments, and we can see this in the Chicago deployment or we could take you to kind of any major US market where we've deployed small cells and show you a similar trend from the carriers. They are using the macro sites because they are cost effective and efficient.

So if you could solve those challenges with macro sites they would end up co-locating equipment on another macro site. But most of the time these small cells are being put in and around and very close to existing macro sites in order to reuse the spectrum and create additional capacity.

And what happens when they do that is it frees up the tower to provide a bit of an umbrella, if you can think of it that way, an umbrella of coverage over an area and enables the user to have a more ubiquitous and uniform experience with their mobile device.

From a practical standpoint the way all of us as users often see this is we will have four or five bars on our phone and yet we can't connect and experience a reasonable data session. The reason for that is that you are receiving or seeing the spectrum and the signal at your handset device, but there is not enough capacity in the spectrum to provide that.

So by putting a small cell in the general facility of where there is an existing macro site you are offloading some of that capacity without changing the coverage or the necessity of the macro site to deliver the data experience to the consumer. And so they are using small cells in essence underneath these micro sites which both improves the consumer experience as well as frees up the macro site to be as efficient and as effective as it was originally designed.

If you were to go -- if we were to go into that -- this example and blow it up to an even greater detail than what we are showing on the slide there, what you would see is macro sites in the very areas that we are putting in small cells. But we don't believe, based on the conversations we have had with the carriers, their continued desire to both lease towers, as we talked about on this call, at the same pace that they are doing throughout 2016 and that continuing into 2017.

We don't see any slowdown in the activity around towers, nor do we see them using small cells as a replacement. It is really just improving the effectiveness of those macro sites that they have.

Do you have towers in that area? Just out of curiosity.

We do.

And do you think that they chose to do this instead of using those towers? Or the point was the towers you have aren't solving the problem that we have right here?

The challenge is that the data usage in the general vicinity of the tower is so significant that the geography that is intended to be covered by the tower can't be covered by that macro site alone. So they go in and offload some of the traffic onto the small cell, which then makes that macro site effective for which it was designed to cover a broader geography.

Okay, got it. Thank you for taking the question.

Phil Cusick, JPMorgan.

Two. One, first on the Chicago example if I can. Can you talk about what the backhaul situation is there? What is the mix of owned versus leased fiber? And what do you see in the cost trends for that?

Yes, we own all of the fiber that we are showing there in the Chicago market.

And it is all you have constructed, that is not IRUs or anything like that?

That is correct. We have constructed the fiber or in some cases we have acquired portions of it over time through the acquisitions that Dan referenced in his comments. But we own it all.

Understood. And can you talk to us more about the expansion of small cell revenue expectations in 2017? We have heard different times of expectations of this business accelerating, the potential of putting more capital into it. How do you think about that and is there potential to put incremental amounts of capital into that each year going forward?

Phil, as we talk about and look at this business, first of all let me answer the simple question, then I will talk the kind of our macro view. The simple answer to your question is, we are expecting more activity around small cells in 2017 than what we are seeing in 2016 and that is the driver of the $15 million of additional growth -- increased growth in 2017 as compared to 2016. That is a result of as we have expanded the asset base we are seeing co-location and continuing to invest in the areas where we are expanding that fiber footprint.

Stepping back from the specifics of the year-over-year change, we have a view that over the next decade there is going to be a significant expansion of small cells utilized by all of the carriers to improve the data experience that consumers enjoy. And as we look across the business that we have today and the $3 billion of capital that we have invested, we think we have barely scratched the surface on what the opportunity is there.

In the near-term, given the amount of time it takes to build these projects in the neighborhood of 12 to 18 months, we have a lot of visibility around what we think that revenue growth is which we are providing in our 2017 outlook. But beyond that we think there is potential for additional upside as both we invest and expand these systems and at the same time co-locate additional tenant nodes on the systems that have already been built.

And the other thing I would say about this is this entire strategy really comes back to sort of the core driver of everything that we think around the business. We are investing capital with the goal of increasing our dividend over the long-term.

And so the opportunities that we see in small cells and the returns that we are talking about, we are excited about them because of their potential to continue to drive the dividend over the long period of time, and potentially even increase the growth rate of that dividend growth over a long period of time.

And everything that we are seeing, as we talked about the business as a whole, you look at the sequential change or the year-over-year change in small cells, we are seeing growth in incremental margins and we are seeing expansion of the revenue. That certainly tells the story. And then if we go all the way down to a single market area we see that thesis that we have had for several years really playing out as we expected and we are pretty excited about what it could mean for us over the long term.

Understood. Thanks, Jay.

Ric Prentiss, Raymond James.

A couple questions still on the prepaid rent side. Can you break out for us a little bit as far as on the prepaid rent received that I think you said 2017 would be similar to 2016? How much of that is going to small cell prepayment received versus the tower side?

Ric, we are not going to give specific guidance to what we see on prepaid rent received. But you can see in our supplement that we break out both the prepaid rent received and the amortization of prepaid rent between small cells and towers.

And you can annualize the nine-month numbers that are in there; they are about similar right now. We will see some continued growth in small cells and, like we are saying, we will see less growth in towers. So the small cells will get a little bigger than the towers over 2017.

Okay. And then the mentioning that you are going to have $25 million less growth in the amortization in 2017 versus 2016. How much was the growth in the amortization of 2016 versus 2015 in the current midpoint?

Yes, again, if you just look at the supplement and take the nine-month numbers and annualize them and then compare them to 2015 you get about $50 million of additional growth from prepaid rent amortization 2016 over 2015.

Okay. And then some questions about -- like Brett's question about cannibalization. Can you update us a little bit on just if a carrier wanted to get off of a macro tower, if their contract was up and they wanted to move to a small cell system or move to somebody that was able to get a new tower built nearby, what is the estimate that you guys have as far as what that switching cost would be?

How much would it cost them to actually physically get off the tower? I know it all varies by how much equipment, but is there a good kind of rule of thumb average that people should think about?

Well, just coming off of the tower is one cost, that would be in the neighborhood of probably $35,000 to $40,000 to take the equipment down. The real issue is the geographic coverage of small cells as compared to a tower. A tower is going to cover anywhere between an eighth and a half a mile of geography around a given macro site. And a small cell is going to cover a couple hundred feet.

And so, if you were trying to replicate with small cells the coverage of towers, the quantity of those would probably be well in excess of 15, 20, 25, maybe 30 plus small cells in order to get anywhere close to the capacity of a macro -- a single macro site. And the economics of a small cell are about a third of that of towers.

So from a cost standpoint it just doesn't ever make sense to utilize small cells as a replacement for macro sites, which is why we don't see that happening. But where they do do it is, as I mentioned in the comments, Ric, earlier to Brett, where they are doing it is in places where it makes that macro site more efficient.

So I just don't -- we are not concerned at all that small cells are going to cannibalize macro sites and the demand for macro sites because of the cost element of that. And if you look at what is the main driver of that cost, it is the cost to construct and build the fiber which makes up the vast majority of the cost of small cell nodes.

And that cost is not going to come down over time. That is a construction cost, it is likely an escalating cost over time rather than a cost that would decline say with the reduction of the cost of technology. So I think the economics, if you look at it on a comparison basis, would suggest macro sites are definitely the preferable approach for the carriers when it will solve the problem.

Makes sense. And one quick one on escalators like David asked. So should we assume that the 85%-15% split as we think about taking escalators into the future, not just 2017, but that 2018 or 2019 should look like 2017 if there were no other changes to CPI then?

Yes, Rick, that is accurate. And you can see that in the supplement. But beyond that it is about 85% fixed and the 15% floating. And so the real variable that will change over time is where does CPI fall out and it could -- given how the CPI is currently I guess it is unlikely that it cuts against us, more likely it cuts for us.

Sure. Great, thanks, guys.

Matthew Niknam, Deutsche Bank.

Just one on fourth-quarter site leasing guidance. So you saw about a $7 million uptick this quarter in site leasing revenues, the midpoint of the guide next quarter implies about a $1.5 million lift sequentially. So maybe if you can explain what is driving the slowdown.

And then second, on the 2017 outlook, just wondering how you get to that 5% core growth next year specifically around how you build in new activity and non renewals split up over the course of a year? Thanks.

Yes, on your first question around sequential growth, it has to do with the timing of leasing and churn. The amount of churn that we have in the fourth quarter is really driving that. As I mentioned, I think we have said a couple of different ways, in terms of the activity that we see we are not seeing any slowdown in the activity and we think the level of 2016 continues into 2017. So it is really around the timing of churn.

Yes, so, on your second question on what we are building into the 5% growth for 2017, we have talked about the push-out of churn from 2016 into 2017. You can see that in the supplement where some of the consolidation churn that we expected to happen in 2016 we have now pushed out to 2017. So what that does is twofold, it increases the 2016 base and decreases what could happen in 2017, so that is baked in.

And the second is what Jay talked about as part of the prepared remarks is that we do see $15 million of incremental growth on small cells, but we talked about historically that being about $55 million of growth in small cell revenues in 2016; we are looking for something in the $70 million neighborhood going into 2017. And then from macro towers we see similar growth from 2016 going into 2017.

So no real slow down at all and a pickup in the small cells, which when you put all that together gets to the growth we have shown, it is just unfortunately it is offset by this prepaid rent that we talked about a bit on the call today that is coming down in growth by $25 million which reduces the growth rate a fair amount. So we see on kind of the core activity growth small cells increasing, towers staying the same and the non-cash portion coming down from a growth perspective.

And just to expand a bit on that under prepaid, it is really related to prior year more than this year because the amortization over a long period of time. And so, really as the leveling off occurs, like I mentioned earlier of the tower business, that is was driving the lower growth as opposed to anything that is impacting this year's activity.

And just to follow up, in terms of pacing of new activity next year, I know traditionally it has been a little bit more second-half weighted. And any change in terms of how we should think about that for 2017?

No, Matt, no change in what is typical. So as we go into the year we would expect next year to be a bit backend loaded.

Okay, great. Thank you.

Amir Rozwadowski, Barclays.

I was wondering if we could just revisit the questions earlier around the escalators. It seems like the way we should think about it is it's really a function of that mix relative to the contribution of fixed versus CPI-related contracts. But clearly there has been a lot of chatter by select carriers trying to express their views on the trajectory of escalators going forward.

Have you seen any push backs or discussions? And would love just sort of the thought process around where those discussions have gone and perhaps why you feel comfortable that the fixed rate that you have in place right now is something that we should consider going forward?

Yes, Amir, appreciate the question. I think one of the -- and this question has obviously come up a number of times over the last few quarters. One of the benefits of the supplement I would say is that you can see clearly and go back over time and see that there hasn't been any change in that escalator, it is set right in where we expected it to several years ago.

So we have not seen any change in our pricing constructs with our tenants and don't expect that we are going to see any change in that pricing construct. Ultimately the conversation around price goes to the very attractive value proposition that we have with the carriers. We are sharing the infrastructure across multiple carriers and the cost of their leasing of these sites or use of these sites is a cost that is sub 3%.

So when they look at including the effects of all of the amendments that they have done over time and adding additional equipment on the site in order to upgrade through multiple technologies, we're still providing that infrastructure at a very attractive cost proposition to them.

And so, when there is conversations around the construct and the lease arrangement between us and the carriers, which there have been dating all the way back to 1999 when we first leased these assets, the conversation always went right to what is the value proposition, what are they receiving in exchange for what the cost otherwise would have been. And that value proposition remains intact and we think it remains very attractive.

The last point I will make about this is the underlying cost of this real estate, as Simon referenced earlier, around municipalities and zoning and other things has driven the cost of the real estate up faster than the carrier's cost to share that infrastructure over time.

And so, the underlying cost of the real estate is very supportive of what we are able to achieve on the escalation side. And we see that remaining intact and we haven't seen and nor do I expect we will see a real change in the pricing construct.

That is very helpful. And if I may, one quick follow up. We are starting to get some data points about potentially some new spectrum deployment starting in 2017, whether it is AWS-3 or even [FirstNet] seemingly getting a lot more traction. What are your thought process around the deployments of those spectrum opportunities? And how should we think about those opportunities in relation to your current expectations for growth?

Yes, we are assuming largely what is in the market today in terms of our expectation for growth. And we are not assuming additional spectrum auctions or additional deployments beyond what is there. So those could be potential upsides to our forecast.

What I would tell you over time, as we have seen multiple auctions done and spectrum in the hands of the operators, that is very likely to extend our runway of growth. So as we talk about our long-term expectation of growth, it supports our view; we believe that we will be able to sustain our 6% to 7% growth over a long period of time.

Excellent. Thank you so much for the incremental color.

Colby Synesael, Cowen and Company.

Two questions if I may. The first one I'm just curious if you are finding that having both a macro portfolio as well as small cells is helping one another. So having small cells has helped your macro business or having macro in a given area has helped you win small cell deals that were otherwise I guess being competitively bid.

And then my second question has to do with the cost of small cells. So at CTIA one of the takeaways I had was that the carriers are pushing that if we are going to build out hundreds of thousands if not millions of these over a period of time versus today -- the typical carrier has I don't know 40,000 or 60,000 or 70,000 macro cell sites out there -- the cost has to come down significantly. And there is talk about is $500 a node or $200 a node the right price.

And one of the ways I've thought about it is that you could actually, as a fiber provider, go and share that cost not just across various wireless carriers but across other industry verticals that might be interested in fiber such as healthcare or cloud companies or financial services or education, etc.

How big of a focus is that for you? And do you see yourself splitting the cost or sharing the cost of fiber in small cells with those other verticals? And does that become an increasing area of focus for you as you go forward? Thanks.

You bet, Colby. On your first question related to does having small cells help the macro tower business. I would start by just affirming as we've done multiple times in the call the tower business is a phenomenal business. And whether you are in the small cell business or not we think there is tremendous value that has been created in the tower business obviously.

But we also believe there is a tremendous amount of value that is going to be created in the tower business. And we don't view being in the small cell business as a requirement to be successful at continued success around the tower business and the growth opportunity there. So I don't think you have to be in the small cell business.

The opportunity that we see in small cells, we are there because we believe that we'll increase and drive dividend growth over a long period of time. So we believe it is a smart allocation of capital that continues to further the growth model. But don't view it as a necessity to be in.

There is certainly -- incrementally at senior levels of the carriers it enables a broader conversation about how they are thinking about network deployment. We think that is helpful and informative to us as we think about the long-term as we can have a broader conversation. But depending on the returns there may be a day when we choose not to invest as heavily in small cells. So it is really a return driven analysis for us. But what it meant there is some incremental value around the conversation with the carriers.

On the second point and your question around do costs need to come down. Again, just like in towers, small cells are a shared infrastructure asset that -- the economics of which are far more attractive to the carriers as a result of us sharing among multiple parties than they would be if they were self performed or built for only one carrier.

So the costs effectively are coming down by us being a third-party provider and the initial carrier not bearing the full cost of the deployment and the ongoing operating cost. We are going to share that across multiple operators.

As we think about the potential growth drivers of revenue and cash flow over time as we think about that as a shareable asset, our primary focus remains on the wireless carriers. And so using this fiber deployment for small cells, we think it is the vast majority of the economic driver for both growth and the opportunity that we see.

We have, as you know from our Sunesys acquisition and in certain geographies, we have used our fiber to provide things like K-12 which is providing broadband to schools, those are long-term contracts that have provisions that are very similar to towers and that of small cells. And that obviously puts fiber in the areas where people are located and where there is a density of population.

And so we have found over time those there are some real synergies of revenues around using that fiber for both schools and then ultimately small cells or small cells and then secondarily to schools. So there may be over time ways for us to continue to [sweat] the assets. That is not our primary driver at this point, but we do look at some specific cases where we see opportunity for us to expand revenue and the cash burn.

Thank you very much.

Matthew Heinz, Stifel.

I was hoping you could kind of give us a little more detail on the approximate split of new construction and co-location activity that is contemplated in your small cell guidance for roughly the $70 million of new leasing, I guess just to get a sense of the capital intensity of that growth.

Sure. About 75%, roughly 3/4 of the growth is being driven by the deployment of new systems in areas where we don't have systems today. And that could be -- when I say new areas that could be like the Chicago example where we are expanding beyond the central business district but still in the Chicago market. And then where about 1/4 roughly of the activity is coming from co-location on existing plant.

Okay and then just to confirm, the new construction activity does go under new leasing, is that correct?

That is correct.

I guess it is a little bit different from maybe how you would account for a new tower?

That is correct. On the tower side that is separated over in the acquisitions and new tower builds. As the Chicago example shows, and the reason for doing that, first of all you are accurate in your description. We have both the new and the co-location in the leasing bucket on the left-hand side of the chart there. And then on the tower side if we build a new tower that goes over into tower builds and acquisitions. So your assumption is accurate.

We do that because we find it very difficult to allocate both capital and revenue of cash flow. As we will build laterals on existing systems at what point is the lateral a new system versus a part of the existing system? And so for ease of use we have just combined those. And again, 3/4 plus of the activity at this point continues to be the deployment of new systems.

Okay. And then you mentioned that potential upside to your guidance was possible? And you think that if that were to occur it would most likely be on the co-location side?

No, Matthew, I would point to the opportunity for us to do it both on the deployment of additional systems as well as co-location. We think those two things will go hand in hand. We are building the early stages of this plant very similar to the early days of towers where towers were being deployed starting at the central areas of urban areas and then expanding out to the suburbs. And that is happening in small cells.

We think that will continue from the markets that we've been in where the vast majority of the activity has been in and around the top 10 markets in the US. We see that expanding outside of the top 10 markets. We also see it expanding beyond the central business districts of those markets. So our expectation today would be that over time we will continue to invest additional capital as the carriers deploy small cells to improve their network.

Okay, that is very helpful, thanks. And then if I could just go over to the non-renewals. It looks like the expected churn from acquired networks in 2017 was down $15 million at the midpoint from your last report. I am just wondering how I should square that up with your comments that churn -- non-renewals have been pushed out from 2016 into 2017.

And then if I sort of split out the acquired network churn from the normal churn I get to about $30 million of, quote, normal churn or 90 basis points, which I think is probably below your historical experience. Just wondering what you are seeing there in the business that is kind of reducing the go-forward churn expectations on the normalized side.

Thanks, Matthew. To your first point, squaring up the 2016 to 2017 move, some churn was pushed out from 2016 to 2017, but other churn was pushed out from 2017 to beyond. So you can see the beyond column in our supplement will go up as well to make up for that.

So, what it looks like is there is lower churn than we expected in 2016. And even though some of that pushed out into 2017, some got pushed even further out. So that is how you square those two things, if that makes sense.

Okay.

And then in terms of normal churn, I think you are right. You back out what we see as consolidation churn and it leaves around 90 basis points is your math, I would say around 1% is what we are seeing right now, yes. And it is a little less than what we have seen historically and talked about historically but 1% is what we are seeing now.

Okay, thanks very much.

Spencer Kurn, New Street Research. Hearing no response we will move on to Nick Del Deo with MoffettNathanson.

I recognize enjoying the process, but can you speak to what you are seeing in terms of [amendment] revenue from AWS-3 upgrades? And more specifically, are the amendment rates consistent with past amendment cycles or are customers trying to put up antennas they can run both AWS-1 and AWS-3 which might garner more modest incremental rents?

Yes, Nick, on the amendment side we are seeing pretty similar both rates on the equipment as well as size of equipment, etc., from amendment cycles. So it looks pretty similar to what we have seen over many years. We would expect, as you referenced, given what is going on in the market, amendment activity as a component of total leasing activity may increase as a percentage of that in 2017 as the carriers focus on some of the activities that you referenced.

Okay, great. And then maybe a longer-term question. If I look at your 2017 outlook, if my math is right, you seem to be guiding to something like an incremental cash EBITDA margin on site leasing revenue of about 80% next year, maybe about 75% if you strip out the amortization of prepaid rent.

Is that something we should be thinking of as sort of a normal incremental margin over the next several years given the tower and small cell mix? Or do you think there is the potential to do better as the small cell platform scales?

Let's separate the two and talk about towers and small cells separately. On the tower side, when you look at the cost structure both at the direct to operating expense side and at the G&A side, we are holding costs effectively flat. We would expect in normal course just the cost of living adjustment to be about 3% increases year-over-year. And we are well inside of that on towers basically holding everything roughly flat year over year. So that drives incremental margins that are extremely high, well over 80%, about 90% would be our expectation as we think about 2017 on the tower side.

On the small cell side we've got a combination of two things going on. On the systems where we are co-locating we're seeing very high incremental returns which expand the margins on the systems where we are seeing co-location. As we put new systems on air, the margins on those new systems are in the neighborhood of about 60% for a new system.

So those immature assets, if you want to think about them that way, those immature assets with slightly lower margins than the overall average basically hold the average down and we don't see the full benefit of the incremental margins that we are seeing on small cells from a co-location standpoint. And how that progresses over time will largely be a function of how many opportunities do we have in front of us, how much do we invest on the deployment of additional systems and then the margins will fall out from there.

If you fast-forward all the way to the end of this when the market becomes more saturated in terms of the deployment of the systems, what we see on individual systems is margins that expand at paces that are in line with what we have seen historically for towers. And so, rolling forward to a very long view of small cells, I think you continue to see the margins expand and approaching levels that we have enjoyed in the tower business.

Yes, Nick, before we move off of that, just if you look at the $35 million increase in OpEx and G&A year-over-year, that is less than 3% of our total cost of structure. And that is increasing activity. So I do think what Jay was talking about is we are seeing some of that operating leverage now.

Okay, terrific. Thanks for the color, guys.

Michael Rollins, Citi Research.

I was wondering if you can refer to your press release where you do the floating bar charts where you break down the leasing activity, the escalators, the renewals, etc. Why not move the impact of prepaid rent amortization from the organic conversation over to the way that you look at straight line and acquisitions? And if you were to do that, can you help us think through what that says about what activity might have been in 2016 versus what activity looks like in 2017? Thanks.

Yes, so the reason is because prepaid rent really is an economic decision that our customers and we are making together. And we are getting that runt either up front or as part of the ongoing operation of the business. So it is really tied on some level to the new leasing activity.

So it is part of the economics we enter into. And if we hadn't gotten the rent up front we would be putting it in new leasing activity revenue. But if we don't put it there or take credit for it from a capital side it gets lost in that. So we think of it as tied to leasing activity.

The reason why it is growing less isn't necessarily reflective of the current activity in 2016 or 2017. Because the activity in 2016 we said is about flat from 2015 and is about flat to 2017, that is why it is actually plateauing and the growth in prepaid rent amortization is coming down.

So what it is speaking to specifically on the tower side is that growth is the same year over year, which is why you are not seeing incremental prepaid amortization. Whereas on the small cell side you are because we are continuing to invest in that business, grow it and get more prepaid rent that we are collecting on that business.

(Multiple speakers).

Michael, you could say -- and maybe just an additional point there which maybe was the second part of your question. Given that we referenced the amortization, what is related to the tower business, basically towers on an apples-to-apples basis is about flat of about $90 million roughly year over year. And then on the small cell side it is up $15 million. So it as gone from about $55 million to about $70 million.

So if you thought about it and you wanted to excluded kind of the prepaid conversation from that analysis, then the growth is up about 100 basis points if you were to take -- as you were saying, take a portion of that graph and move it kind of over to the right-hand side. And that 100 basis points is being driven entirely from small cells.

And then if you look at the $55 million to $70 million for small cells, how much of that is being driven by incremental co-location versus capital spending and deploying new systems?

I would say it is about that same split. So I would put about 3/4 of that to new systems and about 1/4 of that roughly to co-location.

Thanks very much.

Batya Levi, UBS.

A follow-up on the CapEx side. As you look to invest more on the small cells next year can you provide maybe a dollar amount in terms of how much construction CapEx you expect to spend? And more strategically, as you are looking to invest more small cells can you talk about your decision making thought process in terms of building versus acquiring for more fiber assets and maybe leveraging some of your REIT multiple in to acquire some assets that are not getting the REIT multiple. Thank you.

Let me take the first one and I will probably defer to Jay on the second one, which is on the capital side on construction it is similar year-over-year from 2016 to 2017 on the small cell side so you should expect to see generally the same amount of spending.

Batya, on your second question, we will look at and will continue to look at opportunities to acquire fiber assets. But the driver of that, as I mentioned earlier around our strategy, is to lease that fiber to wireless operators primarily for the use of small cells. And so to the extent that we can identify fiber in a given market that we think makes sense on that basis, then we would be interested in the asset.

But we don't spend a lot of time around here thinking about arbitrageurs and multiples and things. Businesses are valued where they are valued. And so our approach is to invest capital in ways that drive the dividend over the long-term and our strategy for doing that is focused on the wireless operators and deploying small cells.

Got it, thank you.

Walter Piecyk, BTIG.

Just a quick question on the co-location. Can you define what that means? Is that just putting another radio in the box that is already in the lamppost? Or does that also include connecting a lamppost that happens to be right above your fiber?

It would be both of those, Walter.

Got it. And then very interesting chart, Chicago, thanks. The five you said gets you to 10% returns, so just two questions on that. In five years how many tenants per mile do you expect to get? You can obviously give a range and best case scenario or worse case, whatever, assuming that we get to like as an industry 1 million plus small cells with that type of backdrop.

And then if you can just give us some sense -- I know this is not easy, but assuming the 75%-25% mix that I think you have referenced now three times on the call, how the returns would increase as that five tenants per mile goes up? Is it like 200 basis points per tenant or any way you can scale that for us as far as the implied returns of adding additional tenants per mile. Thanks.

Yes. I think in Chicago I will punt on how many tenants per -- or nodes per mile of fiber we are going to ultimately get to. These systems -- and just to take you back to kind of how we would think about underwriting Chicago and how we did think about underwriting Chicago -- we would look at return based on the cost of the deployment of the fiber.

So obviously in a central business district, as you would expect, the cost of deploying that fiber would be significantly greater than as we move out to the suburbs, generally speaking in a metro market. And so the number of nodes required to get to an acceptable IRR or yield on that investment is going to differ based on the cost.

What I would tell you generally, Walter, is that the pace of leasing on small cells has been about twice the pace of that of towers that we have seen historically. And the returns, to kind of your last question, the returns track that. So we are three years in on this Chicago system roughly from 2013 to 2016 and then some of that under construction in 2017. So call it three to four years of snapshot there. And the yield in the double-digit area around 10%.

That took us over a decade on the tower side. And that kind of yield on invested capital is well in excess of our cost of capital and provides a terrific return. Now we certainly don't think we are done in Chicago, we are early days on this. And so, we think the yields and return continue to expand well beyond what we underwrote when we invested that capital.

So, but I mean as a rule of thumb, if [5%] went to [10%] presumably your returns would be much higher than 20%, right? It has to scale up from there I would think, right?

That is exactly right.

Okay. And then just one last -- just a little pushback on the urban versus kind of suburban being more expensive. If that is the case why wouldn't every operator just opt for someone that actually already had fiber in the market rather than go with someone that had to build new stuff?

The answer is that there is rarely anything there. So most of those (multiple speakers).

In the urban part of Chicago? I would think that there is fiber running up and down the streets.

Yes, there is -- this fiber is purpose built in locations that there is not a capacity of existing fiber to be able to handle these small cell nodes. There are other places where we have utilized some leased fiber where there was existing fiber in the ground, obviously that is a more cost-effective opportunity. But the vast majority of what we are doing there is not fiber in the ground to provide this. And if there is fiber in the ground there is not nearly enough capacity to be able to handle small cells.

That is helpful, thank you very much.

I would also say that we agree with you, that where we have fiber in the ground, like now you can see in Chicago, we have a great advantage of getting more small cells on that as opposed to somebody coming in and building. Because with all the things Jay just talked about and you just mentioned is we have an advantage at this point and we expect to see leasing on that fiber.

And where we don't and have seen where other people have a tremendous amount of fiber that would be utilized for small cells, we understand the disadvantage we are at and will at times choose not to go into those markets. So I think part of what (multiple speakers).

That makes sense, Dan.

Part of what you are talking about is a competitive dynamic around these things. And we feel like we have a great position competitively both from our capability standpoint but also from our footprint standpoint to really accelerate the growth in this business.

Got it. Thank you very much.

Michael Bowen, Pacific Crest.

A question, I may have missed it, but have you divulged any number of tenants per small cell site or can you give us any thoughts on, if not, kind of where you might see that going in both the near and the distant future and kind of the technology around it and how well that is working?

And then with regard to the carriers, we have been hearing obviously a lot of chatter around pressure from the carriers trying to reduce revenue escalators. Are these current conversations happening, are they not happening? Can you give us any idea on -- I understand -- you talked about the value proposition and I get all of that. But trying to get an idea as to whether this is something that might be eminent. Thanks.

Yes, Michael, appreciate the questions. We haven't disclosed the number of tenants on small cells. We have focused, as he heard from Dan's comments, on yield and return on the investments that we have made. So I think you are going to continue to hear us mention that, that way you can look at the financial statements and see the expansion of that yield.

We think that is the best way to judge it given how much the economics vary. On a market-by-market basis, number of nodes may not be really indicative of the return. So we are focused much more on the capital investment and then what the return of that capital investment is.

On your second question around the carriers and conversations. As I mentioned a little earlier, we have seen no change in the pricing construct with our tenants. And given the value proposition that we provide for those carriers with the cost to lease the assets of something in and around a little bit less than 3%, we think that model continues to hold well into the future.

Okay, thanks.

You bet. Operator, let's take one more question.

Michael Hodel, Morningstar.

I just had a question on the Chicago map again, going back to that. There are certain areas where I would have expected to see maybe a little bit more density actually -- think of a market like Oak Park where you have an affluent suburb that is urban in nature. And I am kind of getting back to the competitive dynamic question.

Is this a situation where the carriers are bidding out kind of neighborhood by neighborhood and there are certain areas where you haven't been able to win the business? Or is this just a matter of progression?

And then also going back to the idea that in some areas the local phone company or the cable company might have existing infrastructure that would be more efficiently leveraged. Is it fair to say that in those markets where a Verizon or an AT&T has built more fiber, like a FiOS type of neighborhood, that takes one potential customer out of the mix for you guys?

On your first question I would just say we are not done yet. So I think you are just think the progression of the build out of wireless networks using small cells. And so we would share your view that over time as we look at that map we think there will be an increase in density. And if you look at some of the suburbs that you pointed out, Oak Park, we would expect that there will be over time greater density of both fiber and nodes in those areas.

On your second question, again, this goes back to the existing fiber and the capacity of that fiber to provide this kind of solution. It takes a lot of capacity of fiber and generally speaking the residential fiber that is used for products like you mentioned is not going to have enough capacity to handle small cells. So if it does it is a great solution and that would be the lowest cost alternative. But typically that is not a competitor really in our space for small cells.

Okay, great. Thank you very much.

You bet. I just want to thank everyone for joining this morning. There may be a few folks that we weren't able to get to your questions. Obviously feel free today to reach out to us, we would be happy to take the questions off-line and answer anything we didn't get to. We are excited about how the business is performing and what we are seeing and look forward to catching up with you early in the New Year to talk about fourth-quarter results.

And thank you very much. That does conclude our conference for today. I would like to thank everyone for your participation and have a great day.