Crown Castle Inc (CCI) 2015 Q3 法說會逐字稿

Good day and welcome to the Crown Castle International Q3 2015 earnings conference call. Today's conference is being recorded. At this time I'd like to turn the conference over to Son Nguyen. Please go ahead.

Thank you, Hannah. And good morning, everyone. Thank you for joining us today as we review our third-quarter 2015 results. With me on the call this morning are Ben Moreland, Crown Castle's Chief Executive Officer; and Jay Brown, 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 risk factors which could affect our results is available in the press release and under the risk factors sections of the Company's SEC filings. Our statement today are made as of October 22, 2015 and we assume no obligation to update any forward-looking statements.

In addition, today's call includes discussions of certain non-GAAP financial measures. Tables reconciling these non-GAAP financial measures are available in the supplemental information package in the investor section of the Company's website at crowncastle.com.

With that, I will turn the call over to Jay.

Thanks, Son. And good morning, everyone. As you can see from our third-quarter results we continue to execute at a consistently high level, exceeding the high-end of our previously provided third-quarter outlook which allowed us to raise the outlook for the full year.

At the mid point our updated full-year 2015 outlook for AFFO per share of $4.30 is up 8% compared to 2014. Similarly, for 2016 we are also expecting to deliver 8% growth in AFFO per share, up from $4.30 per share in 2015 to $4.66 per share in 2016.

Further, our Board has approved a dividend increase, in line with our expected growth in 2016 AFFO per-share, increasing the quarterly dividend by 8% from $3.28 per share to $3.54 per share on an annualized basis. The increase in our common stock dividend reflects our commitment to delivering attractive long-term total returns for our shareholders by growing the dividend over time. The strength of our business model, quality of our portfolio and cash flows, and our strong balance sheet give us the confidence to commit to returning a significant portion of capital to shareholders in the form of a dividend, and at the same time maintaining the flexibility to invest in growth opportunities that we believe will enhance long-term growth in AFFO and dividends per share.

Turning to third-quarter results on slide 4, site rental revenue grew 7% year over year from $718 million to $765 million, an increase of $47 million. The $47 million increase is comprised of $16 million from the Sunesys acquisition which closed during the quarter, and $43 million of organic rental revenue growth, net of $12 million in straight line and other adjustments. The organic site rental revenue growth of $43 million represents growth of 6% year over year, comprised of approximately 10% growth from new leasing activity and cash escalations, net of approximately 4% from tenant nonrenewals.

Moving to slide 5, our third-quarter results for site rental revenue gross margin, adjusted EBITDA and AFFO exceeded the high end of our previously provided third-quarter 2015 outlook, which did not include the Sunesys acquisition. Even after stripping out the contribution from Sunesys, which contributed $11 million in site rental gross margin during the quarter, we would've exceeded the high end of our outlook.

Turning to investment activities as shown on slide 6, during the third quarter we invested $237 million in capital expenditures. These capital expenditures included $31 million in sustaining capital expenditures and $16 million in land purchases. Our proactive approach to achieving long-term control of the ground beneath our sites is core to our business as we look to control our largest operating expense and produce stable and growing cash flow over time.

Today nearly 40% of our site rental gross margin is generated from towers on land we own and approximately 75% on land we own or lease for more than 20 years. Where we have ground leases the average term remaining on our ground leases is approximately 30 years. Additional information regarding the land underneath our towers is available in our supplemental information package posted in the investor section of our website. Of the remaining capital expenditures, we invested $190 million in revenue-generating capital expenditures, consisting of $92 million on existing sites and $98 million on the construction of new sites, primarily small cell construction activity.

As a previously mentioned, during the quarter we also closed on the Sunesys acquisition. Sunesys owns or has the rights to nearly 10,000 miles of fiber in major metro markets across the US where we already have a small cell presence today and would expect to see meaningful amounts of small cell activity in the future. Prior to the Sunesys acquisition we owned or had rights to approximately 7,000 miles of fiber that supported 15,000 small-cell nodes, and growing.

The addition of Sunesys' well-located, high-quality fiber footprint more than doubles our fiber footprint available to support our small-cell network. Year to date, excluding Sunesys, small-cell site rental revenue has grown approximately 30%. While towers will continue to be the most efficient and cost effective way for carriers to deploy their networks, we expect carriers to make investments in small cells to enhance their macro network by bringing cell sites closer to mobile subscribers, supplementing wireless coverage and capacity provided by a tower at a height of 100 to 150 feet that covers one-eighth to one-half of a mile, with small cells deployed at a height of 20 to 30 feet on a traffic light that may cover a city block. We believe small cells represent the natural progression of network densification and cell-splitting by the carriers as they contend with consumer demand for mobile data.

Shifting to financing activities, during the quarter we paid a quarterly common stock dividend of $0.82 per share or $274 million in aggregate. Further, we continued to maintain our long-term target leverage ratio of 4 to 5 times as we remain focused on achieving an investment-grade credit rating.

Moving on to full-year 2015 outlook on slide 7, we've increased our expectations at the midpoint for site rental revenues, adjusted EBITDA and AFFO by $60 million, $37 million and $23 million, respectively. On a per-share basis our updated 2015 outlook provides for AFFO per share growth of 8% year over year compared to 2014, increasing from $3.97 per share to $4.30 per share. The increase in our 2015 outlook includes contribution from Sunesys of $40 million in site rental revenue and $30 million in site rental gross margin.

The increase in the outlook also reflects the strong results from the third quarter, an increase in our expectations for leasing activity for the remainder of the year, and our expectations that a portion of tenant nonrenewals previously expected to occur in 2015 will take place in 2016. It is important to note that our overall expectations for the number of tenant nonrenewals in aggregate remained unchanged.

For 2016, at the midpoint of our outlook, we expect AFFO per-share growth of 8% driven by the continued strength in leasing activity, the contribution of the Sunesys acquisition, and improving operating leverage in our business. On a normalized basis, adjusting for Australia and Sunesys, AFFO growth in 2016 would be between 6.5% and 7%, or towards the high end of our long-term AFFO growth target.

Moving from left to right on slide 8, our expectations for growth of approximately $150 million in site rental revenues is comprised of new leasing activity or incremental revenues from new tenant installations and amendments to existing tenant leases of approximately $170 million, consisting of tower leasing of approximately $115 million and small-cell leasing of approximately $55 million. The approximately $115 million in new leasing contributions on tower leasing is consistent with our long-term expectation of adding approximately one tenant equivalent over 10 years across our portfolio of approximately 40,000 towers, or a 10th of a tenant per tower per year. To put that 10th of a tenant into perspective, over the last six years our annual leasing activity has ranged between 0.09 and 0.13 tenants per tower.

Our organic growth is also supported by annual contracted tenant escalators on our towers and small-cell leases of approximately 3%, which we expect will contribute approximately $95 million in growth during 2016. Growth from new leasing activity and tenant escalation is expected to be offset by tenant nonrenewals of approximately $105 million, to arrive at organic site rental revenue growth of $160 million or 6% growth year over year.

As we previously disclosed, the expectations for tenant nonrenewals during 2016 is higher than our historical 1% to 2% annual nonrenewal rate due to elevated nonrenewals from the decommissioning of a portion of the Leap, Metro, and Clearwire networks. Site rental revenue growth during 2016 also benefited from year-over-year incremental contribution from the Sunesys's acquisition of approximately $60 million. More detailed information regarding our current book of contracted tenant leases, including the impact of straightline accounting and expectations for nonrenewals, is available in our supplemental information package.

Turning to slide 9, I want to spend a moment to walk you through the approximately $130 million of growth in AFFO from 2015 to 2016. Starting on the left-hand side, we have organic site rental revenue growth of approximately $160 million, offset by $35 million increase in normal ongoing operating and G&A expenses. The $35 million increase in expenses compared to the $160 million in organic site rental revenue growth drives incremental margins of approximately 80%, illustrating the improving operating leverage in our business.

For 2016 our expectations for network services gross margin is $230 million to 250 million compared to our 2015 expectations of $280 million to $285 million. The decline in network services gross margin of approximately $40 million is attributable primarily to equipment decommissioning fees generated in 2015 which we do not expect to occur in 2016.

Additionally, 2016 AFFO growth benefits from a reduction in sustaining CapEx of approximately $25 million and other adjustments of approximately $20 million, which includes the incremental contribution from Sunesys. The approximately $130 million in expected growth in AFFO translates to approximately 8% growth in AFFO per share.

As Ben will discuss, we continue to believe that our commitment to returning capital to shareholders with a particular focus on delivering high-quality dividend stream and disciplined capital allocation focused on long-term high-quality growth will drive significant shareholder value over time. And with that I'll turn the call over to Ben.

Thanks, Jay. And thanks to all of you for joining us on the call this morning.

As you heard from Jay's remarks, based on our increased 2015 outlook we expect to finish 2015 on a strong note, and we look to continue this momentum moving into 2016. Consistent with the goal we communicated a year ago to deliver 6% to 7% annual AFFO per-share growth organically, our 2016 outlook for AFFO per-share provides for 8% growth at the midpoint, inclusive of about 150 basis points of benefit from the sale of our Australian subsidiary and acquisition at Sunesys. A year ago we also stated our attention to deliver dividend growth in line with our expected growth in AFFO per share, and I'm pleased to announce that our board has approved an 8% increase in our stock dividend from $3.28 to $3.54 on an annualized basis.

Since we increased our dividend payout a year ago, we have closed the yield differential between the RMZ REIT index by 50 basis point, outperformed the S&P 500 and RMZ year to date, and made good progress attracting the incremental long-term investor. With the latest dividend increase we are continuing to build on our track record of delivering attractive total returns to shareholders. From 2015 to 2016 our total return expectations are 12% comprised of the current 4% dividend yield and 8% growth in dividend.

Looking out beyond 2016, we remain confident in our long-term goal of generating 6% to 7% AFFO growth organically. Combined with the current dividend yield of 4%, we believe the long-term total return opportunity of 10% to 11% should be very compelling for investors given the contracted nature of our business. In fact, we believe a significant catalyst exists as we continue to deliver on expectations and differentiate Crown Castle from its peers in terms of capital allocation and strategic profile.

Compared to the best-in-class REITs with dividend yields in the high 2%s to low 3% range, our dividend yield north of 4% suggests we have an opportunity to help investors appreciate the unique combination of high-quality yield and growth that we represent as an investment. Further, out of 425 dividend paying companies in the S&P 500, Crown Castle has the 383rd highest dividend yield, putting us in the bottom 10 percentile of dividend payers.

While it's up to the market and investors to ultimately determine valuation, I believe there are several compelling points to the Crown Capital story that make the case for a lower dividend yield and with it a higher valuation. First is the quality of our business model and long-term positive industry tailwinds. As has been true since the early days of the wireless industry, network quality continues to be the market differentiator for carrier success, driving continued long-term investments by the carriers.

As such, we believe we're in a multi-year densification cycle that continuously requires all four major US wireless carriers to make significant investments to improve and enhance their networks in order to meet consumer demand for mobile broadband in order to remain competitive. It's also clear that as network capacity and speed increases, new applications and services, such as mobile video, the connected car, and, more broadly, the Internet of Things, are developed that present promising new avenues for carriers to monetize their network investments. Our results to date and our expectations for the balance of 2015 and 2016 further reinforce our view and demonstrate how well-positioned we believe we are to capitalize on the opportunities in front of us.

For some context and highlighting the opportunity the carriers have in front of them, the growth in mobile data traffic in the US from 2016 to 2017 is expected to equal all the mobile data traffic that was carried on cellular networks in 2014. Mobile data is expected to double every two years through 2019, the furthest date that Cisco forecast. In fact, earlier this week on Verizon's earnings call they mentioned they've seen a 75% growth year over year in mobile data usage across their subscribers. As a shared wireless infrastructure provider, we benefit from growing demand for mobile broadband as carriers access our infrastructure to provide service for their subscribers.

Underlying the shared infrastructure model is our ability to generate returns by leveraging our fixed investment and operating expenses over multiple tenants, with each tenant benefiting from the capital efficient model we present. In turn, our value proposition to the carriers is that we provide them with the quickest and most cost-effective access to infrastructure and real estate as they deploy and upgrade their networks.

As a case in point, we were encouraged by the road map that FirstNet has established to seek industry partners in a public-private partnership to facilitate a nationwide public safety network deployment, potentially in 2017. Although still very early we believe our shared infrastructure portfolio will play a meaningful role in the efficient deployment of this network.

This brings me to my second point. We believe that the portfolio we have built and invested in over the last 20 years uniquely positions us to play a critical role in helping carriers with their network needs. In particular, over the last several years we have focused our investments in the US with the acquisitions of T-Mobile and AT&T towers, and continued investments in the small-cell business based on the thesis that there's a long runway of investments that need to be made in the US by wireless carriers and a consumer base that will support it.

Today our portfolio is approximately 40,000 towers and 15,000 small-cell nodes, supported by 16,000 miles of fiber. Makes us the only company that can deliver at scale and provide comprehensive wireless solutions across the US for carriers. Our towers are well located, with three-quarters of our portfolio in the top 100 markets, and have a long runway for additional organic growth with only two tenants per tower on average, suggesting a significant amount of leasing opportunity is yet to come.

On the small-cell front, we are capitalizing on our leadership position to secure some of the best assets where we believe there will be a long-term franchise value that will drive future co-locations, upgrades and co-location opportunities similar to the early days of the tower industry 20 years ago. With small cells we believe we're skating to where the puck is headed, as the pace for network densification continues into the future, and in so doing extend our runway of growth and further reinforcing our leadership position in US wireless infrastructure.

Together, the strength of our business model and the quality of our assets translates into a growing high-quality, long-term recurring cash flow stream. Today our average tenant lease term is about seven years and is backed by committed contractual lease payments of $20 billion, which provides stable and predictable contracted cash flows supporting our dividend payout.

As we look to grow the dividend in the future we benefit from contracted tenant lease escalators, which provides half of our stated goal of organic 6% to 7% growth in AFFO per share. As it relates to the other half, given the need for continued investments by our carriers, customers, and the quality of our portfolio we believe that we can achieve our stated goal of growth to leasing on our portfolio of towers and small cells. Said another way, including the dividend, approximately three-quarters of our expected total return goal is already contracted in any given year.

The quality of our cash flows is further highlighted by the focus in the US where we benefit from what we believe is an attractive operating and investing environment that has a degree of visibility and long runway of growth without foreign currency exposure that can impact returns. Our goal as a management team is to provide our investors with the most compelling, risk-adjusted, long-term total shareholder returns, as measured by dividend growth over time. To close, we're very pleased to be in a position to raise our outlook and dividend by 8% as we approach 2016, which reflects the quality of our assets, the expertise of our team, and positive fundamentals of the US wireless industry from which we benefit.

And with that, operator, I'd be happy to turn the call over to questions.

(Operator Instructions)

David Barden, Bank of America.

Hey, guys, good morning. Thanks. Nice results out this morning. I would say that the big conversation we were having last night after the print was really trying to think about next year's growth. In the first quarter to the second quarter, we saw about $6 million of sequential site leasing and management revenue. In the third quarter, backing out Sunesys, it was probably $12 million.

In the fourth quarter, backing out Sunesys again, the implication is you're looking for $8 million. And then if I annualize the fourth quarter and look at 2016 the implication is that you are only going to be getting $4 million of increment of revenue to get to the midpoint of the future guidance. I don't think people think that the industry is slowing down, but if you could help us figure out how we get from still robust growth to what appears to be a rapidly decelerating sequential revenue growth guide, it would be super helpful.

And then the second question would be, and maybe it's related, relative to 2015 it looks like the straightline revenue is coming down a lot. If you could address what contracts or what is the driver of that function, that would be great. Thanks so much.

Sure, Dave. On the first question, any time I think you go quarterly sequential changes in revenue, particularly in the periods where we've had the disposition of Australia, and then as you think about throwing in Sunesys, that may be a little distracting to look at it quarter to quarter on that. I would point you to full year.

And our underlying assumption in terms of tenant activity, as I mentioned in my comments, is right around 1/10 of a tenant that we would expect to add in 2016. We're expecting basically that same level of leasing activity in 2015.

And one of the ways that you could talk about it in terms of tenant adds per tower is our underlying assumption is that on the tower side we will have about $115 million of tower leasing growth from 2015 to 2016, and virtually that same number from 2014 to 2015. The assumption we're giving you here for the outlook in 2016 driving the top line number, our view is that we're basically holding that assumption the same in 2015 and 2016, and would expect the year to play out that way based on the total activity we can see to date.

In the second question, the straightline revenue number that you see decreasing, in essence what you see is the non-cash portion of that straightline revenue decreasing. And the reason for that is the annual cash escalator that's embedded in our leases. We would expect that number to continue to decrease.

You can roughly approximate that number by taking our adjusted organic site rental revenue and then looking at the escalator of roughly about 3%. Absent any changes -- and, obviously there's always changes because we're assuming additional revenue adds next year -- but absent those changes that number is going to go down by the equivalent of about 3% on an annual basis.

So, straightline is decreasing in essence because of the cash escalator that we have embedded in those leases. We would expect that number to continue to decrease to about 2017, 2018, at which point we would expect cash revenue received from tenants to actually exceed our GAAP revenue. So, I think the trend line that you're seeing in that is something that we would expect to continue for the foreseeable future.

And, Jay, if I could just follow-up real quick, obviously we started out this year pretty soft, and we've been seeing, as you waterfall the number for us, it's been about 5.3% and then 5.9%, and now in this quarter 6.4% organic site leasing revenue growth. And yet you're still next year guiding to the same absolute dollar number of revenue irrespective of the fact that the revenue growth has been accelerating this year. Are you trying to point to some specific deceleration in one of the growth drivers that's been leading to the accelerating revenue year to date or is this just trying to be conservative?

I think, Dave, as we went into calendar year 2015 our working assumption was that tower revenue growth was going to be somewhere in the neighborhood of about $90 million to $100 million of new leasing activity, be that from brand-new leases as well as amendments. And as we've gone through the balance of 2015 we've actually increased that assumption by about 10% to 15% roughly to get to the $115 million of expectation. And then we're going to hold that into our 2016 guide based on the activity that we would expect to continue from the carriers throughout 2016.

Thanks, Jay.

Simon Flannery, Morgan Stanley.

Thank you very much. Good morning. Just continuing on that theme about the 2016 guidance, you're talking about the activity you've seen. Clearly this year we had two carriers very active and two carriers pretty quiet. Are you getting any sense that we can go to a more balanced environment in 2016 or is it just too early for you to actually bake that your guidance? So, could there be some upside as, let's say, Sprint gets some financing or AT&T resumes their normal level of capital spending?

And then I do see you're suggesting services gross margin down 40, with some of that related to the decommissioning coming down. How much of the 40 is the decommissioning versus just another lower level of activity being assumed going forward? Thanks.

Simon, it's Ben. I'll take you back a few years and for many of us on this call I think you'll remember this. We used to really refrain from speaking with, really, any specificity around carrier deployment plans. And over the last couple of years we relaxed that standard and probably spoken a little bit too much occasionally around specific carrier deployment plans.

And, as a result, I've been asked a few of our friends and customers if we would refrain from talking about their specific plans and let them do their own talking about their deployment plans. And I think that's a perfectly reasonable request and, frankly, a pretty good reminder of the way we used to do things. So, unfortunately, the way I'm going to answer your question you may not be completely satisfied with.

But when we give you the $115 million of tower rental growth, again consistent with 2015, that takes into account everything that we expect to see out of the combination of all four carriers, as we know it, as we sit here in late October, of the prior year. And as you certainly know, because you've modeled the Company very thoroughly, if we saw a tick up inside the year next year from a particular carrier or two, that would have a relatively small financial impact inside that year, obviously being more impactful into 2017.

But all within the range of this range that Jay gave you in his prepared remarks of up or down 20% is what we've seen back for the last seven years. And that up or down 20% goes back to when all four carriers were building out LTE to a time when maybe they were a little bit slower. And that up or down 20% would be about 1% change in AFFO per share growth.

So, I think we given you all the detail you need, particularly in the supplement will show you quarter to quarter changes in revenue from the carriers so you can certainly see it on a trailing basis who's doing what. But I'm really going to try to not really speak about specific terms of who's doing what in the marketplace because certainly a lot of that is proprietary information.

Okay. But your basic message is you're currently modeling about a similar level of activity.

That is the absolute message, is that we're modeling off of the activity we've seen this year which, as Jay suggested, has ramped through the year. And we are seeing application volume ramp as we come into the fourth quarter. And that is all taken into account in our expectations. And, again, we try to stay with whole numbers because you lose a few percentage points, few basis points on the larger base every year when you're just looking at percentages.

Then on your service gross margin question, down $40 million, that's one-timers that occurred in 2015 around the pay and walk fees and a few other things, that we don't expect to repeat in 2016. Importantly, our core service contribution in the $240 million to $250 million range is the core contribution we're seeing this year. So, we're looking at it to be comparable to 2015.

Great. Thanks so much.

Philip Cusick, JPMorgan.

Hey, guys, thanks. Two things. One, to follow up on David's question and not any particular carrier, it seems like it's been about five years since you saw in some of the MLAs that we've talked extensively about in the past. Do any of those expire in 2016? And how should we think about the GAAP and cash impact as those roll off?

And then, second, then, can you follow-up on your comments about FirstNet and just expand on the opportunity here and how you see the architecture of shared versus proprietary network that they are planning?

On your first question the structures that we had signed up with the carriers where they were committing to a certain level of leasing and going ahead and paying us for that in advance of actually putting equipment on the towers, those have virtually completely been used up at this point. I think we talked about going into 2015 we were expecting somewhere in the neighborhood of about 10% of the activity in total on amendments that would have already been provided for and were already being paid for in the structure of those agreements.

Today, virtually every lease we have where the carriers are putting additional equipment on the tower is adding additional site rental revenue to the mix. So, we're really past that stage at this point. As we think about the 2016 activity we're not really factoring in any amendments that would not bring additional site rental revenue.

Got it.

Phil, on your comment around FirstNet, look, I'm encouraged. The webinar earlier, the webcast earlier a few days ago, indicated for me a pretty clear road map of time lines and dates to launch a deployment here in the way that we've always felt like it's the most efficient. And from their point of view I think they've suggested today we'll only consume about 5% of the spectral capacity of their spectrum on any given day. So, that's a tremendous opportunity for the industry to use that remaining high-quality spectrum.

As a result, there's going to be a fairly complicated and yet unknown RFP process to go through exactly what will a carrier or collection of carriers -- by market, maybe splitting it out -- pay for and what's it worth to them. And that part is not clear and still very complicated, and I think presents a fair amount of uncertainty still. But I do believe they're on the right track of looking at a shared network that will likely be hosted across one or multiple carriers' networks, which we certainly can accommodate and expect to participate in and use our infrastructure to do so in a very efficient way.

I think in previous call we basically said we don't have any clear direction. I think we now have some clear direction about where they're headed and that is encouraging, albeit it is probably not financially material into, really, 2018 because you're probably starting on deployments mid-2017, wouldn't have any real financial consequence probably until you get into 2018. But we are encouraged.

Thanks, guys.

Ric Prentiss, Raymond James.

Thanks, guys. One quick follow-up to Phil's. Jay, not just the MLAs but you guys mentioned how you bought T-Mobile towers and AT&T towers. I think there was some reserve space or future space in those deals, as well. So, MLAs and M&As, you guys are pretty much through any free space, if you will?

That is correct, Ric, my answer would apply to both situations.

Okay. Great. And then I want to dig a little deeper on Sunesys and small cell, if I could. So, I apologize for going a little deeper dive. On Sunesys, I think you are guiding to about $75 million in gross margins and $60 million or so in EBITDA after G&A. When the deal was announced you thought the first year of Sunesys gross margins might be more $80 million to $85 million and EBITDA might be more $60 million to $65 million. So, what has happened in Sunesys given that we closed the deal early and a little lower number for calendar 2016?

Ric, it just has to do with the way they classify direct operating expenses in G&A as compared to the way we would do it. When we got in and started to figure out how we would reflect the results of the business, we did it consistent with the way we would look at towers and reflected some of those expenses in different categories. So, the net number is unchanged. We just moved some of the expenses from direct operating expenses and G&A in order to match the way we think about it on the tower side to be consistent.

And you mentioned where you hoped there's some meaningful ability to get small cells on top of it. It looks like the guidance is still business as usual at Sunesys as far as their legacy business. Have you assumed any small cell extra growth onto that asset?

Ric, the way we're thinking about the Sunesys asset today is really inventory. I'd be thrilled to come on the call in a couple of quarters and tell you that the activity has ramped beyond what we are seeing today. But the way we think about is we just bought inventory, and it's a cost advantage and a speed advantage for deployment of the pipeline that we currently see and what we expect.

As we've said, our expectation for small-cell revenue growth is about $55 million next year, consistent with what we've seen this year. By implication you can say the pipeline looks steady. If it were to grow from there then terrific, and that's what we are pursuing. And we have a lot more inventory with which we can deploy small cells, again on a very cost-effective basis, because we essentially already own the asset, and on an expedited basis because we don't have to build it. But right now we're not guiding to an incremental increase in revenue contribution from there today.

Your last question is around Sunesys specifically. I would say having owned it a very short period of time there is a very modest amount of revenue growth within the Sunesys traditional business. But that's not core and not something that we would spend a lot of time focusing on with you.

Makes sense. And last question is, when you look at the CapEx that you spent within the quarter, I think it was about $98 million for newbuilds, primarily all of it small cells, should we think about, still, that old rule of thumb of about $100,000 per node if it's a new node? We haven't seen a lot of change in the node count, still 15,000 approximately last quarter and this quarter. We're just trying to get our model correct as far as you spend CapEx on a small cell when does it roll into being nodes on the air and then revenue coming in?

You're right about the $100,000 per node. That's the best way to model the business and consistent with our expectations. As we reflected the nodes, we've shown that number as a combination of nodes under construction and those on air and rounded it. So, you haven't seen it move but that would just be inside of the rounding of 15,000.

Okay. And you know me, I'd love to get some more segment details on small cell as far as revenues, margin and actually some quantities. As you consider now it's 11% of the business pro forma, hopefully someday we'll get some more details.

Yes, Ric, we're continuing to look at that, and provided obviously there's some pretty meaningful detail on the slides that I talked about and then also in the supplement. As we're going through the process of integrating Sunesys we're relooking again at our operating reporting relationships and the best way to operate the business and then the best way to reflect and review the results ourselves. That may result in doing disclosures a little bit differently than what we've done historically. But we are trying as much as we can to provide you the information you need to evaluate how well the business is operating.

We'll take just some decimal points, too, probably. Anyway, thanks, guys.

Michael Rollins, Citi Investment Research.

Hi, thanks for taking the questions. Two, if I could. First, could you go back to the slide where you walk through different revenue drivers for 2016? And can you talk about how much of the internal growth in 2016 is generated by Sunesys versus the amount that you have in the external growth bucket?

And then the second question I had, just taking a step back on leverage, can you talk a little bit about why you think it's so important to drive towards investment grade credit metrics and try to get an investment grade credit rating? Have you modeled the alternative if you didn't do that, maybe what your leverage or AFFO growth per share might look like in the alternative? Thanks.

On your first question, Mike, when we talk about the $115 million in tower leasing and the $55 million in small cell leasing, you should think about that as basically our legacy business prior to the acquisition of Sunesys. The benefit that we will have in Sunesys is over in the acquisition category of roughly $60 million. I assume you are referring to slide 8 in your question there. So, the way to think about that I think is pretty consistent with what we've done in 2015, similar assumptions in 2016.

On your second question around investment grade, we have spent a lot of time working on that and run the exact analysis that you articulated between remaining non investment grade versus investment grade. And I would say for a long period of time Ben and I have held a view that the value creation is driven in our business over the long term by three different areas. One of those is obviously organically leasing on the assets that we already own; secondly by the way that we allocate capital -- and today we're obviously allocating a meaningful portion of capital to what we believe to be growth enhancing around small cell -- and then, thirdly, by reducing and taking actions that reduce the overall expected cost of equity in our business.

There are a lot of aspects of that that have implications in terms of how we operate the business. One of those, which we've obviously spent the last several years talking a lot about, is our focus on the US market and our unwillingness to go to emerging markets.

And as we think about the business that we're running, it is a growing annuity of cash flow payments out of the various wireless carriers. And underlying that is obviously the credit quality and stability and predictability of those cash flows coming from those wireless operators. Our view is that over time, as the story is better understood, the understanding of that risk should necessarily reduce the overall cost of the equity.

Commensurate with that is having a balance sheet that, regardless of market conditions, has free access to the capital markets, and we believe the investment grade credit rating, as much as the rating is, is the indication of how we manage the balance sheet so that in maybe less favorable times than we are currently in the capital markets, we would maintain the access and there would never be a question about whether we would be able to pay the dividend if we were in relatively choppy capital markets. So, it's a combination of factors but we think it has implication longer term as we think about lowering the cost of the equity and driving long-term returns to shareholders.

I would just add one thing, Mike, if you will. If you run out the model long term, which we have, staying levered like we used to, it makes a very insignificant difference to the overall return, and obviously increases the risk profile materially, as Jay just walked through, in times when markets are disrupted.

The other thing I would say, though, just more broadly about our business, as we've all worked on this for a long time we have a fabulous business model around shared infrastructure, where it's generally cheaper for a carrier to occupy one of our facilities and benefit from the capital efficiency of sharing than it is to do it themselves. We can get double benefit out of that to the extent our absolute cost of capital can be less than our carrier customers, where you are less reliant on co-location overtime, to the extent you can demonstrate over time that your absolute cost to capital is less than the carrier customers because of the nature of our cash flow being long-term contracted with escalators out of what we believe has always been the most senior secured cash flow stream coming out of the wireless, industry which is that to operate their network.

That's a long-term aspirational goal. I wouldn't suggest to you we're there yet but it's something that we've had -- we always take a long-term view and I think the next generation of this business will be that we will have an absolute cost of capital, including total return expectations on our equity, that's lower than our customers'. And that's a wonderful place if we can ever get there, and that is what we are pursuing.

Thanks very much.

Brett Feldman, Goldman Sachs.

Thanks. Just two quick follow-ups. With regards to Sunesys, I know you're not really trying to grow the existing business, but could you just remind us about the durability of those existing customer relations, meaning can we just continue to run rate that $100 million-plus of revenue? And then you noted that there's going to be about a $35 million increase in spending. I was hoping we could get a little more detail on what's behind that.

Sure. We think the revenues are very durable. The growth rate there and escalations roughly is about 3% on an annual basis, so we would expect to see some uplift there. And then contracted terms was about five years. So, a little less than what we see on the tower side but still, nonetheless, pretty long term in nature.

And on the increase in spending you're expecting next year?

Do you mean expense growth?

Yes, the expense growth. Yes.

That's a little less than 3%. If you look at total direct operating expenses and G&A it's about 2.8% increase in expenses year over year excluding, obviously, what we're bringing on from Sunesys.

Is some of that related to small cell business or is it just the normal inflation and cost?

Normal inflation and cost. We would see that across the business.

And I'll ask just one quick follow-up along that. How do you feel about the spending growth around the small-cell business? You had noted previously that the spending was growing more rapidly than, say, the rest of your overhead expenses you were growing the business. At what point are we going to start to see that level off and maybe get a bit more operating leverage from the growth in your DASH revenues?

Brett, you are seeing that today, which is a point I made in my prepared remarks, about the 80% incremental margins. For the last several years, since we were really building out the overall infrastructure for small cell, as well as integrating the large tower acquisitions that we did, we saw spending on increasing span of control and management levels, in addition to the more operating level jobs that we put in place in order to operate those assets. Our incremental margins came down from what we've seen historically.

As we think about 2016, though, we think about it in a much more normalized basis as, again, from an operating cost standpoint that coming down much more in line with what we've seen historically in the business, just under 3%. And, therefore, the incremental margins that we see on the incremental revenue being put into the business is coming out at about 80%.

Okay. Thank you for taking the question.

Colby Synesael, Cowen and Company.

Great. Two lines of questioning. You mentioned, I think in response to Simon, that you're seeing some increase in application volume. I was wondering if we could parse through that. Is that fairly even across all four carriers or are you seeing one or two carriers potentially caring the torch there? But you don't have to tell us which ones.

Also, I was trying to get a sense amendments versus new sites. And then, also, has that really been a steady ramp through 2015 or are you seeing some form of slight infection?

And then just another line of questioning had to do with the acquired churn. We saw in the supplemental that it's assumed to be up in 2016 and 2017 versus what you previously had last quarter. Is that simply some of the pushout that you described or is there an actual absolute change in the acquired churn that you're now expecting? Thanks.

Hi, Colby, it's Ben. The churn is no change in the total. It's just sliding from 2015 to 2016, unfortunately, and a little bit into 2017. Obviously we would rather get it done sooner than later but it moves a little bit. And it was our best guess when we originally gave you that number.

And then in terms of activity I'll just stick with what I said before. We've got application volume up in the second half of this year. I'm pleased to see what I'm looking at for fourth quarter. But, again, I want to tell you that that is in our expectations for 2016. So, that $115 million we're giving you is our best view today of what we're looking at.

Again, everyone on the call, even if we started to beat that number, again 20% up would be technically about 0.5% in the first year and a full 1%, or little more than a full 1% in the second year. So, I really think it's a less-than-productive discussion to go down the path of who's doing what to whom, when we tell you that we've got it all in there and we will tell you on a trailing basis exactly by quarter who's doing what in terms of total revenue, so you can see it. Beyond that, I'll really direct it back to our four carrier customers to talk about their specific deployment activity.

Just one thought to that -- any color, though, on whether it's more amendments versus new sites?

It is pretty similar what we expect in 2016 to what we've seen in 2015. I think last quarter we talked about this a little bit, and we talked about the fact that it's returned to a more normalized level where we are in the neighborhood of about 40% or so of amendments and about 60% from new leases. I don't see anything really changing. It obviously moves around a little bit quarter to quarter depending on what work we are working on for which carrier, but somewhere in that neighborhood would be my working assumption for 2016.

Great. Thank you and congrats on the quarter.

Jonathan Atkin, RBC Capital Markets.

Yes, I was interested in the small cell leasing growth that you're expecting next year. Is most of that going to continue to be anchor tenant growth or is there any signs of interest and hopes that you could get additional tenancies in the next couple of quarters on that infrastructure? And then more broadly, talking about fiber assets, I just wondered if there's an opportunity to monetize unused fiber that you wouldn't need in the normal course of your small cell expansions.

John, on your first question, of the activity we're seeing today, about 75% is related to newbuilds. And I'm talking specifically to small cells. About 75% is where we're building new systems and about 25% is leaseup on existing systems. And that would be similar to our assumption for 2016, that that rate will continue.

The other thing I would point to broadly is if we look at the level of leaseup activity and compare that to what we're seeing in towers, we're seeing leaseup on small cells occur at a rate faster than that of towers. Today across all the small cell networks that we've built, we're about a tenant and a half, roughly, across those existing systems. And obviously we've not owned them or built them that long ago. Making the comparison that I made earlier in the conversation in my prepared remarks about adding about 0.1 tenant per tower per year, or one tenant over a 10-year basis, we are ahead of that as you think about the tenancy pads.

The other way to think about it is how are we doing on a yield basis, if you think about the annual recurring cash flow coming off of small cells compared to the total net capital that we put into them. We're building systems, as you've heard us talk about, initial yields of about 6% to 7%. And today the yield across the entire book of business, considering all of the capital that we've spent, including the amount of capital we just spent on the Sunesys acquisition, came in at about a 6% initial yield. The whole book of business is now at about an 8.5% yield. So, we have meaningfully increased the yield on those assets and that's driven almost entirely by the additional leaseup that we've seen on those systems that we've built.

Obviously we're encouraged about how well the co-location has gone. Some of you may have seen the system that we built in Philadelphia recently. That system went on air -- it was a system that we've been working on for about a year or year and a half or so -- it went on air in a densely populated area that got some recent press coverage. It went on air with three tenants on it day one. Obviously that's not our underwriting assumption and it doesn't happen every time, but we are starting to see more and more the carriers want to co-locate and lease up these systems that are being built in phenomenal areas around the country.

And then, John, on your comment about monetizing unused fiber, that's absolutely something we're looking at. Couldn't put that in guidance yet. There's two ways to do that. One would be to contract with third parties, other providers that might want to use it, as we did with Sunesys before we owned them. Or, alternatively, the Sunesys sales force gets turned loose on our fiber that we owned where we can add monetization through additional contracts on that fiber.

We're looking at both of those things. We'll probably pursue them both. That's just prudent to get the return out of the investment. I wouldn't say -- that's certainly not meaningful in the numbers as we look at it having owned it a couple of months, but it's an opportunity for us nonetheless. Not the core business and not why we did the transaction, but it's revenue and margin in the same that I think we can go out and get.

Great. Thanks very much.

Amir Rozwadowski, Barclays.

Thank you very much and good morning, folks. I wanted to build upon the prior question in terms of co-location opportunities when it comes to small cells. There has been some debate among some of the carriers in terms of the strategy at which they're going to deploy small cells, at least some of the commentary that we've been hearing in terms of their strategy. As you sit today it does sound like the opportunity set for a certain number of carriers seems to be improving.

How should we think about the longer-term opportunity? Do you think that some of these technologies are viable over the longer term? Is there going to be a hybrid approach towards some of those small cell deployments whereby you'll still be able to capture your fair share of co-location opportunities? We'd love to hear your thoughts on that.

Amir, this is Ben. I think we've already seen an evolution in some of the technologies with some carriers wanting a lit service, what we think of as a small-cell application, versus a full turnkey dark service, which is a distributed antenna system, which is the majority of what we have today and what we are still deploying today. So, I think that will happen over time depending on what particular issues the carriers are trying to address in that location.

From where we look at it today, though, we're getting long-term contracts were we're starting with a very healthy initial yield, as Jay just walked you through, and co-location opportunities that are today tracking north ahead of where a traditional tower acquisition would be. And we're doing it in areas where -- you've seen some of the maps -- 90% of the activities in the top 25 markets in the US that we think are frankly beach front real estate for purposes of these systems, just like we did when we acquired tower portfolios 15 years ago.

We think it's a terrific long-term objective. But it's location specific. There are places where carriers have needs where we've been a little sheepish about going, honestly, because they have been second- and third-tier markets were one particular carrier might have had a need but it's harder to underwrite when that second or third carrier would have a similar need. And that's just like the tower business, by the way.

So, we're being pretty careful about where we deploy the capital. And so far, to move the needle as we've moved it, as Jay just walked through, on almost $3 billion today capital spend in that business, tracking at about an 8.5% yield today, suggests we're on the right path. It is still very early days. We are going to have the capability -- we already have the capability and I think we'll continue to enhance it -- to accommodate whatever level of architecture or technology the carrier wants to deploy.

And it's very clear that backhaul connection fiber on net is the preferred approach to a durable system that provides scalability in terms of backhaul. And that's what we provide. So, we're thrilled to death and going about as hard as we can go right now, as we've said before.

Thank you very much. And, if I may, one follow-up question. I know we've talked about technology disruption in the industry for some time now. And there's always been percolating conversations around Wi-Fi dislocation and stuff along those lines. Recent news suggests that there is someone of significant heft that may be looking to deploy a service over the near to mid term. How do you think about that in terms of the reality of investment and potential disruption for your market if that was to come to fruition?

I rely on our history, honestly, and I rely on our carrier customers' view this topic, which is we already have more than half of wireless traffic or data on Wi-Fi networks today, and it's a very offload strategy that we all benefit from as consumers. And I think the wireless carriers with license spectrum would have a very hard time supporting all of that if there wasn't an efficient offload strategy.

And I think that's going to continue. Verizon talked about it on their call, though. They don't think that's necessarily the path for them going down with license spectrum to be able to control the network and the network quality. But I think we as consumers have a view that when you jump off of your contracted carrier network and you get on Wi-Fi wherever you happen to be, I don't think you have the same expectation of service. Maybe that's a benefit ultimately to the carriers over time.

In order to accommodate the growth in data services that we're all seeing -- and we see it in the press every single day with new applications and devices -- our view is it's going to take all comers. It's going to take towers, first and foremost, that we, again, have a belief that we will have every tenant on every tower over the long term. We think there's absolutely no reason, unless there's obviously a competing site, which is very rare in the tower business, that you wouldn't have all four carriers on every tower over time, almost by definition. Obviously we don't give you that in our 2016 guidance but that's our long-term thesis.

Behind that comes, then, the small-cell architecture where, as we've talked about now for a couple of years, the benefit of the small-cell capacity offload, is what it is essentially doing, it's dealing with a hotspot where there's a capacity issue in an urban environment. It's also unloading a macro site and making it more efficient. We think it takes that in scale. And we think there will be hundreds of thousands of nodes over time.

And then on top of that, just to finish out your question, I absolutely agree there will be Wi-Fi capability, as there is today. Perhaps we all as consumers will accept the lower threshold of service in a stationary environment, not a mobile environment. But we think that it takes all of the above to accommodate the broadband Internet usage that we're all seeing in the connected devices and, obviously, mobile video.

Thank you very much for taking the questions.

Tim Horan, Oppenheimer.

Thanks, guys. Just a couple of follow-ups. FirstNet, do you think it's going to be material? Can it be 100, 200 basis points increased revenue growth over time? And, secondly, on ATT, it sounds like the reserve capacity got used up a little bit faster than you were expecting, I think definitely faster than what the company was expecting. And I just had one follow-up. Thanks.

Tim, on the FirstNet revenue contribution, I don't how to model that yet. It's just too early. I don't know who is going to be the winner of the RFP and exactly with the economics are. I think for purposes of your thinking, certainly to the extent it's a discrete application on a tower, that would look like a co-location, and I think they've said upwards of 40,000 sites that they need, we'd support that and look at what they are trying to accomplish. It sounds reasonable to potentially a little bit low.

To the extent it's a shared application where they're going to co-locate, it looks more like an amendment, which we're certainly able to price, and it would be an efficient use of the structural capacity on the tower and very quick to deploy. But I don't want to sit here and tell you what that mix is going to be yet. There is no way to know at this stage.

And with respect to where we are on ATT and that particular portfolio, I think Jay's comments I'll just probably leave alone. That's probably about as far as we ought to go.

Great. And then on the small-cell side, are you see any new competitors coming into the market, or maybe just as you get more and more experience your thoughts on the barriers to entry. Thanks.

As I've said before, I think there will be more competitors over time. It's an enormous market when you start thinking of geographies at the city block level. There's a lot to get done and certainly we can't do it all.

We see other competitors in the market and others that are gaining capabilities. And I think that will happen over time. We're getting long-term contracted terms to amortize the investment.

And in terms of the barriers to entry, this is purpose-built fiber. And particularly with what we're doing today it is a very high consumer of strands purpose-built back to a hub location. It's not easily replicated. So, we're very confident that once you get an installed base and it's serving the need of that particular carrier, again on an efficient basis, we think it's a very sticky business. Not too worried about the long-term barriers to entry here.

Thank you.

Michael Bowen, Pacific Crest.

Thank you very much. I think most of my questions have been answered but maybe I'll hit this. How do you think about, you mentioned Cisco continues to talk about mobile data doubling each year through 2019, as you think about how the carriers are going to try to handle that, how do you guys go about modeling that into your assumptions going forward? I realize you're only giving out 2016 at this point but your thoughts on that would be helpful. Thanks.

Michael, we do this one year at a time. It just gave you on the previous question my long-term view on why we bought all these towers and why we think those are still the most attractive and efficient way for carriers to add capacity in a market. And then, as we've talked about, pleased with what we're seeing on the small cell side as a complement to that.

I think it's very clear we all have underestimated, probably forever, what the demand would be on the wireless networks and then what we would need to provide as infrastructure. What I look carefully at -- and I think this is going to be interesting to watch and Verizon talked about it a little bit on their call -- is opportunities to start monetizing the network in a different way than what we all traditionally would think of as subscriber monthly payments. If you want to know what we're watching, that is interesting to us as they generate new revenue streams as a way to further monetize the network. And I realize today that's financially immaterial, but I think that will ultimately drive the incremental investment in the network that obviously we benefit from.

And then maybe last one, I think a few minutes ago you said in 2016 you're assuming a mix of about 40% amendments and 60% leases for 2016. If I'm correct on that, can you remind us what 2015 is? And I believe you said lease apps are up, so should I assume, if 2016 increased at that 60% for leases over 2015, are you at a 70% run rate as a catch-up right now? How can we think about that right now?

Michael, the expectation for 2015 is honestly pretty similar to that of 2016. It may move a few percentage points between new leases and amendments but I didn't mean to imply that we were seeing a big move in the type of applications or the type of activity. Our assumption for 2016 is pretty similar, both in terms of gross dollars as well as the mix.

Okay. Great. Sorry, I ran out of questions, I figured I'd give it a shot.

Jonathan Schildkraut, Evercore ISI.

Good morning. Thanks for taking the questions. Two, if I may. First, on the leverage side I'm seeing that leverage ticked up a little bit quarter over quarter. And maybe that's because we don't have a full quarter of Sunesys in here to reflect on the EBITDA you paid for. But could you give us a little bit of a perspective of how we get down to 4 to 5 times leverage over the next whatever period of time it's going to take?

And then the second question, I just want to circle back on how we could think about revenue-generating CapEx as we go forward. My bad, I didn't do a good job of modeling that this quarter. But would it be fair to maybe think about the $55 million of incremental small-cell leasing you're going to get next year -- I think, Jay, you said 75% of that is going to be on new systems, so take three-quarters of that and then gross it up based on your initial yields of 6% to 7%?

Thanks for the questions. On your first question around Sunesys, you're right about how to do the leverage calculation. It ticked up a little bit just because we didn't have a full quarter of Sunesys in that calculation. We would expect to get down to the 4 to 5 times over time through growth in EBITDA. We're not making an assumption that we're going to allocate capital towards paying down debt but rather just growth in the business as we've laid out our growth forecast. We think the main growth in the business is going to drive that leverage ratio over time.

On the second question, yes, that is the right way to think about CapEx, as you articulated. We think it's going to be about the same as 2015. And as you think about initial yields on invested capital, that would play out real closely to our guide and flow-through of the growth in revenues flowing down to the EBITDA line.

All right. Thanks so much.

Batya Levi, UBS.

Great, thank you. A couple of follow-ups. Throughout this year you saw a delay in the renewal activity and it's pushed out to next year. I was wondering if you are having any conversation with the carriers today to potentially capture that churn for next in maybe new MLAs that you could sign with them.

And then a follow-up on the CapEx side -- can you talk about why this maintenance CapEx is coming down next year? And how should we think about maintenance CapEx for the macro sites versus small cells? Thank you.

In terms of specific conversations around renewals and MLAs I'd tell you there's nothing significant going on at the moment that merits commenting on. And the churn forecast, as we've laid out in the supplement, is our best estimate of where that's going.

On the second question around maintenance CapEx, it's primarily coming down because we saw during 2015 a meaningful portion of that maintenance CapEx being associated with additional office facilities, as we were increasing the employment base for integrating the various assets that we've acquired recently. So, that's the reason for the reduction of about $25 million.

If you think about that reduction to where we're giving guidance for 2016, I would tell you about a third of that is corporate-related, roughly, and then the remaining balance would be on revenue-generating, both towers and small cells. As you think about modeling it long term, though, I would use 2016 as the run rate assumption, and absent any meaningful acquisitions that we do. That's probably a pretty good run rate for forecasting purposes.

Is the maintenance CapEx for small cells a bit higher than macro?

No. It is actually a little bit lower. If you were thinking about it as a percent of revenue, which is probably the best way to think about it, it would actually be a little bit lower. Most of the maintenance, if you think about maintenance, would be in the direct operating expenses on the small cell side.

Okay. Thank you.

Matt Heinz, Stifel.

Thank you, good morning. We certainly appreciate the opportunity to go into your Philly small-cell network about a month ago. Definitely thought that three tenant network was illustrative of the longer-term potential of the business. But I was hoping you could just help us understand the competitive differentiation of your turnkey platform versus others in the market that may be more of just an on-net fiber-oriented provider or just a pure service provider without the fiber.

Yes, Matt. And thanks for the comment on the Philly system. We were quite pleased with that. The majority of the work we're doing today is what we would call full turnkey buildout where we're rebuilding the nodes and the fiber, as we talked about.

Some of our carrier customers, though, different by market, depending on what their individual capabilities are in a particular market, that are wanting to do some of that installation themselves and essentially just contract with us for the fiber on the same long-term contracted basis that we've talked about. And we're okay with that. We're in business to monetize the investment and get a yield off of co-location, and we'll accommodate however they want to go through it.

We believe over time it will become more apparent that we have unique expertise in what is continuing to be a challenging implementation and construction environment. It is not easy to do this stuff and we've probably got the biggest team in the country doing this today.

But, clearly, there are different markets where different carriers want to undertake some of this themselves, and that is completely fine. And we'll, to the extent we can be helpful and let them utilize our component, being the fiber, we will do that. That's how the dynamic is shaping up in the market and either way works for us.

Okay. Thanks. Then maybe just one follow-up on the small cells, I think we have a pretty good understanding now of what the upfront CapEx and returns look like, but I was hoping to get a better sense of what the ongoing capital intensity is and maybe the useful life of the carrier equipment on the nodes in comparison to the traditional tower business.

Sure. Initially when we build one of these systems of the operating expenses represent about 40% of the initial revenue. So, it's about 60% margins out of the gate there. As I mentioned in an earlier question on CapEx, there's very little ongoing maintenance CapEx associated with these systems. Most of the maintenance, if you think about it that way, would be embedded in that 40% of operating cost.

Over the longer term, the asset that we typically own in small cells would be the fiber assets. Most of that, as has been mentioned earlier, is dark fiber. And then the other assets that we would own on the various locations where the tenants are putting their equipment, we would own a cabinet, often referred to as a shroud, at the location. And the carriers would put the electronics inside of that cabinet or shroud.

We have seen already, in the early days of the business, the carriers come back and upgrade the electronics over time but reutilize the cabinets. So there's a pretty good parallel to what we see at the tower site in terms of reusing the infrastructure, the cabinets, if you will, there's a shelter, and just upgrading the electronics inside.

We don't anticipate a meaningful amount of future CapEx or useful life degradation of the assets that we own. The carriers we would expect over time to continue to put the capital and to replace their electronics and upgrade, very similar to what they do on the tower side.

Okay. That's very helpful. Thank you.

Mike McCormack Jefferies.

Hey, guys, thanks. Not to beat the table too hard on the small cell stuff, but just getting a sense from you guys, if you would, on what you are seeing out there on the ground as that gets deployed as far as speed to market? Is it pacing the way the carriers anticipated it would? And maybe any complexities or hurdles they've had to overcome in building that out.

Yes, Mike, it's never fast enough. I'd be quick to speak for our carrier customers and say that they would tell us it's never fast enough. That is the struggle we have every day, is that when they show up and engage us to build a system or co-locate on a system, it needs to be done tomorrow.

The challenge we all have is that it is all new ground. It is new ground for us in certain markets, it is new ground for the carriers. They haven't done this themselves before, either. And we're working through the various permitting and zoning and construction time lines that are required there.

I would say we are getting better. We've done more of it than anybody in the marketplace and we're certainly getting better in finding ways to gain days back in that cycle. But it is still, some of it is outside of our control. And to that end we're continuing to work with municipalities and other jurisdictions to continue to gain access and the right-of-way as quickly as we possibly can.

Is that just a blocking and tackling building by building type thing, then, or is it more widespread where you can do it bigger?

It depends on the jurisdiction, honestly. A lot of it is construction services. Some markets are harder than others, depending upon their affinity for wireless service, honestly, and whether or not they want to be supportive of additional wireless capacity in their market or not. And it is why a couple years ago, three years ago we got federal legislation, 6409 out of Congress, to help us mandate co-location on towers. It is incredible that that had to be a federal law but it is.

We're still dealing with some of that even at the micro level for small cells. It's hard to understand why a community wouldn't want to promote wireless broadband in their community, but we're continuing to encounter that and it's just something that's part of the business and we work through it.

Great. Thanks, guys.

Spencer Kurn, New Street Research.

Hey, guys, thanks for taking the question. Just to take a step back, I was hoping you could articulate how you get from a full per share of 8% growth next year to your long-term targets of 6% to 7%. When I run some calculation, it looks like revenue growth should accelerate by about 100 basis points into 2017 and 2018 just from the churn at PCS, Leap and Clearwire ending. So, any color you can provide on the puts and takes as to your long-term forecast would be really helpful.

Sure, Spencer. Appreciate the question. As you think about the 8% growth that we have going into 2016, as I mentioned in my earlier comments, we expect about #.5% of that is coming from the acquisition that we did in Sunesys. So, on an organic basis, if you will, or a normalized basis, we've got about 6.5% to 6.6% growth coming out of the business.

You're right to note, from the supplement that we provided, that churn comes down. What we attempted to do, when we set the long-term target of 6% to 7% AFFO per share growth rate, we tried to, as best we could, look at the landscape of all the things that could go positively for us as well as negatively into that number, and give you an all-in robust view of what would happen with the AFFO per share. And our best estimate would include that runoff of churn, which we're certainly expecting in nonrenewals that you are illuminating, but we're also, as we think about interest rates in the business, we're running that model assuming the forward interest rate curve.

We would assume that there will be some negative impact as rates rise over time, and that would, in essence, offset some of the benefit that we would expect to see over time from the nonrenewals. We're trying to as best we can give you an all-in view, a robust view, as we think about modeling the business ourselves and running our corporate model. So, there's some of that impact.

There is one other thing that you to think about as you're modeling 2017 and think about the growth rate from 2016 to 2017. The preferred instrument that we did a couple of years back, that actually mandatorily converts to equity at the end of 2016. That has a drag on growth in AFFO per share of about 100 basis points The dividend goes away, if you will, which we're including currently as a deduction to AFFO. And then the shares go into the denominator of the number of shares outstanding, if you will. That probably will help you bridge a little bit maybe some of the puts and takes you should think about to match your forecast closer to what we're thinking about long term.

Thanks.

Walter Piecyk, BTIG.

Thanks. I just wanted to go back to Amir's question. I think that was last hour. Your answer was extremely bullish on data growth. It's actually similar to what Malone has talked about as far as we're going to need cell sites and small cells Wi-Fi, everything. So question is, have you talked to the cable operators or Google about accessing your fiber for their Wi-Fi hotspots? It's a version of a small cell itself. I wonder if that dialogue has begun, because if you believe in that future, which you guys obviously do, and I think others do, as well, shouldn't those conversations be happening about now?

Walter, all I will say is we've talked to a lot of people about our fiber and its gotten attention from a lot of folks that we've got a significant amount of metro fiber holdings now exactly in the places where, as you've identified, people are going to want it. That goes back to the monetization of fiber question. And absolutely something we're looking at. I can't put anything a guidance today and I'm not going to use any proper names about exactly who we are speaking to. But that's something over time that I think there's a revenue opportunity.

So, it's reasonable to assume that those dialogues are -- I'm not asking you whether --?

It is. We'd be crazy if not.

Okay. And then just one other question on small cells -- if you think about the four wireless operators, if there's upside next year in your small cell business specifically, do you suspect that it's going to be more from an existing guy getting more aggressive or that that business model will extend to the operators that maybe are not as embracing of small cells?

I would say both.

Great. Thank you.

Listen, we really wore everybody out this morning and I appreciate those that have hung on the call. We wanted to make sure we got to everybody. Very much appreciate you joining us today.

We're thrilled to death about this first anniversary of raising the dividend, as we have, 8% here today, which is very significant. As we continue to differentiate our business model and our objectives going forward we think it's a very compelling story. We'll talk to you again next quarter. Thank you.

This concludes today's conference. Thank you for participation.