美國電塔 (AMT) 2017 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to American Tower Third Quarter 2017 Earnings Call. (Operator Instructions) As a reminder, this conference is being recorded.

I'd now like to turn the call over to our host, Mr. Igor Khislavsky. Please go ahead.

Thank you, and good morning. Thank you for joining American Tower's Third Quarter 2017 Earnings Conference Call. We've posted a presentation, which we will refer to throughout our prepared remarks, under the Investor Relations tab on our website, www.americantower.com.

Our agenda for this morning's call will be as follows: first, I'll provide some brief highlights from our financial results; second, Tom Bartlett, our Executive Vice President, CFO and Treasurer, will provide a more detailed review of our financial and operational performance for the third quarter of 2017 as well as our updated full year 2017 outlook; and lastly, Jim Taiclet, our Chairman, President and CEO, will provide some brief thoughts on recent wireless trends in technology and our role in the wireless infrastructure industry. After these comments, we'll open up the call for your questions.

Before I begin, I'd like to remind you that this call will contain forward-looking statements that involve a number of risks and uncertainties. Examples of these statements include our expectations regarding future growth, including our 2017 outlook, capital allocation and future operating performance or acquisitions, technology and industry trends and any other statements regarding matters that are not historical facts. You should be aware that certain factors may affect us in the future and could cause actual results to differ materially from those expressed in these forward-looking statements. Such factors include the risk factors set forth in this morning's earnings press release, those set forth in our Form 10-K for the year ended December 31, 2016, and in the other filings we make with the SEC.

We urge you to consider these factors and remind you that we undertake no obligation to update the information contained in this call to reflect subsequent events or circumstances.

Now please turn to Slide 4 of our presentation which highlights our financial results for the third quarter of 2017. During the quarter, our property revenue grew 10.5% to $1.66 billion. Our adjusted EBITDA grew nearly 14% to $1.04 billion, and our consolidated adjusted funds from operations increased by 16.6% to $748 million.

In addition, our consolidated AFFO per share increased by over 16% to $1.73, and net income attributable to American Tower Corporation common stockholders increased by 25.5% to $298 million or $0.69 per diluted common share.

And with that, I'd like to turn the call over to Tom.

Thanks, Igor. Good morning, everyone, and thank you for joining us this morning. As Igor just highlighted, we had another quarter of strong results, posting double-digit growth across our key financial and operational metrics. In fact, it was our 18th consecutive quarter of double-digit growth across property revenue, adjusted EBITDA and consolidated AFFO.

Our consolidated organic tenant billings growth in Q3 was over 7%, including 6.3% United States. We expanded our margins and once again grew our common stock dividend by 20%.

We've also repurchased over $100 million of common stock since the end of the second quarter, bringing our year-to-date total to around $750 million. And in the early stages of Q4, we acquired over 500 sites in the U.S. and signed an agreement to acquire additional assets in Mexico. We expect these transactions to be accretive to AFFO per share, driving additional shareholder value by enhancing our competitive position in these 2 North American markets.

With that, let's dive into the details around our third quarter performance and updated outlook for the year. If you please turn to Slide 6. Our consolidated property revenue grew by nearly 11% in the third quarter, which was driven by tenant billings growth of nearly 10% and organic tenant billings growth of over 7%. Our U.S. property segment revenue growth for the quarter was 8%, including organic tenant billings growth of 6.3%, 60 basis points or so higher than Q3 of last year.

Volume growth from colocations and amendments, which was about 18% more than the prior year quarter, contributed about 5% to the organic growth rate while pricing escalators contributed about 3%. This growth was partially offset by churn of about 1.6% on an annualized basis, just 40 basis points of which was associated with operational churn from the big 4 core networks. Additionally, our average remaining churn with these major tenants stands at nearly 6 years. We believe these positive organic growth trends reflect consistent capital spending by the U.S. wireless carriers in the midst of exponentially growing mobile data usage across the country.

International organic tenant billings growth in Q3 was about 300 basis points higher than that in the U.S. at over 9%, continuing to underscore the benefits of our diversified international portfolio. This included a strong contribution from the in-year benefit of gross new business commencements added primarily at the end of 2016 and earlier in '17 throughout our international markets and in particular, from Reliance Jio's 4G investments in India. Across our international businesses, the total contribution from gross new business in Q3 was more than 20% higher than Q3 of last year. Churn was also slightly lower than expected in India during the quarter. However, we did book some incremental revenue reserves, principally for one tenant in the market that I'll detail a bit later.

In Latin America, Mexico was once again a top performer, generating organic tenant billings growth of nearly 13% as significant 4G deployments by multiple carriers continued. And in Brazil, we're seeing higher levels of new business commencements. In fact, the highest it has been all year as the macro environment and currency continue to stabilize and the carriers reinvigorate their network investment plans. On a combined basis, we expect to generate over 20% more new business in Latin America this year than last year.

And finally, in EMEA, organic tenants billings growth was over 9%, highlighted by a strong quarter of new business in South Africa and solid growth in our other EMEA markets. Volume growth from colocations and amendments contributed more than 7% of our international organic tenant billings in the quarter. And meanwhile, contractual pricing escalators contributed about 4%, representing a sequential decline of nearly 1% and almost 3% versus last year, due primarily to lower CPI-based escalators in Brazil amidst a strengthening economy, and as a result, a stronger currency.

Finally, we had international churn of nearly 2%, primarily driven by slightly elevated churn levels in India, largely attributable to an accelerated consolidation environment.

And finally, the day 1 revenue on the over 5,600 sites we've added since Q3 of last year contributed an additional 2.1% to our consolidated tenant billings growth. Our new build program remained active, constructing over 1,300 sites year-to-date, including nearly 500 sites built during the quarter, primarily in our international markets where average day 1 NOI yields in the quarter were about 9%.

Turning to Slide 7. In addition, we posted our 18th consecutive quarter of double-digit growth in both adjusted EBITDA and consolidated AFFO with growth of nearly 14% and nearly 17%, respectively. Consolidated AFFO per share and AFFO attributable to common stockholders per share grew over 16% and over 15%, respectively. This was driven by strong organic tenant billings growth, proven investments in new assets and disciplined expense management across the business, evidenced by our adjusted EBITDA margin expansion of about 150 basis points versus Q3 of last year.

Now turning to Slide 8. Let's now take a look at our updated expectations for 2017. At the midpoint of our revised full year outlook, we expect property revenue to grow by about 15% versus last year to over $6.5 billion, up $15 million compared to our prior outlook. This reflects about $21 million in positive FX impacts, about $9 million from our recently closed acquisitions, around $7 million from the delay of anticipated churn, primarily in India; and a few million dollars composed of additional straight-line, pass-through and other non-run rate-related revenue. These positive impacts are being partially offset by about $27 million in higher revenue reserves in India, the majority of which are associated with one tenant, Reliance Communications. We currently have just over 3,400 leases with this tenant, which account for about 2% of our property revenues in India. We recorded about $15 million in reserves in Q3, primarily related to Reliance and expect to record an additional $12 million or so in Q4. This does not necessarily mean we don't expect to collect any of this revenue, but for GAAP guidance and our internal accounting policies, we have put these reserves in the books.

Finally, consistent with our past practice, our revised outlook excludes the impact of our pending transactions, including our most recent one in Mexico.

As part of our property revenue outlook for 2017, we continue to expect strong growth in both our U.S. and international businesses, including total tenant billings growth of about 13%. This is based on our expectations that during 2017, we expect to add about $600 million in incremental tenant billings. The majority, approximately 60%, is expected to be organic in nature with the balance coming from acquisitions and new builds. As a result, we continue to expect consolidated organic tenant billings growth in the 7% to 8% range.

Given our expectations of continued consistent network investment by our U.S. tenants in Q4, we are reaffirming our full year U.S. organic tenant billings outlook for growth of over 6%. This reflects the expectation that full year 2017 U.S. colocation and amendment activity will be up 15% to 20% versus 2016 levels. As was the case when we updated outlook last quarter, we have not included any leasing expectations from FirstNet deployments or any material contributions from the expected rollout of the 600 megahertz spectrum.

In our international markets, we are reiterating organic tenant billings growth expectations of about 10% at the midpoint. This is being driven by continued strong gross new business commencements, particularly in India and Latin America. We do expect churn in India to rise to the mid-single digits in Q4, driving international churn as a whole to around 3% in the quarter and just over 2% for the year. In addition, as I just mentioned, both our Q3 results in India as well as our projections for Q4 reflect a higher level of India-associated revenue reserves relating to billings again, primarily with our tenant, Reliance Communications.

We're also raising our outlook for adjusted EBITDA by $15 million or about 40 basis points, which reflects about $17 million in core business outperformance, $13 million in positive FX impacts, about $6 million from recently closed transactions, $4 million in net straight-line and a few million dollars in incremental contribution from our services business. We expect this to be partially offset by the incremental $27 million in revenue reserves we expect to record in India, versus our prior outlook, which flows directly into adjusted EBITDA.

Importantly, our adjusted EBITDA margin, which we expect to be 61.6% for the year, has shown continued steady expansion over the last 1.5 years. Over the long term, we would expect this broad trend to continue as we generate organic growth on our extensive, existing portfolio. We expect to translate this increase in adjusted EBITDA into higher consolidated AFFO, and as a result, are raising our 2017 consolidated AFFO outlook by $30 million or 1% at the midpoint to nearly $2.9 billion. This increase is primarily attributable to about $10 million in higher expected cash-adjusted EBITDA and a $20 million reduction in our expectation for maintenance CapEx versus our prior outlook. I do want to note that due to the timing of certain events, our full year outlook implies that cash taxes and interest expense will be up about $20 million and $25 million, respectively, in Q4 versus the Q3 total. Even with these items, we are reflecting a per-share growth rate of nearly 16% or $6.70 at the midpoint for the full year.

Turning to Slide 9. We've deployed over $600 million of capital in Q3, bringing our year-to-date capital deployment to over $3.6 billion. We've deployed approximately $1.2 billion for acquisitions, which includes the $465 million we spent to acquire over 500 sites in the U.S., subsequent to the end of the third quarter. We also repurchased another $36 million of stock in Q3, which along with the $70 million or so we've spent on share repurchases so far in Q4, brings our total share repurchases this year to around $750 million. We've also spent about $488 million for discretionary CapEx through Q3, primarily for the construction of new sites to enhance the capacity of existing sites and to strategically acquire land under our sites. We've also spent about $90 million on nondiscretionary capital through September 30 for maintenance and corporate CapEx, which represents just 1.8% of our consolidated revenues.

And lastly, we paid over $1.1 billion in common and preferred dividends year-to-date, growing our common stock dividend by around 20% and taking our payout ratio to around 38% this quarter. While continuing to pursue our disciplined and extensive capital allocation program, we've also continued to naturally delever and ended the quarter and the last quarter annualized net leverage ratio of 4.4x. Including the recent transaction in North American markets, we would expect our net leverage at the year-end to be in the range of 4.7 to 4.8x. The success of our capital deployment strategy continues to be evidenced by the expansion in our return on invested capital, both in the U.S. and across our international portfolio. Our consolidated ROIC now stands at 10.6% or about 70 basis points higher than this time last year.

Turning to Slide 10. We have most recently entered into a definitive agreement, subject to certain closing conditions, to acquire a portfolio of Urban Telecommunications assets in Mexico, one of our key global markets. This portfolio includes more than 50,000 concrete poles, conducive to hosting both macro and small cell equipment and approximately 2,100 route miles of fiber. Because Mexico, like many of our international markets, is at an earlier stage of wireless development than the U.S., we believe fiber-related investments are inherently more attractive given there's significantly less competition and far less existing fiber, particularly where we can simultaneously secure distinct franchise real estate rights.

Further, unlike in the U.S, where about 85% of people live outside urban areas and where we have very few towers in cities, the majority of people in Mexico live in urban areas, where we have a significant portion of our tower footprint. So by acquiring fiber transmission real estate points, coupled with presence of fiber in these types of areas, we can immediately participate in a fiber-to-the-tower solution to enhance the value of our existing infrastructure. The fiber we are in agreement to acquire will position us to economically bring fiber to a substantial portion of our portfolio in Mexico. And at the same time, we're positioning ourselves to provide infrastructure for the anticipated urban RAN densification initiatives that we believe will be coming over the next few years.

Taking all those factors into account, our view is that these assets can produce returns that truly are comparable to our towers in the Mexican market. On an annualized first year basis, we estimate this transaction would generate about $75 million of incremental revenue and will drive immediate accretion to consolidated AFFO per share.

And turning to Slide 11 and in summary. We extended our track record of delivering compelling growth across our key financial metrics in the third quarter, highlighted by our 18th consecutive quarter of double-digit growth in property revenue, adjusted EBITDA and consolidated AFFO. At the same time, we continue to delever and have grown our common stock dividend by 20%. We've also simultaneously repurchased about $750 million of our shares this year and have deployed or committed nearly $2 billion to accretive M&A in our U.S. and international markets. Our strong year-to-date results, recently closed transactions and positive foreign exchange rate trends have once again enabled us to raise our outlook for our key financial metrics. Additionally, we've continued to strategically commit capital to attractive acquisitions across our global markets that should help enable us to sustain strong growth well into the future. We believe our increasingly diversified global portfolio is well positioned to capitalize in the secular demand trends, driving exponentially growing global mobile data usage. And further, additional potential catalysts, such as FirstNet in the U.S. and the Red Compartida wholesale network build at Mexico give us incremental optimism relative to the sustainability of our growth path for years to come.

And finally, I did want to briefly mention the impact on our business relative to the natural disasters that occurred in the last quarter. These tragic events did not have a material impact on our portfolio. And we do not expect to see any material impacts to our operating performance as a result. Additionally, we're closely working with local and national authorities and nonprofit organizations to provide tower space for their communications equipment as they continue to assist with rebuilding efforts in the impacted communities.

And with that, I'll turn the call over to Jim. Jim?

Thanks, Tom. As Igor mentioned, I'll begin with a brief update on wireless trends and technology, then I'll outline how we define American Tower's role in the wireless infrastructure industry and how that role positions us to maximize long-term AFFO per share growth and return on invested capital.

Over the years, we found that increases in mobile data usage per subscriber and aggregate wireless industry spending in a given market are the fundamental drivers of our company's property revenue and AFFO growth. As to the first of these growth drivers, mobile data usage in the U.S. and around the world continues to expand rapidly. According to Cisco, the average U.S. smartphone now consumes more than 4.4 gigabits of data per month, a 33% increase from just 1 year ago. This dramatic growth in usage is in large part due to greater mobile video consumption, further bolstered by the unlimited data plans now offered by all 4 U.S. wireless carriers.

Mobile data usage in most of our international markets is growing even faster as a result of increasing smartphone penetration and the adoption of broadband smartphone apps, such as video calling. On average, in Mexico, Brazil, Nigeria, South Africa and India, for example, mobile data consumption's expected to increase at an annual rate of nearly 50% through 2021. To keep pace with this massive growth in mobile data, the large multinational carriers that comprise a substantial majority of our tenant base continue to deploy billions of dollars around the world every year into network investments.

Concurrently, there've been a number of technological advancements, such as software-defined networks, software-optimized networks, cloud RAN and others, that have helped mobile operators to reduce the cost of their core network operations over the last several years. This, in turn, has helped to support sustained investment in the radio access portion of the network that our towers support. In addition, each iteration of mobile technology, from 2G to 3G to 4G has yielded significant benefits from a spectral efficiency and other technical attributes that further drive down the cost for mobile operators to deliver increasingly greater data usage per subscriber. 5G is expected to continue this downward cost trend, while also opening up potential new revenue opportunities for the carriers as wireless services increase from megabit to gigabit speeds with lower latencies. Throughout this technology evolution, macro towers have remained as the most cost and technologically efficient solution for radio signal transmission in the vast majority of areas across our served markets. At the highest level, this has been driven by the physical lines of radio signal propagation coupled with the optimal locations at which towers have been situated.

Therefore, the macro tower real estate asset remains an absolutely critical component of delivering wireless service and will in a 4G world as well as a 5G world as well. The fundamental importance of securing real estate rights on which to locate mobile signal transmission equipment is the basis of our company's primary core competency, and that's our skills and abilities to obtain, aggregate and maximize franchise rights for telecommunications real estate. Our view has always been that these franchise real estate rights and their associated tenant contracts, are the true drivers of value in our industry.

So over our 22-year history, American Tower has secured the real estate franchise rights to nearly 150,000 macro towers and hundreds of indoor and outdoor venues that host small cell transmission systems in 16 select countries around the world. To further enhance the value of these exclusive locations, we've now embarked on a structured, deliberate and institutionalized innovation initiative which is designed to identify, evaluate and pursue opportunities to leverage our macro tower and other real estate assets for emerging technologies and applications. These opportunities might include distributed data center hosting for mobile edge computing functions and a wide range of other next generation and 5G-enabled services. This program is driven out of our 4 operational regions as the U.S., Latin America, EMEA and Asia, and it's coordinated and integrated at the corporate level by our new Chief Technology Officer and our Senior Management's Innovation Counsel. Given the long time line to the technologies being developed, we're aggressively pursuing opportunities across 4 quadrants that we hope can develop into material contributors to ATC's property revenue and AFFO growth within a 5- to 10-year time horizon. These 4 areas are pretty simple. First, on existing assets with existing tenants. Second, opportunities on existing assets with new tenants. And third, on new types of assets with our existing tenants. And finally, fourth, on new types of assets with new tenants. So regardless of the quadrant, geographic location and franchise real estate rights are going to be critical, and we're working to understand that technology is being developed in a number of industries, such as automotive and aviation, and what their transmission siting requirements will be. We believe the rest of the input elements of the signal transmission supply chain are largely just a classic rent-or-own decision. U.S. fiber is an example of this, particularly in the context of small cell deployments. To be successful in that space, we don't believe that we actually need to own fiber in the U.S. Instead, in most cases, we can simply lease it from or partner with one of the many existing fiber providers typically operating in a given U.S. urban geography. This is evident in our recently announced alliance with Philips Lighting, in which we expect to jointly deploy colocatable smartpoles that can generate significant savings for municipalities through LED lighting installations. This solution will initially require fiber connections for communications purposes, but we do not believe that we have to actually own the fiber on the part of American Tower.

Now in markets outside the U.S. where the fiber optic infrastructure is much less developed, such as in Latin America, Africa and India, it may make sense to buy or build select fiber assets. This is especially true given the need for fiber to the tower in 4G and 5G deployments and for small cell installations in cities in these markets which have much higher population densities than in the U.S.

Consequently, we may opt to own fiber assets based on the project requirements and the fiber supply base in some of our overseas markets. As Tom mentioned, our recently signed agreement to acquire urban concrete poles and fiber in Mexico is a good illustration of this. The concrete poles represent a distinctive, large-scale portfolio of franchise real estate rights for signal propagation. And in terms of the fiber market, there is less competition, less fiber in the ground and in the air, and consequently, much more of an opportunity to drive tower-like returns in Mexico. As a result, we expect to add significant value to our existing tower footprint in that market by connecting those towers to the fiber backhaul network.

In addition, the 50,000 urban pole locations coupled with the fiber assets, will position American Tower to roll out 4G and 5G transmission points in the key urban centers in Mexico as a result of this transaction.

So looking forward, both in the U.S. and internationally, our mission is to constantly apply our core competency of securing and maximizing scaled franchise real estate rights from which mobile signals can be transmitted. Fundamentally, it's this core competency that defines our existing business and drives our investment evaluation process. And it's this core competency that's led us to build up our global macro tower business, which we expect to remain as the primary focus of this company well into the future.

Moreover, our latest technical and U.S. mobile market review has reinforced our expectation that 4G, including its advanced versions on a predominantly macro-based network, will remain the primary technology for True Mobility services for many years to come, likely into the early 2030s. And with the sub 6 gigahertz standard for truly mobile 5G service then layered into this network, we also expect that in the suburban transportation corridor and rural topologies, where our towers are located, 5G will still be largely an overlay on existing macro sites, further extending the value of the tower real estate franchise.

As a result of all these factors, we continue to believe that our core competency of securing franchise real estate rights, along with our skills and experience in crafting and managing tenant lease agreements, paves the way for us to continue delivering strong performance for shareholders as technology evolves through an advanced 4G architecture and then toward a 5G world over the coming decades.

So with that, operator, please open the call for questions.

(Operator Instructions) And our first question will come from Michael Rollins with Citi.

I was curious if you could unpack a bit more of what's happening in India and Africa with respect to some consolidation activity in those markets? And if you can just help us understand the puts and takes for the future revenue growth opportunities with what you're seeing from Reliance and Tata and some of the other companies that you provide colocation services to?

Sure, Mike. I'll start with India. We, as we said many times, fully expected significant consolidation among mobile operators in the Indian market. That's coming to fruition now. It's happening fairly rapidly. And at the end of the day, which will probably take a couple of years to play out, we're going to end up with what we consider to be the end game we always expected, a very rational market, both on the demand side among mobile operators and supply-side among tower companies. We expect between 4 and 6 mobile operators in the India market that are going to be well capitalized at significant scale and able to roll out 4G across the entire 1.2 billion population of the country. So that's a positive for us. It will look a lot like the U.S. market does today, for example, which has been very constructive for the tower industry. On the tower side, there are, I think, really 2 good things happening. Again, it'll take a year or 2 to play out. But you're going to end up converting from what is still largely a carrier-captive tower market to one that becomes completely independent and consolidated itself within probably 2 to 3 major national tower companies in that country. Again, look a lot like the U.S. supply-side. So we are going to experience, Mike, a couple years of elevated churn that Tom sort of highlighted when these smaller mobile operators merge into larger ones, which is happening with Tata into Airtel. Ultimately, Reliance will probably follow the same path. We will have churn or revenue -- and/or revenue reserves that go along with that. But we also expect that, that will tee up the consolidators with a much bigger subscriber base, larger financial base to work off of to build-out 4G, which will then reaccelerate our growth opportunity in India. Africa, much less consolidation going on. There are 2 or 3 major players in each of the markets in which we participate. We expect that those players are going to remain independent. They may consolidate against some of the smaller local operators into them, but I don't think that consolidation in Africa is going to be a significant issue for this company.

Michael, I might just add that 3 quarters of our revenue -- over 3 quarters of our revenues is coming from those consolidated entities that Jim referred to. So whether it's the Tata and Airtel, Vodafone and Idea or Jio. So we continue to drive more and more business with those large, well-capitalized players which we've always thought would be the long-term players in the industry. And based upon some of the work that outside consultants have done and ourselves internally, we do believe that we're in the market today that are just 800,000 to 900,000 leases. We do expect that to grow well over 1.1 million leases over the planning period. So there'll be a period of kind of flatness, if you will, I think, on the total leases in the country. There'll be continued growth. If you look at what RGO is doing in terms of the growth that they're having in the marketplace. But there will also be that increase in churn as we referred to. But again, in that 3 to 5 years, when the carriers are rationalized, and we do expect to see some really nice growth at the latter part of that.

And if I could just follow up? Can you just remind us the structure of Viom with respect to how that evolves within, I think, the broader portfolio that you have in India? If I remember correctly, there were some parts to the agreement in terms of opportunities for you to own more but also, I think at one point, there was the possibility of Viom merging into your broader India business. Can you just review where we are with those features and characteristics?

Yes, Mike. So there is roughly 45,000 sites in the Viom JV portfolio and about 15,000 sites in the legacy ATC India portfolio. As a future step in the not-too-distant future we are planning to execute is the merger of those legal entities into one organization. And so it will take our Viom ownership, which today stands at 51%, combine it with our 100% ownership of ATC India, it'll end up with about 66% ownership with the combined entity once that plays through. There are some puts and calls of the existing partners, which is Tata Group and also Macquarie over the next 2 or 3 years. Those puts and calls are optional. So we will play out who -- which is to remain in the entity and who might wish to exit and really work with that partnership, which I think is, frankly, an outstanding combination of parties to be working together. You've got ATC, which we'd like to think as the global leader in our space. Macquarie, which is an outstanding investor across lots of asset classes all over the world, especially in Asia. And also the Tata Group, which is -- which we think potentially the best partner among the best partners in India to have. So we want to perpetuate the JV, and everybody will work their percentage ownership over the next few years.

(Operator Instructions) We'll go to the next line, and the question will come from Amir Rozwadowski.

I was wondering if we could touch upon some of the incremental expansion that you folks are doing in Mexico, particularly around these fiber assets. First and foremost, what was the principal of genesis for wanting to own the fiber assets in the market? I realized in your prepared remarks, you talked about sort of population density. But I'm just trying to understand if this is demand-driven based on what you're seeing from regional customers, are there specific customers that are requiring fiber-linked assets? Because if we think about what's happening in the market, you've got functional separation of American Mobile, which will open up fixed assets. There is a longer-term trend with Red (inaudible) call initiative for wholesale fiber network. So really trying to understand sort of the genesis of that transaction.

Sure, Amir. Let me start with our tower assets. We've got about 9,000 our assets in Mexico today. Only about 20% of them have been served with fiber by the existing supply base, right? So whether it's Telmex or others that are present in the market, we've only got about 20% connectivity with fiber, which, for 4G, is becoming critical. A tower without fiber that's going to convert to 2 or 3 4G tenants is going to be a tougher cell than one that is connected. So it increases the value of the tower in a converting market like Mexico to have fiber connections. And we just weren't getting them quickly enough with existing supply basis there. The -- yes, so the primary motivation of us driving this transaction is our ability to then direct fiber deployments to our sites enhancing the value of the existing tower. This is a reference point. Again, our portfolio is about 20% of the towers connected with fiber. In the U.S., it's 85% or 90%. So completely different situation. We want to get ahead of fiber connectivity to the tower. Secondly, and this is an example of how each market is different. Each, even urban area can be different in its topology and population density. In the U.S., 85% of people live outside of urban markets, in places like Mexico, and especially, say, Argentina. You're looking again at the opposite, 75% to 90% of people in those respective countries live in urban -- dense urban areas, actually. And so there will be a more opportune business plan in those kind of topologies versus the U.S. for fiber-based small and macro cells in a very dense array. And so we're positioning ourselves sort of as a bonus in this particular transaction to have the layout to be able to transmit. So it's more of the transmission points that are supporting the fiber as well as the fiber that actually gives you that second leg of opportunity there. So just to summarize really, it's fiber to the tower in an underserved market. And secondly, very near future 4, 5G transmission points based on the fiber support infrastructure that's out there today.

And just a quick follow-up. Will there be any CapEx requirements to upgrade the lines or anything along those lines that we should consider?

Yes, I mean, the maintenance CapEx, interestingly enough, Amir, is pretty consistent with our general tower. It's in the 2% range. There could be some development CapEx that may get deployed depending upon the success and what the demand looks like. But no, there is -- we don't anticipate any significant startup CapEx, if you will, to upgrade the network.

And our next question will come from Batya Levi with UBS.

Just a follow-up in -- for Mexico again. Can you help us understand what percent of the portfolio will be reached by this asset? And you mentioned that there will be more investments made and if you can help us size that. Also -- do you also get some enterprise exposure with this portfolio? And how do you think about churn or invest or maintaining that in the portfolio?

Batya, as we move forward over the next few years, we'll need to run fiber laterals to its towers, but we've mapped this particular fiber asset out and we've got very good proximity to many of our sites, and so we think -- as Tom sort of suggested, it's a reasonable CapEx for the extensions of the existing fiber networks to get to our towers. And we'd like to take our coverage from, say, 20% of our towers to 40% to 50% over the next few years. And I think that the CapEx required to do that will fit within sort of our normal annual run rates across the markets. As far as enterprise exposure, it's relatively limited for a fiber-based business. Most of the customer base today actually is in the telecom sector. It's the customers that we deal with on the mobility side and some on the landline side that are little newer. But most of the enterprises, if you will, are actually telecom-related.

And the next question will come from Ric Prentiss with Raymond James.

Going back to the U.S. for a second, appreciate your comments. AT&T, on their call, mentioned they would get ramping with FirstNet in January. Can you talk to us a little bit about kind of the timing we could expect in seeing it from your side? And what -- have you seen any applications from them so far?

So Ric, we like to limit our commentary on individual customer transactions and application flows. But I think, first of all, to refer you back to their public statements, there are going to be -- the AT&T is going to be the best indicator of how they're going to roll this out. But our expectation, in general, for any new national network is that they will, first of all, have to meet a time line of 20% geographic coverage annually, starting in April of 2018 and 20% a year after that, up to 100%. The best way to get that kind of geographic coverage in a limited time frame will most likely be, first off, amendment-driven to begin with. And then secondly, and for future phases, there could definitely be some colocations and build-to-suit opportunities as well. So we do think this is going to -- this FirstNet program is -- first of all, is launching in the majority of states and territories that are now kind of accepting and opting into the network. And as you said, the public statement of the carrier is that they're going to start rolling it out. So it's going to be constructive to the tower industry next year and for many years to come, and we'll continue to report to you in our organic growth rates how that may affect it.

If I could put a second quick one in. Tom, what was the maintenance capital reduction in '17? Is it just timing? Is it going to slip into '18? Or what was going on there?

Some of it's timing. Ric, it's been steady at that 1.8% to 2% kind of range. I mean, there are certain programs, certain things that the teams think that they might be able to do. Some of it could slip into '18. It's project-by-project driven. So as I said, I think the way to think about it is in that 1.8%, 2% kind of growth rate.

And our next question will come from David Barden with Bank of America.

Tom, a couple of accounting questions. Just on the Reliance provisions for revenue 3Q, 4Q. Could you kind of talk about what happens there that causes the provision and what might happen that causes that provision to go away? And is this something we need to be thinking about for the '18 AFFO outlook? And then second, on the depreciation change that you called out in the quarter. Just wondering if there is like a cash flow effect to that related to taxes? And if I could throw in a last one, I'm sorry, but just on the 500 towers you bought in the U.S. Could you talk about where those came from because that's a pretty big portfolio by that we haven't seen in a while in the U.S. market?

Yes. Sure, Dave. On the depreciation one, there is no cash flow impact. That was really just some internal work on the asset base. And so there is no cash flow impact on that one. Relative to the revenue reserves, it was really isolated to a specific customer. And so just on an internal and just looking at it from a GAAP perspective, we're continually looking at the ability to collect those types of revenues as we do in every business and for every customer. And so we made the determination that there is some risk to be able to collect this particular revenue and so we booked a reserve, I mean, to the extent then it turns into churn and it ultimately -- we determine that we can't collect it then we'll move it into churn and then it will then be part of the organic growth. So this is just a normal process that we have for every customer in every market that we do business with. And relative to the U.S. assets scattered a bit around the states. I mean, this is work that we've done and we do with certain developers on an ongoing basis. We're really excited about the portfolio, a relatively low tenancy. And so we think it's going to be well positioned, particularly as we start to see the further growth of 4G and even 5G deployments over the next several years.

And the counterparty is a private company in the (inaudible). We'll leave it there, I think.

And the next question comes from Simon Flannery with Morgan Stanley.

Great. Can we talk about Europe for a second? Just update us a little bit on how your portfolio is performing there? And then you had the partnership here a few months ago. I mean, what are your expectations for doing further investments in Europe and certainly a lot of carrier portfolios out there?

Sure, Simon. Our operations in Germany, which have been long-standing over 5 years, and now in France, which is within this year, was acquired, are both performing at or above their business plans. Germany has been a real success for us as 4G has rolled out, a little greater pace than we modeled. In the business in France, we think it's going to perform very well as their regulatory requirements on the mobile operators to expand 4G coverage in these sites that we purchased in both countries, are macro sites, large towers and transportation corridors, suburban/rural areas just like we have essentially in the U.S. So we think these are going to be good growth assets in France. As far as broader opportunities, we have, of course, interest in Europe. It's a great set of market. All the fundamentals for geopolitical and macroeconomics are, of course, there. What we have continued to seek to find are asset prices that match how we think those assets are going to perform from a growth perspective over time, and we especially have not had an opportunity to find assets priced in a way that requires decommissioning to get the return. So until we can make those equations work, we'll continue to be prudent, I'd say, in Europe. But there's certainly appetite there for the right asset price level for us, as we've just demonstrated in France when all the parameters work out for our investment process.

And we'll go to the line of Tim Horan with Oppenheimer.

Jim, kind of a multipart question. But on 5G, any update on when we might see fixed deployments and what do you think it might mean for you guys? Yes, I know it's kind of a constant evolution and the same thing with mobile. And a guess related to this, do you think in the U.S., we can support kind of a wholesale-only kind of wireless build out for either mobile 5G or fixed 5G longer term?

So Tim, for fixed wireless deployments, the standard is we understand the process to be available about 2019. At that point then you can get real scale deployments. Between now and then, they are going to be trial projects and such, demonstration projects that you'll hear about, and they'll be part and parcel of the path to get to real scale deployments under a single global standard. So fixed wireless, which is more or less a cable or fiber to the home competitor, could be in the offing in nearly 2020s, let's say, at scale. We have this initiative as part of our innovation program to say, "Okay, what are the kinds of transmission sites? What are the densities? What are the heights? What are the municipal and/or private owners of those types of sites or access to those kinds of sites that we could get?" And we hope to play a part in that. But it will only play a part in it if we can get serve the macro return potential out of whatever these architectures end up being. But these kinds of projects are coming, dragging fiber cable competitively to individual homes, just proving to be a difficult investment profile for many have sort of tried it. So 5G fix wireless could be the competitive answer here. And what's interesting there is there may be an opportunity, in addition to the last 10th of a mile from the pole, if you will, to the home to add some wireless backhaul solution that could go with this. So we're exploring all of those. Again, it's 5- to 10-year time frame before anything really gets to scale. But we're involved, interested, and we will participate in trials, et cetera. On the mobile side, it's probably a few years beyond that, whereas again, there is a completely globally agreed to mobile standard and spectrum bands that are usable for mobile. So right now -- again we understand that, the max -- the majority effort in the 5G standard-setting process is around 6 gigahertz spectrum and above. Because that's where you can get the largest blocks of spectrum to actually deliver 5G performance characteristics. Narrow bands of spectrum, 5 or 10 megahertz, really aren't going to give you everything you need. And so the large bands at the very high frequencies are the first design standards. Those frequencies, as you know, do not transmit very far. They're great for that sub quarter mile requirement. But they're not great for any kind of true mobile sort of application. So someday down the road, which will be post-2020, the standard for mobile and lower bands will, I'm sure, be written. And that's why we sort of think that for true 5G-based design criteria of a global standard, it's probably not till the 2030s that you see this in mobility substituting for 4G, if you will. The wholesale question, that's going to be for someone else to determine based on the time schedules that I've talked about, how the 4G network evolves. There won't be 4G networks of 5G characteristics, so to speak, that will also come down the road on the 4G standard as we move ahead here. So to me, a very complicated technical space, but what we're focused on is, "Who wants to transmit? What are the radii that they need? What are the heights that they need? And what's the density and actual locations that are required?" We're going to go after those locations because the technology flow will be fairly agnostic if you've got the right locations positioned in the right way and can get the return on those. And that's what we're going to be after.

And our next question will come from Phil Cusick with JPMorgan.

First, should we consider you a competitor for substantial small cell deployments in the U.S. going forward? Or is the Philips deal more about sort of one-off opportunities? And if you looked at RFPs in the past, where it would be attractive to be a small cell vendor while wholesaling fiber?

I'll try to unpack the question, Phil. I think you're asking, do we intend to be a competitor in small cells? Yes, we are a leader in small cells already in the indoor space. And we're seeking to see if there is similar return profiles available to us in the outdoor space. The whole point of agreements like with Phillips is, how do we get large-scale access to real estate rights. In this case, generally, municipalities that will be installing potentially upgrading -- upgraded lighting systems, right? So if there is a sort of midsize city in the U.S. that's got 10,000 light poles, and we get to substitute some proportion of those with the Phillips product, Phillips and ATC product, which is also usable for mobile transmission, we'll have rights to transmit from, theoretically, 10,000 places in a given city. And that real estate right is what we're really after. Everything else, as I sort of suggested earlier in the small spell space, if you will, is our production input, right? It's a factor of production. We or the carriers will buy the radios. We will link the light poles by leasing probably fiber from the local providers, in the U.S., for example, and we'll deliver the solution to the carrier base. So that's what we're really after here. So of course, we're going to compete in the real estate citing side of this and will go to supply base in cases like we've talked about in Mexico when it makes financial sense, we might actually build or buy something. But by large, we're going to be competing in the space we've always competed in, which is gaining franchise real estate rights to transmit mobile signals from. And those could be macro cells, small cells, indoor systems, et cetera. Whatever makes sense to scale a real estate play, that's what we're going to try to do.

That's helpful. And then let me follow up on Mike's questions on Africa. There are a lot of tower assets developing there. Do you think you could scale in Africa like you scaled in India either buying more assets in your current 4 countries or targeting additional markets?

We're starting to get to a position in Africa where there are 2 or 3 large tower companies, maybe 4, if you will, that have aggregated assets. We are one of those. And right now, we don't necessarily see an opportunity to consolidate with any of the others on the immediate time horizon. But that could be out there in the future. And secondarily, there are a few targeted portfolios in some of our current markets that we'd be interested in and there may be a market or 2 that we'd like to bolt-on to the array that we have there now. But I'm not sure it's going to be our biggest investment market in the immediate future.

And we have time for one more question and that will come from the line of Brett Feldman with Goldman Sachs.

And two quick follow-up questions related to the transactions you announced with the release. The first one, on the U.S. portfolio, you mentioned that was a private developer. Some of the private portfolios that have recently been developed have had nontraditional lease terms, particularly with regards to escalators. I'm curious if you can give us some color on whether that applies to this portfolio? And then secondly, it does sound like outside the U.S., you may be interested in doing more fiber deals if you can identify the same type of circumstances you identified in Mexico? And so I'm curious, if you could just talk broadly about how you think about valuation of that asset class versus the tower that you've traditionally focused your investment in. And I'm curious if you can disclose what the multiple you paid for the Mexican fiber portfolio is.

So on the U.S. acquisition, absolutely traditional lease terms that we would have negotiated ourselves throughout that portfolio, which is one of the reasons we went for it. Outside the U.S., our fiber evaluation criteria to be very in line -- much in line with macro tower asset acquisitions in those markets. In other words, we look at the growth rates, the cost of operation, maintenance and redevelopment CapEx, and we're going to try to hit the risk-adjusted return for the market in places like Mexico and Brazil. As you've heard, it's 10% to 13% -- 12%, and we'll use that same kind of standard for -- whether it's a fiber-based asset or a macro tower-based asset. One of the most important differentiator is again is population density in urban areas in proportion of national population in those urban areas. In places like Argentina, as I said, upwards of 90% of the people live in dense urban areas, predominantly in this case, around Buenos Aires. We have an investment there. Because the demand for that fiber and small and macro cell transmission ability that will go with that. And that market is it converts to 4G and it's behind, as you know, because of the geopolitical situation that's been in that country, is going to be, we think, an outsized growth opportunity for us. Same in Mexico. And so we have an offensive option in these kinds of places, where the circumstances happen to come together where it's not been built out yet. 4G is coming, if you will, and you're going to need fiber to really serve the densities that we're talking about. So that -- those are the factors that go into it, Brett. But the investment community process, whether it's an innovation-type project or traditional macro-type project is exactly the same criteria with exactly the same people involved in the same process.

And can you break out what you paid on the Mexican transaction?

No. We don't have that as public.

And if there are no closing comments, I can go ahead and end the call with the replay information?

Perfect. Thank you.

Have a great week. Bye, all.

Thank you, and the call will be available for replay after 10:30 this morning and running through Tuesday, November 14 at midnight. You can access the AT&T teleconference playback service by dialing 1 (800) 475-6701 and entering the access code 431354. International parties may dial (320) 365-3844 with the access code 431354. That does conclude the conference for today. Thanks for your participation and for using AT&T teleconference. You may now disconnect.