How do towercos think about innovation? Into which categories of communications infrastructure are they diversifying? While the source for this report was a panel session and roundtables at the TowerXchange Meetup Americas 2019, the insights are applicable worldwide, as all towercos are ordering from the same menu of innovation. For starters, they’d like some non-traditional tenants on their existing towers, plenty of fibre with perhaps a side of small cells for the entree, edge data centres as the icing on the cake.
This commentary is drawn both from a panel session focusing on new business models for tower companies, held at the 6th annual TowerXchange Meetup Americas (July 9-10, 2019), and also from a round table breakout on a similar topic. As the round table was held under the Chatham House Rule, the sources of insights and quotes are not attributed. However, many remarks from the panel session are attributed to the relevant participant, who we thank for their contributions:
Moderator:
- Isfandiyar Shaheen, CEO, NetEquity Networks, and formerly a Director at MENASA towerco Towershare
Panelists:
- Gonzalo Cornejo, CFO, Mexico Tower Partners (MTP)
- Mariano Gomez, Vice President, Business Development, BTS Towers
- Steven Haymore, Director, M&A and Managing Counsel, Latin America, American Tower
- Don Van Splunteren, Global Vice President of Sales, Phoenix Tower International (PTI)
For the sake of clarity, readers should not assume that non-attributed quotes originated from one or other panellist, because in most cases they didn’t.
Introduction: the parameters that define towercos’ appetite for innovation
Many of the world’s largest towercos have access to substantial cash flows, cash reserves and low cost capital: they have the resources to diversify beyond ‘vertical real estate’. In many cases, the scope of potential diversification is limited to categories of communications infrastructure that are adjacent to towers; typically business models that are still predicated on generating long term revenues through building or buying then leasing communications infrastructure to credit-worthy tenants.
However, large towercos are cautious about diversifying beyond their core tower business for fear of diluting their focus and adversely affecting their robust valuations. If large towercos are to evolve and diversify, they must retain the buy-in of a relatively conservative base of investors, many of which favour investing in the tower industry precisely because of its laser-beam focus on ‘vertical real estate’ – by which we mean the purity of their investment in masts, towers and rooftop sites for the wireless industry. So large towercos have the resources to innovate, and they are innovating, but they are doing so with caution.
In contrast, smaller, privately funded, independent towercos are sometimes free to innovate and act with more agility and flexibility to embrace opportunities to diversify into adjacent categories of communications infrastructure.
The tower industry has emerged as a more efficient business model for the deployment of capital into communications infrastructure than the vertically integrated Mobile Network Operator. The extraction of towers into specialist infrastructure companies has functioned as a remedy for the overlapping tower networks originally built by MNOs. Through organic new build and inorganic acquisitions, towercos now own 68.8% of the world’s investible towers and rooftop sites, creating a new infrastructure asset class with a US$330bn global valuation, creating efficiencies and value by sharing those towers with multiple MNOs, and with other tenants.
The emergence of tower companies has been one of the success stories of the wireless era. But how will towercos sustain their impressive growth and margins? How do they find comparably stable cash flows, which their investors have gotten used to? And how do towercos ensure their proposition, products and services remain indispensable in an era when networks are being transformed as data demand continues to grow exponentially?
Picking up that theme, Gonzalo Cornejo, CFO of MTP, the leading independent towerco in Mexico with 2,500 sites, asked: “Where is growth going to come from? We have seen lots of opportunities to invest in macro towers in recent years, and into the medium term, but that won’t continue in the same rhythm, so where will growth then come from?”
“As wireless operators have evolved in the smart phone era, so our business must evolve,” suggested Mariano Gomez, VP Business Development for BTS Towers. “There’s plenty of money still to be made in macro sites development, and wireless operators’ CTOs have less money than ever for deployment.”
PTI owns and operates over 7,000 towers in 14 countries, from the U.S. to Argentina. PTI Global VP of Sales Don Van Splunteren asked “Where is the technology taking us? Telecom infrastructure depends on how the network needs to be designed, and technology dictates how the network needs to be designed. For example, massive MIMO enabled us to increase capacity, but ultimately created different antenna configurations and different infrastructure needs.”
“The questions we are asking about innovation are: in what sectors and in what business models can American Tower add value for our clients?” Asked Steven Haymore, Director, M&A and Managing Counsel. “We are not always going to be the best company to play in different spaces. So we’re thinking about where we can add value, while respecting our shareholders and maintaining fundamentals of the tower industry.”
PTI’s Van Splunteren echoed Haymore’s view “We like the fundamentals of the tower model very much: long term, non-cancellable contracts, the capex intensity, relatively light opex. We want to respect those fundamentals.”
A glimpse into American Tower’s innovation program
American Tower has invested a billion dollars into their innovation program. Much of that capital has been deployed into acquisitions focused on provisioning fibre to the tower in five countries to date, but they’re not putting investments into U.S. fibre in the same magnitude as Crown Castle and Digital Colony. American Tower are also exploring opportunities in small cells, smart poles, IoT, energy as a service, and edge data centres. In doing so, American Tower is adapting to meet the changing needs of their clients, whilst plotting a roadmap for the potential future evolution of their business.
American Tower is innovating at a carefully controlled velocity and scale – moving from ideas to proofs of concept to regional pilots. By starting early, making small investments, and doing small acquisitions, American Tower is learning in which adjacent categories of communications infrastructure they can add value, where they don’t add more value than other actors, and eliminating innovations that may prove value destructive.
Designed to position the company for continued success in a 5G world, the American Tower innovation program has three key prongs:
- Leveraging existing assets for additional applications
- Evaluating new communications real estate architectures
- Capturing opportunities to serve new tenants beyond their traditional MNO client base
While American Tower is committed to innovation, that doesn’t mean they’re making a radical shift away from a focus on macro towers, indeed they are explicitly targeting macro tower-like returns on innovation investments.
What is the right kind of fibre for towercos to be interested?
PTI’s Van Splunteren continued: “We like fibre. We would prefer to concentrate on dark fibre, getting a contract, constructing a different model. We’re less keen on lit fibre – it’s more of a service than passive infrastructure.”
With upwards of 30% of cell sites in CALA not yet fiberised, PTI are not alone in investing in FTTT – that has been the focus of American Tower’s forays into fibre, albeit focusing to date beyond the U.S., including Mexico and Argentina. During their Q119 earnings call, American Tower Chairman, President and CEO Jim Taiclet said the company had “made some relatively small investments in international fibre… but continue to view U.S. fibre assets as inherently less attractive, due to the extensive availability of competitive fibre supply in the U.S. and the resulting less attractive growth and return characteristics of domestic U.S. fibre.”
Panel moderator Isfandiyar Shaheen of NetEquity Networks emphasised the commonalities between towers and fibre: “fibre is a long term asset – futureproof in terms of bandwidth – which like towers provide long term, predictable cash flows. What are the challenges to applying a similar financing construct as we have to towers to a long term asset like fibre?”
“We’re thinking how we can create contracts and business models more similar to the tower industry than what we see in the fibre industry,” said American Tower’s Steven Haymore. “My personal view is that when we look at wholesale, neutral host, fibre providers with long term contracts, if we can setup contract conditions and margins similar to a tower sale and leaseback transaction, we might be able to unlock some capital and resources for the fibre space.”
BTS Towers’ Mariano Gomez added: “A lot of the capability to deploy fibre at scale is determined at municipal level. If they don’t help by providing rights of way, we’re at an impasse.”
Isfandiyar Shaheen called attention the example of Open Fiber, to a joint venture in Italy between energy company Enel and investment bank CDP. Open Fiber is a wholesale only, FTTH infrastructure-as-a-service play, which last year secured EMEA’s largest fibre project finance package (€3.5bn) to facilitate rollout of their ultrafast network. “My personal thesis is to leverage the electrical grid to provide fibre and bridge the digital divide,” added Shaheen.
Infill sites: small cells, DAS and other alternate site typologies
The rollout of modern wireless networks using lower and mid-band spectrum has enabled radio network planners to rely on a relatively low number of high sites, primarily the macro towers that form the majority of towercos’ portfolios. But as carriers start to use millimetre wave spectrum for 5G, and as legacy macro networks become overloaded, radio network planners need more sites. And as they need more sites, interference becomes an increasing problem. As a result, low height, low power densification sites are being added to networks both outdoors and indoors. How must towercos evolve to meet the changing characteristics of today’s radio networks?
Many towercos have diversified into alternate site typologies. For example, at the end of 2018, PTI acquired Syscom Telecom, a company with 80,000 billboard-mounted wireless sites, primarily in the U.S., but PTI also has 2,000 such sites in Colombia. “Billboards are great infill sites,” said PTI’s Don Van Splunteren, “as they offer low elevation, powered sites, near populations and high trafficked areas. Billboards will be particularly useful as operators start using millimetre wave spectrum and seek to cover smaller distances with smaller search rings.”
As cities become more and more densely populated, and as networks become more crowded, capacities need to increase, and small cells may emerge as a valuable opportunity. But, at an event and in a report focusing on Central and Latin America, it should be noted that there remains a five-plus year lag to the U.S. market, where carriers are actively deploying small cells. CALA is not a completely nascent market though.
“We see demand for small cells, especially in Brazil with Phoenix Tower do Brasil, plus some in Mexico,” said Van Splunteren.
As for who will take the lead when small cells are rolled out in volume in CALA, in the words of one small cell vendor at a round table: “the era of selling small cells to a single MNO has passed. The economics simply don’t make sense if they are not shared.”
DAS continue to be deployed in selected use cases – most CALA towercos have a few DAS in their portolios. But cost and complexity means we’re seeing less investment in DAS by both MNOs and towercos, and often it’s now the facility owner who pays for the DAS.
Edge
“Everyone has heard about moving service to the edge,” said Van Splunteren. “It’s a great opportunity, and it’s a real estate play: saving on the time and cost of transporting data needs secure space, close to fibre. The FANGs (Facebook, Amazon, NetFlix and Google) don’t want one single episode of Game of Thrones sitting middle of the country, they want to bring that closer to the subscriber. We are starting to see the emergence of companies specialising in edge data centres, and we’re starting to get inbound requests regarding our real estate. But it’s still a nascent market. We don’t expect volume for four to five years, until the 5G era.”
“I agree that edge data centres are currently a nascent market,” added MTP CFO Gonzalo Cornejo. “That might give us a couple of years to think which model each towerco would like to see. There’s a model respecting our fundamentals, wherein we provide white floor space with power and security. Then there’s a model when we provide the whole data centre service, but some of the tower fundamentals are not present: there many different contracts, those contracts are of relatively short duration and higher churn, and there is more exposure to technology. We try to follow the fundamentals of the tower business – we don’t want to get involved in providing service to end users – we want to provide service to MNOs and to the FANGs.”
“American Tower has recently acquired Co-lo Atl in Atlanta,” said Steven Haymore. “We see it as an opportunity to learn about the edge. We want to see it as a potential extension of our focus on passive infrastructure, and we feel we can add value both through capital allocation and operationally. We want to meet the evolving needs of MNOs and of FANGs in the 5G era. American Tower believes there’s a real commercial opportunity to build incremental income from new clients from our existing assets, although it might not be for a while in Latin America.”
Experts estimate that adding an eight rack edge data centre to a cell site could add 20KW to the power load – possibly more than the grid can provide. Whilst towercos in the Americas cling to their ‘steel and grass’ model, and seldom provide energy as a service, co-locating edge data centres at cell sites could be problematic. What will be the uptime targets for edge data centres? It is not yet clear. It may be less than five nines, but it will require the site owner to guarantee power against a Service Level Agreement that would make towerco CEOs’ eyes water! And security will have to be similarly robust.
A model response may emerge from China, where Internet giant Alibaba is among the shareholders in China Tower Corporation (CTC). CTC has recognised the concept of the “social computer room” – secure, powered enclosures at cell sites which can enable “intelligent connections” – which sounds a lot like an edge data centre. CTC has carved out a new subsidiary Tower Zhilian to manage social computer rooms, social information, IoT and other new emerging opportunites.
American Tower shares early insights into their thinking on the edge
FierceWireless reported on comments made by American Tower SVP Rod Smith at the Oppenheimer investment conference in August 2019.
Smith intimated that the concept of edge computing wasn’t happening quite yet at the base of the towers, “but we think there could be a day where the latency in a 5G network is so important that having that edge computing right at the base of a tower.” American Tower reportedly has already located a number of small data centres at the base of towers, increasing revenue from the sites, although they don’t think this represents true edge computing yet. Smith continued: “The most important thing there is we’re learning about how to deploy these things, we’re learning about the cost of deploying small data centers,” including the power and cooling needs that goes into them so that if edge computing does land at the base of a tower, American Tower will be ready to execute on that.
Will towercos engage with active infrastructure?
While there are communications infrastructure companies whose scope extends beyond passive infrastructure, they are to date found only in Europe (for example CETIN, MBNL and Cornerstone). While towercos may become fully fledged ‘infracos’ in the long term, deploying both passive and active infrastructure, it is not likely to happen in the near term – at least not in the Americas.
“Getting into active equipment would raise questions about how we’re regulated,” said one towerco.
“The problem towercos are solving on an ongoing basis is to accelerate deployments, help out with permitting, and unlock efficiencies through co-location,” added another towerco. “A broader skillset is required to manage active equipment. If we’re talking about just energy, it may make sense, but managing antennae requires a deeper knowledge of frequency engineering – that’s a core competency of our carrier clients.”
“The potential efficiencies unlocked by managing both passive and active equipment might best be captured at the managed service level,” continued the towerco. “Driving toward one site visit.”
“We check who the MNO is using as their O&M contractor”, added another towerco. “We prefer to have three way conversation and to consolidate site visits for passive and active equipment maintenance where possible.”
“Towercos can offer managed services, and energy as a service, but we still put the network up,” said a tier one OEM. “The barriers are the economics and the contracts we have today.”
“I don’t see us getting into active equipment. Having the active equipment on our books would be an issue for us,” concluded one towerco.
“I agree,” said another towerco. “Our investments are based on long term contracts – I’m unconvinced how long term an investment in active equipment would be.”
RANsharing… or not
One towerco participant at a round table suggested: “we thought would could play a positive disruptive role by enabling active infrastructure sharing. But rent-seeking investors seek bankable contracts with a certain quality of rent – pushing beyond passive to active infrastructure sharing was asking too much from investors. Ultimately our contracts ended up with same penalties against active sharing as other towercos.”
Could RANsharing ever become an opportunity for towercos?
“MNOs don’t want to share anything!” Said one towerco. “It’s been a battle to move the build-to-suit model in Central America – if they were reluctant to share towers, they’re going to be even more reluctant to share antennas.”
“Turning RANsharing into a bankable contract would be difficult,” added another towerco.
Wholesale networks
What role could wholesale 4G and 5G networks play in the future?
ALTÁN Redes is a unique wholesale network – what is it, and how does it make towercos think differently?
Mexico has only three MNOs: dominant market leader Telcel, AT&T and Telefónica, and a lower population coverage than most other CALA countries. Seeking to increase competition and coverage, the government awarded a contract to ALTÁN Redes, which is responsible for the design, deployment, operation and maintenance of Red Compartida, a wholesale network (or ‘carriers carrier’) providing a ‘Connection for All’.
ALTÁN Redes was granted coveted 700MHz spectrum for 4G free of charge, in exchange for an obligation to cover 92.2% of Mexico’s population in five years. Having achieved 30% population coverage by March 2018, Red Compartida commenced commercial operations. The Red Compartida currently covers about 50% of Mexico’s population.
Gonzalo Cornejo, CFO, Mexico Tower Partners explained the implications for towercos. “ALTÁN Redes is effectively a new operator in the Mexican market. While they are a wholesaler, they had no infrastructure at the outset, so initially they’ve leveraged a lot of co-location, particularly in urban areas. They will increasingly focus on build-to-suit as they push into rural areas.”
Many towercos report that ALTÁN Redes represents the lion’s share of both co-location and new build in Mexico. And most towercos have reconciled themselves with ALTÁN Redes’ wholesale business model, which is somewhat competitive to towercos’.
Towercos are thinking about how they need to evolve to solve carriers’ problems, and to support changing radio network designs
Conclusions
Towercos are thinking about how they need to evolve to solve carriers’ problems, and to support changing radio network designs. Whether we’re reporting on TowerXchange Meetups for CALA, Europe, Asia or Africa, there are strong commonalities in towercos’ thinking about innovation. There is a preference to stick close to the proven fundamentals of leasing up communications infrastructure on long term contracts.
But there is also a growing awareness of how cell site owners can leverage their infrastructure to tackle the inefficiencies in cost structures of multiple industries. We’ve seen examples of towercos completing the ‘cold chain’ to bring life-saving vaccines to kids. We’ve seen towercos enable ‘Digital Villages’ and support rural electrification. Reverting to urban environments, the convergence of cell sites and municipal services could enable smart cities: Wi-Fi services, smart lighting, V2I, sensors, CCTV and other public safety applications. PTI enables 911 services in the Caribbean: “the equipment is on 70 towers so far, and it’s usually very light equipment, so it can go on any structures the government wants! We’re conscious that government services and healthcare adds value to our customers, so we’re happy to help,” concluded PTI’s Don Van Splunteren .
