The tower industry a parallel yet distinct segment of the telecom industry leverages long term contracts with credit worthy tenants to generate a predictable and highly investible cash flow.
India is in the midst of a new wave of market restructuring and around 130,000 towers are likely to come to market in the near future.
The Indian Market
State-owned MNOs. Among the latest news, BSNL has received the green light for the carve out of its 65,000 towers into a separate infrastructure unit while the other state-run MNO, MTNL, is considering divesting its 10,000 tower portfolio in an attempt to reduce its debts. BSNL’s carve out could be valued up to Rs 20,100 crore (USD 3 billion) and analysts are excited by the potential of these towers coming to market as many are in prime locations with considerable tenancy ratio growth potential, having not been proactively marketed before.
BSNL has leased out 6,505 of its towers to other telecom operators, suggesting a tenancy ratio around 1.1x. Out of the 6,505 spaces that it has rented out, Bharti Airtel accounted for 2,251 slots. It was followed by Reliance Jio with 1,440 slots and Idea and Vodafone with just above 900 towers each. Reliance Jio has led the rollout of 4G and has opted to self-deploy as many as half the sites in their network, including constructing over 25,000 new structures, most of which are micro sites.
Vodafone and Idea Cellular. Additionally, while Vodafone and Idea Cellular keep pursuing their merger plans, their tower portfolios are up for sale with American Tower and Brookfield still head to head as the most likely buyers. And Brookfield isn’t only eyeing this deal but is also in the process of acquiring a 51 percent stake in Reliance Infratel for Rs 11,390 crore (USD 1.7 billion). The Vodafone-Idea merger could also precipitate a restructuring of the ownership of joint venture giant Indus Towers.
GTL Infrastructure and Tower Vision. GTL Infrastructure is in talks with investors and planning to switch its ownership by March 2018. The company is reportedly been evaluated at Rs 10,050 crore (USD 1.5 billion). There are also rumors that Tower Vision, an independent towerco with 8,400 towers owned by a group of international investors, has been up for sale.
Bharti Infratel. Bharti Infratel Ltd has decided to explore and evaluate acquisition of stake in one or more tranches in Indus Towers Ltd with the aim of making it a subsidiary or wholly owned subsidiary of Bharti Infratel Ltd. Vodafone and Bharti Infratel own 42 percent each while Idea owns 11.15 percent. US private equity firm Providence owns the balance 4.85 percent. In the meantime, Bharti Airtel Ltd may acquire a significant stake in its Bharti Infratel Ltd. It currently owns 58 percent. The markets are speculating that perhaps, Bharti Infratel will first acquire Indus Towers, which will push up its valuation, and that will be followed by a sale of Bharti Infratel itself.
Further rational consolidation is welcomed by towercos who would prefer to see spectrum holdings consolidated into four or five companies with the capital and appetite to rollout. With India’s 3G overlay around half finished and expected to reach 95 percent coverage in the next 18–24 months, the 4G rollout has already started in tier one and tier two cities. In the near term, the 4G rollout is expected to have a marginal impact on the profitability of Indian towercos, whilst the majority of BTS are added through loading – the addition of a second set of antenna by an existing tenant – but when 4G rollout progresses to adding infill sites for densification, expect to see a significant increase in tower cash flow.
The Asian Market
1,992,467 of Asia’s 3,047,847 towers are owned or operated by towercos representing 65 percent of the total inventory of assets.
Exciting news and developments continue to emerge from the region, with the highlight being edotco’s two announced acquisitions in Pakistan months apart from each other, first in June then late August. The Tower Share (Tanzanite Tower) portfolio includes ~700 towers at the price of Rs 600 crore (USD 90 million), while the Pakistan Mobile Communications Ltd (Jazz) sale and leaseback deal includes ~13,000 towers at Rs 6300 crore (USD 940 million) edotco is quickly jumping the global ranks to become not just one of the best, but one of the world’s biggest towercos. Including the acquisitions, edotco would own ~31,600 towers, putting it in eighth spot out of the 272 towercos TowerXchange currently track. It also manages an additional 8,700 towers.
Over in China, the world’s largest towerco China Tower Corporation (CTC) celebrated its third anniversary in July and continues to drive towards a public listing in Hong Kong. It was reported by Reuters that CTC had picked China International Capital Corp Ltd (CICC) and Goldman Sachs to lead its IPO, pending final board approval; more banks could also be added to the final sponsor team. And while CTC is keen to list by the end of the year, it will likely take place in Q1 2018. This summer also saw the formation of the China Independent Tower Alliance, currently with 60+ member organizations. China’s 200+ independent towercos have been lifted by a government document recognizing and legitimising their role in the country’s co-build, co-share system.
The Indonesian tower market remains strong, with new builds and co-locations being added. There’s a lot of business as usual in this mature market, however, also a bit of buzz with the refinancing of KIN and STP. The major players have typically complemented their organic growth with acquisitions and with a good number of smaller towercos in the ecosystem, consolidation will continue to be expected. Myanmar on the other hand is seeing a surge of new entrants rolling out BTS for fourth operator MyTel, while the existing towercos are enjoying a boost in tenancy ratios also through MyTel and increasingly MPT. However, current political turbulence is making it more difficult to attract international investment.
The regulatory environment for towercos and infrastructure sharing varies from mature tower markets such as India and Indonesia, where the regulatory regime is well established, to regulatory environments still drafting policy such as Bangladesh and Nepal, where independent towercos are a relatively new business model.
The tower industry has emerged as a parallel yet distinct segment of the telecom industry. Independent tower companies (towercos) separate infrastructure components of telecommunications from the more volatile retail components – towercos have no subscribers, no spectrum, and have reduced exposure to technology evolution. Towercos leverage long term contracts with credit worthy tenants to generate a predictable and highly investible cash flow.
Inaugurated in the mid 1990s in the USA and Canada, where towercos now own 73 percent of sites, the independent towerco model has evolved for different local market structures and cultures. Towercos now own 68 percent of India’s cell sites, 50 percent of the sites in CALA, 40 percent of Europe’s sites, 39 percent of the sites in SSA, and 100 percent of China’s towers.
Towerco growth worldwide. TowerXchange tracks 264 towercos that between them now own 2,936,086 of the world’s 4.3 million towers (68.2%).
Towercos’ reported EBITDA, based on the results of the top 15 listed towercos, grew at a CAGR of 10.75 percent between FY14-16, while aggregate site counts across the entire tower industry grew at a 14.61 percent CAGR over the last two years, driven by a combination of organic and inorganic growth.
Towercos are building a significant majority of the world’s new cell sites, including ground based towers, rooftops and, increasingly, small cells and DAS. Towercos are also acquiring an increasing proportion of existing cell sites, both through sale and leaseback transactions with MNOs, and through MNOs carving out the towers from their balance sheets to create their own ‘operator-led’ towercos. MNOs are motivated to monetise their towers to release capital to focus on their core business, to reduce debt, to unlock the efficiencies of shared infrastructure, to accelerate rollouts, and to exploit the difference in trading multiples: MNOs typically trade at 4-7x, while towercos trade at 10-21x.
Despite being created only 22 years ago, the telecom towerco asset class should currently be valued at USD 278.8 billion.
Most valuable towercos by market cap in the USD 278.8 billion global tower industry. Market cap of listed entities taken at May 27, 2017, estimates for unlisted entities based on most recent refinancing plus industry sources
Contrasting towerco business models. Only 15 percent of the world’s towers and rooftops are owned and operated by ‘pureplay independent towercos’, yet those towercos represent over 60 percent of the capital value in the asset class. These valuations recognise the absolute independence of third party-owned, pureplay independed towercos as the optimum business model for shareholder value creation.
More than half the world’s towers are held by operator-led towercos, which TowerXchange defines as towercos that are majority owned by one or more MNOs, but which lease their sites to third parties on a non-discriminatory basis. One often finds that the business model for such towercos shares value creation more equitably with MNOs. For example, both of the world’s largest operator-led towercos, China Tower Corporation (1.7 million towers) and Indus Towers (122,730 towers) discount lease rates when additional tenants are added.
58,600 of the world’s towers are owned or operated by joint venture infracos, wherein two or three MNOs pool their towers but share only with each another. Joint venture infracos are currently only found in Europe, but the model is being replicated in Iraq and also potentially Saudi Arabia.
Just under a third of the world’s towers and rooftop sites, 1.36 million, are still owned by MNOs, but relatively few of these remaining captive sites are acquirable through conventional sale and leaseback transactions – at least, not if towercos comply with the investment thesis they’ve used to date.
Diversification of business model. The business of building, buying and leasing up macro towers is a proven, investible proposition which has created companies that are among the largest in both the telecom infrastructure and the Real Estate Investment Trust (REIT) asset classes. And some towercos have little appetite to diversify beyond that proven business model. Other towercos embrace innovation to evolve the business model, positioning themselves as neutral hosts of a shared Future Network.
Diversification models open to towercos. In emerging markets, towercos can expand their service proposition beyond vertical real estate to include provisioning primary and backup power. This innovation is well established by the powerco towercos of SSA and Southern Asia, but in many instances such towercos are only harvesting the low hanging fruit in energy efficiency, such as upgrading CDC batteries – unlocking the maximum energy efficiencies would require substantial investment in more capitally intensive renewables, which could be facilitated by partnerships with ESCOs, an option hitherto largely unexplored by towercos.
The second innovation cluster involves adding alternate site typologies to towerco catalogues. Towercos already offer the most familiar alternate site typologies by building new city poles or acquiring and leveraging existing street furniture. Many towercos already offer in building solutions, most frequently DAS – indeed in the US and UK towercos own many more DAS than MNOs. Early adopter towercos are already adding small cells to their inventory of site typologies – some functioning as site acquirers, with the small cells themselves owned by MNOs, while other towercos acquire and own their small cells – a segment which will be boosted as the first bona fide multi-operator small cells come to market.
The third cluster of innovations are towercos exploring diversification into provision of fibre for shared backhaul. Crown Castle has acquired over 26,500 miles of fibre, Protelindo and STP in Indonesia have both acquired fibrecos, while IHS has acquired a license to rollout fibre in one region of Nigeria. TowerXchange are aware of several towercos with as much appetite to acquire data centre as tower portfolios. Data centres are not a significant departure – complex hub sites have similar operational requirements to data centres. Nonetheless, exploration of both fibre and data centres remains confined to early adopter towercos at this time.
The evolution of site design. The era of the single tenant tower is drawing to a close. Even MNOs with no plans to divest their towers frequently build new sites with the structural capacity to accommodate multiple tenants to facilitate bi-lateral swaps, or commercial leasing of excess capacity.
While structures need to be more robust to accommodate multiple tenants, the wind load, and power load, on sites is generally declining as successive generations of radio technology are lighter and as indoor sites are outdoorised. Cell sites in the future network must be modular and scalable.
30m+ ground based towers make up a declining proportion of new sites deployed by towercos. Lower, lighter infill sites make up the majority of orders for many towercos, and an increasing number of these infill ‘city poles’ are smart towers, with a variety of applications, from EV charging to advertising and information screens. Don’t assume that all new telecom structures will be made from steel – the leading pan-Asian towerco edotco have been pioneers in exploring carbon fibre towers and even a rooftop structure built from bamboo! Many new cell sites don’t look like cell sites at all – JCDecaux Link aims to leverage over a million bus stops and billboards as potential connectivity sites.
The innovations in site design have only just begun. If 5G is going to need 10x as many sites as 4G, then wireless sites must be delivered for a tenth of today’s capital outlay. An era of mass production, or mass installation at least, of cell sites is approaching. An urban cell site needs to be installed in hours not days. And maintenance visits must be kept to an absolute minimum to address the challenge of spiralling opex costs.
When is a tower not a tower? Alphabet’s X (sister company to Google) Project Loon proposes a network of balloons over 100 km apart, which effectively function as a tower in the sky, capable of providing coverage to up to 5,000km². X reportedly has a queue of prospective MNO partners keen to deploy the innovation, and will soon have a viable commercial product. Meanwhile Facebook’s Project Acquila, which uses solar-powered drones, has also been piloted (without pilots!), while SpaceX are pushing a solution to provide high speed Internet from over 4,000 satellites. Will these innovations displace ground based sites as the de facto means of provisioning mobile coverage? Only time will tell, but innovators and incumbents alike are more inclined to label such innovations complimentary rather than competitive to the existing tower network, perhaps functioning to fix the economics of rural coverage.
Compiled from various latest research studies, drawing on TowerXchange’s unique proprietary research and data. TowerXchange is a research firm and community host focusing on the telecom tower industry.