If towercos are so great, why wouldn’t an MNO want their own one?

As we know, the Middle East and North Africa is the region least penetrated by the towerco business model globally, and decisions made by MNOs over the next few years will dictate who ends up as the ultimate owner of MENA’s 275,104 towers. Zain has opted to sell and leaseback its towers with IHS Towers, but the Saudi Telecom Company (STC) is exploring the creation of an STC-owned, but independently managed, towerco for the kingdom. In Europe, joint ventures like CTIL are common, with Vodafone and Telefónica sharing their assets since 2012. Our four panellists discuss the advantages and disadvantages of different tower ownership models and what the future holds for towers in MENA. Capitel’s Atirek Gupta chairs.

Rabin Sen has had over 30 years of experience in the telecoms and tower business with experience in India, Southeast Asia, South Asia and GCC countries and he is currently setting up a towerco for an operator in the region.

Nihat Narin is the CEO of Global Tower. Global Tower is the first and only standalone tower company in Turkey and was formed in 2006 by Turkcell. With Turkcell as an anchor tenant, Global Tower is also leasing space to other mobile operators, TV and radio broadcasters, public institutions and other service providers, including tower sites, rooftops and in-building systems.

Chuck Green is a non-Executive Director at edotco, Co-Founder of Helios Towers and has experience of independent towercos, MNO-captive towercos with minority-MNO ownership and majority-MNO ownership.

Finally we had Marc Perusat who is Managing Partner of TASC Towers. TASC towers has developed a modest tower portfolio in Jordan but has been involved in potential tower sales in the Middle East for a number of years.

If TowerXchange has an editorial evangelical objective, it is to promote that independent towercos offer enormous value to the telecoms industry and regional economies generally. But we don’t align ourselves with any belief that one towerco business model is inherently superior to another. Despite being advocates for the independent towerco model, we agree with our panel that there is no one right model for tower sales or carve outs, and that the economic and political case will vary from market to market and from operator to operator. 

Chuck Green pointed to Crown Castle’s original sale and leaseback deal with Verizon, where Verizon retained a minority equity stake in their towers. So even back to the early days of the industry there has always been a balance between independent operations and MNO’s retaining an interest in their towers. The same was true for Helios Towers in their early sale and leasebacks with Vodacom and Millicom, where both MNOs retained minority interests. So it is little wonder that MNOs examining tower sales or carve outs are exploring options for retaining an interest in their towers in MENA. 

The debate is ongoing at STC about the future of their network. The Saudi government wants to rapidly deploy 5G and that will require new site designs and typographies, with an optimisation of the country’s network. Although MENA MNOs are not as capital constrained as they are in Europe, Saudi Arabia needs to manage its capex for 5G roll-out and network densification and find a way to reduce opex in a network with significant operational challenges. Therefore a dedicated tower manager makes sense, but the debate over the ownership structure of that towerco remains live.

There is a contradiction between MNOs’ desire to professionalise, commercialise and optimise infrastructure sharing and their desire to still own towers after the sale. MNOs can be envious of the substantial gross margins towercos enjoy, but the question laid down by our panel was not whether these margins were justified, but whether towercos were adding value for MNOs, and the simple response was that towercos do add value in ways which are elusive for MNOs.

The financial pressures MNOs face are real. Competitive pressure on MNOs is coming from new entrants, ARPU remains under pressure, which has pushed their EBITDA down. At the same time regulators are insisting in increased investments to meet coverage targets. That in turn pushes up capex even while the spectrum required for 5G remains expensive. This leads some MNOs to trade at multiples in the region of 4-5x, while American Tower trades at a multiple north of 20x.

However, MNOs are still well placed to capture consumer demand for digitalisation and for that MNOs need to transition from asset owners to digital service providers. Towercos can add value by taking those passive assets off their balance sheet and by managing them better, and through lease up, add value to the industry. But because MNOs are so competitive and must focus on digitalisation, not tower management, that value add is harder for them to capture. There is a risk that MNOs trying to latch onto the towerco business will kill the goose that lays the golden egg.

Broadly speaking MNOs are not good at sharing their infrastructure, which is one of the principle ways towercos have been able to add value. However, MNO ownership itself has never really been the problem, it has been MNO control and the dynamics of competitive markets which held back infrastructure sharing. With the right governance, our panel agreed there was relatively little preventing an MNO-majority owned towerco from acting independently and delivering the benefits of infrastructure sharing to the industry and the benefits of towerco ownership to an MNO’s shareholders.

edotco is majority owned by the Axiata Group, a Malaysian telecoms conglomerate, but Axiata only takes two of the ten seats on board. At edotco there are limits on what majority shareholders can see, as towercos are privy to confidential information which could impact competitor plans and Axiata’s own planning decisions. There is also a Related Parties Transactions Committee which helps to keep things clean and inspects transactions which could suffer from any conflicts of interests. These additional governance structures can become cumbersome, but also show that risk of interference can be overcome. Each towerco and MNO needs to find a model that works, and it was suggested that STC might ultimately be following this model.

Turkey’s Global Tower owns 11-12,000 towers in Turkey, Ukraine, Cypress and Belarus and has increased its top line by 2.5x in last years, it’s EBIDTA is up 2.7x and its tenancy ratio now sites at 1.65x, but it is 100% MNO-owned by Turkcell. It is a completely captive-MNO but has an economic performance to rival many wholly independent towercos. The key is, just like at edotco, good governance, with a focus on communicating the company’s commercial mind-set to customers. They work with VEON, Vodafone, Telenor and Turkcell, and none of these would be happy if the reality of tower management wasn’t creating value for each tenant.

There is a transition from captive-towerco to independent-but-still-captive-towerco. Initially any carve-out is single tenant, or is at best limited to sites previously shared on an MNO to MNO basis. From a practical level that means that even with the best of intentions a captive towerco will lack the institutional memory of acting like an independent towerco and it will lack the operational experience of working with second and third (and fourth) tenants. But with time this will change and a willing captive towerco can upskill quickly.

However, unlike independent towercos, a captive-MNO may be predisposed to lean towards the preferences of their owner MNO. If the towerco is well managed these differences will be minor, our panel offered the example of a difference of preference over cabinet temperature: if the owning MNO prefers a different cabinet temperature to another tenant then the owning MNO may prevail. That will have operational knock-on effects and reduce the flexibility of the towerco. But the overall effect should be hard to detect.

Chuck Green discussed the valuation differentials between captive towercos, independent towercos, and best-in-class towercos like American Tower. Few companies enjoy the 20x+ multiple that American Tower enjoys, but many independent towercos still trade at more than 10x earnings. The captive towercos are typically valued around 7x which suggests financial markets and investors recognise that some value is destroyed by keeping the towerco captive, for example spending on extra governance is essential, but still increases the cost of doing business or missing out on potential tenancies because your governance quality is not easily marketable.

Another potential financial downside to captivity is access to capital. Of course, a well-capitalised towerco with opportunities for lease up, good quality assets, good governance in place, but MNO-owned will still have access to different capital to the parent MNO. But the capital structure will be limited by what the MNO is comfortable with or capable of carrying on its own balance sheet. edotco is levered at a ratio of 2.5x because of limitations imposed by its ownership by Axiata, but an optimal leverage level for a towerco is generally thought to be about 4x.

Especially in markets where towercos are in deployment mode, limitations of leverage will negatively affect the scope of strategy. Marc Perusat said that in extremis a towerco might not be able to access capital if its owner had a weak balance sheet. But in Saudi Arabia this is unlikely to be a problem because STC’s sister company Saudi Aramco enjoys a net income of US$111bn per year after having paid taxes and dividends to the Saudi state to the tune of US$159bn.

Our panel agreed that the future of communications infrastructure will include more sharing, more wholesale models and more neutral host networks. For that reason it is probably both tempting and counterproductive for MNOs to want to play in these markets. It is tempting because the potential returns look substantial, but it could be counterproductive because it is harder for MNO-captive infracos to act in an independent way. Already in North America you see Crown Castle investing in fibre, likewise, in Europe and North America Digital Bridge are betting big on digital infrastructure like data centres too. Nihat Narin argued that in five to ten years all towercos will become multi-asset infracos, operating more infrastructure than just vertical real estate. As these new, diversified infracos develop, MNO-captive towercos will face difficult decisions and will likely become more independent as they also seek to broaden their service propositions and asset portfolios.