African towercos’ strategies to date and plans for the future

This article summarises talking points from the keynote towerco CXO panel session at the TowerXchange Meetup Africa 2014, featuring Chuck Green, CEO of Helios Towers Africa; Terry Rhodes, then Co-founder and Director of Eaton Towers (now interim CEO); American Tower’s Africa CEO Pieter Nel; and Abhulime Ehiagwina, CFO of Helios Towers Nigeria. The panel was moderated by Marco Cardoni of Analysys Mason.

The dominoes have fallen in the sub-Saharan African tower industry; over the last four years independent towercos’ ownership of SSA towers has risen from 0% to 29%. Given that the towerco business model depends on the credit worthiness of the anchor tenants, and towercos have little appetite to acquire towers with overlapping coverage, Africa’s towercos do not want to acquire all the towers; 55% penetration might represent market saturation.

If Africa’s largest towercos have indeed acquired over half of the towers they targeted, that was reflected in the comments of the panelists. “We’re in five or six markets now, operating 7,800-8,300 towers in SSA – I hope we’re at least halfway!” Commented Chuck Green.

“We’ve got a long way to go,” said American Tower’s Pieter Nel. “We’re looking at every country and environment in terms of the competitive nature of operators, maturity of sharing and existing barter agreements, which informs our view of the lifecycle. The infrastructure sharing and towerco concept is so well established and embraced on the African continent that it bodes well for tenancy ratio growth potential and for organic growth. With lots of holes appearing in the network, there are lots of opportunities for build to fill and build to suit. For example, in Ghana we’re seeing 12-15% organic growth,” concluded American Tower’s African CEO.

There will be ongoing M&A, felt the panel, both from MNOs divesting towers and trade acquisitions.

“If you look across a broad horizon at the African telecoms industry,” commented Chuck Green. “Multiple SIMs exaggerates reported penetration – there is hidden growth potential. Unique subscriber penetration rates are still relatively low but rising rapidly. There is need for coverage expansion, while regulators are increasingly concerned about QoS. African mobile networks are operating with 4,000-10,000 subscribers per site compared to less than 1,000 in the US and Europe. So there is growth potential in both coverage and capacity. As disposable incomes increase, as smart phones get cheaper and more widely rolled out, growth in SSA will be on the same track as developed markets.”

“Africa’s towercos have established footprints strategically across the continent – the Airtel transaction has helped a lot in that respect. The liquidity events to which we’re all building, either strategic sales or IPOs, should attract multiples which will reflect the rate of growth,” concluded Green.

Eaton Towers’ Terry Rhodes added: “Penetration is already over 120% here in South Africa. Even with American Tower already here, Eaton saw a market opportunity to build towers in areas like Gauteng where we felt the networks could do with improvement. So we built towers focusing on capacity and infill, generating tenancy ratios of more than two. If we can achieve that in the richest region of the continent, we’re confident of the need for capacity and coverage elsewhere in SSA.”

“There are more similarities than differences between the SSA and US tower markets,” said Helios Towers Africa CEO Chuck Green. “The fundamental model is the same: robust tower sharing, long term recurring cash flows, high leverage – the same as in the US. The primary differences are that operations and maintenance is a commodity for developed markets, whereas in SSA there’s more exposure to power and security, and more forex exposure. It’s about becoming more efficient – doing everything we can to reduce reliance on DG runtime; converting from AC to DC power so we have more control over power systems. So whereas in the US operating leverage is limited to putting co-locations on towers, in Africa we also have opportunity to increase efficiency at an opex level.”

American Tower’s Pieter Nel added: “it’s a very simple concept: lease up the towers and everyone is happy – we get scale, the MNOs get efficiencies, savings can be passed to the consumer. American Tower are in 12 countries on five continents. In other markets such as the US and Latin America, many costs are passed through to the tenant – energy, ground rent, maintenance. The business model is more complex in Africa, but at least we own it and control it.”

“We see the same growth in consumer demand in Africa as anywhere else in the world,” added Eaton’s Terry Rhodes. “Give an African consumer a smart phone she’ll use it. The penetration curve may be behind, but we still see a big wave of data growth – operators are going need every bit of help we can give them.”

Abhulime Ehiagwina, CFO of Helios Towers Nigeria added: “infrastructure is not as mature in Africa, so you have to devote time to look at innovative power solutions. You have to ensure backups are up to date. You have to follow through your contracts. You have to ensure the sites are always up.”

Tower companies’ well-informed counterparts at African MNOs have learned a lot from how the tower business works overseas, so in Africa standard equipment configurations are usually bigger than in the US, giving the towerco less flexibility. A menu pricing formula is typically used, with the addition of 3G and 4G antenna often accounted for. Once the anchor tenant’s requirement is outside that standard equipment “bucket”, the towerco has a similar opportunity for growth. And of course there is the same opportunity to lease extra space to other tenants. Towercos still get significant additional revenue from additional equipment in SSA, and they don’t count that in their tenancy ratios, but it does of course contribute to Tower Cash Flow, which is a more inclusive metric to measure towerco performance.

When their process started, apparently Airtel considered keeping 26% equity in each market, but so far African towercos have confirmed 100% ownership of Airtel towers in 13 markets. “If it’s a good business, we want to own it, and make the independence as clear as possible,” said Terry Rhodes.

The other panelists shared the view that SLB is the preferred model. Under a SLB agreement, a minority interest can be retained by MNOs. As long as that remains below 49% and is a passive investment to protect the independence notion of this model, investors are comfortable. Anchor tenants retaining more than 50% equity can impact the perceived value to investors, as there is a risk that the model loses it’s notion of independence, making other operators reluctant to share. So there should be a significant discount for that kind of deal structure. In fairness it should be noted that IHS, who recently announced a tower transaction in which 51% of the equity was retained by anchor tenant MTN Nigeria, had to withdraw from this panel at the last minute, and thus were not present to make the case for their innovative deal structure – doubtless they would have asserted that the independence of the towerco was enshrined in the terms of the agreement.

American Tower tends to stay away from MLL deals unless there’s a good strategic reason, feeling they create unnecessary complexity. Initial concerns were swiftly assuaged about how market would react to their joint venture deal structures in Ghana and Uganda – the market understood that their MNO counterparty was a passive investor.

African towercos’ appetites for managed services are declining – MLL and managed services represented 31% of African towerco’s portfolios at the end of 2013, that had reduced to 10% by the end of 2015 as a function of the acquisition of thousands of managed towers, the cancellation of MLL and Managed Services contracts in Kenya and South Sudan respectively, and the dilution of the MLL share of market in a year where all the new deals were SLBs.

One participant outlined their reservations about Managed Services and MLL deal structures. He felt Managed Services was not seen as a great business relative to the towerco model – margins are relatively skinny, early termination provisions often have to be made flexible, which further reduces the value perception by the market. If the MNO can buy back the customers it also degrades value. In summary, the valuation of a Managed Services or MLL deal depends on the term, the termination provision, who owns the co-locations, who is responsible for investing capex, and the pricing of the deal.

Another towerco defended the MLL deal structure, explaining that it was critical to help investors understand what could happen after the term of the agreement. Tower Cash Flow is what delivers the value, and excellent TCF growth can still be created under an MLL deal, suggested the same panellist.

The TowerXchange Meetup Africa 2014 took place after the announcement of IHS’s acquisition of 9,151 and 2,136 Nigerian towers from MTN and Etisalat respectively, but shortly before American Tower announced their acquisition of 4,800 towers from Airtel Nigeria. Understandably, American Tower weren’t at liberty to contribute to this section of the discussion.

Abhulime Ehiagwina, CFO of Helios Towers Nigeria, said “what we’re looking at in terms of demand for capacity now in Nigeria is tip of iceberg. There is congestion at base stations across networks. It’s almost impossible to stream a YouTube video on a smartphone in Nigeria during peak traffic hours, and with smartphones for under US$100 coming onto the market, the number of subscribers who can afford mobile broadband is increasing. Nigeria currently has a little over 24,000 towers, yet the Ministry of Communications believe we should have at least 60,000.”

The panel seemed to agree that with four substantial MNOs in Nigeria, there was room for two large towercos and a handful of niche players – although there was speculation that a lot of smaller towercos would get picked up.

“I don’t think there’s a one size fits all concept – there are quite a few of us in Ghana, for example,” said one of our panelists. “Being second to market is workable, but if you’re the third to arrive and you don’t have unique locations, you’re in trouble.”

“There was something of a mad rush in Ghana where the first three operators to outsource ended up with three different towercos,” said another. “However, in Ghana we’re on a tenancy ratio right where we expected to be in our acquisition economics. We were realistic about the fact that there’s some overlap. Ghana should probably have no more than two towercos. Countries with 30-40mn population can tolerate two towercos, but with the exception of negative FX performance, Ghana has been fine.”

“Ghana was first market to liberalise, and there were six MNOs, plus 4G new entrants,” said another panellist. “Would we all go in again now? Probably not, but there’s just so much more opportunity now. Our business is also doing well in Ghana, although the currency has had more impact than the competition. The only competition is where towercos have overlapping locations, and that’s in 20-40% of cases at the most. Whether it’s Ghana or another market in SSA, if we can meet service level agreements at a price better than it costs MNOs to build towers themselves, then the model works. It’s that simple.”

“For us it’s all about optimum return on capital invested (ROCI) – making selective decisions on capex to get optimum value for ourselves and our customers,” added American Tower’s Pieter Nel.

“After a deal closes, there follows a natural cascade of capex priorities,” said Chuck Green. “First we shore up towers that are already overloaded, and make improvements on to comply with our Health and Safety standards. Then we look at strengthening towers for co-location and investing in opex reduction. In reality this all happens concurrently.”

“When you acquire a portfolio, some of it’s older, some of it is newer and in better condition,” added Eaton’s Terry Rhodes. “After you’ve made sure everything is safe, after that it’s about individual tower profitability. If we can improve cash flow by spending capex, we’ll do it. If capex improves EBITDA it looks good to the outside world, just like in any industry.”

Of course, it takes a while for a towerco to close a deal then get their hands around the state of the portfolio before it’s brought up to the operating performance they’d like to achieve.

“In Nigeria one of our principal challenges is that we get only four hours per day of usable grid power,” added Abhulime Ehiagwina. “The government has privatised the grid, so looking ahead, our improvement capex budgets will depend on how quickly power reforms improve uptime from the grid. If we’ve moved from four to 18 hours of good grid, capex will decline, and we’ll focus more on upgrades for co-locations.”

Q&A is when the panel sessions at the TowerXchange Meetup really come alive, and this one was no exception. The first question came from Mecc Alte, who wanted to advice on how to proposition towercos with innovations to reduce energy opex.

“All the tower companies are interested in schemes that improve efficiency and reduce costs. We see a lot of good looking proposals – but we need to see a proof of concept,” said Abhulime Ehiagwe of HTN. “Pilot it at your own cost to prove it works and delivers value. Successful pilot schemes are easier to ramp up to scale.”

Helios Towers Africa’s Chuck Green added: “As this industry has emerged from the aggressive land grab phase to where we are now integrating towers, we’re now more open to innovations. Our natural question are: is it fit for Africa? Will it work in the field? Can it be maintained easily enough?”

“The lifecycle of any innovation is going to be important,” added Terry Rhodes. “We’ve recruited experienced tower people from India – our COO comes from American Tower in India – what he wants to see is a solution that works well there and comes at an Indian cost base, then we’re interested.”

Pieter Nel echoed the same sentiment: “As our operations reach maturity, there is more time to spend running and improving networks. Come with a reference site and/or proof of concept. Reducing carbon footprints is absolutely a priority,” continued Pieter, “but this can be complicated such that every site needs it’s own design.”

“The consideration of renewables is on the same path as the economic and security situation in terms of reducing DG runtime,” added Chuck. “When we’re looking at alternatives to DG – CDC batteries, solar, wind et cetera - it’s driven by economic reality than by being green per se. We have framework for finding situations where it makes economic sense.”

“Reducing energy opex has an important administrative / security aspect to combat shrinkage – it’s more about controls than technology,” added Terry Rhodes. “We’re interested in capex models as long as the African technician on the ground is capable of making it working. We’ve seen too many complex solutions – it’s got be simple, robust and economical.”

With regard to the potential of ESCO or powerco business models, the towerco business leaders seemed more enthusiastic than at the TowerXchange Meetup 2013. However Pieter Nel voiced their primary continuing concern “it’s not about the technology used to deliver power as a service – it’s all about operational execution. It’s so important that you understand what it’s going to take to successfully execute.”

Chuck Green continued the perspective: “Helios Towers Africa have considered ESCO type models,” said Chuck. “We have a direct responsibility to our tenants for uptime, governed by strict SLAs and penalties apply. So far we’ve not been comfortable with counterparty credit and capacity such that we’re not prepared to delegate the heartbeat of the service we provide to our customers. It doesn’t matter if the party we’re outsourcing to takes on penalties, what matters is that SLAs are met.”

The separation of power and non-power costs, and packages inclusive of fixed energy prices per site are commonplace now in SSA, but an evolution to towercos buying and selling energy by the kWh may be on the horizon.

However, towercos’ first line of offence in the battle against energy opex is always going to be to connect sites to the grid, where it makes economic sense. That’s tough in Nigeria with its four hours of usable grid power, but it’s a more attractive option in a market like Tanzania where cell sites on the grid get perhaps 19-20 usable hours. Towercos back off from that grid connection preference, looking next at CDC batteries, then solar and other renewable sources according to local resource availability to reduce DG runtime.

“There’s a misconception that American Tower are not responsible for power equipment in SSA,” said Pieter Nel. “Many African MNOs expect service at DC level. We’re providing DC service level in Uganda, and migrating to a DC service level in Ghana. In South Africa, energy remains a pass through cost. We’re embracing owning and delivering power, and have achieved some great successes. For example, in Uganda we’ve been able to reduce fuel consumed per site by >30% in 24 months. Power is now a core competency of American Tower in Africa.”

During later round table discussions, we learned that American Tower typically invests US$10-30,000 per site to migrate to DC service provision.

“Growth has to come with ancillary extensions along the value chain from where we are now as passive equipment providers for largely 2G operators,” said Chuck Green. “We are in the early stages of rolling out IBS in developed residential and commercial locations in Ghana and Tanzania, a precursor to small cells. As broadband take up increases, DAS and IBS will become increasingly important in Africa’s major city centres.”

While active infrastructure sharing is more in the interests of MNOs than towercos, a lot of regulators don’t allow it, and it tends to become an option for more mature telecoms markets than SSA. The current levels of congestion across Africa’s networks means the opportunity for and economics of active infrastructure sharing remain quite limited. While in early towerco contracts, addressing potential future active infrastructure sharing was too challenging, Africa’s leading towercos now have preliminary pricing in the event of active sharing built into contracts. In that sense it’s not that different from sharing the benefits of energy reduction. While the towerco CXOs don’t see active infrastructure sharing as a near-term issue in Africa, they are anticipating it becoming a factor over the ten-plus year lifecycle of contracts.

While the towerco leaders were obviously aware of satellite providers seeking to compete with land based cellular, and Google trying to get into the business using balloons, the prevailing view was that such approaches could change the economics of rural coverage, “but right now we’re happy being in tower business rather than the balloon business!” Concluded one of the towerco executives.

Pieter Nel, American Tower: “As much as providing energy is a core competency of American Tower and other African towercos, there is significant opportunity for suppliers to come up with a model for selling power as a service. If we get a credible consortium or partner offering a complete service with SLAs enabling us to focus on our core business of leasing up our towers, we’ll be interested.”

Abhulime Ehiagwina: “Governments need to harmonise the different layers of taxes under a unified tax scheme. MNOs have robust regulatory teams, the tower industry needs to build up to those levels.”

Chuck Green: “Until and unless there is universally reliable grid we all have to deal with power and security issues. We’re all searching for a silver bullet giving us reliable, easy to maintain, fit for Africa solutions so we can monitor in real time what’s going on at our sites…. And talent development is a huge issue across the continent from the top management team to the field level staff we employ.”

Terry Rhodes: “There are 400-500mn Africans without communications. We must help to address that gap, improve penetration and extend networks to where it never made sense for one company to go alone.”

The MENA market remains a relatively fallow field for towercos, although Zain recently admitted they are currently reviewing their tower strategy. The imminent Egyptian tower sale will be a bellwether for the tower industry and their prospective MNO counterparts across the region. However, the structure of any tower market in MENA is likely to be different from SSA, consisting as the region does of quasi-rural developing markets and prosperous developed markets in which leveraged buyouts might be a more common deal structure than sale and leaseback. In the near-term, the PE-backed towercos don’t see as many opportunities in MENA with the promise to deliver returns comparable to those they see in SSA. “Ask us again next year and we’ll let you know!” Said one of the panelists, whose towerco is rumored to be in pole position to secure MobiNil’s Egyptian towers.

IHS smashed through their five year target to acquire 20,000 towers in SSA in just 30 months. With the company’s recent addition of the Airtel towers to those already acquired from MTN in Rwanda and Zambia, as well as IHS’s acquisitions from MTN and Etisalat in Nigeria, where IHS now owns 14,222 or 56% of Nigeria’s towers, IHS has reaffirmed their position as Africa’s largest independent towerco, with over 21,000 towers.

IHS has always been a Nigeria-centric West African tower play, with country risk diversification provided by their acquisitions in Cameroon, Cote d’Ivoire, Rwanda and Zambia. With ~80% of Nigeria’s towers now transferred from operator-captive to independent towercos, and IHS’s initial growth plan achieved, IHS may be refocusing on driving toward profitability and a future IPO.

That’s not to say IHS wouldn’t look at further opportunities that meet their investment criteria. IHS’s original wish list of targeted countries included several countries in North, East, Central and Southern Africa, and they’re believed to have looked at opportunities as distant as Afghanistan. The current Sonatel process in Senegal, Mali, Guinea Bissau and Guinea Conakry may be of interest to IHS as an extension of their West African footprint and relationship with Orange. If and when the North African tower market opens, we’d expect IHS to be bidders. And it would be interesting to see whether they would go toe-to-toe with American Tower should an opportunity of scale come to market in South Africa – whether the current Telkom process triggers such a showdown remains to be seen.

However, elsewhere in SSA, IHS are not believed to have bid for the Airtel towers acquired by Helios Towers Africa, which reinforced their strongholds in Central and East Africa, nor for the Airtel towers which added to Eaton Towers’ footprint in both East and West Africa. With the towerco business model best suited to a ‘natural monopoly’, IHS may be disincentivised and disinclined to enter previously targeted markets where competitors are now active, unless they have an appetite to try to acquire one of the other PE-backed members of Africa’s ‘Big Four’ towercos.

IHS has raised record-breaking capital; over US$4.5bn since 2012, the largest equity raise in Africa since 2007. In seven tower transactions to date, IHS has deployed ~US$2bn, with initial acquisition expenditure phasing into improvement capex programmes and efficiency programmes in their more mature markets. IHS has funds for more acquisitions, and their enthusiastic backers, led by French family fund Wendel Group, have been only too happy to re-invest and to help add new members to the IHS’s growing tribe of PE and sovereign wealth fund backers.

In terms of procurement strategy, IHS are believed to be considering moving from a conventional capex programme to an opex-oriented procurement strategy in which they simply pay for energy by the kWh – a move which may be mirrored by other African towercos, as hinted at the TowerXchange Meetup Africa 2014.

In conclusion, IHS has achieved scale in SSA. After adding almost 13,000 towers and almost trebling the size of the company in 2014, we think IHS’s acquisitive growth may slow as the pipeline of transactions slows in SSA, refocusing their existing opcos on organic growth and driving TCF. But IHS will doubtless remain active and aggressive bidders for any further opportunities that meet their investment criteria.