Proof of Concept

Chuck Green needs no introduction. The former Crown Castle CFO and now CEO of Helios Towers Africa was the pioneer behind the first sale and leaseback deal in Africa between Helios and Millicom in Ghana. Chuck looks back on the deal, shares valuable lessons learned, and gives his verdict on whether three towercos is too many for one market.

TowerXchange: Thanks Chuck for generously giving up your time to share some insights with TowerXchange readers. Let me start by asking what characteristics of the Ghanaian telecoms market meant it leant itself to infrastructure sharing?

As ever, the opportunity in Ghana came about as a result of decisions made by a specific operator– the initial impetus to share infrastructure is not only driven by the attractiveness of the market in every case. One or more operators have to take the initiative.

Ghana represented a clear opportunity for us in Africa. The country represents a big, relatively sophisticated market with mobile penetration at less than 60% (probably 40-45% adjusted for multiple SIM cards) and rising when we first assessed the opportunity in 2009-10. With five operators, reasonably high GDP per capita and personal disposable income, market conditions were favourable for infrastructure sharing. While there were, and are, other markets with similar characteristics on the continent, the first sale and leaseback in Africa took place in Ghana because Millicom took a decision to outsource tower assets to create shareholder value.

Millicom initiated a competitive process in 2009, and we signed the deal in January 2010 making Helios the first licensed independent tower company in the market. As is often the case, other operators quickly followed suit. MTN quickly entered into a marketing agreement with American Tower, as we aggressively leased up Millicom’s towers, and subsequently entered into a sale and leaseback deal with American Tower. Vodafone ran a lengthy process first looking at sale and leaseback, then a managed service deal structure. Ultimately, Vodafone Ghana agreed a managed service deal with Eaton Towers, who market and lease their portfolio (see the interview with the then CTO of Vodafone Ghana, Tony Dolton).

TowerXchange: What are you able to tell us about the final terms of the deal in Ghana?

We announced the acquisition of 750 towers in Ghana for $54m. Across three transactions with Millicom, with whom we subsequently did deals in Tanzania and DRC, we acquired just under two thirds of their towers, a total of 2,500. Millicom announced that the three deals created $400m of value for the group, inclusive of the $180m in cash we paid them, plus the value of their retained 40% stake in the venture.

TowerXchange: Can you help readers who are new to infrastructure sharing transactions understand a little about what they can and can’t tell from the details of these transactions?

You can calculate the headline price per tower – such as $72,000 per tower in the Helios/Millicom Ghana transaction and $235,000 per tower paid by American Tower for 2,000 of MTN’s towers in Ghana. However, TowerXchange readers should be wary of over-reliance on price per tower data which is meaningless without knowing the term of agreement and rental rate, which is commercially sensitive information. There should be a direct relationship between price per tower and rental rate, and you need to know if is it’s a conventional 10-12 year anchor lease tenure; any changes here also affect valuation.

American Towers’ price per tower suggests a higher anchor tenant rental rate, while ours reflects Millicom’s preference for achieving a lower cost structure over the release of stranded capital.

TowerXchange: What were Helios’ objectives in Ghana and to what extent have they been realized?

Ghana was our first step in a diversified strategy for building an independent tower company for Africa. Helios Investment Partners and I established the first independent tower company in Africa with the formation of our sister company, Helios Towers Nigeria in 2005. By the time we started working with Millicom in Ghana, we had almost five years experience building, operating and growing an independent tower company in what was at times a hostile commercial and operating environment in Nigeria. We wanted to maintain the first mover advantage and Ghana was an attractive market for the aforementioned reasons. To date our business in Ghana has performed at or above our expectations at the time we executed the deal.

TowerXchange: What were Millicom’s objectives during negotiations?

Millicom prioritized driving down and stabilizing operating expenditure, with some monetization of stranded capital. Furthermore, they were concerned with continuing to own a residual stake, a priority important enough that they were prepared to make compromises on up the front fee and rental rate.

There are five potential operator priorities motivating any prospective infrastructure sharing deal.

• Up front cash

• Minimization of operating expenditure

• Retention of stake

• Capital preservation

• Focus on core business

Each operator will rate each of those priorities differently. For example MTN took a minority stake in Ghana, paid a higher rental rate, and took more money out upfront. Millicom did the opposite. As buyers, we’re mostly indifferent to the balance of price per tower paid versus the rental rate and we’ll agree a deal anywhere along a “curve of indifference.”

Millicom understood the shareholder value they could create by sharing infrastructure in Africa. They wanted to be owner of a diversified tower company and exploit the favourable relative multiple arbitrage that can see tower company valuations at fifteen times EBITDA.

On the other hand, some operators resist minority stakes. Some are less motivated by long term monetization than by the need for cash to fund immediate rollouts.

TowerXchange: So if there’s a curve of indifference at the balance between price per tower and rental rate, is there a threshold of equity stake Helios Towers Africa are comfortable for operators to retain?

The integrity of the independent tower business model relies on an operator’s retained interests being less than 50%. The operator shareholder has minority shareholder rights with no access to competitor information. Operators are fiercely competitive and only share as a matter of necessity, so the independence of the tower company is sacrosanct.

If you look at the tower swaps operators do, they are usually just one-for-one barter exchanges, where they often don’t even share power and have no Service Level Agreements (SLAs) – they’re very cautious. Understandably, operators don’t want competitors having control over their network.

The Service Level Agreements between independent tower companies and their customers are tough. We’re required to deliver better quality of service, and that is our core business.

TowerXchange: As pioneers leading the way into Africa’s first sale and leaseback transaction, what were the key lessons learned you took from the experience in Ghana, particularly in terms of the transfer of assets post-deal close?

My experience in tower acquisitions around the world gave us a clear plan for the integration and transition of tower assets.

There are two key challenges to be overcome in the integration and migration of assets. The first is ensuring major conditions precedent are resolved expeditiously, the most important of which requires going to every landlord to negotiate the assignment of ground leases – the novation of financeable, sub-leasable terms. We had already put in place controls over the financial impact of those negotiations on the transaction. And of course we had to confirm building, environmental and civil aviation permits, which is always a condition of purchase for Helios.

The second challenge concerns due diligence and assessing the quality of what we’re buying. During due diligence we are given access to a sample of sites to determine the quality of assets, but inevitably there is a certain amount of risk that you can only quantify once you own all the sites and can complete engineering site surveys. We can then remediate sites to be structurally sound and stand up to our Health & Safety requirements, to meet the requirements of the SLA, and to be ready for co-locations. All those investments have to be built into the capital economics of transition.

While the towers we acquired in Ghana were a pretty good portfolio and had been well maintained, they weren’t all in the condition we would need to run for multiple operators.

The transition of assets we’re talking about here accounts for the period between the announcement of signing and the actual initial close. In Ghana, we closed more than the contractual minimum number of sites at initial closing, novating leases over the subsequent 12-18 months, to complete the seamless transition of ownership from Millicom to Helios Towers Africa.

TowerXchange: Interesting – so thinking back to before the deal closed, or indeed before any infrastructure sharing deal closes, how do the buyers and sellers agree on who carries the risk revealed during transition?

Competitive tension generally determines the risk sharing dynamics of any transaction.

A lot of the asset risk depends on the sample of towers we evaluate. We often take bigger samples than proposed by the seller, evaluating 10% or more of the sites. We try to pick the sites and cover a variety of different regions, but sellers offer few reps and warranties so due diligence is key to mitigating, or at least taking calculated, risks.

Ultimately, the seller expects a fairly standard set of terms to be agreed, relatively few of which are subject to significant negotiation. The auction process gets prospective buyers to compete and improve upon negotiable terms.

I feel the biggest risk is not in the tower portfolio itself, but in potential deviation in transaction terms from the proven independent tower company business model that creates value for all parties. Some of our competitors are more relaxed about certain terms than Helios. For example, overly generous termination clauses destroy value; a 90-day cancellation clause turns it from a 10-12 year deal into a three month deal, which kills the value for the tower company, even if there are penalties involved. There are better steps operators can take to retain flexibility and control.

TowerXchange: How reliable and extensive is grid power in Ghana?

Grid power is available for 18 to 20 hours per day on average across Ghana, and is of variable quality. This of course can have a huge impact on network availability (uptime) unless you have adequate reserves in terms of backup generators and batteries.

TowerXchange: Is the involvement of a towerco good news or bad news for energy service companies and equipment manufacturers? Infrastructure sharing deals may result in an incentive, or indeed a requirement, to invest in generator and battery upgrades et cetera, but towercos are canny buyers with a reputation for driving a tough deal…

Chuck Green, CEO, Helios Towers Africa: For energy service companies and equipment manufacturers, this is an opportunity. There is often deferred maintenance leading up to an asset transfer, but it’s mainly an opportunity because operators had been running sites for a single power user, whilst tower companies are designed to load more tenants. That means generators need to be enhanced, resized and upgraded. We’ve also invested in technologies such as deep cycle batteries that reduce generator run time and associated fuel costs.

Our SLAs are very strict; we can afford little downtime and few faults that require material time to deal with, yet we need to drive down opex. As such, I have to agree that tower companies must be canny buyers. We’ll tender maintenance contracts annually or bi-annually to be sure we’re getting a competitive cost structure, and of course we’re able to order in bulk. At Helios we have a centralised procurement process, and a Procurement Committee that includes members of the Board. It’s a disciplined process with lots of internal controls, and that drives a highly competitive spirit in negotiations with suppliers.

TowerXchange: Does the involvement of a towerco necessarily mean a consolidation of suppliers?

Not always. We want multiple suppliers, sometimes regionally focused, driven to operate sites economically. Ultimately it’s important that we don’t squeeze our suppliers so hard they can’t stay in business! We’re simply looking for business partners who can help us achieve those strict SLAs at a competitive cost.

TowerXchange: Can you share any success stories in terms of the opex savings Millicom realized in Ghana?

In each of the transactions with Millicom the anchor tenant rental rate agreed was below Millicom’s all-in opex cost so they secured an immediate improvement in cost structure. They also reduced uncertainty about their future tower operating cost structure, don’t have to worry about the vagaries of fuel theft and maintenance costs, and can focus on their core business. Helios is incentivized to invest in the reduction of power consumption, for example by installing Remote Monitoring Systems and using deep cycle batteries, and solar in some instances, as alternatives to diesel power. We’re also reducing the carbon footprint, so there is a positive environmental impact.

Currently, we are making a $40-50m investment in Remote Monitoring Systems and deep cycle batteries across our three existing markets. We install remote monitoring systems to track site conditions, such as access, fuel delivery, generator run time, utility availability, etc. This generates efficiencies in controlling the fuel supply chain. We’re replacing station batteries with deep cycle batteries so we can get more hours of battery support, reduce generator run time and reduce the risk of downtime. Such a strategy makes sense only where we are managing DC power, so that is an important consideration for tower companies and sellers to agree.

TowerXchange: Is Ghana an urban fill-in driven play, or a rural network extension focused market?

Ghana follows a fairly usual pattern: there are still deficiencies in 2G voice coverage, and there has been an initial deployment of 3G in urban centres. There has been activity to extend rural coverage, albeit modest to date.

General coverage is still a big issue in Ghana. There were lots of co-locations after the tower companies entered the market in 2010-11, but few builds. The network planning priorities were largely motivated by coverage, with some 3G rollouts.

TowerXchange: To what extent is capacity a concern?

Operators’ concerns about capacity are accentuated when regulators and ministers complain about quality of service! With capacity for data, and even for 2G voice, restricted, regulators are increasingly applying quality of service assessments across Africa. There will always be pressure from regulators to push forward with the expanded coverage requirements within operators’ licences, and to improve quality of service generally, which means operators will need more in-fill sites for 2G, let alone 3G and, eventually, 4G LTE.

The continuing growth of voice and data traffic puts a strain on networks, which needs to be improved and expanded. Towercos are there to preserve capital, expand networks and satisfy quality of service requirements without burdening the balance sheets of operators.

TowerXchange: Are towercos paying a premium for first mover advantage in Africa?

Chuck Green, CEO, Helios Towers Africa: The strategic interests of four competing tower companies have played to benefit of sellers. But we’re happy with our investments in Africa, and we’re achieving our objectives.

The proliferation of tower companies may not be sustainable over time. There are only so many tower companies can operate in any given market.

TowerXchange: Based on your experiences in Ghana, would it be difficult for many individual markets to sustain more than two towercos?

The larger markets can probably tolerate two tower companies but three is not ideal.

Having said that, demand in Ghana has been significant enough to spread among three towercos with different footprints. We have the market share we targeted and we’ve already had a reasonably decent lease up experience but we never expected to be the only player in Ghana.

The independent tower company model lends itself to a common set of terms and pricing assumptions. We don’t have that much flexibility on price, and we haven’t experienced predatory pricing despite three towercos being in Ghana. There’s only a 30-40% overlap in footprints at outset, so direct competition only occurs when an operator wants to locate on a site right next to another tower company’s tower. Operators have a specific rollout plan, and generally whoever has the optimum site they need gets the tenancy.

Rollout demand from Airtel, Vodafone and to a certain extent MTN has been almost entirely satisfied to date by sharing with tower companies in Ghana. Over the last 30 months or so there have been very few builds, apart from Glo. Operators will generally favour the sharing or co-location option before they build – furthermore, the market has a self-adjusting governance: if independent tower companies charge too much for the co-location lease rate, operators would build rather than co-locating.

TowerXchange: So having three towercos active in Ghana slightly moderates upside, lowering the glass ceiling on the achievable tenancy ratio?

The achievable tenancy ratio might be 0.1-0.2 lower, but it’s not of a huge magnitude, especially given the required densification needed by the data demand associated with 3G, and subsequently LTE.

“In 2010, we made significant progress in the area of asset optimization and network efficiency, having come to the conclusion that owning passive infrastructure no longer confers a competitive advantage and that it makes sense, where possible to share tower networks with other operators with similar coverage to ours.

In 2010 we signed tower deals with Helios Towers Africa in Ghana, Tanzania and DRC through which we have committed to sell the majority of our towers to a Helios subsidiary company in each of these countries. We now have almost 2,500 or two thirds of our towers in Africa committed to be outsourced which enables us to focus our efforts on areas of real differentiation from our competitors, namely: sales, marketing, branding, distribution, service innovation and customer care.

We believe that the value created from these deals exceeds $400 million through a combination of more than $180 million of cash to be received for the sale of the towers, the 40% stake that we have in each tower company, expected future cost savings and significantly reduced capex for towers in the three countries. In 2011 we expect that these transactions will produce an additional percentage point of EBITDA in Africa, which could be reinvested in sales and marketing.”