Why building rather than buying towers creates more value

Brazil Tower Company (BTC) has built over 300 towers, primarily in Northeast Brazil, and is ahead of their growth schedule, forecasting having a portfolio of 1,000 towers by year end 2015. In TowerXchange’s exclusive interview Dr Chahram Zolfaghari, BTC’s CEO, discusses building, buying and ultimately selling towers in the high demand Brazilian market.

TowerXchange: Please tell us about the origins of Brazil Tower Company (BTC).

Brazil Tower Company was conceived of in 2011 in Miami between myself and three lawyer friends of mine. We wrote our first business plan on a napkin, and named ourselves Brazil Tower Company because we wanted to do for Brazil what American Tower Company had done for the USA! We secured seed capital from 1848 Capital while my management team and I also put some money in. We added development finance from CAF to our shareholder structure, and have deployed close to US$50mn to date in equity, supplemented by a further US$70mn we’ve been able to raise in debt.

BTC is focused on build-to-suit. We are a developer. Like all the other Brazilian tower developers, we have an exit strategy to transfer assets, probably to one of the big three listed US towercos, American Tower, SBA Communications or Crown Castle.

TowerXchange: Talk to us about the thinking behind your capital raising to date.

Like any young business, we don’t want to dilute equity, so our first preference is raising debt. We have plenty of investors willing to buy more equity, but as the value of our business increases, it becomes cheaper to raise debt. With the collateral we’ll be able to put on the table in a year’s time, we should be able to refinance at half the interest rate we have today with the same bank.

Debt is always cheaper than equity. That said, our shareholders are very knowledgeable in the field – they understand the business, they have been involved from beginning in our strategy, and they are patient – there is no pressure to exit before the optimum time. We’re had three offers to buy BTC already, but we want to create more value first.

TowerXchange: What do you see as the relative merits of building rather than buying towers?

Given the premiums paid for Brazilian towers today, I’d rather build than buy towers. Besides which, there’s not much left to buy – Grupo Torresur is the only major portfolio left!

Let me explain BTC’s roadmap. We have 308 towers today. We had targeted building a total of 1,000 towers by the end of 2016, but now we’re on track to have 1,000 by the end of 2015! The volume of instructions we’ve received has increased, we’ve raised more money, and ramped up our team to get a year ahead of schedule.

We’re building our targeted 1,000 towers for an average of about US$100,000 per site. However, in order to acquire 1,000 existing sites we would have to pay at least two to three times as much as it costs to build. In addition to that capital outlay you have to invest improvement capex to reinforce towers, many of which carriers built for a single tenant, sometimes even just a single technology. Acquiring towers is further complicated by legal issues; land leases are not always readily transferrable. All these tower transactions take many months, even a full year to close because when the new owners get to a site, the owner might not be willing to transfer the lease to the towerco.

Under our build-to-suit business model, we do everything with a future exit strategy in mind. So from the get-go we’re building towers with structural capacity for four tenants. All our contracts are designed so that leases can easily be transferred to another entity – so a future buyer doesn’t have to talk to the owner for approval to transfer the lease.

Older towers need more maintenance – ours and the other build-to-suit towercos’ towers are all new – and new towers require minimal maintenance for the first two years.

TowerXchange: What quantities of new towers are Brazil’s operators investing in in 2015? And how much of that build-to-suit business is BTC going to win?

In 2015 TIM are forecasting the addition of 2,000 towers in 2015. Vivo will add 1,200 in their first batch alone. Claro are planning to deploy at least 4,000 new sites in 2015 through a combination of co-locations and new build out. Oi’s financial difficulties mean they are not investing in new towers currently.

The 1,000 sites BTC will have in our portfolio by the end of 2015 are all either already built, or are in the process of licensing or in construction. We’ve won RFPs for over 300 towers in 2015 with Vivo, TIM and Claro, so we know there are additional sites coming to us.

TowerXchange: Please put that into context for our readers – how many towers are there in Brazil and how many do there need to be to achieve Anatel’s coverage and QoS objectives?

There are around 70,000 towers in Brazil now. In order to meet coverage objectives (never mind optimising QoS), you need to multiply that number by four, but that’s not going to happen in the near term.

In a city like São Paulo you have one antenna per 15,000 subscribers, whereas in other major cities worldwide you might have one antenna for 4,000 – that illustrates the need.

Never mind the long term need for four times as many towers, Brazil doesn’t have the capacity to double the number of towers we have now. Between all the suppliers in Brazil today we can’t build that volume – there are not enough suppliers, and not enough resources – we can’t go out and buy 70,000 macro tower structures for example. So there’s still plenty of room for new market entrant towercos and suppliers in Brazil.

TowerXchange: Does BTC focus on macro towers, or have you looked at lamp-posts, DAS, small cells and other infill solutions?

Never say never, but BTC is going to focus only on macro towers, plus a few rooftops, in 2015.

Is there a need for more specialised small cells and DAS? Yes. But at BTC we have created a production machine to produce macro towers. It took time to build relationships and find suppliers, and now that machine is working full speed, we don’t want to get distracted by other activities, even if they are lucrative. We’ll focus on what we do the best.

I envision that at some point the market will start stagnating, macro tower rollout numbers won’t be as crazy as today. We always have to show growth, so at that time I’ll have to find new areas, and at that time we might move into indoor systems and outdoor DAS. There are several high density cities in Brazil – vertical cities –  which creates unique coverage and capacity requirements ideally suited to heterogeneous networks. The need is there today but there’s no-one really attending that need – macro sites remain carriers’ priorities, so if we speak again in 2016, I’ll be telling a different story! For now, the low hanging fruit is to build macro towers.

TowerXchange: How might the restructuring of the carrier market, and potential consolidation of operators, affect the tower market?

There are different rumors of consolidation every day! Will there be consolidation? Yes. There will be less carriers – the same thing happened in Europe and in the US. But in terms of timing, it’s not going to happen tomorrow. Oi is not deploying capex – so something needs to change there. But in Brazil the government will take a huge interest in and operator consolidation as these companies employ tens of thousands of people. With Brazil now in recession, the Brazilian government won’t make operator consolidation easy – it’s still a free market, but the Brazilian government will put obstacles in place to delay consolidation, or if the result is the dismantling of a carrier, they will do their utmost to ensure jobs are preserved.

There will be consolidation – the market can’t support all these carriers – but it’s not going to happen tomorrow, it will take a couple of years with the exception of Oi either “merging” with TIM or being purchased by a new entrant.

How will operator consolidation effect BTC?  Not much. We’re not doing any build to suit for Oi – that would be too high risk. We also do not build sites close to any existing site so even in case of consolidation the “new consolidated entity” will still need our site. So our exposure to operator consolidation risk is almost zero.

TowerXchange: What is BTC’s tenancy ratio?

Our tenancy ratio hasn’t increased from 1.3 since I spoke at the TowerXchange Meetup Americas in May 2014 because our portfolio keeps growing! We’re building towers even faster than we can attract new tenants!

TowerXchange: What proportion of your tenancy revenues derive from non-traditional carriers?

A lot of non-traditional carriers are our clients; they represent around 15% of our revenues. There are a lot of small companies in Brazil offering high speed broadband Internet and cable services. They tend to co-locate rather than building their own towers, and there is enough volume in this segment that it will not disappear in the event of consolidation.

The majority of our sites are in northeastern Brazil, which is less developed than the South. So the non-traditional carriers have less fixed infrastructure available so have to use wireless for high speed internet, cable TV, even fixed line communications.

Some of our competitors, particularly those focused on São Paulo, have a lower ratio of non-traditional tenants where the availability of fixed infrastructure makes the probability of securing non-traditional tenants lower.

However, all BTC’s anchor tenants are either Vivo, Claro or TIM. We don’t build-to-suit for non-traditional carriers because it could have an adverse effect on our credit rating – when your investment comes from outside Brazil it can be difficult to explain building for a carrier no-one has heard of.

TowerXchange: What effect will consolidation among Brazil’s towercos - the tower acquisitions by American Tower and SBA Communications - have on Brazil’s smaller tower developers?

The reality in Brazil is that everyone else has exit strategy to sell to American Tower and SBA Communications (or Crown Castle). For example, BR Towers’ people were telling everyone their objective was to grow to 10,000 sites, and three months later they were sold to American Tower! We know there will be consolidation among Brazil’s towercos - I wouldn’t even call this consolidation! We’re all in the market to sell!

TowerXchange: But does the growing inventory of towers marketed for co-location have any impact?

When American Tower and SBA are growing exponentially in Brazil, it’s not a threat, it’s an opportunity for companies like BTC.

A company like American Tower is not a competitor to BTC – we even share our plans with them. I’d be flattered if they considered us a competitor – BTC is a little drop in the ocean compared to their portfolio!

We do a simple analysis to ensure 100% of our sites are where there is no existing site for a couple of kilometres. You can download an excel spreadsheet showing our site locations from our website and you’ll see that there is no overlap between our existing sites and those of American Tower and SBA. So we believe in creating value for a future sale by ensuring they’re not buying duplicate sites.

The funds that invest in American Tower, SBA and Crown Castle are seeking growth. With the US tower market almost saturated, Brazil is the most attractive of the sizeable international markets where they can expand.

TowerXchange: Do you think Crown Castle will join American Tower and SBA in Brazil eventually?

I think Crown Castle will be active in Brazil soon. As last market entrant they will have to pay a premium, just like in any other business.

TowerXchange: As the carriers sell most of their towers and divest much of their passive infrastructure building and maintenance capabilities, who will build the single tenant towers as networks extend into lower population density areas?

Good question! Like other towercos, BTC build sites with the potential to add another tenant within months for a second carrier who needs coverage at a similar location.

In rural locations where carriers are only going because Anatel insist, one solution would be if the carriers could accept a running partner to invest in both the tower and antenna under an opex and revenue sharing model. Other than that, no-one is going to accept building those single tenant sites.

TowerXchange: How would you sum up the state of the tower market in Brazil?

Three years ago when we started we felt like pioneers. We had to convince carriers of the merits of the build-to-suit model. Today that model is fully accepted by all Brazil’s major carriers. The mobile market is booming, and the need for new sites is tremendous. The current build-to-suit companies cannot keep up with demand – there is a huge gap in capacity, which means room for new entrants. I forsee the market will remain this buoyant for at least the next two to three years.