Coverage versus capacity

Mott MacDonald has extensive experience of studying and evaluating digital infrastructure and has advised on several emerging market tower deals. The consultancy was recently appointed by a potential investor to evaluate an opportunity with one of Myanmar’s towercos. TowerXchange spoke to Andy Grey, a telecommunications consultant in Mott MacDonald’s digital infrastructure team, to get a first-hand account of tower rollout in the country.

TowerXchange: Please introduce yourself, Mott MacDonald and your experience in Myanmar.

Mott MacDonald is a diverse £1.3 billion global management, engineering and development consultancy. We are one of the world’s largest employee-owned companies, with over 16,000 staff, 180 offices in nearly 50 countries and projects in 150. We were named Global Consultant of the Year 2014 and International Consultant of the Decade 2003-2013 by the UK’s New Civil Engineer/Association for Consultancy and Engineering and we were also named Global Technical Advisor of the Year 2014 by Europe’s Infrastructure Journal for the fifth time in six years.

I’ve been part of Mott MacDonald’s digital infrastructure team for 18 months and have worked on a number of tower deals. Overall we’ve been involved in over 30 tower deals across nearly 20 countries over the past few years.

Based on our widespread experience of analysing tower opportunities in similar developing markets, we were engaged by a potential investor that was considering investing into one of the Myanmar towercos. This would be their first investment in the country and into the tower sector.

TowerXchange: Having recently returned from studying the Myanmar tower rollout, please share some of your observations of the country’s under-developed transport infrastructure.

There are plenty of paved roads in Yangon and dual carriageways on some arterial routes, however congestion and a lack of effective traffic management systems means traffic moves slowly quite often, even outside of rush hour. The congestion isn’t helped by the infrequent circular train line or the fact that motorcycles have been banned from the road in Yangon, except for government officials. Locals told us that the increase in traffic has been notable over the last couple of years and is illustrative of increasing disposable income and lowering prices.

We visited a variety of locations during our work, including a site in the middle of a marketplace where locals had to collapse several stalls to get our 4x4 past. It must have taken us over half an hour to drive 200m to the tower!

Outside the big three cities and off the main routes roads quickly give way to dirt tracks and you see far less traffic. One of the sites, which serves a small community, was only accessible down a lengthy, narrow dirt road. A 4x4 was almost mandatory!

TowerXchange: As well as under-developed transport infrastructure, another logistical challenge to the rollout is the monsoon season which affects the second half of the calendar year.

It seems that the towercos have encountered problems in working during the rainy season and lessons were quickly learnt. For example, there are challenges in pouring concrete foundations which mean that these are being scheduled around the rainy season, although steel erection can continue provided the steel is delivered to site before the rain hits. There is also a requirement to build on raised platforms in areas where flooding is susceptible.

TowerXchange: It must be challenging to forecast demand in Myanmar with such a paucity of existing performance data and reliable third party sources.

There wasn’t even a recent census when rollout commenced. Although significant demand from cities was expected and formed the focus of the initial roll-out, network planners often deployed single cells in outlying areas to see what happened and learned on the job, in several cases having to roll-out cell-on-wheels to supplement demand in the short-term. The census has now been completed but the lack of detail means it has limited value beyond population statistics.

What isn’t always apparent from the limited data available is that Myanmar is dotted with hidden towns. When you are on the ground you might think you’re in the middle of nowhere but as you drive along a dirt track you come across houses and even villages that seem to go on forever hidden amongst the trees. These seemingly rural are in reality good-sized towns and in some cases operators told us that within a day of putting up a tower the capacity is maxed out.

Under another contract, Mott MacDonald is building a tower/tenancy demand model for Myanmar for which there is a lack of data and a lack of benchmarks. We normally model one new market entrant in relatively established markets, not several in a greenfield market. So we have had to adapt our model to modify inputs and anticipate different scenarios. We’ve had to rely on various data sources, from the US government’s night-time light levels to Google Maps, however each month more data is becoming available and we have a good grounding in terms of what’s been done so far.

TowerXchange: Most people we’ve spoken to have suggested that tower sharing has been minimal during the first two phases of the rollout, concentrated in the big three cities, Yangon, Mandalay and Napyidaw and along the Irrawaddy River. We understand this has much to do with the incompatibility of Telenor and Ooredoo’s power strategies. The coming third phase of rollout may start to push deeper into rural areas.

It depends on your definition of rural. Myanmar is a relatively rural country, with only a small percentage of the population living in urban areas. The next rollout phase will include a lot more of the smaller towns.

The first two phases of rollout were concentrated on the big three cities and the routes out of them. Ooredoo and Telenor have largely proceeded with separate orders to two different towercos each. However, I understand they are sharing site databases to meet demand for infill sites.

The grid is not very extensive in Myanmar. Where it is available, it can vary between 90 and 240v, so can be very unreliable. In the initial phases, Ooredoo sought a tower solution that didn’t include power, using its own solution instead, while Telenor’s did. This clearly introduced an additional challenge to sharing and it is not clear how this will be resolved. Ooredoo may seek to offload to an energy service company or towerco the energy solutions that they retained in the initial phase.

We believe that Myanmar’s fourth mobile network operator may be more likely to co-locate than build their own sites, although much depends on whether they can secure a financial partner.

TowerXchange: So what are the most pressing drivers, coverage or capacity?

A greenfield market creates pressures on deploying coverage, however there is also pressure on capacity, with a need for infill sites where capacity is quickly exceeded. Therefore there is a demand for infill sites and a need to continue rolling out the network, pulling the operators in both directions.

TowerXchange: What have been the main challenges around zoning and permitting in Myanmar?

There is inevitably an element of uncertainty about the enforcement of restrictions on the distances between towers, particularly in situations such as in Myanmar where a new Ministry has been created to regulate the industry.

However, I didn’t see any blatant examples of parallel infrastructure in Myanmar. I believe the restriction on urban permitting requires that sites be 160m apart.

Getting permit applications processed in a timely manner may have been a bottleneck early on in the rollout, however the authorities seem to be getting quicker at processing permits. The number of sites we studied which ran into issues with permitting wasn’t unusually high.

There might have been some initial concern about whether it would be possible to build towers on land adjacent to Myanmar’s many religious sites, yet that does not seem to have had an impact on the tower build. We actually visited a site under construction within the grounds of a pagoda. Ooredoo has had to be sensitive about dealing with some landlords and subscribers given their origins from a Muslim country. There have also been a few isolated instances where permits have been issued but the military has vetoed the sites. I recall one example of a site in proximity to a military base which led to an objection.

Identifying who holds the lease on a given parcel of land can also be a challenge. However, the bigger challenge is that local landlords are now catching on to the value of their property, particularly in city centres and demanding lease prices beyond operator expectations. Those expectations may have to be tempered as the market settles down.

Nevertheless, perhaps the biggest challenge has been that the towerco license applications have taken a remarkably long time to be processed. The towercos appear to have secured the required authorisations, yet it’s another illustration that with a new Ministry things can take time.

TowerXchange: How are the citizens of Myanmar responding to the rollout - what anecdotes can you share illustrating demand for and take up of mobile devices?

Affordability had been a concern. However the price of SIMs has plummeted from a reported US$200 to US$2 and we’ve consistently found that citizens are prepared to spend a larger percentage of their income on mobile than Westerners are used to. There has been voracious demand for mobile and, particularly in the big cities, technology has skipped a couple of generations and people have moved directly to smartphones.

It will be interesting to see how Ooredoo’s 3G focused play is received in rural areas where mobile may be more focused on necessity as 2G could provide all the required functionality.

One thing everyone who visits Yangon notices is that everywhere is plastered with Telenor and Ooredoo adverts, with only a couple of adverts for the incumbent operator.

TowerXchange: Summing up your experiences, what do you think the future will be for the six towercos vying for business in Myanmar?

Current demand for towers promises continuing growth, however that may be confined to fewer towercos. We expect a degree of consolidation among Myanmar’s towercos, each has their exit strategies, which could be played out successfully. In the long-term we believe that Myanmar might sustain two towercos of scale, perhaps keeping one or two small local towercos to focus on more remote areas like Shan State.