Pace Group: When will towercos need Energy Services?

There are few people in the world who know more about planning and powering shared tower networks than Sairam Prasad, former CTO of Bharti Infratel, edotco group and WTTIL. Sairam also has many years of experience with mobile operators like Tata Communication (JV of TATA, Bell Canada International), Birla Tata AT&T and Idea Cellular. For the last couple of months, Sairam has been getting comfortable behind a new desk, as CEO of Pace Group’s Lineage Power. Sairam shared his vision of the energy services proposition, and explained why, where and when towercos will want to partner with energy services providers.

TowerXchange: Good to speak to you again and congratulations on your new role – please introduce us to Pace Group.

Pace Group is arranged into three verticals.

1) Our product-focused business Lineage Power is an acquisition from GE Power Electronics, which has good technology, systems, processes, people and manufacturing facilities. Thanks to this acquisition, we could significantly strengthen our product portfolio and offer end-to-end site solutions.

2) Pace Power, which is handling our services business, has scaled from nowhere in the O&M space in 2012 to provide comprehensive site O&M services to 36,500 towers as we speak. Services being manpower intensive, today we are proud to say we have a workforce of more than 600 for product services and more than 2,000 for O&M services.

3) Pace Renewables is our energy services business wherein we have invested capex with close to 500 solar sites on long term contracts under an opex model from Indus Towers and Vodafone. In addition to this, we are a “unique energy services provider in the telecom space offering end to end energy management solutions like solar hybrid, battery hybrid et cetera on capex, semi opex and full opex models” to a variety of customers like ATC, Bharti and Viom.

Pace Group has entered Myanmar and Africa with our products and services businesses. We entered the African market in 2011-12, when Airtel introduced a few selected Indian vendors with the right equipment capabilities into the Africa market for their rollout. Since then we strengthened our relationships with other customers as well. In 2014, we expanded our operations to Myanmar when telecom licenses were issued and towercos were being formed. I consider Myanmar to still be a virgin market with very high growth potential in years to come.

We now see good drivers in all three markets to scale up in to energy services.

TowerXchange: What has changed to make energy services more compelling?

When energy services were first trialled in India in 2009, it was difficult to drive to scale. There were no regulatory norms. Technology risk was not yet fully understood. Lithium-ion was at a high price point of over US$1,000 per kWh and the diesel price was very low – US$0.50 in India. Diesel costs have since exceeded to US$1, Lithium-ion now costs US$500-600 per kWh, the price of grid power has increased and the price of renewables have fallen. While solar previously costed US$1.50 per watt, it’s now around US$0.70 per watt. Tenancy ratios in India have increased from around 1.35 in 2008 when large towercos were formed, inching towards 2.0 today. This is increasing the electrical load on sites, and we know how technology can be scaled as load has increased further with the 3G overlay.

We are now reaching an inflection point toward energy services wherein all the aforementioned drivers are converging. According to some reports, Indian telcos are burning around two billion litres of diesel every year, prompting regulations to reduce carbon footprints with aggressive targets to be met by 2020, and with the first major milestone compliance scheduled this year.

Fortunately, the misconception that solar could achieve high penetration has been set aside. We now understand that the addressable market for solar in India is around 15% of sites. Many sites simply don’t have shadow free space for a solar array. Solar is better suited to off grid sites with outdoor equipment and load within a given band.

TowerXchange: What is the “given band” within which solar might be the optimum energy source? And is solar an option for multi-tenant sites?

Any load beyond 3kW is generally a challenge for solar. But one can accommodate one or two, or sometimes even three tenants with 3kW, depending on the type of equipment.

TowerXchange: You’ve emphasised the importance of the affordability of lithium-ion energy storage – how would you break down the balance of sites where VRLA is the best option compared to those where the extra investment in lithium-ion batteries can be justified?

Given that the price of lithium-ion may not fall much further, the main trade offs between VRLA and lithium-ion remain on price, charge discharge rates, cycle life, and power density.

The superior power density of lithium-ion enables light weight energy storage for the metro cells providing street level coverage on small poles. Metro cells and micro cells are becoming a market requirement as users migrate from voice to data and as new technologies increase speed, network planners need street level coverage in dense urban areas. You cannot put mammoth battery banks on concealed sites without compromising aesthetics and making those sites more difficult to permit. Everything has to be integrated into the pole – including limited energy storage. Metro cells typically have a 500W equipment load, and these are typically located in urban areas where the grid is generally good. Considering the scale required, and the need for perhaps two to four hours of backup power, the metro cell market will be a good fit for lithium-ion.

At good grid and medium quality grid macro sites, VRLA batteries remain a good option purely on a commercial basis, because of their price point and longevity when used in float. At off grid and poor grid sites, VRLA batteries need 8-12 hours to charge, which limits the DG runtime reduction that can be achieved. In comparison, lithium-ion batteries charge much faster, can cycle for multiple charges in a day, and still offer a good life.

In summary, VRLA often remains the best choice in good grid and medium grid environments. For metro cells, poor grid and off-grid cell sites lithium-ion may offer the best energy storage technology. To express that in terms of addressable markets, approximately 15% of India’s cell sites are off grid, 25% are on poor grids and the rest are medium and good grids. So the addressable market for lithium-ion is around 40% of India’s 350,000 macro towers (400,000+ sharable structures). And there are perhaps 50,000 sites that are candidates for solar power.

TowerXchange: Are the majority of India’s new build sites likely to be urban infill sites, like those metro cells you mentioned, or are there further rural network extensions to come?

The new build drivers are more focused on enabling data growth, rural penetration is more or less achieved in India. The USO has awarded some funds for rural builds, but most of the MNO capex is concentrated in urban areas.

TowerXchange: You were slightly less bullish about energy services when I last spoke to you as CTO of edotco group – was that just the cautious procurement strategy of a new company, or do you feel edotco’s footprint is currently less suited to energy services?

edotco currently operates in several markets where energy is a less immediate concern. 97% of Malaysia’s cell sites are on the grid – at the time I left, edotco had very few off grid sites in Malaysia. Similarly well over 90% of Sri Lanka’s cell sites are on grid. Microgrids are predominant in Cambodia, and those microgrids tend to burn a lot of diesel.

The most attractive market for energy services among edotco’s current footprint is Bangladesh, where around 15 to 20% of cell sites are off grid, and the grid is very unreliable. However, the towerco concept is still at nascent stages in the country and towercos are still focusing on operational efficiency before they progress to focus on energy efficiency.

TowerXchange: Given that the Bangladesh tower industry is too immature for energy services, please explain the phases of maturation until tower markets are ready for energy services.

I see the maturation of the tower industry as a five step process.

1. MNOs focus on their own rollout, retaining towers for as long as they are seen as a competitive differentiator.

2. Once the rollout is complete, the MNO optimises costs through infrastructure sharing, initially through bi-lateral swaps before divesting or carving out towers into a towerco.

3. The towerco’s initial priority is to optimise efficiency in network operations – implementing systems and procedures to improve network uptime. In the early stages for towercos, power and fuel costs often start as a pass through.

4. In an intermediate phase you might move to an indexed power pass through or semi fixed energy models like in India.

5. Eventually, often under pressure from their MNO tenants who want fixed cost energy contracts from towercos, the towerco starts exploring energy services, moving from proofs of concept to full scale pilots and then to a shift to the energy services model.

This maturation has already taken place in India. After the complete rollout, we saw the formation of several carve out and joint venture towercos. Power costs were initially passed through to the tenant, before the fixed cost energy contracts were proposed, leading to energy services pilots. Network uptimes and other operational efficiencies have been realised to the maximum level over last few years. Further optimisation in costs can come only through effective use of technology services. From this context, India is now ready to shift to the energy services business model.

Myanmar is also approaching readiness for energy services. The towercos have been in Myanmar since day one unlike other markets – the Myanmar tower industry effectively started from phase three being a late entrant to mobile services. Most of Myanmar’s towercos are private equity backed firms deploying the majority of their capital to build more towers required for coverage. Given that most of Myanmar’s towercos are new to this field, they have finite expertise in terms of optimising field operations and energy costs. However, Myanmar’s towercos remain subject to stringent SLAs on costs and uptime. Added to this, the grid situation in Myanmar tilts to more off grid and poor grid with very little good grid scenario. So the tower industry in Myanmar has every reason and incentive to embrace energy services faster than other markets.

TowerXchange: In a recent editorial “How to resolve the Mexican standoff on tower power in Myanmar”, TowerXchange noted that the incompatibility of Telenor and Ooredoo’s power strategy in phases one and two of the rollout compromised efficiencies. It now seems that phase three of the rollout will be led in a more co-ordinated manner, with Telenor issuing an RFP for over 1,100 sites with the power to be provided by the towerco and with Oordeoo to co-locate on many of those sites.

Ooredoo initially retained and managed power through their active equipment managed services partner, but seem to have realised that separate focus drew attention away from their core business. So while Ooredoo is looking for partners who can deliver good O&M and energy services for their sites from phases one and two of the rollout, it will be interesting to see how the new co-ordinated third phase of the rollout takes shape.

For example, there needs to be a commonality in lease price structures; at the moment Telenor benefits from more aggressive pricing with their towercos, whereas Ooredoo is paying a higher pricing but asking for higher scope. I feel Ooredoo may need to come closer to Telenor’s price. So the towercos may find it difficult to adapt to this change. There is also the complexity of identifying over 1,100 shared sites in a virgin market where both mobile operators are fighting fiercely for market share.

Despite these challenges, I feel that the co-ordinated rollout is a good move to achieve the objective to work together and speed up the rollout while managing sites effectively. This also helps to avoid duplication of towers in the nearby vicinity.

One of the challenges for energy services companies to reach scale in Myanmar is the fragmentation of the local ecosystem of O&M subcontractors. However, I feel the shift to an energy services model is inevitable in Myanmar because a high percentage of the cell sites are going to be off grid or on poor grids. Also, Myanmar’s operators already know the towerco model – in India, for example, Telenor has leveraged co-location extensively, so they know the benefits of setting stringent energy uptime SLAs – as knowledgeable customers, they may reject the power pass through model within the next two to three years.

TowerXchange: What’s your view of the maturation of the sub-Saharan tower market toward energy services, particularly in markets like Nigeria and Tanzania where towercos now own the majority of tower assets?

The proportion of cell sites that are off grid or on bad grids in SSA will increase with the rollouts over next six years. African cell sites have a long way to go to reach their energy efficiency potential. There are some renewable energy, and even a couple of energy services, pilots and trials. But most African MNOs and towercos have done little more than rollout CDC batteries – the first efficiencies are being created but there is still a long way to go before efficiency is optimised.

Nigeria is an interesting market as it is both SSA’s biggest market and has a high proportion of off and poor grid sites. Although Tanzania has a smaller footprint compared to Nigeria, both markets have good potential for energy services as both have tower companies in consolidation mode. Even in a small country like Burundi, with a high percentage of cell sites off grid, virtually the entire network could be converted to energy services.

It is simply not sustainable for African MNOs and towercos to spend over US$2-3,000 per month per site on O&M, power and fuel alone – that’s too high by any standards. So African MNOs and towercos are interested in energy services, but they want to see an African proof of concept to demonstrate execution excellence and a business case backed by the funds to scale the energy services company.

TowerXchange: What are the key ingredients of a credible energy services proposition?

There are four key ingredients of the energy services proposition.

1. Technical solutions: a credible energy services company needs to be able to install and maintain their own or third party products including solar, lithium-ion or VRLA batteries as appropriate, integrated power management, high efficiency, variable speed DC generators, and free cooling systems that can be seamlessly combined using a controller into at least four or five different solutions according to grid and load conditions.

2. Ground and field operations knowledge is critical to ensure execution excellence.

3. Solutions must be scalable and modular. One achieves scale by applying energy management to the whole network in clusters, rather than in scattered bits and pieces.

4. And the energy services company must have the financial capability to invest in capitally intensive technologies.

Pace Group is an aspiring energy services company. We have grown as a product company until 2012, since when we have made a strategic shift to provide and scale our O&M services. We have completed successful proof of concept pilots in energy services, and now we’re in a position to provide the aforementioned four key ingredients. The Lineage acquisition gives us a DC power solution, Pace Group has it’s own AC power products. We have partnerships for lithium-ion batteries, for RMS and solar with quality brands. And we have our own controller and network management system that can connect to other products to offer complete site solutions.

TowerXchange: One of your key ingredients was having the financial capability to invest in capitally intensive solutions, how does the blended capex requirement for energy services compare on a per site basis between India, Myanmar and SSA?

Obviously the capital investment requirement varies significantly according to the quality of grid available and the load on the site. For example, in India we might invest around US$10,000 to bring an off grid site up to standard for energy services, compared to US$5,000 for an on grid site.

There are two capex models for Myanmar: for the greenfield sites being rolled out over the next two to three years, if they were built for energy services from day one, the capital invested into energy equipment might be around US$25-30,000 per site, whereas adding solar or additional batteries to legacy sites might require an investment of US$5,000 to US$10,000 per site at a blended rate.

In contrast, the capital investment in an African green field site, where you may still need to buy a DG, could be as high as US$40,000 per site, whereas legacy sites might need an investment of under US$25,000.

TowerXchange: Given the substantial capex required to scale energy services, I must conclude by asking how Pace Group is financed?

We will be touching around US$ 100mn revenue this year on a consolidated level.  Pace Group today is a privately owned organisation with healthy blend of EBITDA. We have a good mix of products, solutions and services with visible individual growth drivers coupled with presence in three geographies. We are currently updating our vision and concluding our strategic action plan, then will look into our resource requirements, including investment, in the next quarter.