We have spent the past five years researching and advising mobile network operators in India and Africa, and like many other industry observers we have followed Bharti Airtel Limited’s recent acquisition of the sub-Saharan African assets of Zain Group with great interest. The acquisition raises many questions, both about the impact that the entry of a major Indian mobile network operator (MNO) will have upon competitive dynamics on the African continent, and the business model that Bharti will choose for its African operations. Manoj Kohli, CEO and joint managing director of Bharti Airtel, once declared that the Indian model “is great for all emerging markets,” and in this article we explore the fundamental aspects of Airtel’s approach. We then turn to an analysis of the degree to which the Indian MNO model can (or should) be transferred to the sub-Saharan Africa context.
Up until the turn of the new millennium, the traditional view in the wireless industry had been based on the “ARPU Model,” where business viability was assessed primarily in terms of ARPU (average revenue per user) and cost per subscriber (combining acquisition and ongoing care costs). The entry of Reliance Infocomm to the Indian market in the early 2000s saw a totally different perspective emerge — Reliance introduced a “volume” perspective based on metrics such as net margin per minute, cost per minute, and capacity utilization on airtime capacity created. This intense focus on volume and margin was an ideal fit for the geographically large Indian market in which there were literally hundreds of millions of very low-income consumers awaiting mobile telephony, but was also accelerated by the industry regulatory regime which fostered the entry of multiple MNOs and the push towards operational efficiency as a key lever for competitive advantage. India’s mobile telecom sector has one of the lowest Herfindahl indices in the world, and the highly fragmented and competitive nature of the industry has resulted in hyper-competition that has only intensified in recent years.
The Indian “volume” model involves the aggressive management of capex and opex per minute, and aims to maximize the net margin per minute (also referred to as the AMPU model – average margin per user). Total usage, and therefore capacity utilization, have become critical operational parameters in the Indian market, and typically, the Indian MNOs have spent less on marketing when compared to operators in many other parts of the world. The focus has been upon maintaining a low tariff rate to attract and retain customers, and costs associated with joining (for example SIM card costs) have been designed so as to cover the costs of acquisition. The lower variable price for usage has driven higher minutes of use (MOU) per subscriber, with operators adopting strategies to ensure that the average capacity utilization is kept at as high a level as possible.
Bharti Airtel adopted the “volume” model wholeheartedly from 2003, and was one of the first operators in the world to scale the approach through the aggressive outsourcing of five critical value chain activities: network (active infrastructure), mobile towers (passive infrastructure), information technology (IT), call center support, and distribution. Manoj Kohli and his senior management had the brilliant foresight that no integrated MNO 2 could possibly scale fast enough to keep pace with the rapid growth of the Indian mobile market. Kohli led the strategy to build a complementary network of vendors and partners to meet the challenges of adding millions of new subscribers every month, and Airtel’s success in implementing this approach propelled the company into industry leadership.
It should be acknowledged, however, that Airtel’s success in scaling the “volume” operator model occurred in the context of several unique aspects of the Indian market. Each of these aspects will be discussed in turn.
The huge size and massive capital investments in the Indian market enabled large MNOs to leverage their scale to drive down procurement costs as well as generate operating scale benefits in back-office operations. Operators such as Airtel also benefited from the existence of a large domestic base of Indian companies in areas such as call center operations and other areas of business process outsourcing. The scale of the Indian market presented a relatively attractive proposition for companies such as Nokia Siemens, Ericsson, and IBM to enter in to partnerships with operators to deliver network and back-office outsourcing. These companies already had significant representation in the country and therefore had existing structures which they could grow organically, or into which they could incorporate the respective functional departments and personnel of the operators themselves.
Within just a few years, India became the global center of excellence for outsourcing of key elements of the MNO value chain, and a reference point for cost efficiency for telecom operators around the globe. There emerged clusters of co-specialized suppliers who worked with mobile network operators to deliver discrete activities within the mobile value chain — whether it was network outsourcing vendors, back-office systems suppliers, call center specialists, or smaller-scale regional distributors. The core competence of many MNOs shifted from operating a mobile network (and supporting infrastructure) to managing vendor/partner relationships and a few key areas such as customer management, brand, regulatory affairs, financing, and people management (HR). Today, the lion’s share of a large Indian MNO’s value network is delivered by an interlinked ecosystem of complementary partners. In the words of Manoj Kohli: “Overall what we have done in this new business model is that we have outsourced expertise areas to people that are better than us... and we have kept to ourselves our core competence... Everything else we don’t do.”
The economies of scale benefits from the technology and back-office outsourcing in India enabled a lowering of costs per minute of airtime capacity created. The lowering of costs per minute was also enabled by the relatively low costs of customer acquisition in the early phases of mobile market growth. There was huge latent demand in India for mobile telephony, and the initial growth phase of the industry was focused upon urban and semi-urban areas with high population densities. Mobile network operators leveraged established distributor networks that had been created by the fast-moving consumer goods (FMCG) industry, and operators were frequently able to piggy-back on this existing distribution at relatively low incremental cost. There were real advantages for first movers, especially in semi-urban areas, whereby operators could enter into exclusive dealer relationships with established distributors.
Furthermore, Airtel and other Indian MNOs directly benefited from the focus of Nokia, the world’s largest handset vendor, in building a pervasive distribution network across the country for affordable mobile handsets. In India, most handsets were sold by retailers which were not affiliated with the network operators, and Nokia had acted to strengthen distribution relationships with these retailers, not just in larger urban areas such as Mumbai and Calcutta, but also in smaller towns and rural areas. Within just a few years Nokia handsets were being sold through more than 80.000 retail outlets, and the company offered models that had interfaces in local languages like Hindi and Marathi. Nokia’s activities boosted key complementary drivers of mobile penetration and usage — the acceptability, affordability, and availability of low-cost handsets.
The combination of these factors meant that the relative cost of technology versus customer acquisition costs was very favorable in the Indian market, boosting the success and expansion of the “volume” operator model. But as Bharti Airtel moves into Africa it is necessary to ask if these factors are present.
As has been mentioned above, India’s massive scale in terms of geographic area and population was a key enabler of the emergence of the “volume” operator model, and Bharti Airtel will need to understand the degree to which the key metric of population density matters for the applicability of its model to Africa. India has a population density of approximately 324 inhabitants per square kilometer, but no sub-Saharan African country in which Zain has an operation comes close to this. Nigeria has more than 150 million inhabitants and a population density of 141 inhabitants per square kilometer. Uganda and Malawi have population densities of greater than 120 inhabitants per square kilometer, but just 31 million and 15 million inhabitants respectively. Congo DRC (68 million), Tanzania (42 million), Kenya (38 million) and Ghana (23 million) have relatively large populations, but population densities of between just 25 and 60 inhabitants per square kilometer or less – not even close to the population densities of even the least populous Indian states of Jammu & Kashmir and Arunachal Pradesh (see Exhibit A).
One could argue that scale economies could be achieved in sub-Saharan Africa through the creation of regional network outsourcing contracts rather than national arrangements, but there are significant regulatory and socio-political barriers to such cross-border integration in most of the continent. For example, existing African operators have reported challenges in leveraging resources such as network engineering personnel between some neighboring African states due to restrictive labor laws or complex visa requirements. Similarly, the free flow of technical equipment and spare parts is frequently complicated by customs requirements, import duties, or taxes.
The presence of co-specialized suppliers (.e.g. outsourced network vendors, tower companies, IT vendors, call center operations, and exclusive distributor/dealer networks, etc.) has been critical to the success of the Indian model, as too has been the role of complementary players such as Nokia in the area of handset distribution. But in many parts of sub-Saharan Africa, these co-specialized suppliers and ‘partners’ are either completely missing, or in early phases of establishing local capability and capacity; therefore, the ability of an operator to internally manage and scale core functions such as network, IT, and customer care has been critical to competitive advantage. Quite simply, even though both the Indian and African mobile telecom industries deliver the same core “products” (i.e. voice and text messaging) to millions of relatively low-income consumers, the way that the mobile value chain has produced and distributed these “products” has evolved in distinctly different ways in the two regions. This has serious implications for the entry of Bharti Airtel into the African continent.
The lack of co-specialized suppliers is compounded by relatively low population densities and poor infrastructure (especially roads and electrification) in many sub-Saharan African countries. Fewer than half of all African countries have more than 20% of their population electrified, which means that MNOs in most African markets not only run a mobile network — they also run decentralized power plants. The three leading MNOs in Nigeria alone are estimated to consume more than 250 million liters of diesel fuel a year to run generators on their BTS sites, and collectively employ more than 20.000 security personnel to protect their assets. Paved roads are the exception rather than the rule in sub-Saharan Africa — just 5% of Nigeria’s roadways are paved, while in Congo DRC the share of paved roads is just 2%. In India, paved roads represent 30% of the total. Recruiting and retaining skilled personnel can also be a challenge for MNOs in some sub-Saharan African countries due to a lack of strong technical training institutes, while access to skilled labor in India is much less of a barrier. There are admittedly some very challenging markets in which to operate in India — Bihar and Jharkhand for example — but these truly ‘complex’ regional markets still represent a relatively small slice of total Indian operator revenues.
The complexities of running mobile networks in Africa have resulted in relatively high technology and distribution expenses versus customer costs when compared to large parts of India. This divergence from the Indian mobile market scenario suggests two options for an operator that might wish to launch the Indian “volume” operator model in Africa:
a. An operator could bring in the full spectrum of co-specialized partners from India (or other parts of the world). But this could be expensive and time-consuming as partners develop local expertise and capacity, and could be impossible for some activities. This is especially true for distribution as established dealer networks can be stubbornly difficult to transform, especially in situations where the traditional dealer model has been non-exclusive and/or the operator is a challenger (e.g. Zain in markets such as Nigeria and Kenya).
b. An operator could use a hybrid approach to deliver the “volume” model. This would involve bringing in some vendors (e.g. network outsourcers and IT), developing local partners (e.g. distribution), and/or building own local competence for areas that are still lacking (e.g. call centers). Early indications suggest this is the approach that Bharti Airtel has adopted for its entry into Sri Lanka, where local competitive conditions and the structure of the mobile value chain differ quite markedly from India — despite the geographic proximity to the home market.
There is no doubt that Bharti Airtel is a highly professional and well-managed company — its track record of success attests to this fact. If its “volume” operator model can be adapted for sub-Saharan Africa, it offers the potential to significantly disrupt the status quo, especially in large and increasingly competitive markets where Airtel is the challenger — as was the case in India after the entry of Reliance Infocomm in the early 2000s. But if the company decides to migrate its home-market model to Africa, it will need to acknowledge that the way skills and activities have evolved and been divided between different firms within the mobile telecom industry value chain has occurred in a locally-specific manner in India and sub-Saharan Africa, underpinned by significantly different socio-demographic, regulatory, and competitive conditions.
While Bharti Airtel’s Indian model has the potential to work in some large African markets (Nigeria being the most likely case), success would still depend upon the degree to which co-specialized partners could be brought in, already exist, or could be developed. It would also depend crucially upon the ability of Bharti to transform established distribution approaches — no small task in markets such as Nigeria and Kenya, where it faces negative network externalities in its battle against MTN and Safaricom respectively. But the “volume” model would likely struggle in markets with small populations and/or very low population densities, or in markets where it is prohibitively expensive, time consuming, or socio-politically difficult to build the complementary ecosystem of vendors and partners that underpin the Indian approach. In the latter case, Bharti Airtel would have to demonstrate that it is better than the established firms at playing their own game, and that could well prove a daunting proposition.
Jamie Anderson is professor of Strategic Management at Antwerp Management School and visiting professor at INSEAD. Jean-Paul Evrard is partner and Ronan Moaligou is former head of the Telecom Practice with Globalpraxis.