Na Imprensa

GSM growing pains - Next wave of subscriber growth to come from low-ARPU markets

Telecommunications Magazine

By Ouida Taaffe

If you sell jam doughnuts, you can probably expect regular, on-going demand for your product. If, on the other hand, you sell GSM infrastructure, revenues may well be a little 'lumpier' - even if your customers are not. This is, analysts argue, one of the reasons why developing markets are of great interest to infrastructure vendors.

However, ARPUs in developing regions can look very small from a Western European perspective. Where an operator in the UK or Germany would expect a blended post-paid and pre-paid ARPU of over £20 a month - O2, for example had a Q3, 2005 annualised ARPU of £272 (€397) - operators in developing countries might expect an ARPU that is not that much more than a tenth of that.

The reason for this, of course, is that gross national income (GNI) per capita in developing countries can be extremely modest. According to World Bank figures for 2004, Luxembourg (with around 468,000 people) led the global earnings pack with an annual GNI of US$56,230 per head (it was also first in terms of purchasing power parity (PPP) with US$61,220 per person, per annum). In contrast, India with a population of around 1.1 billion, had an annual GNI of US$620 per head (PPP was $3,300) and Nigeria, which has around 130 million people, had a GNI of US$390 (PPP of $930).
Despite such yawning gaps in purchasing power, there is - vendors argue - still good telecom business to be had in developing countries. "Most [developing region] operators are profitable down to ARPU levels of US$6-7 a month," says Johan Bergendahl, vice president of marketing at Ericsson. "That means incomes need to be in the range of around US$2,000 a year for people to be able to afford to spend around four per cent of their income on telecom. We aim, over the next 2-3 years, to lower the profitable ARPU range so that those earning US$600-800 a year can also be targeted."

Further, Bergendahl says that if operators hand over full network management to Ericsson, the vendor can - depending on country - already bring that profitable ARPU number down to US$4 a month.

Assuming, however, that operators run the networks themselves - and taking World Bank GNI figures - this would mean that of the 208 nations surveyed by the bank in 2004, around 160 of them would have a GNI per capita that could support high GSM penetration. Ericsson expects that around 80 per cent of subscriber growth over the next two years will come from emerging markets and that, by 2008, around three billion people could be GSM subscribers. How can this be done in practice?

According to Sylvain Fabre, an analyst with Gartner Group, the equipment used in the developed world is "lower price and lower spec". Bergendahl, however, says that the GSM infrastructure being rolled out in developing countries is not necessarily a cheaper version of that used elsewhere. "When rollouts first started, the aim was to get the price down. Now, the cost of ownership issues are not so different in Sweden as opposed to, say, India," says Bergendahl. "Competitive forces are pretty strong in Western Europe."

It could, of course, be argued that Chinese vendors like Huawei, which have targeted developing markets and are rumoured to pass on access to low-cost capital as advantageous vendor financing, concentrate minds at least as much as big Western European operators do.

That said, Bergendahl also suggests that cutting corners in developing markets can have unfortunate consequences. "Delivering cheap equipment with low functionality or capacity is a bad move in these countries," he says. "80-90 per cent of new subscribers will come from areas that already have coverage, so the issue for operators will be 'how do I increase capacity in a cost-effective way'." This, he adds, means upgrading - rather than stripping out - old boxes. Bergendahl argues that having a base station with advanced functionality, as opposed to the bare minimum, will allow operators to increase capacity at 30 per cent of the cost seen over the past two years.

However, developing markets do not just have subscribers with limited budgets, operators can also be severely financially constrained, points out Alvarion, the Israeli-based equipment vendor. "Companies like Ericsson don't really target the tier three cellular operators," argues Rudy Leser, vice president of marketing at Alvarion. Alvarion has what it calls a 'network in a box' GSM solution that, he says, can provide a 'full network' giving coverage of around 10,000 subscribers for around US$2 m. The product is not specifically aimed at developing markets, and does not aim to serve low-ARPU customers per se. For example, it also finds deployment in environments such as airports, cruise ships, or islands.

Many developing countries are rolling out mobile networks because they lack telecom infrastructure of any sort. Of course, where there is no fixed network there is generally no broadband, which means a yawning digital divide. Mobile will, Bergendahl believes, be able to bridge this gap.

"Operators definitely have mobile data services on their roadmaps. There is also a big interest in the next step beyond traditional data to services like IPTV," he says. Operators in Nigeria, according to Bergendahl, see a business case mobile TV at a cost of around US$20 a month. This is, of course, not cheap by Nigerian standards, but satellite TV, which has no local content, costs around three times as much. Further, video services - even if they are priced beyond mass-market levels - can have a strong draw in countries where many people are illiterate.

There have often been suggestions that WiMAX, which promises high and long-distance capacity from a more limited number of base stations than GSM, could be the right solution for the provision of data, and later, mobile services in developing countries. The mobile iteration of the WiMAX standard, 802.16e, is due to be released in 2007.

"The difference in cost between GSM and WiMAX is not that enormous," says Bergendahl of this. He argues that, by the time mobile WiMAX is ready for deployment in 2008-2010, technologies such as HSDPA will be widespread and able to command economies of scale. "The WiMAX business case is not that clear," says Bergendahl.

Alvarion, which bases much of its business on WiMAX equipment, agrees that WiMAX is unlikely to replace GSM and that it will not be the cost-killer that has sometimes been envisaged. "It will take a while for WiMAX to get to the same kind of price," says Leser. "Mobile WiMAX will go first to developed countries."

However, Leser does believe that WiMAX has a strong future as a broadband technology in developing countries where fixed infrastructure is limited. "There is a clear need for broadband at an addressable cost," he says - though this cost will be well above that commanded by 'low-ARPU' GSM operations. According to Leser, Reliance Infocomm and 'other operators' in India aim to roll out WiMAX in 80 Indian cities. The reason for this is that there are two main segments in India that promise to be very lucrative. The first is the SME market. Indian SMEs are considered likely to spend upwards of several hundred dollars a month on a high-speed and reliable broadband connection. The second segment is the high-end consumer market, where ARPUs could be in the US$10-US$30 a month range, Leser says. Demand for WiMAX in China, which is often perceived to be key, will be a reflection of regulatory developments that are still unclear he adds

« Full Story »

« Back to List