Where is Africa in the great data race and will cellular connectivity be the way the continent bridges the digital divide?

Where South African users have begun to embrace the power of mobile Internet connectivity with telecommunication companies starting to look beyond HSPA and HSPA Plus technologies towards fourth generation solutions such as LTE and Wimax, the rest of Africa is at the point where 3G connectivity is only now becoming an option. And depending on who you speak to, the story varies between being extremely positive and downright disheartening. While on paper a number of telcos have technically rolled 3G out in their regions of operation, that should by no means imply that 3G is available across the entire reach of their coverage area – or even in areas with a healthy saturation of users. It seems like, 3G is still reserved for the most built-up urban areas and set up as an overlay of a blanket GPRS or EDGE network. But it’s not because the telcos don’t see any potential. It’s among other things because Africa is a market with an extremely low average revenue per user (ARPU) and in conditions such as these telcos need to gauge what the uptake and revenue generating capability of a new service is likely to be before they launch that service. It really isn’t a case of ‘build it and they will come’.

Expensive licenses, limited returns

Douglas Lubbe, Group Executive for Vodacom Group’s International Business says the first hurdle to get through is acquiring a license and justifying the associated capital expense. Acquiring a 3G license can be a costly affair, so costly in fact that often there’s little hope of recovering that expense through adding 3G revenues to the mix over the short-to-medium term. That could be why the Vodacom Group has 3G coverage in Mozambique, Lesotho and Tanzania, but not in the Democratic Republic of Congo, where the government is asking $55m for a license to deliver 3G services in its country. Bertus Ehmke, Senior manager for technology strategy and the MTN Group agrees. He says that telcos paid ridiculous sums for 3G licenses ten years ago and when the technology under-promised somewhat in terms of the revenue it delivered, prices came sliding down. But, now they’re on their way back up to a ridiculous level. “Take another emerging market like India for example,” he says. “The base price for the country’s 3G spectrum auction is already at the $2bn market,” he says. It seems like the market sentiment is that regulators want too much for a decent size chunk of 3G spectrum and aren’t willing to consider the limited revenue generation opportunities for telcos, which in turn hinge on the disposable income in a country, the level of cellular penetration and lastly, the telco’s ability to backhaul Internet traffic into and out of the country. One positive Ehmke points out is that the growth and adoption of HSPA on 900MHz is going well. “We’re especially excited by the ability it grants telcos to serve a three times larger coverage area at the same price and with the same capabilities,” he says. And Ehmke says this service is already being trialed by MTN’s operations in Uganda, Zambia, Ivory Coast, Cameroon and Botswana. Once the license has been acquired, it’s also not realistic to expect the 3G coverage area to be vast.

Conservative rollout planning

“When we embark on a 3G rollout in any market, we start with the major urban areas i.e. where it makes most financial and business sense, and then extend into other areas based whether there’s a business and commercial rationale to do so,” Lubbe says. He says that for many customers, it’s more about having some form of reliable connectivity than having enhanced speed of connectivity, and Vodacom is not convinced that by extending 3G capabilities to areas that already have EDGE or GPRS coverage, there will be a massive increase in revenues. “In Mozambique for example, some of the biggest revenue generating data connections we have are in the rural areas, where a handful of game lodges require Internet connectivity and GPRS or EDGE suits fine,” he says. That said however, Lubbe says that Vodacom is focused on constantly expanding its 3G network in these markets. “But we have to make sure there’s a market for 3G services before we deploy into an area,” he says. Furthermore, Lubbe says there’s no hard and fast rule for when a region or a coverage area needs 3G. The company conducts extensive research into a piece of geography and its population’s usage patterns, and then makes a decision from there. Lubbe says the Vodacom Group is also not stopping with 3G. “We do envisage rolling out nextgeneration technologies such as HSPA and LTE out into Africa, but it’s not likely to be in the near future,” he says.

Is there backhaul?

Another big factor Vodacom weighs up when gauging the readiness of a market for enhanced data services such as 3G is whether or not there’s adequate backhaul infrastructure in place to feed users what they need – whether that’s access to a website in the U.S. or Europe, or a piece of content or functionality that is available on a local server. While the international bandwidth issue is in the process of being solved, right now there’s no redundancy of which to speak, as was demonstrated by the massive outages companies across the continent faced while the Seacom cable was being repaired a month ago. He says the local loop in many African countries is also a massive issue, since there’s not enough working fibre cabling in the ground to connect major business centres together and ultimately provide a speedy, reliable link to the much needed undersea cables carrying traffic to and from the more developed world. Lubbe says that Vodacom is making investments into fibre principally at local loop level. But it’s important to note that these investments are not at the same level as Vodacom’s investments in South Africa, both in terms of the volume of capital being spent or their long-haul nature. 

Customer equipment costs

Once the economies of the network are out of the way, Lubbe says there’s also the availability and price of terminal equipment to consider. “3G is still a relative niche and the client-side equipment that enables 3G tends to be more expensive than 2G equipment,” he says. “The prevalence of notebooks equipped with internal 3G modems is also not as significant in Africa as it is in other parts of the world,” he says. “We are seeing encouraging growth here however,” he says. Given time, these markets will develop. “It’s contingent on factors like the reduction of terminal and handset pricing and since many of the markets Vodacom is interested in have upwards of 98.5% on prepaid plans, there’s little that can be done to bring terminal pricing down with subsidies,” he says. “That will change, though. “A couple of years ago we were dreaming about handsets that cost less than $20 a unit and that’s now a reality. The next step is for 3G handsets to get to that level and we’re hoping it can happen in the next three to five years,” he says. MTN’s Ehmke says that 3G terminal equipment is already in the region of $50 for a USB modem-type device and that it’s already much less of a barrier than what is used to be. “And with the industry’s help, the price of 3G equipment is getting driven down even further,” he adds. While Lubbe says that businesses currently have the core drive for mobile data requirements in these regions, the consumer market is catching up and this is where the momentum will come from. But the point is, enhanced third generation and new fourth generation technologies are still some way off.

Looking beyond 3G, HSPA

Ehmke says that internationally LTE has grown in maturity rapidly over the past 18 months and this is largely because mobile Wimax, another fourth generation technology is being relatively successful in emerging markets. The recent maturation of LTE is largely because the GSMA has been encouraging mobile operators to hold off on Wimax and instead start investigating LTE. “The problem in Africa,” he says, “is although LTE is hugely diverse and adaptable, it does have a ‘favourite spectrum’, namely the 2.5GHz to 2.6GHz band. This means it’s subject to higher levels of attenuation (noise) and a lower range than 3G/HSPA and other technologies.” It also means the coverage area of LTE is smaller, making it increasingly difficult for telcos to build a financial model that works. While Ehmke says LTE can work in the 700-900MHz band, because there’s less traction for the technology between those frequencies, the equipment capable of operating within that spectrum will bear a price premium. “Chances are this will be a very different discussion in two years’ time,” he says.

Similar issues to 3G

Another reason LTE is not the perfect solution for Africa right now is that Telcos all recognise that voice revenues are under pressure and remain a massive portion of income made in the Africa markets. “Voice is an afterthought when it comes to LTE,” Ehmke says, “and having solid voice capabilities available in Africa is key. It’s far more difficult to build a business model based on a purely data service,” The terminal equipment is also too expensive. Ehmke says customers need to contend with average price tags of $150 for normal USB-type data modem for LTE and closer to the $300 mark for a router-type modem for a small office. “All of these factors make LTE too early for Africa,” he says, “and besides, there’s a huge amount of headroom that can still be extracted from 3G and HSPA.”

The headroom is there

Ehmke says that this is also a more logical route for telcos to follow, since the majority of the 3G radio equipment available and being installed today is already capable of providing HSPA services. “All it takes is the telco licensing itself for that technology and bolting a handful of enhancements onto its infrastructure, like taking advantage of MIMO by adding more antennae elements,” he says. HSPA is also very capable of offering the kind of speeds and services that allow telcos to deliver ‘actual value’ to their markets. “The next billion customers being connected to the Internet will not necessarily be using the next billion PCs to be connected to the Internet,” he says. “Many of them will be engaging and experiencing the Internet over a mobile device – whether the experience is as trivial as a WAP browser embedded on a low-end cellular phone or a 3G-enabled iPad type device. “And this leads me to believe that once the network has reached the 7.2Mbps mark – as is comfortably offered by HSDPA today – it reaches an inflection point in terms of what one can appreciate on a device that’s not a PC or a notebook. “I mean, we’re talking about devices with a maximum of ten inches worth of screen – and that’s too small a screen to appreciate or even need high definition visuals or sound,” he adds. “Considering this, even from an multimedia perspective, there’s little need for much faster connectivity than what our current networks have the headroom to provide,” he says. This substantially changes the way we need to think about things on the African continent. It makes it less about ‘keeping up with the Joneses’ than making sure today’s technology is capable of scaling linearly with what future demands are. And from the looks of things, the technology being installed by telcos is capable of just that.