In the past year a number of infrastructure sharing deals and outsourcing of network management from firms like Ericsson, Nokia Siemens Networks and Huawei have been announced across a wide spectrum of developing countries. For operators seeking to reduce expenditures as ARPUs relentlessly decline, these various techniques for cost sharing and organizational optimization can be very attractive. But is it also possible that such practices lower the barriers to entry or to expansion to the extent that competition in a given market is effectively enhanced?

See "Can Outsourcing, Infrastructure Sharing and Managed Services Lead to More Competition?"

On the occasion of America’s great festival of consumption and the official start of the riot of gift buying it seems appropriate to pause for a moment to consider Sharing.

I’m not talking about the last piece of pecan pie or the game controller. I refer to network sharing, which is on my mind because of a couple of recent stories:

• A Bloomberg report last week mooting interest on Ericsson’s part in owning networks and providing capacity to multiple operators, with Africa mentioned as a possible target area.
• A report in the Business Daily (Nairobi) indicating that Kenya was planning not to issue LTE licenses to private operators but instead create a public-private partnership to own and operate the network, leasing capacity to all service providers.

Africa has had some experience with infrastructure sharing already, and the pace may be picking up: American Towers just purchased 3,200 towers from Cell C, intending to lease back tower capacity to Cell C and provide smaller operators access to a national footprint. But what Ericsson and the Kenyan government are talking about goes a step beyond that, into a business model where every operator is a MVNO. Worried about becoming a dumb pipe? – forget it. We’ll be the pipe.

This model has attractions for the developing world. For one thing, it gets around the difficulty that many operators have in disengaging from the expensive arms race of network coverage. And it would probably speed rural development.

But it doesn’t change the fact that it will cost a lot of money to provide Africa with mobile infrastructure, and it doesn’t create that money. Both the Ericsson and Kenya proposals – sketchy as they are at this point – include unnamed partners who are presumably putting up some or all of the capital: Valter D’Avino, Ericcson’s VP of managed services, characterized the endeavor as “Ericsson plus a financial company,” and the private half of Kenya’s “public-private partnership” is likely to be asked to put up some cash.

So the question then is – and I apologize for the lack of holiday spirit – “What’s in it for me if I’m the money guy?” Governments may be able to justify infrastructure investment on the basis of business development; Ericsson obviously would see ongoing revenue from network management and equipment sales. Making the case for private investment in African mobile infrastructure may be a bit more challenging.

The mobile market potential of the “bottom of the pyramid” has rightly aroused a lot of interest – including some from Strategy Analytics. The ability to profitably service the huge untapped market of the poor and rural will have a major effect on the long term emerging market prospects of operators like Orange, Vodafone, and Telenor, device manufacturers like Huawei and LG, and infrastructure vendors like NSN, Alcatel Lucent, and Ericsson.

But let’s not forget about the developing world’s middle class, which may number as many as 2.6 billion people. It is middle class consumers who, having some measure of economic security, are in a position to stretch just a little on their handset purchases, or sign up for that slightly frivolous entertainment service. The revenue from each of those stretches, times 2.6 billion, will help fund rural roll-outs and maybe, just maybe, add to the bottom line.

The 2.6 billion comes from Martin Revallion, an economist at the World Bank, who estimated the number of people in developing countries in 2005 whose daily consumption was between $2 and $13 a day, measured on a Purchasing Power Parity basis.

Now, if you are reading this in an industrialized Western nation you might be saying “$13 a day is my idea of poor, not middle class! And $2 a day is very poor.” And in your context you would be right, because that range does not describe the middle class where you live. (In fact, $13 a day was the US poverty line in 2005, which Revallion somewhat arbitrarily chose as his upper bound. $2 was the average officially reported poverty line in 70 developing countries he looked at.)  

The question is whether daily per capita consumption of $2-$13 supports a middle class life in the developing world. This in turn raises the question of what is “a middle class life,” which even those of us who live them find difficult to define. A Middle Class (Perhaps) Shopkeeper in Liberia

Fortunately, sociological precision is not always required to gauge mobile demand. The key element for our purposes is whether in the local context there is enough of a middle class sense of security at $2 a day – or $3 or $7 or some other number – to free up discretionary spending for non-basic communications services. This is a course of research that we will be pursuing in the next year, and we would welcome comments and suggestions.