[By Duncan McLeod]

More than 17 years after SA’s first democratic elections, politicians are still indecisive over how to extend connectivity into rural areas and bridge the so-called “digital divide”. Government continues to concoct ideologically confused plans. Instead, it should just get out of the way.

The country’s first attempt at extending telephony in rural areas, using Telkom as the vehicle, was a hopeless failure. The company was given a five-year monopoly (it became a de facto decade) to install fixed-wireless services in areas that didn’t have coverage. On paper it looked good, but in practice the voracious foreign management team leading Telkom at the time hiked prices, making access unaffordable.

Instead, it was the mobile operators — led by Vodacom and MTN — that extended basic telephony to some (but by no means all) outlying parts of the country. You’ll be hardpressed to find Telkom services outside the big cities and towns, but millions of rural people use prepaid mobile phones.

Now government has turned its attention to taking broadband, and not just basic telephony, into underserviced parts of the country. Government is right to focus its attention on this: lifting millions of South Africans out of poverty and allowing them to compete in the global information-driven economy is as much dependent on cheap, ubiquitous and fast Internet access as the railways and harbours were in the industrial age.

But government is talking — and has been talking for a long time — about using largely dysfunctional state-owned enterprises to do this. It’s the wrong approach, and will fail for several reasons, not least among them that state-owned companies, especially those that have been “corporatised ”, don’t have the motivation to do it.

Instead, government should be crafting policies that help private enterprise to make money by delivering broadband to everyone. It’s generally assumed that broadband can’t be delivered on a financially sustainable basis to rural areas. That’s an assumption that hasn’t been tested.

Here’s what we know. It’s too expensive to build fixed-line infrastructure across SA’s vast distances into every home and business. But building next-generation wireless broadband networks is not rocket science. What’s required is high-speed fibre-optic backhaul to towers and wireless “last-mile” access using the spectrum at present occupied by television broadcasters because of its physical ability to carry signals over long distances. Fewer towers means lower costs for operators to serve more customers, which translates into profits at lower revenue per user than is possible now.

Instead of inventing elaborate plans for companies like Sentech, with its dismal track record, to take the Internet to underserviced areas, what’s required is a policy rethink that would give spectrum to privately owned operators to build these networks.

Government needs to pay attention to Steve Song, a former fellow at the Shuttleworth Foundation. Song argues that SA needs to ensure spectrum is suitably allocated to encourage new entrants to use the analogue television broadcasting bands. He argues this spectrum should be offered to newcomers, even before SA’s broadcasters vacate the bands when they move to digital technology in December 2013.

He ’s right. SA’s incumbent operators aren’t the best suited to connecting rural areas: they’ve been living off the fat of the land for too long to care. It’s time to encourage new entrants. The incumbents will put up a fight, as they always do, but if government is serious about addressing the digital divide it will put an end to its ideological nonsense about relying on state-owned enterprises and free the market to compete.