Captured by the incumbents
By taking too long to issue policy decisions about next-generation broadband, government is allowing the broadband future to be dictated by the incumbent mobile operators. By Dirk de Vos.
The introduction by South African mobile operators of next-generation long-term evolution (LTE) networks does not amount to much. But we can’t ignore it because these networks are the first bit of evidence of a future of lost opportunities to deliver cheap, properly fast and ubiquitous broadband services to all South Africans.
It is hard to fault the operators. They are simply taking the gap left wide open by a complete absence of policy from the department of communications and industry regulator Icasa.
It’s apparent that not having any discernible policy on LTE does not mean that operators will defer decisions on LTE. As we have seen, they will forge ahead and deploy it in a manner that best suits their objectives, knowing that once any policy and associated regulations are in place a more-or-less complete LTE edifice would have been built.
Then, using that well-worn South African regulatory device, these investments become “grandfathered”. There is precedent for this. For example, MultiChoice, which owns DStv, for many years operated in a policy vacuum on pay television delivered via satellite. By the time Icasa got around to licensing new operators, MultiChoice had sewn up the market, both in terms of content and subscribers. TopTV, the only one of five new licensees to launch a commercial service, had its back against the wall from the outset.
LTE has the potential to transform South Africa’s broadband landscape. World Bank studies show that a 10% increase in broadband penetration generates a multiplier that increases GDP in developing countries by 1,7%.
In practical terms, LTE addresses the somewhat disappointing performance of 3G networks. One of the main problems has been throughput latency — the end-to-end delay from the time a packet of data is sent from say a server to the subscriber’s handset. Another problem is the issue of interference with all other communications on a cell and the great variability in performance. Finally, the 3G standard is fraught with numerous intellectual property rights issues in relation to 3G-compatible handsets, which has driven up the cost of these handsets.
At the beginning of 2012, there were at least 33 commercial LTE deployments and at least 75 LTE networks should be in commercial service by the end of 2012.
Yet we do not have any discernible policy in place. This means that we do not know how or on what basis the “digital dividend” in the 800MHz radio frequency spectrum band or, indeed, the spectrum around 2,6GHz will be licensed. For South Africa, especially, knowing how the digital dividend will be licensed feeds directly into the costs of any LTE network. The most direct cost benefits come from the ability to use fewer cell sites to provide blanket coverage geographically.
Fewer cell sites mean lower deployment costs, fewer towers and facilities, and lower operational costs. These economics allow the very real prospect of bringing broadband to even the poorest and most underserved areas of the country.
For now, South Africa’s operators are “refarming” (reallocating) some of the spectrum traditionally reserved for legacy voice and data services. While LTE infrastructure has been designed to operate on spectrum of as little as 1,25MHz in size, 20MHz of contiguous spectrum is required for optimum connection speeds. Refarming, therefore, is far from ideal.
Obviously, the operators will want to retain their 2G cash cows for as long as possible. Also, they have poured billions of rand into 3G networks and will need to see a return on that investment. But operating three simultaneous networks is not a long-term option.
The fact that they’re proceeding with LTE regardless ultimately undermines the future benefits of the technology. One of the key benefits of GSM networks was the use of harmonised spectrum. This permitted the introduction of cheap handsets. In the absence of harmonised spectrum, the burden shifts to handsets and other end-user mobile devices to support multiple frequency bands. This results in more expense, complexity and inefficiency. LTE devices are likely to be either very expensive or provide a poor quality of service and not be compatible for use on other LTE networks.
LTE can be transformative in South Africa. We should understand that not making policy decisions around this technology right now does not mean a deferral of whatever emerges in the future. It just means that the important decisions are handed to others — in our case, the incumbent mobile network operators.