The withdrawal of South Africa’s wireless open access network (WOAN) licensing direction has understandably revived the debate about what should happen to the spectrum that had been reserved for it.
Much of the discussion has focused on who may eventually gain access to that spectrum, particularly in the sub-5GHz bands, and whether this could create room for new market entrants. Those are important questions, but they are not the most interesting part of the policy shift.
What deserves more attention is that government has retained several of the principles that underpinned the original WOAN idea, including obligations around leasing networks and facilities, providing wholesale capacity to other licensees, and improving access in rural and underserved areas.
In other words, the WOAN may be gone, but the case for better infrastructure utilisation has not disappeared with it.
South Africa’s telecommunications market has spent years debating access to spectrum. The next phase needs to focus much more closely on how efficiently both spectrum and network infrastructure are used.
ICASA has become increasingly firm about spectrum that is licensed but not deployed. That is understandable. Spectrum is scarce, and it has little value to the economy if it is simply held rather than put to work.
Perhaps the mistake was assuming that open access required one centrally designed national network.
For operators already using their allocation, the principle should not be controversial. Spectrum rights need to support investment, but they also come with a responsibility to deploy.
The same thinking should extend beyond spectrum.
Building a network is expensive. High sites, radios, fibre, backhaul, power, security and maintenance all require capital, and the economics become harder outside major urban centres where customer density is lower.
That is where infrastructure sharing becomes more than a regulatory idea.
South African operators already rely on one another in many parts of the network. Towers are shared. Fibre and backhaul are leased. Wholesale capacity is bought and sold. ISPs build services over infrastructure they do not own.
The question is whether we can make more intelligent use of that model, particularly in areas where duplicating infrastructure does not make commercial sense.
If one operator already has a viable high site, fibre route or backhaul connection into an area, there should be practical and commercially sustainable ways for other operators to make use of it rather than automatically rebuilding the same layer of infrastructure.
That does not mean infrastructure should be handed over cheaply, or that companies that invest in networks should be penalised for having taken the risk.
Quite the opposite. Sharing only works if the company that invested can earn an appropriate return, while the company buying access can enter a market more economically than it could by building every component itself.
Get that balance right and the benefit extends beyond the operators involved. More service providers can enter a geography, customers gain greater choice and network investment can reach further.
This matters particularly in underserved areas, where the issue is often not a lack of technology but difficult economics.
We have several technologies capable of connecting remote towns, farms, mines and businesses. Fibre, mobile, fixed wireless and satellite all have a role. The difficulty is making the numbers work when population density is low and distances are high.
That is why the conversation around spectrum should not become another argument about which technology is best.
Different frequency bands and network types solve different problems.
Lower-frequency spectrum can provide broader geographic coverage. Higher-frequency licensed spectrum can deliver significant capacity over more targeted distances. Fibre remains essential in many environments, while satellite is creating new options in places that were previously extremely difficult to reach.
The opportunity lies in using these networks together rather than expecting one model to solve every connectivity problem.
The WOAN was intended to broaden market participation and improve access to network infrastructure. It never achieved that ambition, but that does not mean the underlying objective was wrong.
Perhaps the mistake was assuming that open access required one centrally designed national network.
A more practical model is already emerging: commercial operators investing in their own infrastructure, while making better use of each other’s networks where there is a sound technical and commercial case.
The Electronic Communications Amendment Bill’s proposed “use it or share it” approach to unused spectrum points in much the same direction.
For South Africa, the opportunity is bigger than releasing another block of spectrum.
It is about creating an environment where spectrum is used, infrastructure is properly sweated, and operators can collaborate without weakening the incentive to invest.
That may ultimately do more to extend connectivity than the WOAN ever could.
Share
