Spectrum trading will make all Icasa’s problems disappear

It has been 14 years since the last time South Africa’s telecommunications operators received any new radio frequency spectrum.

At the time, 3G was still fast, the very first iPhone was not yet available in South Africa, and 4G was only a prototype in Japan.

Operators desperately need additional spectrum to provide new-generation high-speed mobile data services.

Yet the road to auctioning high-demand spectrum in the International Mobile Telecommunications (IMT) radio frequencies, which includes bands that are supposed to be freed by the endlessly-delayed transition from analogue to digital terrestrial television, has been long and arduous.

The regulator – the Independent Communications Authority of South Africa (Icasa) – is trying to balance a number of conflicting interests, not only among market participants, but also those imposed by the government’s SA Connect broadband policy.

This policy is couched in terms of grand central-planning buzzwords, from ‘active citizens’ to ‘decent work’, from ‘social cohesion’ to ‘nation-building’, and from ‘job creation’ to ‘prioritising rural and under-serviced areas and stimulating economic growth’.

Key among these ambitions is to establish a government-run wireless open-access network (WOAN), as a response to the perceived failure of markets to deliver widespread access to broadband services at ‘affordable’ prices.

Of course, what constitutes an affordable price is entirely subjective, achieving affordable prices is not something governments know how to do, the market managed to offer 3G to 99.7% of the population and 4G to almost 90% of the population entirely without government help, and it would have got a long way towards offering even more widespread access to even better broadband services at even more affordable prices if it had received the radio spectrum with which to do so a decade ago.

The entire WOAN enterprise is unnecessary, is flawed, has failed everywhere it has been tried, and should be scrapped entirely.

The SA Connect policy proposes both ‘demand- and supply-side policy interventions’ in a ‘four-pronged strategy’ aimed at achieving fast, cheap and universal broadband.

It is similar to the government policy for delivering high-quality, cheap and universal electricity, drivers’ licences, and postal services – and will meet with similar levels of success.

Government intervention hurts service delivery

The problem is that, no matter how many prongs it has, government intervention has never improved on the efficiency of markets.

In the South African telecommunications sector in particular it has always led to reduced competition, exploitative value-extraction by hand-picked cronies, and low-quality, over-priced services for businesses and consumers.

Because of all these competing policy and commercial interests, Icasa’s plan to auction off spectrum isn’t a simple matter of defining it and putting it up for sale. No, it’s far more complicated than that.

It has put the WOAN on the back burner for now, where all useless government projects go to moulder for five or ten years, but this has raised valid concerns among existing operators that they cannot predict the WOAN’s future impact on the market, and therefore cannot properly value the spectrum bands that are being auctioned.

In an effort to appear ‘fair’, Icasa has determined that it should distribute spectrum not to those companies who would use it most effectively, but among as many market participants as possible.

To this end, it has capped the maximum amount of spectrum any operator may own and has created a preferential round of bidding for small players or new entrants to get their hands on spectrum, without the risk of being outbid by ‘tier one’ operators.

Only MTN and Vodacom were designated as ‘tier one’ operators, and would have to wait their turn at the auction until the rats and mice have finished their feast.

Among these rats and mice, one can find Telkom which, despite being a mobile minnow, actually owns a third more spectrum than Vodacom or MTN.


MTN challenged this so-called ‘opt-in’ round in court claiming, quite reasonably, that all the best spectrum would have been cornered by the time they got to bid.

Most of those rats and mice will end up hawking their spectrum to the large mobile operators for roaming agreements regardless, while running bargain bin network infrastructure themselves.

This creates a layer of middle-men profiteers that reduce the quality and raise the price of products that will ultimately reach end-users.

Telkom has been the most consistently obstructive over the years and has been raising complaints on the exact opposite grounds: that the auction process will entrench the power of its larger competitors, that its larger competitors will be unfairly advantaged, and might even – heaven forbid! – get to upgrade their networks and improve their service to customers.

The former state-owned monopolist does not deserve any sympathy, but it does raise another important point: that the 700MHz and 800MHz bands, which are to be freed by the final switch-off of analogue television broadcasts, are currently encumbered by legal action by e.TV – which after being given notice a hundred years ago is not yet ready for the transition to digital broadcasts.

The provisions in the invitation to apply for spectrum run to 91 pages of complex bureaucratese, and even so, everyone has complaints, some of which are entirely valid.

How tradeability will help

The solution to all this central-planning madness is simple: make all spectrum in the IMT bands tradeable.

There’s probably no way one could convince the government not to auction it off at a fat profit to pay for VIP security and ‘security’ upgrades at politicians’ houses, but if the spectrum is tradeable, at least an auction isn’t the final, irrevocable commitment that it otherwise would be, and wouldn’t entrench any operator at the expense of another.

If spectrum sold at auction, either through outright ownership or long-term leases, remains tradeable, then it does not matter how spectrum is initially allocated.

Market forces will in due course produce the most efficient allocation.

Instead of everyone suing the regulator, they could just fight it out among themselves.

If that strengthens major players in the market, so be it.

Ultimately, it is the interest of customers that matters, and customer interests are not served by rigging the market or artificially sustaining smaller operators who exist only to lease their networks to the major players in any case.

Tradeability of spectrum also incentivises spectrum owners to use their spectrum more efficiently, to minimise interference with neighbouring bands, and to maximise the network performance extracted out of any particular sliver of spectrum.

Most of all, it would remove from Icasa’s shoulders the impossible burden of keeping everyone including the ruling party socialists happy.

It would invalidate most of the challenges against its auction process and turn its ‘invitation to apply’ from a lengthy book of lawyer-bait into a simple notice it could publish as a small newspaper advert.

Ultimately, businesses and consumers would benefit by putting a swift end to the endless rounds of litigation and not creating a layer of middle-men in the name of ‘competition’.

Competition is not something a government can create. It is what happens when you set the market free.

This is an opinion piece.

Ivo Vegter, a former technology journalist and a columnist for the IRR, writes in defence of free markets and individual liberty. This article was commissioned by the Free Market Foundation.

Now read: Telkom’s shady deals — “We already investigated and went after the culprits”

Latest news

Partner Content

Show comments


Share this article
Spectrum trading will make all Icasa’s problems disappear