After Poison Ivy

I'm not going to indulge in festive-season feel-good rumination, trying to look on the bright side of Ivy Matsepe-Casaburri's decade as Minister of Communications. I've heard her own department's PR on the subject. Most of it is false, and none of it is worth gushing about.

Someone else will be taking over her chair in 2009, however. Speculation (including my own) about who it might be aside, this person might want to consider a few options to make a new beginning, blow out some cobwebs, or distance themselves from the reputation the DOC has made for itself.

This won't be easy. It's worth recalling that Poison Ivy's policy of 'managed liberalisation' (which neither liberalised much nor was well managed) predates her own incumbency. It predates even her predecessor, Jay Naidoo. It was, after all, Pallo Jordan, who memorably got up in front of the world's financial media at the Carlton Hotel in 1995, to say: "I wouldn't even consider competition for Telkom." Representatives of very large amounts of capital got up and walked out, to go lay tons of fibre in friendlier climes.

Here's what a new cabinet minister could do to fix this legacy.

Think long and hard about what it means to govern. As often as not, it means stepping out of the way, rather than controlling everything.

Ivo Vegter, ITWeb contributor

Think long and hard about what it means to govern. As often as not, it means stepping out of the way, rather than controlling everything. This is hard for an executive minister to do, perhaps, just as it is hard for a regulator to contemplate the case against regulation. Action instinctively seems braver, wiser, and more productive than inaction. But a wise governor, especially of a business sector, knows when not to govern; when to pay attention to just the few things that really can improve the performance of that sector, for the benefit not of the businesses themselves, but for their customers, the citizens of South Africa.

In this spirit, I would like to propose two major policy changes a new minister could make.

First, remove the restrictions on how many telecoms licences can be awarded. Issue a standing 'invitation to apply', to anyone. If an applicant meets licence requirements (the means by which a minister governs the actions of private entities), then have the regulator issue a licence. Revoke it only if those licence conditions are breached.

This works for most other sectors, and it permits adequate policing of abuse in an industry. Compare with liquor sales, for example. A licence is simply a means to ensure they don't break liquor-related laws. Finding or creating space in the market, and making a profit, is the liquor store's own problem.

Why should it work differently in telecoms? Limiting numbers of licences is not 'liberalisation', it is an attempt to 'manage'. Less kindly, it creates opportunities for patronage, by creating cartels. It rewards and protects favoured companies at the expense of would-be competitors, and ultimately at the expense of consumers.

If the minister fears chaos, they could attach a fee to major licences sufficient to prevent their abuse by just anyone. This isn't ideal - there is no reason why a licence should cost more than its administrative cost - but it might make the new minister feel a little more comfortable about such a dramatic change in policy.

Second, make spectrum tradable. Auction it off to the highest bidder, and leave its allocation and use up to the private sector. Land is also a physically limited resource, but there's no government control to make sure only those who "deserve" land get only the acreage they "need". All the government needs to do about private ownership is protect the property rights of owners: make sure their land isn't unjustly occupied, and that their land use doesn't infringe on the rights of those around them.

Same with spectrum. Police spectrum use, based on who owns the spectrum, and leave it at that.

By making spectrum tradable, you create all sorts of opportunities for innovative and more efficient use of this scarce resource. A large company might buy a slice, and sublet slices to smaller operators who use frequency-hopping to share it. This could allow cheap access to limited spectrum for small, regional or emerging operators, just as land developers maximise land-use efficiency by building high-density complexes with certain shared amenities. Bigger companies, with more demanding needs, would pay more for dedicated slices, or develop alternatives.

At a stroke, spectrum management is no longer a complex duty performed by a tax-funded bureaucracy. At a stroke, those who feel they can put spectrum to the most profitable use can buy what they need - remembering that profitability is dictated not only by efficiency, but by delivering what consumers want, as indicated by their willingness to pay for it. No longer will the CEO of a major telco complain that the slice of spectrum they got for free just happens to be completely useless to them. Instead, he'd buy the slice he feels he needs, at a price the market thinks is fair.

Again, the minister might have fears about this. Although most of them are unfounded, training wheels might offer comfort.

One-time auctions, for example, have bankrupted major companies elsewhere. While this is not the government's problem, and others will surely benefit from their rash misallocation of capital, the regulator might instead develop a different scheme for initial allocation of spectrum before making it tradable. A new minister might fear spectrum hogging by incumbents. Although the market will ultimately offer the right price to prevent this, a use-it-or-lose it rule could be used to mitigate this perceived risk.

If these two things, unlimited numbers of licences and spectrum tradability, are the only policy projects a new communications minister attempts, their term will be celebrated as one of true service to the South African public, rich and poor. And their term will be easy, to boot. Let's hope in 2009, everyone's a winner.