The strong response by renewable energy consultant Dirk de Vos to my column on South Africa’s nuclear power plans betrays a certain idealism, although I have to commend him on an excellent headline: “Groucho Marx and the Nuclear Energy Sector”. But before we get to my alleged similarity to a dead comedian, let’s tackle the meat of his critique.

De Vos correctly points out that I am a libertarian – that is, a proponent of small government and free markets – of the Austrian School persuasion. In my ideal world, government would do very little beyond protecting basic human rights like life, limb and property. Naturally, then, I agree with him about the root causes of South Africa’s electricity crisis, namely that Eskom is a vertically-integrated state-owned monopoly which provided widespread electrification and cheap electricity at the expense of necessary maintenance and upgrades to its infrastructure.

However, when discussing questions of national policy, such as whether or not to support the idea of building nuclear power stations, it doesn’t help to be a hopeless idealist. It also is of little use to conflate different issues. What forms of new generation capacity ought to be licenced in the short term is only weakly related to the question of how to structure our electricity grid in the long term.

If I had my way (and De Vos advocates this, too), Eskom would be broken up. Generation would be separated from transmission (the grid), local distribution, and customer supply. Competition would immediately be introduced in the generation sector, by privatising existing power stations and issuing new licences only to private investors, as has been done with the 92 renewable energy licences issued to independent power producers. What technology is used by new private power stations would be entirely up to the investors concerned. The only concern of the transmission and distribution system would be how much electricity these power stations can supply, at what times, and at what price. In the short term, Eskom would be responsible for maintaining and upgrading the transmission grid. This forms what is often described as a “natural monopoly”, and to accommodate distributed, variable generation technologies like wind and solar requires an improved grid that can better match supply with demand.

Ultimately, I’d like to see transmission, distribution and supply in private hands too, but since turning a state-owned monopoly into a private monopoly is jumping from the frying pan into the fire, privatisation is not enough. It would be important to introduce competition into these sectors.

Both transmission and distribution are capital-intensive sectors, with complicated policy issues surrounding rights-of-way and ownership of existing infrastructure. With distribution, the process can be compared to local loop unbundling in the telecommunications sector, granting multiple suppliers access to the same connection to customer premises. This process is fraught with difficulty and delay, however. For example, in 2006, the late Communications Minister, Ivy Matsepe-Cassaburri, set Telkom a five-year deadline for unbundling the local loop. Ten years later, Telkom is only in the testing phase.

With the transmission grid itself, competition would operate on the principle that two competing lines provide both generation and distribution customers with better options. As a thought experiment, consider which is cheaper: one railway line between Johannesburg and Cape Town, or two? A government planner would say one, since a second line would double the capital expenditure. However, the correct answer from an economic policy perspective ought to be two, since competing lines will result in cheaper fares to passengers and freight customers. Again, however, getting to a point at which monopoly infrastructure is either shared or duplicated will take many years, and probably decades.

There are few case studies involving successful privatisation of electricity transmission, distribution and supply. Most continue to require massive subsidies to attract investment in infrastructure. One of the main reasons is that transmission and distribution typically remain regulated national or regional monopolies, so while competition in the generation sector can lower or raise wholesale prices, retail prices do not necessarily follow suit.

Getting grid-level policy decisions wrong can have catastrophic consequences. California offered a spectacular example in 2000, when wholesale prices were allowed to float freely, retail tariffs remained fixed, market manipulators like Enron had a field day, and customers suffered widespread blackouts.

South Africa doesn’t exactly have a great record for decisive, responsive policy decisions. Even if we were to anticipate such a radical transformation of the energy sector, the policy and regulatory issues are complex, and it would take decades to become reality. Yet De Vos thinks it is realistic to expect, “big changes in our thinking about energy, including the role of Eskom, the role of markets, the need to separate electricity generation from transmission and distribution, closer integration with our neighboring countries and more.”

Sure, that’d be nice, and I’d be the first to support fixing our dysfunctional government electricity monopoly to replace it with a model which “will be flexible, distributed and market orientated”, as De Vos wants. But if this is not a pipe dream, it certainly is a separate discussion about long-term policies that has little bearing on the matter at hand: whether or not to support building nuclear power stations in South Africa now.

Restoring our supply to the abundance needed to revive the mining and manufacturing sectors is a matter of considerable urgency. Electricity privatisation, even just in the generation sector, will not happen in South Africa on anywhere near the scale that is required in the next decade or two. Large-scale renewable energy, which is the only part of the generation sector that is mostly private, will remain fairly marginal, and even under the best of circumstances will only be a part of the long-term solution to our energy crisis. De Vos himself wrote that it will constitute only 9% of our energy mix by 2030.

Gas could contribute most of our energy needs, thanks to its flexibility and low price, as that article also points out. I’m a great fan of natural gas, and I entirely agree with De Vos. However, we’re limited to imports on a relatively small scale until we develop a significant domestic gas resource with its attendant infrastructure. Partly thanks to regulatory delays and partly because of private sector risk aversion in the face of a stubbornly low oil price, that isn’t going to happen in a hurry either.

For all our ideals, we need more than either renewables or gas can offer, and we need it sooner. That doesn’t mean rejecting either, but it does mean continuing to build coal-fired power stations, as we’re doing, and possibly adding nuclear power to the mix. One is unsafe and polluting. The other is safe and environmentally friendly. Perhaps we’re talking about the lesser of two evils in the context of an idealised distributed power grid, but I know which I prefer until that ideal becomes reality.

In my ideal world, the private sector would be building liquid fluoride thorium reactors by the dozen. They are superior in almost every respect to uranium-fuelled reactors, which thanks to proliferation fears are always going to remain strictly state-controlled. In fact De Vos also claims to support such next-generation reactors, but since they’re nothing but vaporware at the moment, they’re not a realistic option to add to our energy mix in the short term. Supporting the development of fourth-generation nuclear reactors is not mutually exclusive with agreeing to third-generation reactors today.

In an amusing point about whether nuclear energy has a future, globally, De Vos compared me to Groucho Marx, who asked a cuckold, “Who are you going to believe, me or your eyes?”

My alleged error was to cite the projections of the World Nuclear Association on how many reactor projects are ongoing and planned. Presumably, relying on hypothetical future plans which could be scuppered by financial challenges, public opinion, or technical hurdles, isn’t smart. I’m not sure how relying on the plans of the nuclear industry is any different from relying on the speculative expectations of a renewable energy consultant about future smart grids, but in any case, those projections are not essential to my point that nuclear power is undergoing a modest revival, and not only in developing countries.

De Vos himself cited a chart that illustrates the nascent recovery in nuclear power builds. While new construction is still low, and suffered a setback in the wake of Fukushima, it is clear that after a two-decade hiatus the recent trend is strongly upwards. The numbers given by the World Nuclear Association are not only hopeful projections, but also cover ongoing construction in Switzerland, the USA, Spain, Finland, Sweden, France, Japan and India.

Yet De Vos wrote: “Those part of the world that procure electricity using democratic, transparent and market-based methods do not build new reactors.” You don’t have to believe anyone’s projections to see, with your own eyes, that this claim is overstated, as I wrote.

I understand that De Vos opposes nuclear power, for the most part in favour of renewable energy. Perhaps being a consultant to the renewable energy sector has something to do with that. I agree with his ideal vision of a smart grid supporting a mix of distributed generation companies. That is indeed, as he calls it, “an electricity utility model appropriate for the 21st century.” But as the link he provided in support of this vision says, this is only a “re-imagining” of electric utilities. It’s an ideal which exists nowhere, let alone being appropriate for South Africa in the short term.

Call me Groucho, but at least third-generation nuclear power stations are real. DM