Nuclear for South Africa – a spectacularly kak idea

The environment minister announced authorization for a new Nuclear plant in Duynefontein.

https://www.reuters.com/sustainability/boards-policy-regulation/south-africa-pushes-ahead-with-new-cape-nuclear-plant-2025-08-08/

There are a number of reasons why this is a spectacularly bad idea.

The short reader for the lazy is this – we don’t have the money, and new build nuclear needs a lot of that. We also have entrenched corruption both in terms of government, and in terms of BEE tax, which will double or triple the eventual costs. Solar and storage makes more sense, especially in a sun drenched country like South Africa.

Some key points below:

High Costs and Length Build Times: Nuclear projects are expensive. There is a substantial investment required for nuclear power construction, in both CAPEX and OPEX. The low costs of solar energy paired with low cost storage makes renewables far more cost effective vs Nuclear. Nuclear builds take a decade or longer to build, vs Renewables, which can be generating in months.

Funding, Corruption and Political Challenges: The impact of the B-BEE policy on nuclear investment, corruption within the ANC government, and the ANC’s unbelievably stupid financial policies make this a definitive no-go. We can’t afford it. We definitely can’t afford it with B-BEE costs and corruption costs added on top.

Transmission Infrastructure: We lag in high-voltage direct current (HVDC) transmission infrastructure, which would be needed to transport the power to where it’s needed.

Grid Reliability and Flexibility: Nuclear power is inflexible generation, and doesn’t play nicely with others; it requires continuous high-capacity operation.


If Nuclear goes ahead it will both bankrupt the country, and saddle Eskom with insane costs. 
The government on the other hand will be keen, as there will be enough theft to make the Gupta’s look like petty thieves.

Yeah yeah, you just don’t like Nuclear, or <insert some other deflectionary excuse> here, I hear you say.

Well… the longer answer follows:

Nuclear is inflexible generation, which is why it’s used for “baseload” (i.e. to cover minimal load).
Ah, but we need baseload to cover X demand. No, you need enough generation to cover X demand. The type of generation is moot (other than factors such as cost), as long as you have some available.

Nuclear is best used for baseload, as you need to run Nuclear at its highest capacity factor in order to make it financially feasible. i.e. you want to run your Nuclear plant at 100% capacity all the time otherwise Nuclear electricity costs are too expensive. You also need sufficient dispatchable generation alongside your Nuclear in case of issues, as the grid will be rather unhappy if 900MW of power suddenly drops off, and nothing is there to back it up. Inflexible generation pairs well with more inflexible generation (the general term is dispatchable, but I prefer inflexible, as it makes things clearer).

See Page 6 onward here for more detail on what Nuclear needs are in a grid, and how its typically implemented – https://www.iaea.org/sites/default/files/gc/gc53inf-3-att5_en.pdf

Dispatchable generation is needed in a modern grid, as you do need a mixed grid to cover various scenarios. It unfortunately isn’t the cheapest thing to build if you’re looking to use the traditional dispatchables of Coal or Nuclear. Solar, Wind, or even the dispatchable Solar + battery are cheaper than Coal, and much much cheaper than new build nuclear.

Dispatchable generation is good, as you can rely on it, but it’s a bit of a tease. It looks hot in that red dress, or tight tuxedo, but is going to cost you a fortune to keep.

The thing we actually want is reliability – yes, your wife or husband may not be as exciting at the hottie or stud, but they’re there for you.

Our real needs are met by the middle guy in the image below:

(Image stolen from https://rmi.org/reality-check-dispatchability-and-reliability-are-not-the-same-thing/)

Nuclear is supposed to be reliable. Ours isn’t. Koeberg has an EAF (Energy Availability Factor) of 65%, which is similar to our Coal EAF. That means that close to 40% of the time, we can’t rely on South Africa’s dispatchable generation.


More reading on dispatchability vs reliability here.
https://rmi.org/reality-check-dispatchability-and-reliability-are-not-the-same-thing/

“Yes, but a new build will be reliable” – I hear you say. Well, I put it to you, that a new build nuclear will be 0% EAF for at least 10 years, likely longer, as Nuclear doesn’t generate anything until it’s built.
Other cheaper generation can be up and running in months (Solar, Solar + Battery, Wind, Gas Turbines etc), and will have much better EAF.

Ok, so we don’t need inflexible generation, we need reliable generation, and we need various types of that.

So…, what else is wrong with Nuclear?

Well, build time is our next set of problems, as is eventual cost.

Nuclear takes at least 10 years to build, even in western countries. As it stands, Nuclear is already more expensive than other generation. Add in a minimum of 10 years of financial payments before you even make a kw in generation, and it starts looking much much worse.

We need more generation now, and at costs that are affordable for us, and for future electricity consumers. Nuclear is already too expensive, and current costs differentials for battery storage are making new nuclear uneconomical. Future cost differentials will be downright silly. Yes, in 10 years (if we’re lucky), we may have a nuclear plant, but running costs will make it even sillier to use. We have issues with current electricity pricing. Imagine having to subsidise an uneconomical Nuclear build on top of that. This, is one of the reasons Nuclear builds are always government subsidised.

The private sector will never build nuclear, as it doesn’t make financial sense.
Worldwide – government needs to fund and insure Nuclear, as the private sector isn’t stupid. Private companies are happy to build, but no-one is going to build and run a plant without government backing, as you’ll lose money. New renewable capacity is cheaper and produces returns far quicker, so the money follows that route. 

Build costs in South Africa, will be a complete disaster with so much embedded corruption, that it will make the Gupta’s look like modest thieves. In South Africa, a large Nuclear build will be prime theft time for the ANC. I don’t have to look too far to explicitly prove this – see https://www.onlinetenders.co.za/news/uproar-over-eskom-bribes for some coverage of the ANC doing exactly what they always do.

We will also have the added “benefit” of B-BEE, and associated government policies. B-BEE is going to add at least 30%- x00% to Nuclear build costs, for zero benefit to the poor. Don’t like my take?

William Gumede is quoted as saying: “Black Economic Empowerment has led to increased poverty, unemployment and inequality in South Africa. Over R1 trillion has moved between fewer than 100 (politically connected) individuals since 1994. The same people have been empowered and re-empowered over and over again for decades.” 

https://irr.org.za/media/unmasking-south-africa2019s-bee-billionaires-who2019s-in-the-power-circle-viv-vermaak-biznews

Worse still, the ANC doubles down on bad ideas..
https://www.sakeliga.org.za/en/the-third-wave-of-bee-what-it-is-and-why-it-must-be-stopped/

That is the truth of B-BEE. I don’t want to go into race politics, here though.

Nuclear is already super expensive to build, even without B-BEE – See UK’s new build nuclear for an example. The price keeps going up and up and up, and… up. UK haven’t even built the thing, and it’s already projected to be uneconomical to run. UK’s already exhorbitant build costs will be double or likely triple, which makes Nuclear even worse than it already is in terms of costs.
https://www.reuters.com/business/energy/uks-sizewell-c-nuclear-project-cost-rise-near-50-billion-ft-reports-2025-01-14/


Even “normal” Nuclear build costs are just not sustainable, I’m lazy, so I’ll just link the below for further reading.

Build costs and interplay with other generation –
https://www.worldnuclearreport.org/Power-Play-The-Economics-Of-Nuclear-Vs-Renewables

Running costs aren’t even that cheap – https://www.nei.org/resources/reports-briefs/nuclear-costs-in-context 

Nuclear running costs overseas are more expensive than renewable with generation. Battery pricing is dropping year on year, so that pricing discrepancy is only going to increase, as Nuclear becomes even worse cost wise.

Volts has a good writeup on that here – https://transcripts.volts.wtf/solar-storage-is-so-much-farther-along-than-you-think

If we look toward Australia, which is extremely similar to South Africa in terms of sun – even existing Coal plants are going the Solar + Storage route.

https://reneweconomy.com.au/half-the-price-a-lot-more-dense-but-still-very-smart-why-baseload-giant-has-doubled-down-on-big-batteries/

https://reneweconomy.com.au/agl-says-batteries-will-be-its-new-cash-cow-as-coal-fades-in-switch-from-baseload-to-grid-flexibility/

Makes sense – as you have existing grid capacity, and it’s cheaper to run vs coal.

What should we be doing – well…, not Nuclear, thats for sure!

I’ll briefly go into what we should be investing in. I have covered this before, so this is a light rehash.

Daytime use is peak use.

We should be adding GW of solar, even without storage.
This will get rid of daytime generation requirements to run coal.

Some battery storage can be added to handle evening peak, and daytime intermittency, but not that much is needed. Oz has some good studies on that, as they started with battery for peaker plants, then expanded past that.

With sufficient generation, we would then have enough headroom to use pumped storage properly, instead of the current misuse. We should be priming pumped storage for morning and evening peaks…

Coal can then be relegated to standby generation, or converted to Solar + Battery as is happening overseas. Coal would just be needed for peaking, or overnight generation – i.e. morning and evening peaks (if we still need to cover and pumped storage doesn’t).

We have Coal, so it would be a relevant mechanism in South Africa’s energy system, especially during transition to other generation.

Our next issue is transmission – we sorely need more HVDC lines being built. Ironically we were industry leaders in that, but now… well, bring China in and have them build our infrastructure. They are the world leaders now.

Eskom is actually fairly good at the transmission part, but investment is needed.

We would be better off having 1-2 years of load shedding and spending that diesel money on infrastructure and solar. Mantashe and the rest of the ANC make far too much money siphoning off from shitty solutions to make that happen.

I could go on for a long long time, suffice to say, issues are solvable, but we need a harsh mistress in place. Minister Ramokgopa is not that.

Further reading

https://news.bloomberglaw.com/environment-and-energy/why-nuclear-power-is-bad-for-your-wallet-and-the-climate

Key Excerpt:

Does climate protection need more nuclear power? No—just the opposite. Saving the most carbon per dollar and per year requires not just generators that burn no fossil fuel, but also those deployable with the least cost and time. Those aren’t nuclear.

Loading