The consumers can't afford it

The Integrated Resource Plan 2019 (IRP 2019), the government's electricity plan, looked at various energy scenarios. The least-cost option (Scenario 1) allowed for an unlimited amount of renewable energy generation to be built. In Scenario 1, the cost of electricity was estimated at R1.35/kWh, compared to R1.51/kWh for Scenario 7 which included nuclear power generation.

Thus, limiting the renewables and adding in nuclear would result in an increase of 12% in the electricity price.

The country can't afford it

The Minister's determination states that Eskom or another organ of state, or the state in combination with a business, will generate the electricity, and procure the new power station(s). This means the state will be funding the building of nuclear power plants. As nuclear has the most expensive capital costs of any generation technology, OUTA questions why the state must bear these costs.

In the last attempt at a nuclear build in 2015, the government's agreement with Russia stated that the new build would go ahead without any agreement about financing, which would leave South Africa at risk of a mega-infrastructure capital-intensive project it could not afford. OUTA is concerned that a similar scenario is happening now.

What's "affordable"?

The government has repeatedly said that nuclear power will be considered "at a pace and scale that the country can afford". But nowhere in that determination has government given any indication of what it considers "affordable" or even how much nuclear power will cost.

Planning a nuclear new build takes no account of the fiscal crisis.

National Treasury DG Dondo Mogojane described the situation in October 2020, in the medium-term Budget:

“For several years, the National Treasury has been warning that an absence of fiscal space would leave South Africa vulnerable to external shocks. That risk is now a reality. At the time of the 2020 Budget, economic growth was already low and the fiscal position had deteriorated significantly. South Africa has begun heading into a debt spiral.

“Government is spending far more than it collects in revenue. As a result, debt has mushroomed. A failure to halt and reverse this pattern will harm the livelihoods of South Africans for many years to come. Left unchecked, the interest payments on that debt will become one of government’s largest expenditure items over the medium term. An ever-increasing share of tax revenue will not go to hospitals, schools or social grants – instead, it will be transferred to bondholders.

“Cabinet has resolved to reverse this pattern."

Treasury predicted that gross national government debt would increase from R3.6 trillion (63.5 percent of GDP) in 2019/20 to R3.97 trillion (81.8 percent of GDP) in 2020/21 and to R4.83 trillion (86 percent of GDP) in 2022/23.

Debt-service costs are expected to reach R301.1 billion in 2022/23.

South Africa needs to see government plans to control spending, not to increase it by introducing an unaffordable vanity project. 

South Africa has a disturbing history of mega-projects which get out of control

Eskom's two coal-fired power stations are a sobering example: Medupi was supposed to cost R80 billion but is now estimated as reaching R234 billion (about three times the original estimate), and Kusile was supposed to cost R69.1 billion but now is expected to be R160 billion (about 2.3 times the original estimate). Neither are finished, both have design defects which have added to the costs, and both are years behind schedule.

What's this about Koeberg being the cheapest electricity?

OUTA questions the claims that Koeberg provides the cheapest electricity. This narrative compares Koeberg's electricity, calculated by excluding the build costs as these are long paid off, to the cost of renewables which are still paying off the building costs. Such comparisons also use the most expensive renewables costs, from the early renewables, although this is technology which is reducing in price. This is an unrealistic comparison.

The government isn't saying how much it thinks nuclear power would cost. 

But even using government's 2015 numbers, building just 2 500 MW could cost up to R281.25 billion, excluding finance costs.

IRP 2019 used a study carried out in 2013 (by consultants Ingérop) for its nuclear prices. The wind and solar PV costs were from the South African Renewable Energy Independent Power Producers Procurement Programme. These are old costs and need updating.

Nuclear power stations take a long time to build. To build one, the country or company takes out a loan and so has to pay back the interest on the loan. The longer a power station takes to build, the more expensive it will be. The cost of building nuclear power stations increases substantially as they take a minimum of 10 years from placing the reactor order to commercial operation, whereas wind and solar power stations take only a couple of years to build.

The Cabinet calculations

In December 2015, Cabinet approved plans to go ahead with buying 9 600 MW of nuclear power, apparently envisaged as a series of small power stations. The Cabinet memorandum on which this decision was based used a lowest-scenario cost of $2 500 per kilowatt (kW) of build and a highest-scenario cost of $7 500/kW. This gave them an estimate of R240 billion to R720 billion to build 9 600 MW. But these figures are the overnight capital cost only, which is the cost calculated as though the power station was built overnight and no funds had to be borrowed, so they excluded finance costs. The Cabinet also used an exchange rate of $1 to R10, which was two years out of date at the time.

 The Cabinet memo emerged in the Zondo Commission hearings, as an annexure to evidence given by former Finance Minister Nhlanhla Nene. This annexure is here.

So how much could it cost?

Cabinet's best guess, in December 2015: R240bn to R720bn for 9 600 MW.

Using Cabinet's 2015 numbers: R62.5bn to R187.5bn for 2 500 MW.

Using Cabinet's 2015 costs per kW to build, with current exchange rates ($1 to R15, January 2021): R93.75bn to R281.25bn for 2 500 MW.

Using Cabinet's 2015 costs per kW to build, with current exchange rates, plus 25% finance costs and 10% owners' development costs: up to R379.688bn for 2500 MW.

It seems that government intends to build Small Modular Reactors (SMRs).

No details are available from government.

Globally, there are a lack of commercially available examples to follow, mean that the build cost trajectory is likely to follow that of a NASA rocket launch.

According to the World Nuclear Industry Status Report 2020 (WNISR 2020): “Small Modular Reactors or SMRs have made little progress ever since the first WNISR assessment in 2015, as this edition’s update concludes: ‘delays, poor economics, and the increased availability of low-carbon alternatives at rapidly decreasing cost plague these technologies as well, and there is no need to wait with bated breath for SMRs to be deployed’."

The Integrated Resource Plan 2019 is the government's energy plan. It was published in October 2019.

It's the plan for the electricity generation needed up to 2030. It doesn't include a nuclear power station.

This is government's own plan, which went through public consultation and was supposed to be based on what South Africa can afford.

What the IRP 2019 says on nuclear

This plan says this on nuclear power generation:

Koeberg nuclear power station reaches the end of its 40-year design life in 2024. Work will be done to extend Koeberg's design life and licence for another 20 years.

This is the decision on any new nuclear build:

"Decision 8: Commence preparations for a nuclear build programme to the extent of 2 500 MW at a pace and scale that the country can afford because it is a no-regret option."

The IRP 2019 is here.

The plan: preparation, not procurement

This plan talks about "preparations" for building nuclear power, not starting procurement.

OUTA believes government is manipulating this plan to suits its own agenda.

Where are the feasibility studies?

How can government consider such an enormous programme without presenting detailed feasibility studies and costing?

In December 2019, the Southern African Faith Communities' Environment Institute (SAFCEI), which had been part of overturning the 2015 nuclear deal, asked Eskom and the Department of Mineral Resources and Energy for any such affordability or economic feasibility studies. Nearly two months later there has been no meaningful response.

It is thus unclear if such studies are being kept secret or if they don't exist.

South Africa should build more renewable energy generation.

IRP 2019 makes it clear that renewable energy is the cheaper option for South Africa. Thus the most affordable way of increasing generation capacity is to issue additional section 34 determinations for additional renewable energy capacity. This was proposed by a previous minister to allow renewable energy power plants in the IRP to be commissioned earlier than planned to address system constraints.

Blocking the unrestrained building of renewables will keep South Africa short of power, which may be used as a justification for a nuclear new build.

The baseload argument no longer applies

There is a narrative that nuclear build is essential to provide baseload power, which is a constant minimum supply of power for the grid, and that solar and wind cannot provide this.

But this is an out-of-date notion.

What is needed is a constant minimum supply of power, but this does not mean it must come from a single source. Renewables can be scattered around the country, taking advantage of different weather conditions. In the same way that a factory operating 24/7 does not require its entire workforce to be on site 24/7, power stations do not have to all be generating at the same time.

As long ago as 2015, the International Renewable Energy Agency wrote this:

"An oft-heard critique of renewable power generation is that renewable options are unsuitable for baseload supply, therefore fossil power and nuclear power are needed. This critique is misleading. Baseload is a demand characteristic, not a supply technology characteristic. Nuclear or coal power plants are operated in baseload mode simply because: i) they are not technically capable of operating in a more variable mode and ii) they must rely on high utilisation to recover their high investment costs.

"In the future power system, the value of baseload will decrease. With higher shares of renewable power, particularly from variable sources such as wind and solar, supply and demand will be matched in a much more concerted and flexible way. Variable renewable power generation can ideally be combined with smart-grid technologies, demand response, energy storage and more flexible generation technologies, including gas power plants and dispatchable renewable power supply options. A flexible, renewables-based power system is not only reliable, but also economically efficient."

Assuming a cost of about R100bn to R280bn to build 2 500 MW of nuclear power (see #NoNewNuclear: How much would it cost?), we looked at what else South Africa could spend this money on.

This is the estimated cost excluding finance costs and owners' development costs.

What could South Africa do with this money instead?

R100bn

• At R200 000 per low-cost house, R100bn could provide 500 000 families with homes. (In September 2018, the Minister of Human Settlements told Parliament that provinces spent up to R184 000 to build a fully subsidised home: see here.)

• Pay off the municipal arrear debt of R28bn to Eskom (see Eskom's 2020 financial results here) and have enough left to install solar water geysers in 2.88 million homes (assuming a cost of R25 000 per geyser).

• Get the schools fixed, so children have safe, usable schools with proper toilets, by adding R10bn a year for ten years in funding to the Education Infrastructure Grant and the School Infrastructure Backlogs Grant. These grants were supposed to get R40.6bn over three years (before the Covid-19 reallocations cut these grants), less than R15bn a year combined. See the Budget Review 2020 here.

• Add another R10bn a year to the budgets for the central hospitals for ten years (about 20% extra a year). The joint budget for the central hospitals (these are the big hospitals) is about R47bn in 2021/22. See the Budget Review 2020 here.

• Many people still do not have access to safe drinking water or adequate sewerage. The Department of Water and Sanitation's biggest projects have an average total cost of R2.825bn each (we used estimates in the department's budgets over the years for the total cost of the 17 mega projects). That R100bn could build 35 big water and sewerage projects.

• R100bn could provide 2 million households with water tanks at a generous cost of R50 000.

R280bn

This would really help Eskom pay towards its unaffordable debt, run up mainly for Medupi and Kusile. Eskom's debt by March 2020 was R483.682bn (see Eskom's 2020 financial results here). If Eskom can't pay this existing debt on power station builds, how will South Africa pay for a new nuclear power station? 

The state-owned National Nuclear Regulator is responsible for overseeing the nuclear sector.

But the Minister of Mineral Resources and Energy, who appoints the NNR board, has failed to include a civil society representative. OUTA believes this is an ominous move to reduce transparency.

The process for appointing the NNR board is legally required to be transparent, to strengthen oversight in this potentially dangerous industry. But the Cabinet has failed to honour this requirement. OUTA queried this and officials were slow to respond, then said that the civil society representatives legally required as part of the board will be identified in a restricted process. OUTA objects to this.

For more on this issue, see here.

The government previously tried to set up a nuclear power deal in secret.

The secrecy underlines the problems of affordability and corruption linked to that deal, and the dangers to the country of such a deal.

On 26 April 2017, this secret deal was overturned by the Western Cape High Court, as a result of legal action by Earthlife Africa and the Southern African Faith Communities' Environment Institute (SAFCEI). OUTA had supported SAFCEI in this application. 

A copy of this judgment is here.

OUTA's summary of this judgment is here.

NERSA has set aside just five days to consider and analyse public comment on the proposed nuclear build.

How is this sufficient?

If NERSA does not have sufficient time to adequately consider the submissions, how will it come to a rational decision?