WWF-SA, the autonomous member of the World Wide Fund for Nature, welcomes EE Publishers’ recent interview with Eskom’s David Nicholls and appreciated his candid response particularly on the question of costs.
At the outset a more general comment must be made: that the advisability of the proposed nuclear new-build programme must be situated in a broader debate about the future of Eskom, and the ability of the South African economy to generate the levels of growth and investment needed to justify a build of this scale.
Power utilities which continue to rely on the traditional business model of large centralised power generation, as Eskom does, are coming under increased pressure. The uptake of renewable energy technologies, the shift towards more decentralised power generation, and changing energy policies are just some of the factors driving utilities to reassess their business offering. PwC’s 13th Annual Global Power & Utilities Survey reports that 94% of the 53 power and utility companies surveyed predict a complete transformation of, or important changes to, the power utility business model.
Eskom’s inability or unwillingness to accept these trends, its insistence on maintaining the status quo and its reluctance to explore new models is most evident from its gung-ho commitment to the nuclear programme in its full 9,6 GW glory. This is problematic not only for Eskom but for the country as a whole. Given Eskom’s wide-ranging influence over power generation and by extension the economy, prudent decision making, not informed by the desire to maintain a stronghold over power generation, is critical. Investing in a technology that we may be tied into for the next 60 years, which is particularly susceptible to construction delays and cost overruns and which ignores global trends both in the nuclear industry and in the deployment of energy infrastructure, puts the country at serious risk if things go wrong.
This must also be viewed in the context of the current and emerging constraints facing Eskom. Inefficiencies within the power utility, political interference, politicised procurement processes, the difficulties it faces in managing its coal fleet, dealing with environmental challenges, its constant battle to recover full cost tariffs from customers and of course, its inability to finish large infrastructure projects on time are but a few of the constraints the utility currently faces. These are major red flags which cannot be ignored when considering the prudence of entering into an energy infrastructure project on the scale of the nuclear new-build programme.
It is encouraging that Nicholls acknowledges that it is only once bids have been submitted that realistic cost estimates will be available. His acknowledgement confirms that government itself simply does not know how much a build on this scale is going to cost. This highlights the critical importance of robust engagement with stakeholders across all sectors and with the public. It is these engagements that allow us to weigh up all the possible benefits or risks associated with the programme.
The overnight capital costs and the levelised cost of electricity (LCOE) that Nicholls is working with appear to be overly optimistic and do not take into account the potential risks associated with a build of this complexity, or what we should have learned from Medupi and Kusile. Again, Nicholls earlier statement, that we simply do not know how much it costs until bids are received, is important to keep in mind when interrogating his assumptions. Equally important is the fact that construction delays, cost overruns and the effect of inflation and interest rates significantly alter the equation as Medupi and Kusile, as well as numerous nuclear builds around the world have demonstrated. The World Nuclear Association asserts that capital costs, including accrued interest, account for around 65 to 85% of the levelised cost of a new nuclear plant.
Although it is notoriously difficult to anticipate these variables and factor them into calculations, prudence dictates that when dealing with an investment this large that we err on the side of caution or at the very least do not completely overestimate our own capabilities or plan for the lowest costs.
In this regard, the overnight capital cost of US$4500/kWh suggested by Nicholls is significantly lower than the
$5945/kWh that the US Energy Information Administration’s Capital Cost Estimates for Utility Scale Electricity Generation Plants (2016) suggests. Similarly, the R0,80 to R1,00 that Nicholls assumes for the LCOE is, with due respect, completely unrealistic if one looks at examples like Hinkley Point C in the UK or even the LCOE for new coal in South Africa. EE Publishers and the CSIR calculations of around R1,20 to R1,30/kWh are far more accuate.
Furthermore, to assume a construction period of six years, as Nicholls does, flies in the face of global experience. According to the 2016 World Nuclear Industry Status Report, of the 46 units that started up in ten countries since 2006, the average construction time was 10,4 years. South Africa’s experiences with the construction of Medupi and Kusile show that it seems overly optimistic to think we will be able to beat the world average.
Finally, there is absolutely no indication from the interview with Nicholls if and how Eskom is factoring in the full costs of a nuclear build. This would include long term waste management and decommissioning costs of the plants.
As to the localisation argument, history has consistently shown that the devil is in the detail. In South Korea foreign companies dominated and the localisation benefits were very low. For localisation to be achieved at any scale in South Africa we would need to build a very large fleet of reactors that can be sustained over a 15 to 20 year period. We would then be looking to compete in a highly consolidated market that may already have reached its plateau in terms of market penetration with existing Generation 3 technologies. The idea that we will become immediately export competitive after building two or three reactors is, with all due respect, spurious.
All of this aside, the bottom line is that there are a host of variables that must be considered when deciding on the advisability of a nuclear programme. These are not factored into overnight capital costs or LCOE calculations, but will be crucial in determining the success of the programme and the implications for the country. The cost of finance, particularly in the context of Eskom’s BB+ credit rating and the country’s BBB- credit rating, the possibility of cost overruns and delays, the current difficulties Eskom is facing and the changing nature of energy production globally are serious risks. Not to mention the complexity of the contractual arrangements associated with builds of this nature and the potential for long, costly legal disputes as we have seen with Areva and TVO Finland.
In determining our country’s energy future, these are matters which must be discussed. While indicative numbers are useful as a basis for debate, space must be created for these risks to be unpacked. Unfortunately, this has not happened. We are encouraged by Nicholls’ willingness to engage in this discussion with EE Publishers and we appeal to Eskom to keep this dialogue going. We also appeal to Eskom to consider more modular builds, which have far less risk for the country, maintain Eskom’s role, and keep it abreast of global developments. South Africa simply cannot afford to throw all of its eggs into one basket.
Send your comments to firstname.lastname@example.org
Source: EE plublishers