Nuclear and gas ‘forced in' SA's draft updated electricity masterplan

Daily Maverick

08-05-2025

South Africa's draft Integrated Resource Plan 2024 (IRP) is 'deeply flawed' in both process and outcome, according to a panel of independent energy economists, climate policy specialists and engineers.
A group of eminent South African technical experts have roundly criticised the draft Integrated Resource Plan 2024 (IRP) – the country's electricity blueprint for the future – as being flawed in both process and outcome, with at least one expert saying that nuclear energy and gas were 'forced in' despite there being no techno-economic rationale for their inclusion.
These extra costs could be 'socialised' and filter through to taxpayers and electricity tariffs.
The Electricity Regulation Act defines the Integrated Resource Plan (IRP) as 'an indicative, forward-looking plan [established by the national sphere of government to give effect to] for electricity generation, which reflects national policy on electricity planning, which plan specifies the types of energy sources and technologies from which electricity may be generated and indicates the amount of electricity that is to be generated from each of such sources or technologies.'
The current draft of the IRP is being discussed at the National Economic Development and Labour Council (Nedlac) before Cabinet considers a final version later this year.
Subesh Pillay, acting director-general of the Ministry of Electricity and Energy, gave the opening address at a webinar focused on policy and planning considerations for energy and electricity in South Africa in the years to 2050.
He said conversations around the IRP were happening as the country was at an inflexion point from an energy perspective. Pillay said past energy debates were framed by scarcity and load-shedding, but improved generation performance now allowed 'long-term thinking' and 'evidence-based planning'.
However, other attendees would later suggest that the document best exemplifying this long-term thinking – the IRP – was anything but a display of 'evidence-based planning'.
Dr Grové Steyn is MD at Meridian Economics. He specialises in infrastructure regulation, policy and restructuring.
He said executive policy-making was bound by the constitutional principle of legality, which essentially required a minimum level of rationality and that least-cost power-system planning mattered greatly – tariffs were likely to increase as old Eskom coal plants were retired.
'Securing a reliable supply at the lowest possible economic and environmental cost is one of the most critical challenges in supporting our long-term development. This challenge is immensely complex and requires rigorous analysis.'
A credible IRP, Steyn said, was one where, in terms of process, 'outcomes must follow logically from the results of proper analysis, outcomes have to be evidence-based, it should be grounded in sound, transparent and well-documented techno-economic analysis.'
The IRP should also be consistent with other policy frameworks such as South Africa's climate commitments, local air quality regulations and broader socioeconomic goals.
Based on these considerations and others, he described the draft IRP as deeply flawed.
Integrated Resource Plan 2024 by Ethan van Diemen on Scribd
Nuclear 'forced in'
'So the IRP, as in earlier years, develops essentially a 'modelled optimised reference case'. In this case, this includes the existing committed public procurement projects and also the private sector plans that are very far advanced,' said Steyn.
He noted that the optimiser's 'emerging technology mix' pointed only to wind, solar PV, gas and storage, with no new nuclear, coal or pumped storage in the least-cost case.
'The modelled scenarios using the actual optimiser in the model does not build nuclear so the IRP modellers wanted to have a nuclear scenario and I presume the policy makers – the government – wanted that investigated and the only way they could do that was essentially to take away the other generation options that [were] available to the model to build the power system to meet demand.'
He said modellers disabled new gas options after 2030, in effect, forcing the model to 'build nuclear.'.
'That is how the nuclear scenario was built. So the nuclear case is not an outcome of the model's optimisation, it's a forced-in scenario.'
Gaps in the data
Steyn highlighted more flaws he said undermined the credibility of the IRP.
'It's astonishing that the document does not show data on the technology capacity, energy mix costs or emissions for the modelled scenarios or the actual proposed balanced plan. Technically, of course, this means that it does not qualify as an IRP in terms of the legal definition.'
'It's not clear to us how Nedlac can participate in a meaningful consultation process if this information is simply not in the document.'
'The assumptions about nuclear technology are overly optimistic… The analysis does not test compliance or alignment with other policies, whether it is our net-zero emissions objectives or our local air quality objectives, etc, no information is available on how we meet those requirements.'
'The key concern here is that the proposed balanced plan is not logically derived from the underlying analysis presented in the IRP document. The conclusions and the plan are not the result of the analysis presented in the IRP document,' Steyn said.
The IRP creates an illusion of science-based planning when its recommendations do not follow the evidence, Steyn argues.
'We have an IRP that neither went through a meaningful public consultation process nor benefited from review by independent experts, an IRP that contains lots of technical descriptions of our power system modelling, but does not adequately investigate the most critical uncertainties affecting future outcomes, an IRP that presents recommendations that do not follow logically from the analysis or are not otherwise substantiated or for which the technical and cost implications are not presented and, in fact, creates an illusion that the recommendations are based on scientific and economic principles when they are clearly not.'
'The technologies that are uneconomically forced into the plan, especially gas and nuclear, will actually rely on public procurement with limited competition, where risks and costs of these investments will be socialised,' Steyn warned.
'Unfortunately, as we all know, South Africa's recent empirical experience with these types of projects provides stark warnings about the enormous risk for cost overruns, opportunities for rent-seeking and even corruption.'
Lebogang Mulaisi, executive manager responsible for policy and research at the Presidential Climate Commission (PCC), questioned whether short notice for stakeholder meetings constituted 'meaningful consultation', noting no public record of how inputs changed the document had been made available to her.
Policy-based evidence-making
Professor Harald Winkler with the Policy Research in International Services and Manufacturing (Prism) unit in the School of Economics at the University of Cape Town (UCT) summarised the views of others and shared his thoughts about the draft plan.
'I don't think the current draft of the IRP meets the standards of a good IRP in terms of its outcome, and in terms of process.'
'The fundamental approach should be one of evidence-based policy making based on clearly stated data and assumptions,' said Winkler.
Winkler, whose research includes equitable transitions away from fossil fuels and low-emission development strategies, continued, 'It's remarkably clear that this is not a least-cost plan, but even more in terms of process, the balanced plan is not logically derived from the underlying analysis.
'It's more like an exercise in policy-based evidence-making… which is the opposite of what we need, which means policy determines outcome.'
'It must be the other way around, policy must be evidence-based, plans must be based on strong evidence, and I think that was not the case.'
Chris Yelland, managing director at EE Business Intelligence, said, 'The current draft IRP currently before Nedlac is deeply flawed both in terms of process and in terms of substance. In terms of collaboration, in terms of consultation and the way it's been put together, I think it's deeply lacking.'
'I take heart from one thing… We're moving from a time of prescriptive IRPs into a time of indicative IRPs because, ultimately, Eskom and government don't have the balance sheet to finance this, it relies on external finance and government and National Treasury do not have the appetite to give government guarantees, and therefore everything hinges upon the business case of the different proposals.
'It's easy to put something in an IRP like 10,000 megawatts of new nuclear power that was in IRP 2010 that never happened, and we're moving to a stage where if it does not make business sense, it will not fly. And because it will have to fly without government guarantees, and external investment will insist on a business case that makes sense for their investment… I just don't see some of the ideas put in the IRP as flying.'
'They'll be in there, but they will not happen. So, in a way, the situation is moving to a stage where it just has to make sense as opposed to being ideologically driven,' said Yelland.
'It worries me when a Minister of Energy and Electricity says long before the finalisation of the IRP that 'there will be significantly new nuclear and gas in the IRP' because it presents an ideological position as opposed to a rational position one has taken after doing a socio- and techno-economic study like the IRP should be.' DM