It took 33 days of continuous loadshedding in June and July this year, to finally spur president Cyril Ramaphosa into action.
“After more than a decade without a reliable electricity supply, South Africans are justifiably frustrated and angry. They are fed up,” he said in a televised address. “The crisis that we are facing requires that we should take bold, courageous and decisive action to close the electricity gap.”
In the days that followed, the newly-created National Energy Crisis Committee (Necom) announced concrete steps intended to bring a permanent end to loadshedding.
In many ways, the plan is revolutionary: it will bring 15 GW of new wind, solar and storage online, and force energy minister Gwede Mantashe to loosen his grip on energy planning.
In some ways though, it is painfully familiar: “Government will focus on: First, improving the performance of Eskom’s existing power stations,” minister in the presidency, Mondli Gungubele, announced.
In a follow up response, Mantashe said: “[T]he belief is improving the operational efficiency on power plants already on the ground can get us out of the woods much quicker than any other option.”
The problem is that this has been “the plan” for years, but despite billions poured into the aging coal fleet, its performance has continued to decline, reaching an all-time low of 55% availability in 2022.
The decline, as we detailed in part 1 and part 2 of this series, has left a gaping hole in our energy supply and is the primary cause of loadshedding today.
So, the question we set out to answer in the final part of our Eskom trilogy is: does it make sense to reinvest in the coal fleet, or is it time to let the coal fleet die?
The top six
Currently, Eskom’s fleet of 25 power stations is only available to produce electricity 56% of the time (12-month rolling average: 58.7%).
In a bid to reverse this decline, Eskom plans to spend R34-billion a year on maintenance over the next five years. At R170-billion, this is significantly more than the R144-billion spent in the previous five-year period.
Most of the extra budget will be spent on six power stations – dubbed the “top six” – which will receive VIP treatment, meaning the money, skills and time they need to carry out in-depth, restorative maintenance.
The “top six” coal stations – Tutuka, Majuba, Matla, Kusile, Kendal and Duvha – are amongst the worst performers in Eskom’s fleet, with an average energy availability factor (EAF) of 40%.
Tutuka, for instance, has an EAF of only 28%, meaning it produces about as much electricity as a wind or solar plant of the same size, but on a far more unpredictable schedule.
The six VIP stations are all scheduled to keep running after 2030. Tutuka, for instance, is supposed to shut down its last unit in 2040, but the likelihood it will last another 18 years is slim.
Professor Mark Swilling, the chair of the Development Bank and a seasoned energy expert, recently described it as “probably irredeemably corrupt and therefore damaged beyond repair”.
However, Eskom’s hope is that spending an additional R5.2-billion a year on maintenance – mostly at these six stations – can rein in breakdowns and put desperately-needed electrons back on the grid: “[T]his would improve EAF by approximately 7%,” Eskom told us in a written response.
Not overnight, however: a presentation to Parliament in late August indicates that Eskom anticipates that the EAF of the fleet could recover to 61% in 6 months, 63% in a year and so on, only reaching 66% by 2025.
But this is a gamble. With independent power producers we know the cost of the electricity we are buying in rands and cents per gigawatt hour. With the Eskom coal fleet, we are handing over billions, with a hope and a prayer that we are not just throwing good money after bad.
And if the plan fails, Eskom will be deeper in debt and no closer to providing the country with a stable supply of electricity.
The cost of coal
Eskom chief executive André de Ruyter has made the Reliability Maintenance Recovery programme a key part of his strategy to end loadshedding, but the return on investment is yet to materialise.
Initially, chief operating officer Jan Oberholzer predicted that the EAF would rebound within 18 months. Now, 24 months later, he says it will likely take another 18 months to show results.
If the plan works and Eskom can squeeze another 7% availability out of the existing fleet, it should add electricity to the grid at between 40c and 50c/kWh, on par with prices paid for new wind and solar.
But those figures are misleading because they are subsidised by the large, sunk costs of the existing coal fleet, which generates power at over R1/kWh. And while these cheaper electrons would bring down the cost of coal-fired power, this would not be enough to make coal a competitive source of electricity.
AmaBhungane’s own calculations suggest that shutting down under-performing stations like Tutuka, Duvha and Kendal and replacing their capacity with renewables would be a far cheaper and more reliable option.
Eskom’s R170-billion budget also has no room for the technology needed to curb air pollutants, such as SOx, NOx and particulate matter. At R300-billion those may sound like nice-to-haves but nine coal power plants will soon lose their licence to operate unless Eskom installs abatement technology or secures another “legal indulgence” from the regulator.
For financial reasons, if nothing else, coal is on a path to extinction.
Stage-managed decline
Knowing this, why would we spend another R170-billion on the coal fleet?
We put this to Eskom chief executive André De Ruyter when we interviewed him in July. His view is that the transition to a renewables-dominant system is “inevitable” but that for now Eskom must invest more in maintaining the coal fleet because it cannot pull the plug overnight: “[W]e simply can’t deploy enough renewable energy capacity quickly enough,” he explained.
Under Eskom’s current plan, half of its 46-gigawatt (GW) fleet will be shut down by 2035, including nine coal-fired power stations. This is slightly more aggressive than government’s official plan that would shut down eight by 2035.
De Ruyter has long urged government to embrace the global transition away from high carbon fossil fuels like coal. He sees his role as shepherding Eskom through a break-up that will leave it smaller – 12 coal power plants and thousands of employees lighter – but alive.
To others, De Ruyter is a grim reaper bringing death and the destruction of livelihoods to the Mpumalanga coal fields and, significantly, wiping out roughly R229-billion in future sales for a new generation of coal barons.
“I’m accused of being deployed to break Eskom, so that Patrice Motsepe can make money from renewable energy to line the President’s pockets and all sorts of conspiracy theories – which I can tell you are absolutely unfounded,” he told us.
Eskom has looked at all the options, De Ruyter said, including whether it is cost-effective to extend the life of the coal fleet. “[T]he one that we’ve landed on … is this staged-managed deliberate decline, not ‘run it until it blows up’, which comes with its own risks.”
To restore Hendrina power station, for instance, and run it for five years past its 2026 shutdown date would cost roughly R12-billion, De Ruyter told us.
“If you do the math, it doesn’t make sense. And therefore it’s far better to spend the R12-billion on enabling cleaner and greener generation capacity to be added by private sector by investing in the grid, … but also by ensuring that from a system operator perspective that we run a grid that can supply the 50 hertz 24/7, which is a huge challenge, significantly exacerbated by … renewable energies.”
In terms of value for money, “this is the best strategy … that we have been able to come up with,” De Ruyter said.
‘The uncertainty of the plan must be understood’
However, executing this maintenance-for-GW strategy means accepting the risk of more severe loadshedding in the short-term.
Eskom desperately needs another 4 to 6GW of additional capacity – new power plants, in other words – to shoulder the burden of providing electricity to the country while it conducts maintenance.
But thanks to the Department of Mineral Resources and Energy (DMRE) slow-walking the rollout of renewables, those are, at best, two years away.
“Maintenance is not a solution to meaningfully shift the dial on loadshedding…” analyst Peter Attard Montalto of Intellidex wrote in a recent briefing note to clients. “Eskom’s strategy of doing just enough to try and stabilise the system whilst not exacerbating loadshedding is the correct one given the age of the fleet and historic lack of maintenance as well as financial constraints.”
In the absence of the 4 to 6GW of additional capacity, Eskom is pressing ahead with their strategy anyway.
“The uncertainty of the plan must be clearly communicated and understood by all stakeholders including government and the public,” Eskom officials warned Parliament in late August.
“The plan is ‘tight’ and any significant outage slips” – where maintenance runs over schedule – “will have a knock-on effect that will influence the plan from that point forward.”
To give the plan the best chance of success, Eskom built in a bigger buffer zone for breakdowns, assuming that 13 of the system’s 49 GW would be lost to breakdowns at any one time, up from the previous assumption of 10GW.
Eskom assured Parliament that if this new base-case scenario materialised, there should be no loadshedding in the upcoming months of September and October.
Instead, September would deliver 25 days of loadshedding and October 27 days of up to stage 6 loadshedding.
When we interviewed De Ruyter back in July, his Reliability Maintenance strategy seemed to have the support of government. But an energy crisis years in the making was not going to be solved overnight. And as loadshedding intensified, so too did calls for De Ruyter’s head.
Marching orders
“This is not about any political instruction, this is about what the country needs,” public enterprises minister Pravin Gordhan told journalists on a Friday afternoon in late September.
What the country needed, in government’s view, was to end loadshedding and restore Eskom’s aging fleet to 75% EAF, far more than the 66% in De Ruyter’s already ambitious plan.
“[I]f experts tell me and others that it is possible to get there, then every effort must be made … to reach as close to 75% … as is possible,” Gordhan said.
Government has a long and inglorious history of setting unrealistic EAF targets for Eskom. The 2019 Integrated Resource Plan – the country’s official energy roadmap – is based on the flawed assumption that Eskom’s fleet would perform at 75%.
“I think privately many engineers in Eskom were shaking their heads at this,” De Ruyter told us when we asked him about the 75% figure. “[But] this was the answer that government wanted, and therefore that was the answer that government got.”
- Read more candid comments from De Ruyter in The collapse of old king coal: part 2.
It is easy to see why improving the EAF is so alluring: an EAF of 75% would be like adding 12.5GW of new capacity to the grid – far more than the 6GW shortfall causing the current energy crisis.
However, there is very little evidence that this can be achieved. In 2017, Eskom predicted that the country would have an 8GW surplus of capacity by 2022 and an EAF of 80%. Each year since, the EAF number has been revised down but, without fail, has always been too optimistic.
Yet despite the bruising experience of the last five years, government seems determined to plough ahead with the same strategy.
“Normally, shareholder compacts are not just imposed – there’s also arm-wrestling and debates and discussions about what is realistically possible – but the job of a shareholder is not to accept at face value the lowest common denominator,” Gordhan told journalists.
“If it is possible, to get more megawatts out of the system … then clearly, both the board and the management and the shareholder have a responsibility to this country to ensure that the optimal is done.”
As for De Ruyter, his fate would be left in the hands of Eskom’s new board, Gordhan said.
The comeback
Within days of being appointed, the new Eskom chair Mpho Makwana weighed in. Makwana is an Eskom veteran: he served as chair from 2009 to 2011 and briefly as CEO.
“The immediate priority is to keep the lights on,” Makwana said in an interview a few days later. “We have to grapple with how to return … the EAF to healthy levels. Under normal conditions, the EAF is 86%.”
It is worth noting that the EAF has not been at 86% since 2007, before loadshedding began. During Makwana’s tenure, Eskom implemented the notorious ‘keeping the lights on’ policy, which resulted in critical maintenance being sacrificed in order to avoid loadshedding.
- Read The collapse of old king coal, part 1: “[B]ack in the day you never had to run a unit for extended periods with a boiler tube leak … you shut it down [with] very little consequential damage … Post ‘keeping the lights on’ or once we adopted that, that all changed.”
Although Makwana calls government’s 75% target “a tall order”, he evidently believes it is achievable. (Makwana did not respond to our request for an interview.)
“[T]he challenge is to reignite in Eskom employees a passion for serving their country and its economy. Related to this is the idea of reigniting a sense of internal competitiveness between power stations … to see who maintains the highest EAF levels. This would get us well on the way to maintaining healthy [EAF] across our operations,” he said.
Open revolt
Others disagree and are saying so in public.
“There’s a danger here of chasing unrealistic EAF targets,” outgoing Eskom board member Busisiwe Mavuso wrote in a recent op-ed.
Mavuso, who is also the chief executive of Business Leadership South Africa, warned: “Eskom doesn’t have the money that would be required to undertake a high EAF strategy … This wouldn’t be a bit of maintenance around the edges mut would be paying to rebuild the old plants because they’re in such a poor condition.”
Swilling, the DBSA chair and head of the Centre for Sustainable Transitions at Stellenbosch University, echoed this: “I have an uneasy feeling that some recent decisions in the appointment of the new board are inappropriate, and I’m referring specifically to the mandate to this board to achieve a 75% energy availability.”
He compares it to trying to fix an aeroplane: “If you want to fix an aeroplane, you can’t do it while flying. You can only fix an aeroplane by landing it, putting it into a hangar and opening it up. That takes time. But that’s what you have to do [with the coal fleet]. You have to take them off the grid in a systematic way, shut them down, repair them, rehabilitate them, and bring them back onto the grid.
“But you can’t do that because you would plunge [South Africa] into permanent loadshedding.”
Asked what it would cost in additional maintenance to get the EAF to 75%, Eskom said this figure “has not been accurately determined”.
The coal lobby
De Ruyter his been coy about what he thinks the fleet can realistically achieve, but has said enough to put him on a collision course with the new board.
Last year, he told TechCentral that “given the age of the fleet it would be unrealistic to expect a sustained performance above 75%”.
Of its more modest target of getting to 67%, Eskom recently told us it is “committed to doing everything we can to achieve this. However, without adequate space (and funding) … any performance improvement will be a challenge.”
Eskom’s system adequacy report, released over the weekend, says a “more likely” scenario is that we will see an average EAF of 58% over the next five years: “South Africa has been faced with chronic power supply constraints for over 10 years. The current year has been the worst yet, and it is evident from this study that the situation will worsen as the plant performance of Eskom’s fleet continue to trend downwards, power stations shut down, and demand grows.”
When we asked De Ruyter in July if it was time to give up the dream of improving the EAF, he said: “I think that that’s a fair challenge. But you but you also need to think about the reality of the environment in which Eskom operates. [If] we go out there and we say ‘all is lost, it’s all over’, not only will there be significant pressure on the shareholder to remove management – because that’s not an acceptable message – [but] there’s a very strong coal lobby.
“So we need to maintain a balancing act between, on the one hand, trying to arrest the rate of decline as much as we can. But on the other hand … deploying renewable energy [to] scale as quickly as possible.”
The EAF matters to the coal industry. The difference between 58% and 75% is potentially R7-billion a year in coal sales. Add in the accelerated shutdown of Tutuka, and Eskom’s current strategy could cost the coal industry R279-billion by 2050.
Mantashe, the self-proclaimed “coal fundamentalist”, has made it clear he opposes any attempt to shutdown coal-fired power stations, even a station like Tutuka which has an abysmal EAF of 29%.
“[Tutuka] has had a much higher EAF than 29%. The [Tutuka] problem must be found and resolved ASAP to improve its EAF … Until we understand the real reason behind EAF decline (which we are expecting the Necom work to uncover) it is wrong to assume this is an impossible task,” he told us in August.
There is no indication that either Gordhan or Makwana are trying to protect the coal industry with the 75% EAF target, but even the of strategy of shutting down 22GW of coal by 2035, including Tutuka – which De Ruyter and Oberholzer have trumpeted in public – is now in question.
Asked for a copy of the shutdown schedule, Eskom declined, saying the strategy “is due for a review by the newly-appointed board … the Board will in the coming months either confirm or revise the current Strategy.”
De Ruyter was remarkably frank with us when we interviewed him in July. But he was also aware of how much is at stake.
“[T]here’s a fine balancing act you need to play in transitioning to this new electricity industry,” he said. Investor confidence, employee morale, a repeat of the wildcat strikes that shut down power plants in June – these were all the things that played on his mind.
“[M]aybe you’ll say, ‘you’re playing your cards very close your chest and you’re not being fully transparent’, but I’m more than happy to let informed commentators make the conclusions on our behalf rather than me sticking my head above the parapet and getting it chopped off.”
He is not speaking metaphorically either. De Ruyter has received death threats and now travels with a bodyguard. Every time he talks about closing a coal-fired power station, he adds a new target to his back.
The war over coal begins
Shooting for 75% is not the problem. The problem is that Eskom and government have, in the past, put all their chips on achieving a 75% EAF and failed to plan for any other eventuality.
As we have tried to demonstrate with this series, we are in an energy crisis today because both refused to acknowledge that the coal fleet was collapsing, and clung to unrealistic projections of how much electricity it could produce.
As a result, South Africa has a 6GW shortfall today and an official energy plan that is full of holes.
It is assumed, for instance, that the 1.5GW of new coal in the IRP will never be built. Eskom estimates that new coal would be two to four times more expensive than renewables, and take 10 to 12 years to build, “resulting in more years of loadshedding”.
The emergency plan, announced by Ramaphosa, will bring forward the timeline for all the rest of the new capacity in the Integrated Resource Plan (IRP): instead of gradually building 15GW of renewables, 3GW of gas and 0.5GW of storage by 2030, we will now add it as soon as is logistically possible.
That might solve the short-term problem, but in the long-term the IRP still falls woefully short. And it is here that a battle is brewing.
On Friday, Eskom released its annual transmission plan, which looks at how the grid needs to expand over the next decade. Normally, the plan takes its cue from the IRP, but this year Eskom went off-script: ”[A] substantial amount of additional generation capacity, over and above what is reflected in the IRP2019, would be required by 2032 to meet the country’s needs,” Segomoco Scheppers, the managing director of Eskom Transmission, warned.
South Africa currently has around 6.5GW of renewables on the grid. Scheppers estimates that we will need another 53GW by 2032, mostly from renewables to keep the lights on. Those projects will largely be funded and built by the private sector, but Scheppers says Eskom Transmission will need at least R72-billion to expand the grid in the parts of the country with the best renewable resources. (De Ruyter puts the figures for transmission and distribution at R186-billion by 2030.)
If we want to accelerate the shutdown of coal – or if the performance of the coal fleet deteriorates further – we will need more, faster.
When Ramaphosa announced his emergency energy plan in July, the expectation was that government would adopt a single-minded approach to solving the energy crisis. “[W]e as a nation will have to grit our teeth as things get worse, while we all pull together to achieve this one unifying strategic mission” of ending loadshedding, Swilling wrote in an op-ed in May.
But instead of unity, what we are seeing is open warfare.
On Monday, finance minister Enoch Godongwana told a group of business leaders that Treasury would only agree to take over R400-billion of Eskom’s debt if the utility agreed to invest more money in “old reliable technologies” like nuclear and gas.
The DMRE has committed to drafting a new IRP to respond to the energy crisis, but it is unlikely to give up on coal. Instead, Mantashe wants to up the government’s high stakes gamble on coal by investing in carbon capture technology.
The hope, he said during the recent Council for Geosciences Summit, is that “our Just Energy Transition programme can be attained with one of our most valuable commodit[ies]”: “Our commitment to the international protocol on climate change remains resolute. Our transition from high to low carbon emissions might be achieved with coal as part of the solution, if the hypothesis of clean coal is proven.”
That is a big “if” and the consequences for the country would be unthinkable if the experiment fails: Eskom’s own research says that failing to cut carbon emissions will soon put 46% of South Africa’s exports at risk as other countries impose carbon taxes at their borders.
Even if we can resuscitate the ailing coal fleet, and even if we develop as-yet-unproven carbon capture technology, what will we achieve?
Coal is already more expensive than new generation technology. Add the cost of carbon capture and the cost of air pollution abatement technology and we may as well burn R100 notes in Eskom’s furnaces to keep the lights on.