Maverick Citizen


We can’t just wave goodbye to Eskom – we need to fix it, and it’s possible 

We can’t just wave goodbye to Eskom – we need to fix it, and it’s possible 
The author questions why South Africa is not harnessing more wind and solar energy at speed and instead remains fixated on fixing Eskom. (Photo: iStock)

Although many South Africans yearn to dump the state-owned power utility, fixing and maintaining Eskom’s power stations to boost energy availability is the quickest way to stabilise supply in the short to medium term. Although not cheap (or fast), it’s probably the most affordable means of providing the dispatchable energy needed to create space for long-overdue maintenance. This is part 2 of a series of three articles that address South Africa’s energy crisis.

“Why are we not harnessing more wind and solar energy at speed? Instead, we remain fixated on fixing Eskom. It cannot be fixed. It, like every other state-owned entity, is just about beyond repair — tragically so.” 

This quote, from a recent Daily Maverick article by Freedom Under Law executive officer Judith February, reflects the sentiment of many who wish we could just say goodbye to Eskom as the solution to rolling blackouts. However, Eskom generates almost 90% of SA’s electricity and, while it may be an inconvenient truth, Eskom will be required to play a substantial role in South Africa’s energy sector for the foreseeable future. 

What is possibly even more surprising, however, is that fixing Eskom is probably the quickest and most affordable way (albeit not cheap) to address SA’s energy crisis. 

The energy crisis

In 2022, South Africa experienced rolling blackouts on more than 150 days, a 100% increase in the amount of electricity cut from 2021. It is anticipated that 2023 will be even worse, with energy expert Clyde Mallinson predicting “permanent load shedding at stage 4 levels through 2023”, as there is no short way of providing the dispatchable energy required to stabilise the power system and create the space for maintenance. It will probably take at least 12 to 24 months to achieve this. 

Against this dark reality, the growing calls for the private sector to save the day are easily understandable. So too are the demands for the accelerated rollout of rooftop solar, both to allow municipalities to procure energy from independent power producers (IPPs) and to incentivise private households to go solar. 

Read in Daily Maverick:Informal settlements could drive South Africa’s rooftop solar revolution” 

There are two primary ways for the private sector to add generative capacity — through government’s Renewable Energy Independent Power Producers Procurement Programme (REIPPPP) and via the rollout of private embedded generation. Let’s evaluate the prospects of these options and their implications. 

Failures of REIPPPP

REIPPPP, based on the auction model, was established in 2011. This model requires private renewable energy producers to submit bids in a bid window, with the cheapest bids being awarded contracts in the form of 20-year power purchase agreements (PPAs). 

These private producers do not use their own capital, but turn to banks for financing. Banks only provide financing to those projects deemed to be sufficiently profitable. 

Over the past decade, just more than 90 of these projects — also known as IPPs — have been connected to the grid. By mid-2022, they accounted for only 8% of installed capacity, yet they accounted for more than 30% of primary energy costs. Of the 2,600 MW meant to be generated by bid window 5 (BW5), less than 1,000 MW has reached financial closure and BW6’s expectation of 5200 MW is proving to be even more unsatisfactory. 

Cut-throat competition means that most of the preferred bids of BW5 are no longer sufficiently profitable. This is due to declining profit margins, as input costs are not falling at the same rate as the bidding prices. 

Nothing will be built if it’s not profitable enough. To reduce input costs, the government has slashed the local sourcing requirement for solar photovoltaic (PV) systems from 100% to 35%. 

BW6 is facing challenges too. Less than 20% of the generative capacity expected to be procured from this bid window can proceed, given grid constraints in areas where renewable energy is generated. Eskom’s assessment of the capacity of the transmission network (March 2022) indicated that the Northern Cape is saturated, and the Western and Eastern Cape have less than 3,5 GW of capacity available in total. It is unlikely that any of the preferred bid winners from BW5 or BW6 will start building before the end of 2024, and additional generative capacity from the IPPs will probably be brought on board only sometime in 2025, at best.

Accelerating Eskom’s death spiral

It is possible for 1,597 MW of energy generated from private embedded generation (where companies produce energy, mainly for their own use) to come on board before the end of the year, and it has been announced that there’s a commitment to produce 9,000 MW in total, with 5,200 MW made available by 2027. 

The energy-intensive user group, which accounts for 40% of total energy consumed in the country, will embark on producing this private embedded generation. 

This raises some issues. Eskom has been responsible for enabling the connection of embedded generation to the grid, and for handling the wheeling of electricity (through the usage of the transmission grid) to customers. Eskom (or the new national transmission company) incurs direct costs from this responsibility. Wheeling costs money. The utility’s tariff proposals note that lost revenue may have reached R4.1-billion by 2021 due to the entry of embedded generators into the market, an estimate obtained before the lifting of the 100MW cap on embedded generators. 

In terms of rooftop solar, given that a minority of the population is able to go off grid (even with subsidies), not much energy will be generated. The estimates vary from 250-400 MW at the conservative end to a more ambitious target of 1500 MW per annum.

These are not renewable energy’s (RE) only challenges. It is important to recognise that 1 GW of RE-generated energy is not equal to 1 GW of energy generated from coal. This is because RE (solar PV and wind — onshore and offshore) is intermittent energy — in other words, not available all the time. 

While we can predict solar output and wind speeds quite accurately, we cannot create them. As a result, the capacity factor (energy output over time) of energy from solar PV and wind is lower than that of dispatchable sources of energy — energy that is available on demand.

The technical challenge is not just how much energy is available, but when it is available. As Eskom’s medium-term system adequacy outlook points out, “variable generation resources will reduce the energy gap if the degree of their energy production and the demand shortfall coincide. There may be instances where the degree of coincidence between the two is low or non-existent.”

Unbundling Eskom and the establishment of the National Transmission Company of South Africa is part of a strategy for the increased liberalisation of SA’s energy market. By design, the new independent transmission company would purchase electricity indiscriminately and not prioritise energy generated by Eskom.

Should we not just take all the energy we can get? 

The combination of unbundling and the increased privatisation of the energy sector will necessitate further tariff hikes and bailouts. Besides increasing costs, Eskom faces the much greater indirect cost of falling sales volumes and the sales it is losing are from its best-paying customers. The reduction in its market share and declining sales volumes will accelerate its death spiral and drive its collapse. The rollout of rooftop solar and growing municipal procurement of electricity from IPPs contribute to the problem. 

An Eskom collapse will put the country’s energy supply in even greater danger. If meeting the immediate capacity shortfall and stabilising energy supply in the shortest time is the primary concern, servicing Eskom’s fleet of power stations and improving its plant performance must be prioritised. As the medium-term system adequacy outlook points out: 

“An improvement in EAF [energy availability factor] provides more energy to support higher electricity growth with quicker turnaround times.” 

Visit Daily Maverick’s home page for more news, analysis and investigations

No matter which way you cut it, certain power stations have to be taken offline for extended periods of time for maintenance and refurbishment. 

Fixing and maintaining the existing fleet to improve Eskom’s energy availability factor (the proportion of installed capacity that is available at any one time) is the quickest way to stabilise energy supply in the short to medium term. To rush this needed maintenance for the sake of claiming a quick victory would be to repeat one of the key mistakes that has led to the current round of heavy rolling blackouts. 

As Zamani Mathe, CEO of government energy think-tank Sanedi, wrote, the lack of consistent planned maintenance between 2018 and 2020 has now caught up with us, leading to a critical spike in unplanned breakdowns, which have doubled in frequency over the past few years. This is behind the massive decline in Eskom’s energy availability factor: how much energy is generated as a percentage of total installed capacity. 

Currently, Eskom’s EAF is 58%. The coal-fired power stations that are working are also being run extremely hard, and the energy utilisation factor (EUF) — how hard the fleet is being run — is rising. Already, 14 out of the 15 coal-fired power stations recorded an EUF of above 90% and the average EUF for the coal fleet is estimated to be just below 94%. It should be running at an EUF of 75% given the average age of the fleet. This abuse will lead to further technical problems. 

Eskom chair Mpho Makwana announced that the utility would improve the EAF to 60% by March 2023, 65% in March 2024 and 70% in 2025. This can be a realistic plan if the necessary time and resources are provided. But first the energy gap (the difference between energy supply and energy demand) must be reduced to allow space for servicing the fleet.  

What is to be done?

  • Plug the energy gap

Government must guarantee the resources and licensing for Eskom to procure the diesel needed to operate the open-cycle gas turbines (OCGTs) at the cheapest possible price. Eskom has two significantly capacitated OGCTs — Ankerlig in Atlantis near Cape Town and Gourikwa in Mossel Bay, both in the Western Cape. Running the OGCTs is not a long-term solution, and they are already being run 800% harder than they ought to be. 

However, it is an important interim solution. Together, they have a generative capacity of just more than 2000MW (1327MW and 740MW), helping to prevent higher stages of rolling blackouts. Nevertheless, much more is required. 

The first of Koeberg’s units that have been shut down for refuelling and maintenance is anticipated to be operational by June this year, which will add 2 GW of capacity. 

However, unit 2 is scheduled for servicing shortly after. While this will mean a reduction in generated capacity, once revamped it is expected that Koeberg’s lifespan will be extended by 20 years. It is important to expedite the revamp to bring both units on board as soon as possible. 

To improve the performance of Eskom’s coal-fired power stations requires addressing major design and construction defects at new-build power stations. The Presidency’s January 2023 update on the roadmap to deal with rolling blackouts clarifies that the biggest amount of energy that can be generated in the shortest time is through fixing build defects at Kusile units 1, 2, 3 and 5. This can add an estimated 2880 MW of energy to the grid before the end of the year. Fixing Medupi units 5 and 6 would add a further 1520MW.  

In June last year, it was reported that the energy availability factor at Medupi and Kusile almost doubled after fixing boiler defects. According to a senior engineer, 80% of the boiler defects have been addressed and Eskom claims that the process will be complete by the end of 2023.

Over and above fixing the boiler defects, one of the chimneys at Kusile Unit 1 has collapsed, making units 1, 2 and 3 inoperable and removing more than 2,000 MW of capacity. Fixing the chimney could take up to two years, while a temporary solution could be in place within 12 months. The latest total estimated cost for the correction of defects at all of Medupi and Kusile units ranges between R5.6-billion and R7.2-billion  according to Eskom’s 2022 annual financial statements. This money must be provided to ensure that the process can be finalised by the end of 2023 as scheduled. 

Together, the use of OCGT, re-operationalising Koeberg and fixing build defects at Kusile and Medupi will account for approximately 8 GW of generated capacity, as illustrated below, which is enough to meet the immediate energy gap and allow for the required deep maintenance of the fleet. 

Eskom sources of energy

There is no need to turn to the expensive and potentially corrupt Karpowerships deal that will also take between 12 to 18 months to be operational, while being extremely detrimental to the climate. 

  1. Stabilise energy supply

Once the energy gap is plugged, it is possible to undertake deep maintenance of the coal-fired power stations. Besides Medupi and Kusile, the ones listed in the table below have a large generative capacity but a low energy availability factor. Eskom must prioritise deep maintenance of these plants for the stabilisation of energy generation. 

For the effective fixing of the fleet several important steps must be undertaken, including ensuring Eskom has sufficient resources to cover maintenance costs and to insource added maintenance capacity. The procurement process for maintenance must be made efficient. Eskom needs to be able to procure maintenance parts and services directly, rather than having to go via the National Treasury. The procurement process must be made transparent to safeguard against corruption.  

  1. Public pathway to a low-carbon economy 

Once the energy gap is closed and we have stabilised the energy supply, Eskom must look to develop renewable energy generative capacity as part of an essential decarbonising programme. It is essential to develop RE capacity to solve the problem of the declining attractiveness of RE projects to the private sector and secure Eskom’s revenue and prevent the death spiral. In this process, it is essential that its mandate to deliver electricity as a public good is restored. 

Electricity is far too fundamental to the economy and society to leave to the vagaries of the market.

This is not a defence of Eskom in its current form. It follows that Eskom must be de-corporatised so that it is no longer required to operate in the same way as a private company. Eskom’s financing model must be shifted to end the dependence on revenue and debt for maintenance and investment, and ensure its sustainable financing. Moreover, its governance structure needs to be reformed to end bad incentives and networks of patronage and corruption. 

This would require a change in culture, but it is possible. The material base for a change in culture is an end to outsourcing, to eliminate the proliferation of opportunities for corruption by bringing servicing and maintenance work in-house. And all those responsible for corruption must be held rapidly accountable and punished. DM/MC

Read part 1: If Eskom’s tariff increase is to go, its financing mode…  

Dominic Brown works for the Alternative Information & Development Centre (AIDC). 


Comments - Please in order to comment.

  • Johan Buys says:

    “ It is important to recognise that 1 GW of RE-generated energy is not equal to 1 GW of energy generated from coal. This is because RE (solar PV and wind — onshore and offshore) is intermittent energy — in other words, not available all the time”

    Not true. Coal EAF is below 50% so half is not available. Without the 6GW from these intermittent IPP we would be beyond stage 8 right now. I can more accurately predict what my factory’s output will be tomorrow than Eskom can predict its coal output tomorrow.

    When I add storage (factory size not mountain cottage), Eskom will be less than 10% of my energy, playing fourth fiddle to solar-direct, solar-stored, generator. The generator role above grid is because tariffs mean I am better off running it 50 times a year than hitting big draw from the grid – if it happens to be available, which it isn’t 8h a day. I’ll welcome Eskom back to second spot if it sorts out tariffs and breaks 8h a month instead of 8h a day.

  • R S says:

    “The reduction in its market share and declining sales volumes will accelerate its death spiral and drive its collapse.”

    Yes, this is inevitable. What we essentially have to do is save the bits that are worth saving and are affordable to save, and slowly let the rest die off.

    Anyone with vision can see the future will be with decentralised and smaller scale energy generation, so essentially we have to put Eskom (at least we know it today) on life support until we can get there.

  • R S says:

    “The first of Koeberg’s units that have been shut down for refuelling and maintenance is anticipated to be operational by June this year, which will add 2 GW of capacity. ”

    Aren’t Koeberg’s units each about 1 GW?

  • R S says:

    “This is not a defence of Eskom in its current form. It follows that Eskom must be de-corporatised so that it is no longer required to operate in the same way as a private company.”

    I never understood why Eskom was moved away from producing the cheapest electricity possible to a “for profit” entity…

    • sue fry says:

      Neoliberal theory and practice? The market rules. Isn’t that the way the game works?

    • Roelf Pretorius says:

      We should keep in mind where the problem is. It is not with the grid (except that some upgrading is needed for the change in location of generation) OR with the distribution side as far as I know. The whole problem in total is with the generation side. So once these three are seperated as the President has indicated, then the generation side can be reformed in any way to increase its’ capacity. And the way I understand it, the idea is that the grid will remain where it is (under government control) and it will also control the electricity sold onto it and its’ quality. The distribution side will probably go to regional companies which will also be part of the public sector; whether they will be controlled by national, provincial or district/metro governments is not clear at this stage. So it is only the generation that is going to be privatised, and Eskom may even still be controlling some of it; at least the storage facilities will probably have to remain part of the grid (for instance the hydro-electrical pump power stations). And I can’t see why it should NOT work.

  • Richard Baker says:

    A sensible and informed overall analysis Dominic. Within that, I have been writing for months that the quickest and cheapest way to bring the required bulk power back on to the grid-where it’s needed-is to fix the failed coal-fired units at the 3.6 GW 6-pack stations( per your list and others). This requires return of technical and operational skills that have been forced out in the past 20 years. There has been a list of available personnel for nearly 3 years which the government have strongly vetoed.
    If nothing else, the State of Disaster is used to by-pass cadre deployment and provide protection for those teams plus address the criminality that is plaguing Eskom at all levels and permit direct engagement of the OEM’s, then it could serve a purpose. That said a SOD is not needed to do those things. The problem is that a political solution( a “minister of electricity” and all that implies) will not solve technical and operational problems. ANC -Get out of the way and let the right people and companies in to fix the mess you have created!

  • Dennis Bailey says:

    Key takeaways:
    1. There is no need to turn to the expensive and potentially corrupt Karpowerships deal that will also take between 12 to 18 months to be operational while being extremely detrimental to the climate.
    2. The procurement process for maintenance must be made transparent to safeguard against corruption.
    3. The material base for a change in culture is an end to outsourcing, to eliminate the proliferation of opportunities for corruption by bringing servicing and maintenance work in-house. And all those responsible for corruption must be held rapidly accountable and punished.

    How simple must it get mister president for you to understand what’s possible, when and how?

  • Michael Bellis says:

    An excellent and balance summation of where we are today on the electricity state of disaster. If we work through all the issues (and stop making up fast turnaround and even better five point plans) then the problems should be solved.
    So what can go wrong?
    The coal supply in terms of quality and quantity has not been mentioned. The underperforming new build boilers designed for high grade coals are showing some improvement but there still lurks potential problems of ensuring consistent coal prepping.
    The hiccups along the way, such as blowing up a multi billion rand generator set during its commissioning and the collapse of the critical flue gas duct serving three generator sets boggles the mind. They say there is never only one cockroach under the fridge.
    Beyond those small things, we have seen a concerted effort by DMRE to use the REIPPP projects as gratuitous BBEEE sources of rewarding ANC benefactors and which have actually hindered the progression of useful additional and much need generation capacity.
    The current state of disaster has been self made, either as a result of incompetence in some quarters, or deliberate greed.
    The credibility gap between informed citizens and the government is too big to believe.

Please peer review 3 community comments before your comment can be posted


This article is free to read.

Sign up for free or sign in to continue reading.

Unlike our competitors, we don’t force you to pay to read the news but we do need your email address to make your experience better.

Nearly there! Create a password to finish signing up with us:

Please enter your password or get a sign in link if you’ve forgotten

Open Sesame! Thanks for signing up.

We would like our readers to start paying for Daily Maverick...

…but we are not going to force you to. Over 10 million users come to us each month for the news. We have not put it behind a paywall because the truth should not be a luxury.

Instead we ask our readers who can afford to contribute, even a small amount each month, to do so.

If you appreciate it and want to see us keep going then please consider contributing whatever you can.

Support Daily Maverick→
Payment options

Daily Maverick Elections Toolbox

Feeling powerless in politics?

Equip yourself with the tools you need for an informed decision this election. Get the Elections Toolbox with shareable party manifesto guide.