Key topics
- Solar and renewable energy adoption is rapidly displacing Eskom’s power.
- Eskom faces financial decline as solar reduces grid dependency and usage.
- By 2030, Eskom’s grid may become unsustainable, risking service disruptions.
Sign up for your early morning brew of the BizNews Insider to keep you up to speed with the content that matters. The newsletter will land in your inbox at 5:30am weekdays. Register here.
The seventh BizNews Conference, BNC#7, is to be held in Hermanus from March 11 to 13, 2025. The 2025 BizNews Conference is designed to provide an excellent opportunity for members of the BizNews community to interact directly with the keynote speakers, old (and new) friends from previous BNC events – and to interact with members of the BizNews team. Register for BNC#7 here.
If you prefer WhatsApp for updates, sign up for the BizNews channel here.
By Ir. Wietze Post*
Opinion
Last week, I shared data on Eskom’s falling residual demand. We South Africans are gradually breaking free from Eskom’s grip on electricity supply. This doesn’t mean we’re consuming less; instead, we’re sourcing more power from alternatives like solar panels on our roofs.
I invite you to read last week’s article first. That contains all the data, definitions, assumptions, and explanations. This piece is the sequel. Here, I’ll unveil several observations and trends that have captured my attention.
Exponential Growth Rate and the “S” Curve
Solar power and other renewable energy resources are eating into Eskom’s market share. However, most people don’t notice renewables’ impact. That’s because the initial steps are tiny, yet they’re growing exponentially.
When a new technology is introduced, it is adopted at an exponential growth rate. The changes can be illustrated with “S” curves that track the adoption rates.
Figure 6. S-curve phases and diffusion process
The following graph shows many new-technology items vs time. For instance, the adoption of telephones, washing machines, colour TVs, and cell phones. More recent inventions spread faster.
Figure 7. We adopt recent inventions quicker.
Rate of decline of Eskom’s power supply
S-curves also describe the demise of legacy technology—a descending S-curve. With this in mind, I studied the rate of decline of Eskom’s Total Annual Demand. I looked at the year-on-year decrease and the gaps between successive years. These gaps follow a growth sequence.
I also sought the maximum hourly demand for each year. The yearly maxima are gradually declining, and the gaps between them follow a growth sequence too.
The same trends emerged for the average hourly demand and daytime minima. The annual and overnight minimum demands initially tracked downwards for a few years. Yet, they rose in 2024 after load shedding ended.
Typically, we consume more electricity during the day and less at night. However, solar power displaces Eskom power. As a result, Eskom is asked for less energy during the day. This can cause diurnal demand inversion for Eskom, where more power is drawn at night than during the day (see my previous article).
So I checked daytime and overnight demand minima. Before recent events, the last time we had a demand inversion was New Year’s Day 2021. Yet, from 12 October to the end of 2024, we had eight such events. In January 2025, we had four inversions. These are a clear sign of solar power’s impact.
The trends are all correlated, and we only have four years of data points. The sample size of annual consumption declines is small. Yet the decline can develop quickly and has far-reaching consequences.
My Thesis – a shift from Eskom’s power
My thesis is that we have begun a definitive shift away from Eskom power. This shift is proceeding exponentially. It’s driven partly by the rollout of renewable energy plants, especially solar. Besides, Eskom’s price increases are boosting solar uptake. The growth rate of the inter-year gap is approximately 150%. The South African energy sector is on the cusp of a rapid, fundamental change. This will severely affect Eskom’s Generation, Transmission, and Distribution Divisions.
Consider the sequence of the Total Residual Demand (see Part 1 of this article here). The data from 2021 through 2024 shows inter-year gaps of 0.82 TWh, 3.94 TWh, and 5.95 TWh.
Thus, the following sequence appears: 4, 6, 9, 13.5, 20.25, 30.4, 45.6, and 68.3 TWh. These figures represent the differences between years at a 150% growth rate. Note that the last number, 68.3 TWh, is likely to comprise the tail-end from 2029. The sum of these numbers is 197 TWh, which is close to the 201 TWh we consumed in 2024.
An exponential growth rate is characterised by small initial changes – 4, 6, 9, and 13.5 seem insignificant. However, they are followed by substantial steps. To most people, the final steps appear sudden and unexpected. Yet, I would not be surprised by a dramatic demand drop in 2029 and 2030, especially in the latter half of 2029 – just 5 years away.
If this sequence unfolds as an S-curve, the NTCSA will transmit ±32 TWh during 2031. Wheeling contracts may supply a substantial part of this. NTCSA is the National Transmission Company of South Africa.
What could be driving this change? How can such significant changes occur in such a short time?
SWB is displacing Eskom
SWB (solar, wind, and battery) power is replacing grid energy. Consumers are learning to work with solar, wind and batteries. They’re finding they can provide sufficient uptime. As users adapt, their demand for renewable sources increases.
Solar and wind plants, with batteries, generate power on-site at the consumption points. Renewables will not immediately displace all of Eskom’s demand. But, they can replace a significant portion.
Where applicable, the battery inverter provides the baseload power. If there is not enough solar charging, Eskom can supplement the customer’s battery.
Off-grid consumers install sufficient solar capacity to meet winter demand. If needed, they’ll use a small genset to top up their battery.
Figure 8. SAPVIA – Renewables installed
By the end of 2024, South Africa had about 7000 MW of solar plants installed. These are all 100kW and larger. An unknown number of smaller plants were installed. 2894 MW of solar plants, larger than 100kW, were installed during 2024 (Click on the image for updated data).
Cost of renewable plants
Renewable plant costs are gradually decreasing. Today, equipment prices in USD are about 20% of what they were a decade ago. That’s helping to speed up SA’s transition.
Solar panels and lithium-ion batteries have seen steep price declines. Since 1991, their prices have fallen by around 97%. On average, prices drop by 19% for every doubling of capacity. This promising reduction rate continues apace (as shown in the following charts).
Figures 9 and 10 illustrate Wright’s Law, also known as the Learning Curve. This principle states that for every doubling of manufactured goods, their cost will fall by a constant percentage. Now, the cost for batteries is around $100 per kWh at the cell level, although prices are significantly lower in China.
Figure 9. Solar (photovoltaic) panel prices vs. cumulative capacity
Figure 10. The price of lithium-ion batteries fell by 97% since 1991.
Eskom is experiencing steady decline
Figure 11. Annual USA electricity generation from all sectors (1950-2020).
Renewables became the second-most prevalent U.S. electricity source in 2020
Source: U.S. Energy Information Administration (EIA), Monthly Energy Review
Note: This graph shows electricity net generation in all sectors (electric power, industrial, commercial, and residential) and includes both utility-scale and small-scale (customer-sited, less than 1 megawatt) solar.
In 2008 the USA consumed a total of ±4 trillion kWh of electricity. At its peak, coal-fired plants generated 50% of this. From 2008 to 2020, coal lost its dominance and retracted to a 19% share. Thus coal lost 60% of its market share in the USA in the 12 years up to 2020.
In a previous section, I mentioned the sequence of the Total Residual Demand and inter-year gaps. That trend leads to accelerating decreases from year to year. I believe Eskom has passed its apex generation level. It could now face a rapid decline as coal power did in the USA since 2008.
For a decade, Eskom has been an unreliable supplier. It has also forced rapidly rising prices on South Africa. Thus renewable energy, especially solar power, has stepped into the breach.
Solar power’s lead can be seen in the change of day and night electricity consumption. Usually, we use more grid-supplied energy during the day than at night. I expect Eskom to experience more days of diurnal demand inversion, which are markers of solar power’s impact.
Furthermore, Eskom could experience a near zero-demand hour by late 2028 and many zero-demand hours in 2029.
South Africans will add more solar power plants in 2025. For Eskom’s energy consumption to grow, we need economic growth using their energy. Moreover, Eskom’s demand growth rate would have to exceed the solar addition rate. That’s unlikely to happen. Now, solar plants are an integral part of many developments. Besides, Eskom will keep on raising prices, which puts consumers off.
Some say we must build more power capacity to meet the increased demand when the economy grows again. Yet, developers already add renewable plants to their projects. Thus they avert the need for centralised generation plants.
The NTCSA will almost certainly dispatch less electricity in 2025 than in 2024. The question is: How much less? I expect Eskom to supply 8-9TWh less in 2025 vs 2024.
Advice for Eskom
Daytime hours should become off-peak tariff hours. Eskom should change its rate structure to charge higher tariffs at night, at least during summer. During the summer, Eskom will sell more electricity at night than during the day. So it will earn more, overall, with higher overnight tariffs. Besides, low daytime tariffs will allow Eskom to compete better with solar pricing.
Eskom must not raise connection and energy costs. Raising the connection (service) costs will chase customers away. Instead, to improve sales, they should remove the connection costs and roll them into the energy tariff. However, I realise that the consultation process will delay a decision. Thus, it will be too late when they’re at last allowed to make the changes.
Winter consumption will become a significant problem. Eskom must maintain generation capacity to supply this. Meanwhile, summer consumption will vaporise. Thus, Eskom must amortise the annual cost over quickly receding energy units. Eskom needs recompense to maintain capacity even while many plants are idle most of the year.
Eskom should earnestly consider how to lower costs. Alternatively, it could raise prices. However, with higher prices, customers will use less Eskom energy, and then it will need to raise prices again. That would create a death spiral.
A bleak future for transmission lines
Solar with batteries is a distributed, decentralised system. You can install it on your roof or yard and go off-grid. You can run your operation from your battery inverter. Install a small generator to trickle-charge your battery during overcast days.
When a solar plant provides a significant part of a consumer’s energy needs, the grid connection costs become irksome. The fixed cost pushes the cost of a unit of electricity up, at times to ridiculous heights. This drives customers to disconnect from the grid.
In contrast, coal, nuclear, wind, and solar farms need a centralised system to distribute their energy. However, centralisation and decentralisation are opposing concepts. Solar power plants are decentralised generators, which don’t need a grid.
Combining centralised and decentralised systems raises costs. Eskom’s plans to increase the connection costs for low-consumption clients is helpful. It clarifies the actual cost of the centralised system. It will enable customers to choose between systems with a precise price for both.
Solar and battery prices are constantly decreasing. Thus, solar power plants will eventually generate most of South Africa’s energy. Only a tiny fraction of today’s energy consumption will remain to be transmitted, which will not justify the cost of maintaining the grid.
As more solar is installed, Eskom’s outlying lines on the grid’s fringes will be used less. This is most visible in rural areas, but even urban lines occasionally break down. Then, Eskom will not replace them as they’re not financially viable. The reduced demand will be felt in the core of the grid. Gradually, the HV transmission lines will carry less electricity.
Maintaining the current grid will be too costly if distributed solar plants erode grid transmission. Low traffic on the grid depresses income from energy sales. Thus maintaining the grid is not financially justifiable.
If the sketched trends hold, the grid will be used much less by 2030 than now.
However, South Africa’s new Electricity Regulation Amendment Act (ERA Act) came into effect on 1 January, 2025. The law aims to create a more competitive electricity market. Besides, it aims to help the country transition to clean energy. Whilst the ERA Act probably won’t boost Eskom’s generation fortunes, it may help to save the grid.
Risks for IPPs
IPPs are Independent Power Producers.
The expected grid deterioration increases risk for any contract that depends on Eskom’s services after 2030. Eskom may not be able to deliver the expected service. It’s worth keeping this in mind when concluding contracts.
IPPs distributing through the grid will find Transmission & Distribution costs significantly increase their energy’s price. They will find it hard to compete with distributed, on-site, solar plants. Their noon-generated energy will be curtailed. However, battery plants adjoining the solar farms could make the difference.
Cannibalisation of solar farms is a known phenomenon. Rooftop solar cannibalises the solar farms because rooftop solar is always used first. At the moment, solar farms compete with Eskom pricing, and thus the farms win. But this will end in 4 to 5 years when distributed solar plants become an overwhelming presence. Then, solar farms will be competing with rooftop solar. Solar farms’ financial position will rapidly deteriorate.
Generating electricity to sell through the grid may not be sustainable. It’s helpful to put your renewable plant close to your client to minimise transmission risks. Thus, you can maintain the transmission lines yourself if need be.
Large consumers
Eskom believes it cannot significantly reduce electricity costs. About half of the cost of a unit of electricity is in transmission and distribution. These costs are part of the expensive consumer-level pricing. So, even if Eskom can generate cheaper electricity, it won’t make much difference.
Grid-connected clients are reducing their consumption. Many can generate cheap power on their roofs.
Some companies may have several local facilities. A particular facility may have little capacity to generate energy on-site, while another has excess space. These companies may wheel energy between facilities.
Large power users may consider establishing their facilities near a solar or wind farm. For example, consider putting your factory next to a solar plant in the Northern Cape if you want power from a solar farm.
Companies with high demand but little space can use their neighbours’ roofs to generate energy. They can enter into supply agreements with their neighbours. It’s helpful when the suppliers are downstream from the same sub-main as the company. Thus, the company can create a mini-grid or Virtual Power Plant (VPP). Companies should consider covering the grid connection costs for their suppliers. Those neighbours might otherwise opt to disconnect from the grid while the company is considering their options. For many consumers, cheap solar panels and batteries already make it worth disconnecting.
Conclusion
Early indications project a steadily decreasing demand for Eskom energy over the next 6-8 years. Solar power will displace Eskom’s demand. Yet, in the coming year, Eskom may recover some demand lost during load shedding. I will regularly update my projections.
From summer 2028/9, near-zero-demand hours will occur. This may destabilise the grid and cause supply interruptions.
Supply interruptions provide an opportunity for theft. If sections of a low-traffic line are stolen, Eskom will likely not replace them. Damage from any other cause will also not be repaired. Fixing the line would not be financially justified. This will force late-adopting consumers along the line to go off-grid.
Beware of agreements that depend on Eskom services after 2030. Eskom may not be able to provide the services.
Check your exposure to any Eskom service beyond 2030 – take appropriate steps.
Hopefully I have alerted all concerned to the scenario described in this article. Being prepared should help us avert the worst consequences.
Read also:
- Eskom Kusile fraud: R450mn after-hours scandal exposed
- Here’s Cyril’s full SONA record – Trollip nailed it: “delusional, fanciful, disconnected.”
- Last chance to “Let SARS fund your investment”
*Ir. Wietze Post has broad research, engineering, and business experience. Wietze currently operates as a project leader. He has advised C&I and NPO renewable energy buyers for several years. The title of Ir is an abbreviation of ‘Ingenieur’, which is awarded to MSc graduates at technical/engineering universities in the Netherlands. Wietze’s alma mater is Wageningen University. Wietze avidly follows energy developments around the world and in South Africa. In this series, he considers how global trends may play out in South Africa.
