Eskom’s residual demand is collapsing. The MTSAO still assumes it will grow

South Africa's new 10-year grid adequacy outlook assumes electricity demand on Eskom will keep rising, backed by new gas plants and thousands of kilometres of transmission lines. Engineer Wietze Post argues the data point the other way. The energy Eskom must still supply has dropped from 201 TWh to 179 TWh in under two years, while rooftop solar now outstrips Medupi and Kusile combined and coal units sit in cold reserve. If he is right, the risk for consumers and taxpayers is not a shortage of power but billions spent on infrastructure the market has already moved past..By Ir. Wietze Post*.Response to the NTCSA Medium-Term System Adequacy Outlook 2027–2036 and Chris Yelland’s analysis.The National Transmission Company South Africa published its Medium-Term System Adequacy Outlook for 2027–2036 on 30 October 2026. Chris Yelland of EE Business Intelligence reviewed it the next day. Both treat rising underlying demand, new gas capacity, and large transmission expansions as central to keeping the lights on. The residual-demand data, and the shape of the S-curve they form, tell a different story.The data firstEskom’s residual demand is the energy it must still supply after contracted renewables and behind-the-meter generation. The 12-month rolling totals have been falling for years. The decline is now accelerating. That acceleration signals the rapidly approaching inflection of a descending S-curve..The drop from 201 TWh to 179 TWh in less than two years is not a gentle taper. It is the steep middle section of an S-curve. Early declines were small. They have grown. The next steps, if the pattern holds, will be larger still.Rooftop and behind-the-meter solar has already reached roughly 8.3–9.1 GW. That is more capacity than Medupi and Kusile combined. Most of it sits on commercial roofs, warehouses, and farms. Household numbers are rising too. Eskom’s sales volumes fell to about 178 TWh in the year to March 2026. That is the lowest level since the early 2000s. Industrial sales dropped 22.5 percent in the same year..Eskom has repeatedly placed several gigawatts of plant into cold reserve because available capacity exceeds residual demand. Peak residual demand is lower than a year earlier. Operating reserve margins have reached 50–70 percent on some days. Diesel use has collapsed.The system is not short of generation capacity. It is short of customers who still need Eskom’s electrons. Cold reserve is an operational status, not a visible feature of any particular power station. The plant shown below is an example of the coal fleet that is increasingly under-utilised..What the numbers meanThe MTSAO models underlying sent-out energy. Under its High trajectory this rises from 238 TWh in 2026 to 330 TWh in 2036. Peak demand is projected to reach 49 GW. That forecast is not wrong as a description of possible total electricity use. Instead, it is the wrong basis for assessing how much energy Eskom Generation must still supply. The residual-demand S-curve is already past its early, slow phase. The rapidly approaching inflection is visible in the quarterly data.If Eskom continues on its declining S-curve trajectory, it will only generate ±540TWh during the rest of its existence. Those 540TWh will have to pay off Eskom’s debts. Eskom provided for an Annual Residual Demand of 201TWh in 2024.Most new load will not be met by Eskom’s power stations. Data centres, electrified process heat, and industrial recovery can be served by co-located solar and batteries. Alternately, from independent generators connected by short, privately financed lines.Heat pumps and thermal storage already offer practical alternatives to gas for many industrial heating applications. The failed Gas IPP Bid Window 1 on 28 October 2026 confirmed that long-term gas supply assurances are not available. An LNG terminal does not exist. The infrastructure cost is high and the committed offtake is uncertain.Transmission follows the same logic. The MTSAO assumes roughly 14 500 km of new lines and 144 GVA of transformers. Delivery of that scale is unlikely. Shorter dedicated lines between generation sites and large consumers are already appearing in other markets. They will be cheaper and faster here too.As existing Eskom lines reach end of life, each will face an economic test. The result is more likely a spiderweb of thinner links than a continued set of thick corridors from a few large coal plants.On Dr Frans Cronje’s viewDr Frans Cronje has argued that Eskom is “coming right,” that fixing existing coal plants can add 10–15 GW of reliable capacity, and that South Africa needs 35–45 GW of dispatchable output to support 4–5 percent growth. He has warned against a rapid coal-to-solar transition and has said the grid cannot support the IRP’s renewable ambitions..Read more:.Wietze Post: Your electricity costs from 2026 to 2030 – the great divide.The residual-demand S-curve contradicts the premise. Eskom’s problem is no longer insufficient capacity. Its problem is excess capacity relative to the demand that still arrives at its plants. Refurbishing idle coal units does not create new customers. Those customers are installing their own generation or buying from independent producers. Capacity factors on solar and wind are lower than coal, but the energy is produced where it is used. It does not require the long-distance transmission that the old model assumed. The “easy electricity solution” of simply restoring coal plants ignores the fact that the market has already moved past the inflection.Cronje’s growth arithmetic also treats all electricity as interchangeable. It is not. Electricity generated next to a data centre or a factory does not need to travel across the national grid. Electricity generated by Eskom’s coal fleet does. Residual demand will continue down the descending S-curve. The megawatts that matter for economic growth will increasingly be the ones that never appear in Eskom’s residual-demand statistics.ConclusionThe MTSAO and Yelland’s analysis correctly identify delivery risks around gas, transmission, and fleet performance. Those risks are real. The deeper structural risk is that both documents still treat Eskom Generation as the default supplier of a growing national load.The residual-demand numbers, the clear acceleration in the descending S-curve, the rooftop capacity already installed, the cold-reserve volumes, and the failure of the first gas procurement round all point the other way. Eskom Generation is entering a period of permanent overcapacity. Most new electricity will be produced closer to the load by solar, storage, and independent generators. Policy that assumes otherwise will invest in infrastructure the market no longer needs..*Ir. Wietze Post is an engineer and the owner of East West Solar (Pty) Ltd. He has written previously on BizNews about Eskom’s residual demand, the duck curve, and the descending S-curve of grid demand..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 every morning on weekdays. Register here.Support South Africa's bastion of independent journalism, offering balanced insights on investments, business, and the political economy, by joining BizNews Premium. Register here.If you prefer WhatsApp for updates, sign up to the BizNews channel here.

Eskom’s residual demand is collapsing. The MTSAO still assumes it will grow

Related Stories

SA’s businesses stuck with solar systems for loadshedding, not savings

Jaltech creates exit route for solar owners - targeting R500 million in acquisitions

Every unit of power South Africans used in 2024 carried a bigger carbon cost

Andy Calitz on Eskom, the job he didn't get, and what came next