When the last stage of loadshedding ended in South Africa in late March 2024, analysts were sceptical. Eskom had made similar announcements before, only for the blackouts to return when a coal unit tripped or demand spiked on a winter morning. The utility had cried wolf enough times that the milestone was treated with appropriate suspicion.

By March 2026, that suspicion has given way to a more careful question: what actually changed? The answer is not one thing. It is the convergence of four separate structural shifts — two supply-side, one demand-side, and one on the balance of risk — that produced sustained grid stability for the first time since the mid-2000s.

Understanding those four shifts matters well beyond South Africa. As the continent’s largest and most industrialised grid, the South African energy system is the canary in the coal mine — or, more precisely, the canary in the transition away from coal. If the analysis holds, it contains lessons for Nigeria, Kenya, and Egypt about what it actually takes to break a chronic power crisis. If it does not hold, the continent’s most advanced grid will have provided a cautionary lesson at scale.

Shift One: 6,000 Megawatts of New Renewable Supply

The Renewable Energy Independent Power Producer Procurement Programme (REIPPPP) was the policy mechanism that began adding utility-scale renewable capacity to South Africa’s grid in 2012. By 2023, it had delivered approximately 3,500MW. Between 2023 and the end of 2025, an additional 6,000MW+ came online — solar, wind, and a small number of battery storage projects — as the backlog of approved but delayed projects finally reached commercial operation.

That 6,000MW figure matters in context. South Africa’s average grid demand is approximately 25,000–27,000MW. Eskom’s operational coal fleet had been providing roughly 22,000MW at peak, degraded from a nameplate capacity of 42,000MW — degradation being the product of chronic underinvestment in maintenance over more than a decade. Adding 6,000MW of reliable renewable capacity in two years did not solve the grid. It changed the margin of error. When three coal units fail simultaneously — the scenario that used to trigger Stage 4 or Stage 6 loadshedding — the system now has renewable headroom that it previously lacked.

The battery storage component of the REIPPPP — specifically BESIPPPP (Battery Energy Storage IPP Procurement Programme) Window 3’s 612MW — has added a dispatchable stabilisation layer. Battery storage does not generate continuous power, but it absorbs frequency deviations and covers the brief periods when coal units cycle off and renewables fluctuate. At the margin, that buffering capacity has prevented cascading grid events that the system previously could not absorb. See: Africa BESS 2026: Battery Storage Turns Solar Into Reliable Power.

Shift Two: 7 Gigawatts of Embedded Solar

The more quietly consequential shift happened not in Eskom’s fleet but in homes, commercial properties, and factories across South Africa. Between 2022 and early 2026, South Africa’s rooftop and embedded solar capacity grew from approximately 1GW to an estimated 7GW+. This is the direct product of loadshedding: when the grid becomes unreliable, the economic calculation for solar-plus-storage at a household or business level changes completely.

The effect on the grid has been profound. Peak demand — historically the morning rush of industrial and commercial load between 06:00 and 09:00, and the evening residential peak between 17:00 and 20:00 — has reduced by an estimated 1,500–2,000MW from embedded solar generation during daylight hours. What this means in practice: the system demand that Eskom has to serve with its dispatchable fleet is materially lower than it was three years ago. The grid is not under less strain because Eskom is performing better. It is partly under less strain because South Africans — at considerable personal cost — have reduced their dependence on Eskom.

This is a structural change. Unless installed solar capacity is decommissioned (which does not happen), the grid will continue to receive this demand-reduction benefit. It will grow as battery prices fall and more households add storage alongside their panels.

Shift Three: Industrial Load Migration

South Africa’s largest industrial energy consumers — Sasol (chemicals and fuels), ArcelorMittal (steel), several large mining groups — have progressively migrated load off the Eskom grid, either through embedded generation, wheeling agreements with independent power producers, or operational curtailments. The regulatory changes that allowed wheeling across the Eskom transmission network — changes that companies like SolarAfrica and SOLA Group structured their financing around — directly reduced peak grid demand.

This is not entirely positive from an economic perspective. Some of the load migration is not replacement; it is destruction. ArcelorMittal’s South Africa operations have been curtailed or closed in part because electricity costs have made production uneconomic. That is not a stability win — it is a sign that the grid crisis extracted permanent economic damage before the structural fixes arrived. The grid is more stable partly because some of the industrial activity that used to run on it no longer exists.

What Has Not Changed: The Three Structural Risks

The 335-day milestone is real. The risks it obscures are also real.

The most immediate is the coal retirement timeline. Eskom’s aging coal fleet — units with 40–50 years of operational life now approaching end-of-life — is due for progressive retirement through the 2020s and 2030s. Each retirement removes dispatchable generating capacity that the grid currently relies on. The Medupi and Kusile plants (4,764MW and 4,800MW respectively) are newer and will not retire soon, but both have experienced years of construction delays and technical problems. The Komati and Camden stations have been decommissioned; each decommissioning removed baseload that needed to be replaced. The just energy transition was supposed to fund a managed replacement programme. With the US having withdrawn from the JETP and Germany carrying a larger share, the concessional financing gap is real and the timeline is compressed.

The second risk is the transmission bottleneck. An estimated 4,300MW of approved renewable IPP projects are waiting for grid connection — held up not by financing or technology but by Eskom’s transmission infrastructure, which has not been upgraded at the pace required by the generation buildout. REIPPPP projects that have financial close and are ready to build cannot always connect. This is the next constraint: the grid can absorb the supply it has. Adding the supply it needs is a different and harder problem.

The third risk is Eskom’s balance sheet. South Africa’s state-owned utility still carries more than R400 billion in government-guaranteed debt, the legacy of the Medupi and Kusile construction programmes and years of deferred maintenance funded by borrowing. The government has taken on a portion of this debt through the debt-relief arrangement announced in 2023, but the restructuring is still in progress. Eskom’s ability to invest in the grid — new transmission capacity, maintenance of existing plant, the operational transformation needed for a renewable-dominated system — is constrained by its financial position.

The Business Dividend

The economic cost of loadshedding was not abstract. At peak in 2023, the South African Reserve Bank estimated grid instability was reducing GDP growth by 2 percentage points annually. Businesses running diesel generators were paying an effective electricity cost four to six times the Eskom tariff. The South African Chamber of Commerce and Industry’s Business Confidence Index showed generator-related costs as the primary constraint on small business expansion for three consecutive quarters.

Eskom Group CEO Dan Marokane, in the utility’s March 2026 operational briefing, acknowledged the milestone and attributed it to a combination of improved plant maintenance and new renewable integration. “The Energy Availability Factor of our generation fleet has reached 62 percent — up from 54 percent at the trough — and we have maintained that level consistently for three quarters,” Marokane said. “This is a structural improvement, not a seasonal one.” At 62 percent EAF, Eskom’s dispatchable fleet can absorb the routine failures that previously cascaded into loadshedding — a failure of two or three units no longer tips the system into deficit when renewable generation and embedded solar are factored in.

The manufacturing sector has registered the shift. South Africa’s Absa Purchasing Managers Index, which tracks factory-floor production conditions, recorded its highest monthly average in four years during Q4 2025 — the first full quarter in which the grid was stable for the entire period. The causal link between grid stability and manufacturing output is not linear, but the correlation is consistent: businesses that had mothballed energy-intensive equipment during the loadshedding years are returning to production capacity.

What the Milestone Means for Investment

The most immediate consequence of sustained grid stability is a shift in South Africa’s investment risk profile for energy-intensive industries. Green hydrogen production — one of the most capital-intensive and electricity-dependent industrial propositions — becomes materially more credible as an investment when the electricity supply is reliable. The data centre boom, already visible in South Africa’s Cape Town corridor, accelerates when uptime is guaranteed. Business process outsourcing, cold chain logistics, pharmaceutical manufacturing: all of these carry electricity reliability as a baseline investment condition.

This is not a theoretical claim. The Coega Special Economic Zone’s green hydrogen project and SOLA Group’s Naos-1 industrial wheeling deal with Sasol and Air Liquide — announced within weeks of each other — are both predicated on a functioning South African grid. Neither would have reached financial close under Stage 6 loadshedding conditions. See: SOLA Naos-1: South Africa’s First Major Industrial Solar-BESS Wheeling Deal.

For the rest of the continent, the South African grid stabilisation story offers a structured lesson rather than a simple template: renewables added supply, embedded generation reduced peak demand, industrial efficiency reduced load, and BESIPPPP provided frequency balancing. Nigeria has the gas reserves. Kenya has the geothermal and wind resources. Egypt has the solar irradiation and the DFI capital flows. Each market has different inputs. The output — a grid that can sustain investment — requires structural policy choices, not just technology deployment.

Three hundred and thirty-five days is a meaningful number. Whether it becomes 1,000 depends on decisions about coal retirements, transmission investment, and the financing of the just energy transition that are currently in progress — not yet resolved.