South Africa Has Too Much Clean Electricity at the Wrong Hours. So It Is Buying Batteries First.

The most consequential energy announcement of the past week did not involve building a single new solar farm or wind turbine. On Wednesday, October 7, South Africa's Electricity and Energy Minister, Dr Kgosientsho Ramokgopa, told a media briefing in Pretoria that the government's next round of electricity procurement would prioritise 4,600 megawatts of battery energy storage systems and 5,000 megawatts of gas-to-power capacity, covering the 2026 to 2037 planning horizon. It is the first determination under the country's Integrated Resource Plan 2025, the formal step that starts carrying that plan into actual procurement.

The determination totals roughly 9.6 gigawatts of new capacity. What it leaves out is as important as what it includes: there is no allocation for new wind or solar generation. Ramokgopa said a subsequent determination will cover variable renewables, hybrid projects that pair renewables with storage, and longer-term pumped storage. But this round is about flexibility first, and new generation later.

For the first time in South Africa's energy transition, the binding constraint is not generating enough electricity, it is having somewhere to put it until the evening.

The problem is curtailment, not scarcity

To understand why a country once synonymous with rolling blackouts now has surplus power, start with what has changed. Ramokgopa credited the turnaround to what he called the clinical implementation of Eskom's Generation Recovery Plan, which has brought long-broken coal units back to reliability. Combined with the country's growing fleet of independent wind and solar plants, the system now regularly records daytime electricity surpluses that government planning data puts above 4,000 megawatts.

That surplus is the new crisis. When the grid cannot absorb available electricity, either because transmission lines are full or because supply runs ahead of demand, the system operator turns generation away. The industry term is curtailment, and it has risen materially, according to reporting by Engineering News. Curtailment is not just wasted clean power. The National Transmission Company of South Africa pays compensation to renewable independent power producers whose output is curtailed under their contracts, so unused electricity still costs the system money. Ramokgopa put it bluntly at the briefing: the country is addressing an immediate curtailment problem, and left unmanaged, it could push financing costs so high that future private power projects become unbankable.

The logic of the minister's sequencing follows from that diagnosis. Adding more solar panels to a grid that already has to throw away midday solar does nothing except make the curtailment problem more expensive. Storage changes the equation. Batteries charge from electricity that would otherwise be curtailed and discharge it during the evening peak, when demand is highest and solar output is zero. As the government department put it in its statement, battery storage is central to mitigating curtailment.

What 4,600 megawatts of storage actually does

The scale of the proposed battery allocation is striking. South Africa currently has five grid-scale battery projects totalling 513 megawatts, all of which reached commercial close by June 2025 and entered construction, attracting 15.4 billion rand in investment, according to the Department of Electricity and Energy. The country's first storage project, the 20-megawatt, 100-megawatt-hour Hex facility, went online in 2023 under a contract with Eskom. The new 4,600-megawatt tranche is roughly nine times the entire existing pipeline: a step change, not an incremental top-up.

But the design of the procurement will matter more than the headline number, and the government has sketched an unusually specific vision for how the batteries should operate. The storage procurement is to be aligned with the System Operator's charging and discharge requirements, backed by enforceable availability and performance obligations. According to reporting by the Energy News Network, batteries in the programme are meant to charge from electricity that would otherwise be curtailed, sit where they can reach that surplus and discharge without reproducing the network constraint, and be judged on the electricity they actually move into useful hours and the reliability services they deliver. That is a subtle but important shift: storage is being procured as a managed service, measured on output and availability, not as a box of cells rated by megawatt.

The programme is also linked to transmission expansion, new industrial demand, and regional electricity trade, according to the official SAnews report of the briefing. The sequencing reads as deliberate: first make the grid flexible enough to absorb what is already being generated, then expand the transmission that lets more generation reach the places that need it.

The gas half of the deal

The other 5,000 megawatts goes to gas-to-power, a more contentious choice for a country trying to reduce its dependence on coal. The government's framing is that gas adds flexibility and dependable supply: turbines that can ramp quickly when renewable output dips or demand surges, smoothing the same variability the batteries are meant to absorb. Last year's power plan, the Integrated Resource Plan that this determination begins to implement, made provision for 16 gigawatts of gas in the energy mix by 2039, and Reuters reports the government is evaluating proposals from four short-listed consortia bidding to provide 2 gigawatts of gas-to-power capacity.

Seen in context, the gas allocation is the smaller part of a much larger plan. Reuters reports that Africa's biggest economy plans to add more than 105 gigawatts of new generation capacity by 2039, with renewable energy accounting for more than half of that total. The 9.6-gigawatt determination announced this week is the first down payment. Whether gas remains a bridge fuel or calcifies into a permanent fixture will depend on how fast the storage and transmission investments catch up, and on gas prices, which will decide whether those turbines are cheap insurance or an expensive habit.

Perspective: the grid's problem has moved from megawatts to hours

Strip away the policy machinery, and South Africa's announcement captures one of the defining technology stories of the energy transition. For a century, the electricity business was about building enough generators. The binding constraint was always capacity: could the system make enough power at the peak? Batteries invert that logic. South Africa now has moments where it makes more clean power than it can use, and its scarcest resource is not the electron but the hour it is needed in. A battery is, in effect, a time machine for electricity, and the minister's determination treats it as exactly that.

This is not a South African eccentricity. Grids from California to Spain are learning the same lesson: the cheap years of the transition were about installing solar and wind; the expensive years are about installing everything around them, storage, transmission, and software, so that variable generation behaves like the dependable supply the old grid took for granted. South Africa's version is unusually honest about the sequencing, putting the absorptive capacity before the next wave of generation. If the procurement documents carry binding commercial terms, as tender specialists expect, the market will decide whether batteries can actually be delivered at the promised scale. But the diagnosis itself is hard to argue with: the country is drowning in midday megawatts and starving at dinner time. That is a storage problem wearing a generation problem's clothes.