Eskom Wants to Sell Surplus Power to Bitcoin Miners at R1.20/kWh. The Details Still Matter.

South Africa’s state power utility Eskom has a surplus electricity problem and a proposed solution involving Bitcoin miners. Energy regulator Nersa is currently consulting on a two-year pilot tariff that would sell surplus power to Bitcoin miners and other flexible-load customers at a benchmark price of approximately R1.20 per kilowatt-hour. Comments closed on October 23. Public hearings are planned for November 3. Nersa’s decision is due on December 7.

The idea is straightforward. Eskom estimates it has between 5GW and 7GW of spare capacity at times, driven by daytime solar generation and low overnight demand. More than 5TWh could be curtailed in 2028 under current projections. Bitcoin miners, who can switch off instantly when grid demand rises, are a natural buyer of that surplus. They provide flexible, interruptible load that grid operators need. Eskom gets revenue from power it would otherwise waste.

However, as one industry expert put it plainly: the idea is a no-brainer. The devil lies in the details.

What the tariff actually offers

The R1.20/kWh benchmark applies only to the highest-voltage customers in Eskom’s southern region. Customers in other transmission zones or lower voltage categories would pay rates escalated in line with Megaflex, Eskom’s existing time-of-use tariff for large customers. Customers supplied below 500V are excluded entirely.

The operating window under the pilot would be roughly 12 to 16 hours per day, with most weekday daytime hours that are normally standard-rated under Megaflex reclassified as off-peak. However, the full schedule of demand, capacity, and network charges has not been published. Therefore, the complete cost of participation cannot yet be assessed by prospective miners.

Two Bitcoin mining companies have already expressed interest. Together they represent combined initial demand of approximately 10MW and potential expansion of up to 500MW.

The price problem

One prospective operator indicated that around R1.30/kWh, with a 14 to 16-hour daily window, “could be commercially viable and sustainable.” However, not everyone agrees the proposed price is low enough.

An anonymous industry insider told TechCentral the pricing is not yet economical. Specifically, they said miners need sub-$0.05 per kWh to operate profitably. At the rand’s exchange rate of approximately R16.36 to the dollar at the time of the report, $0.05 translates to roughly 82 cents per kWh. The proposed R1.20/kWh benchmark sits significantly above that threshold.

Furthermore, the insider noted that other key terms remain unresolved. Specifically, they questioned whether miners would switch off during peak hours voluntarily or through automated telemetry, and whether Eskom’s stated discount represents a genuine competitive advantage or simply a repackaging of existing tariff structures.

Steven Boykey Sidley, a partner at Bridge Capital and a professor of practice at the University of Johannesburg who has written two books on Bitcoin, was more constructive. “The idea behind this is a no-brainer,” he said. “Sell electricity at a discount when no one else wants it.” However, he echoed the same concern about details: discount size, other eligible off-takers, and the mechanism for switching miners off during peak demand.

The switching question nobody has answered

The consultation paper does not settle the curtailment question. Participants “may be required to participate in demand-response programmes”, according to Nersa’s published paper. The regulator has asked stakeholders whether “mandatory demand-response capability, minimum curtailment requirements, response times, telemetry, metering, dispatch instructions and penalties for non-performance should form part of the pilot.”

That is a significant open question. Bitcoin mining rigs built on Application-Specific Integrated Circuits (ASICs) can indeed be switched on and off almost instantly, as the industry insider told TechCentral. However, the pilot is a tariff arrangement, not a contract for grid-balancing services. The National Transmission Company South Africa (NTCSA) sets separate technical requirements for frequency reserves, with instantaneous reserve needing to be fully available within 10 seconds. Whether the Eskom pilot meets that standard remains unclear.

The Texas comparison

Sidley pointed to Texas as the most relevant international precedent. Riot Platforms, one of the world’s largest Bitcoin miners, earned $21 million in power credits in the first quarter of 2026 alone. Of that, $13.5 million came from selling contracted power back to the grid rather than using it, and $7.5 million from demand-response programmes run by ERCOT and its Midwest counterpart MISO. After those credits, Riot’s all-in power cost was just 3 US cents per kilowatt-hour.

In August 2023, Riot had already demonstrated the same model by curtailing its power usage by more than 95% during peak demand periods and earning an estimated $31.7 million in credits.

The key difference between Texas and South Africa is that ERCOT has a mature, transparent demand-response market with clear pricing, contracts, and automated dispatch. South Africa’s proposed pilot has none of those features yet. Nersa’s December decision will determine how many of those gaps get filled.

The Ethiopia lesson

Africa has seen this kind of arrangement before. Ethiopia’s experience is the most instructive parallel, and it is not entirely encouraging.

As ABN has covered, Ethiopia aggressively courted Bitcoin miners to monetise surplus hydroelectric power from the Grand Ethiopian Renaissance Dam. Miners signed power purchase agreements promising them 98% of contracted supply. For years, the model worked. Miners paid Ethiopian Electric Power in hard currency, and EEP used the revenue to build grid infrastructure.

Then, in September 2026, EEP cut miner power supply to just 23% of contracted capacity after El Niño reduced reservoir inflows by more than 20%. The gap between contractual guarantee and operational reality was enormous. Miners who had invested in rigs and logistics based on the promise of affordable, reliable power found themselves running at a fraction of capacity.

The lesson is not that stranded energy deals are bad for Bitcoin miners. The lesson is that the contract terms govern who carries the risk when conditions change. Eskom’s pilot, as currently structured, does not clearly answer that question. Therefore, miners entering the South African market will need the December framework to be far more specific about curtailment compensation, contract duration guarantees, and what happens when Eskom’s surplus narrows.

Nersa’s own concerns

Nersa’s consultation paper is notably cautious. Specifically, the regulator warned that if unmanaged, cryptocurrency mining “may place additional pressure on system capacity, increase emissions and create pricing distortions.” Furthermore, Nersa flagged risks of “unintended cross-subsidisation or costs being shifted to other customer categories” if the preferential tariff is not carefully structured.

Eskom says the pilot should be revenue-neutral in its 2027 financial year. However, there is an established precedent for protecting other customers from tariff concessions. When Nersa approved a 62 cent per kWh concessionary price for ferrochrome smelters Samancor Chrome and Glencore-Merafe in May, it ordered any revenue shortfall to be ring-fenced within Eskom and not recovered from standard tariff customers. A similar protection for Bitcoin miners would be an important signal of good faith to other electricity users.

The bigger picture for African Bitcoin mining

South Africa’s Eskom pilot, if approved and structured correctly, could be a meaningful addition to the African Bitcoin mining landscape. Specifically, South Africa offers several advantages that Ethiopia lacked: a more mature regulatory environment, a sophisticated commercial electricity market, and existing Bitcoin industry infrastructure through companies like Africa Bitcoin Corporation.

Furthermore, the solar curtailment dynamic driving Eskom’s surplus is structural, not seasonal. Unlike Ethiopia’s hydro dependency, which fluctuates with rainfall and reservoir levels, South Africa’s solar surplus is predictable and growing as rooftop solar installations accelerate across the country. Eskom’s own data shows curtailment instructions rising from roughly 100 per month earlier this year to more than 1,000 per month recently.

That predictability is exactly what Bitcoin miners need. A surplus that disappears when it rains is a risk. A surplus that grows as the country installs more solar panels is an opportunity.

However, the December decision needs to answer the questions the October consultation paper left open. Specifically: what is the full all-in tariff across all voltage categories and regions? What are the curtailment terms and penalties? How does the tariff interact with the NTCSA’s demand-response requirements? And how are other customer categories protected from cross-subsidisation?

If Nersa answers those questions clearly, South Africa could offer African Bitcoin mining a more durable model than Ethiopia managed to build. If it does not, the pilot risks repeating Ethiopia’s experience on a different continent.

The public hearings on November 3 will be the next signal worth watching.

Sources

• • TechCentral — Eskom has a R1.20/kWh offer for bitcoin miners — https://techcentral.co.za/eskom-bitcoin-power-deal-nersa/286838/
  • Nersa — Eskom Crypto Pilot Consultation Paper — https://www.nersa.org.za/files/files/2026/09/Annexure_A_ESKOM_Crypto_Pilot_Consultation_paperEdited_LTV-vm2.pdf
  • Riot Platforms — Q1 2026 Production and Operations Update — https://www.riotplatforms.com/riot-announces-first-quarter-2026-production-and-operations-updates/