In June this year, Eskom group CEO Dan Marokane made clear Eskom’s financial woes and hinted at possible “alternative” revenue opportunities. “The business has to reinvent itself and use part of this baseload in a manner that can help it deal with the remainder of its debt pile that is sitting around our necks,” he said. “We have to be looking at alternatives, and there are exciting opportunities around AI and data centres, but also within the space of bitcoin.”
The layperson’s first question: Isn’t Eskom struggling to generate enough power without adding more strain by plugging bitcoin miners into the grid?
It’s actually far more complex than simply meeting demand with more supply, given the intricacies of generating sufficient power for peak demand and then keeping the grid idling at just the right capacity during off-peak periods. But it seems that bitcoin could, hypothetically for now, be a practical way for Eskom to stabilise the grid and generate income from it.
In 2018, Carel de Jager, a former Eskom engineer turned blockchain consultant, floated an idea about the South African power utility mining bitcoin. He posed a question in an essay, Bitcoin Mining as a Revenue Increasing Mechanism for Eskom, and again during a presentation at the 2025 Adopting Bitcoin conference in Cape Town, about whether Eskom could effectively contract the bitcoin network as a client.
Bitcoin pays in real time and stabilises the grid in the same way that Eskom would throttle or shut down power to steel mines
For the uninitiated, bitcoin miners are essentially computers built to handle huge amounts of computation. These miners guess numbers over and over, calculating trillions of possibilities to find one that fits bitcoin’s rules. The winner earns the right to add the next batch of transactions to the network, and earns rewards paid in bitcoin for their service.
It takes a lot of energy to run these computations. The energy demand isn’t a byproduct but an intentional feature. The high cost of mining is what secures the network, since overwhelming the network would require one miner to amass an almost unthinkable amount of energy and capital.
So, how can taking on bitcoin as a client benefit Eskom? The gist of it is that Eskom is falling short of supply during peak times and losing money during off-peak times, as it costs roughly the same at any time to keep power plants running.
De Jager argues that mining operations can use the oversupply during off-peak periods and when demand exceeds capacity, they can be shut down within seconds without incurring any operational costs — as opposed to when Eskom shuts off power to a heavy energy user such as a steel plant during peak demand. It often takes days to power the plant back up, leading to the power utility having to pay the plant for lost production.
The bitcoin network uses as much energy as Eskom is prepared to provide, and the energy provider gets paid about R2 per kilowatt-hour. It pays in real time and stabilises the grid in the same way that Eskom would throttle or shut down power to steel plants.
Using hourly data, De Jager shows that with 2,000MW of surplus power from coal-fired power stations, Eskom could have been mining about 75% of the time over the past five years and mined about 406,794 bitcoin during that period. Had they sold it in real time, total revenue today would have exceeded $12bn.
This simulation assumes that the logistics of setting up such operations go smoothly.
And what if the price of bitcoin had moved the other way in the past few years, adding to Eskom’s financial worries? At the very least, it seems a worthy exploration of time and energy for Eskom. Dan Marokane and his team at Eskom appear to agree, given his recent remarks.
De Wit is country manager for Luno South Africa
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.