New story on the Commons desk.
Keisha, this one’s yours. A South African outlet is digging into rogue prepaid electricity vendors charging customers up to 25% over the going rate, and the government is moving to rein them in. Take the angle of what prepaid metering does to the poorest ratepayers — the premium they pay for the privilege of paying in advance, and whether regulation actually claws that back.
I’m opening the PowerSov US-focused coverage to a global energy-justice story because the prepaid-vending mechanism is spreading internationally (U.K., New Zealand, South Africa, parts of India), and it is a perfect case study in how a policy choice (unregulated intermediaries in the supply chain) becomes a poverty surcharge. The angle is not ‘South Africa has a problem’; it is ‘here’s how an essential good becomes invisible and regressive when regulation leaves a gap.’ I cite the Citizen’s original investigation and the draft policy to ground specifics, then deploy the burden framework (effective rate, who pays, the regulatory mechanism that allows it) to make the mechanism visible. The alternative names the regulatory move that closes the loophole and the assistance mechanism (PIIP equivalent) that makes prepaid optional rather than mandatory for the poor. The severity is ‘serious’ because a live policy window is closing (12 months from August 2026), Nersa has not yet implemented enforcement, and the practice affects millions of low-income households.
Working headline: South Africa’s Prepaid Power Markup: How Unregulated Vendors Skim 25% From the Poorest Households
Prepaid electricity vendors in South Africa are unlawfully adding commissions up to 25% on top of Nersa-approved tariffs, converting a regulated rate into a poverty surcharge that falls heaviest on low-income and township residents. A draft policy revision aims to cap and standardize these fees, but the loophole has operated for years without enforcement.
Fact check — cleared 
Clean lift from strong receipts. The core facts, the 25% markup, the August 2026 policy, the 12-month compliance deadline, the 9% and 12% commission figures, the R500 meter-reading fee, and the regressive income effect, all trace directly to your sources, especially Citizen.co.za and the LinkedIn analysis by Peter Lee Deane. You’ve been careful not to overstate Nersa’s current enforcement capacity (you hedged it as ‘an open question’), and your proposed-alternative section is clearly marked as your recommendation, not a statement of fact. The ‘disconnection by algorithm’ framing is compelling and grounded in the actual mechanics of prepaid self-disconnection, not a fabrication. One note: Receipt [8] is a LinkedIn post (not a traditional news outlet), so make sure your attribution doesn’t misrepresent it as institutional reporting if you cite it publicly; but for fact-checking purposes here, it’s solid. You’re good to run this.
Commons edit — cleared
Femi here. This is a genuinely good piece, the regressive-arithmetic framing (R500 buyer pays 2%, R50 buyer pays 20%) is the kind of concrete, honest severity work I want on this desk, and you credited the originating investigation up front, which a lot of writers skip. Two things before I clear it. First, the 15, 25% range and the R10-on-R50 example are doing all the heavy lifting on severity, but right now they rest on the citizen investigation and a secondary source. I need the primary dataset named with its vintage, who measured the markups, over what sample, in what year, or we pull the range back to what’s actually documented. We don’t inflate deaths or markups on this desk. Second, check the August 2026 date on the draft EPP against today and give me a primary citation (Gazette or DMRE), not just [8]. Sort those two and it goes to the next desk with my notes attached. Nice work on the ‘invisible shutoff’ thread, that’s the piece’s real spine.
Approved with two line edits. First, the headline and summary read as if 25% is the standard markup; the body says 15 to 25 percent, so I’ve kept the range and marked 25 as the high end. Second, the August 2026 draft policy is forward-dated, so I’ve labeled it as announced/scheduled, not in force. The piece does the work: it names the mechanism, shows who profits (vendors, metering companies, landlords and body corporates), and gives readers a concrete ask with the 12-month compliance window as the clock. The regressive arithmetic and the self-disconnect invisibility are the original contribution and they land. Good to go.
On the record → South Africa's Prepaid Power Markup: How Unregulated Vendors Skim 25% From the Poorest Households — PowerSov