Africa’s smartphone market just posted its first decline in three years, and South Africa grew 17% straight through it. If you read that as a boom, I understand the instinct. It isn’t one. The number measures something else entirely: a market climbing the price ladder while the bottom rung gets quietly sawn off.
What the 17% actually counts
Omdia’s tracker puts SA shipments up 17% year on year in Q2 while the continent fell 7% to 17.8 million units. Nigeria dropped 11%, Kenya 15% and Egypt down a brutal 26%. Samsung grew 15% to 3.9 million units and now holds 22% of African shipments, with Honor up 13% on the back of mid-range 5G stock. On paper, dominance.
Then look at what happened at the tills. NielsenIQ’s half-year retail read shows the telecoms category up just 1.1% in value while unit sales fell 7.9%, with average selling prices climbing 9.8%. Fewer phones left the shops, and each one cost more. That doesn’t contradict the 17%. It’s the same premiumisation seen from the other end, and MTN has been blunt that its customers are shifting into higher-value 5G devices rather than buying more of them.
The rand did real work here, stretching purchasing power just as the 5G upgrade cycle matured. Treasury’s 2025 decision to scrap the 9% ad valorem duty on phones under R2,500 kept the entry tier stocked while the continent’s sub-$100 segment collapsed 34%. We’ve been watching entry-level listings locally and the pattern is hard to miss. The cheapest Galaxy and Honor models cluster suspiciously close to that R2,500 line, like they all know exactly where the cliff is.
Rentals quietly rebuilt the mid-market
Here’s the part most coverage treats as a footnote. Pep’s FoneYam, alongside PayJoy and MyFlex, has turned rent-to-own into the main route millions of South Africans use to get a proper phone, and MTN’s rental book alone hit R2.6 billion by March. I’ve watched that aisle change over the past year. The devices moved from basic handsets to genuinely good mid-rangers, paid off in weekly instalments that fit a grant-cycle or weekly-wage budget.
It’s the old furniture-account model reborn for the smartphone era, and it explains something the shipment data can’t. Households are stretched, borrowing costs are keeping bond repayments heavy, and the only way a phone budget moves is when it’s spread across twelve months instead of one payday.
There’s a catch in the fine print I wish more renters clocked before signing. Rental devices run remote lock software, usually Samsung’s Knox, and a missed payment can disable the phone from the other side of the country. Until the last instalment clears you’re renting a device with a landlord who holds the keys. In a country where that phone is often the only computer in the house, that’s not a small trade-off.
And the alternative is telling. I’ve spent time in local buy-and-sell groups recently and the secondhand phone trade has become risky enough that sellers arrange handovers at police stations by default. When offloading your old handset feels like a security risk, rent-to-own stops looking like desperation and starts looking like the sensible option.
The cheap phone is disappearing
Now the uncomfortable bit. The price pressure isn’t local. AI data centres are hoovering up the world’s DRAM and NAND supply, memory costs have surged, and phone makers are passing the pain straight down the chain. Vendors are quietly deprioritising the ultra-cheap models that got much of Africa online in the first place, and the continent’s sub-$100 segment is where it bites hardest.
Which brings me to the quiet flaw in SA’s success. That R2,500 exemption is a fixed line in a market where prices only move one way. Handsets that sat at R2,199 last year are drifting past the ceiling and losing the benefit, so the policy that insulated the entry tier is being outflanked by inflation. Even Apple’s local answer to rising prices is a lease programme rather than a cheaper phone, which tells you where the market’s head is at.
So my read on the 17% is this. It’s real growth, genuinely impressive against a continent in retreat, but it’s growth financed in instalments and propped up by a tax break with an expiry date the memory makers control. Shipments measure phones, not people connected, and if the sub-R2,000 handset keeps fading the next wave of first-time buyers doesn’t get cheaper devices. It gets a longer wait. Watch three things into 2027. Whether Treasury extends the exemption, whether the rand holds its ground, and whether anyone publishes a number for who’s actually buying their first phone. I’d happily trade a point of headline growth for that answer.
