Vodacom discontinues M-Pesa in South Africa: why it failed there
Vodacom discontinued M-Pesa in South Africa, closing the local version of the mobile money service that had become the most cited success story in financial inclusion. The same brand, the same parent group and a similar technical platform produced a transformation in one market and a withdrawal in another, which makes the comparison one of the most instructive in the field.
The conditions that made M-Pesa work elsewhere
In Kenya, mobile money arrived where formal banking had left a wide gap. Branches were few and far from most people, opening an account was slow and expensive, and moving money to a relative in another region meant handing cash to a bus driver. A phone-based transfer was not a better version of an existing service — it was the first version of any service.
Two further conditions did the rest. One operator held a dominant share of mobile subscriptions, which made a network effect possible almost immediately; and the regulator allowed a telecommunications company to hold customer funds and operate through small retail agents while supervision caught up. Agents mattered most: a dense grid of shops where cash could be turned into electronic value and back is the part of mobile money that is hard to build and easy to underestimate.
Why South Africa was a different country in the relevant sense
Bank penetration was already high, driven partly by the payment of social grants into accounts, and the existing system worked: debit cards were widely accepted, ATMs were common, and interbank transfers functioned. A person with an account and a card gains little from a wallet that does less.
The remaining unbanked population was, by definition, the hardest to serve — lower incomes, informal work, identity documentation problems — and reaching it required exactly what regulation made expensive. Strict know-your-customer requirements raised the cost of signing up precisely the customers whose transactions would be smallest. The agent network was harder to build in a market where retail is more concentrated and cash handling more tightly controlled, and no single operator held the dominant subscriber share that gives a wallet instant reach.
The general lesson
Mobile money succeeds against the alternative, not in the abstract. Its value is the difference between what exists and what it offers, and in a country with functioning cards, accounts and ATMs that difference is small for most people and concentrated among those who are costly to serve.
That is why later attempts in middle-income markets have gone after specific frictions rather than the whole payment system: remittance corridors, transport fares, merchant acceptance for tiny businesses, cross-border payments. The winning products are narrow.
What the withdrawal did not prove
It did not prove that mobile money is a niche product, and the subsequent decade demonstrated the opposite across much of Africa and South Asia. What it established is a rule that applies to any financial-inclusion product: transplanting a success requires reproducing the conditions that made it one — the gap in provision, the distribution network, the regulatory space and the concentration that lets a network form.
Remove any one of those and the same product, run by the same company, becomes a line item that is eventually closed.
