When Vodacom relaunched M-PESA in South Africa two years ago, some questioned the wisdom of yet another relaunch of a service that had failed to gain traction. Neither of South Africa's mobile money operators (the other being MTN Mobile Money) had succeeded in gaining a significant active user base, and the underlying fundamentals hadn't changed. So what made Vodacom believe that another relaunch would light a fire under M-PESA South Africa, and where did it all go wrong?

Market Opportunity

Given Vodacom's success with M-PESA in Kenya (the company is the largest shareholder in Safaricom), it is easy to see how the size and value of the South African market was an irresistible proposition. Research carried out by Mondato estimated the value of South Africa's P2P market would grow to US $18bn by 2018. Vodacom had clearly learned from its own mistakes in previous iterations of M-PESA South Africa, and concluded that distribution networks had been a significant cause of its previous failure. Therefore at the relaunch in 2014, it had 8,000 agents ready to go, more than 10 times the number that had been available during its first launch in 2010. Nevertheless, this fell substantially short of one of the secrets for mobile money success that former Safaricom CEO and latterly Director of Mobile Money for Vodafone Group, Michael Joseph, told Mondato Summit Africa 2014:

"You need a mobile money agent on every corner. It's like Starbucks. That's what you have to have."

In a country of South Africa's size, 8,000 agents was barely a drop in a bucket in comparison to Kenya and Tanzania, countries smaller in size and population than South Africa, but which have a combined 150,000 M-PESA agents. Nor would the need to establish a robust and extensive agent network be satisfied by its new banking partner, Bidvest Bank, which had barely 140 branches across South Africa. This was an even lower number than M-PESA's previous bank partner, Nedbank, who had more than 500 branches across the country.

Regulatory Environment

Vodacom's choice of bank partners point to a more fundamental problem for mobile money in South Africa. By law, any mobile money service must be bank-led. Quite why a bank would want to allow an MNO to muscle into their business was never immediately clear, and it is thought that a lack of trust between Nedbank and Vodacom lead to the demise of the previous iteration of M-PESA in the country. Its new partner, Bidvest, however, is a niche bank with a relatively small retail section in its portfolio, meaning that M-PESA was not going to eat its lunch. (MTN partners with The South African Bank of Athens, another niche bank that specializes in partnership banking). Bidvest, however, was even less well-known to low-income, un-banked South Africans than Nedbank was (with its reputation -- and branch locations -- for being a bank for the well-heeled and wealthy).

This speaks to a fundamental, existential challenge to the success of any mobile money operation in South Africa: the 'Big Four' banks (who between them have 90% of the retail banking market) essentially control the National Payment System, access to which is necessary to be in the "e-money" business in the country. The regulatory burden, including requirements for Anti-Money Laundering procedures, is that of the banking sector, with no dedicated regulatory framework for mobile money. This makes it inherently extremely challenging for any MNO or other non-bank entity to create a scalable business model in South Africa.

Value Proposition

Moreover, regulatory requirements imposed by the South African Reserve Bank's requirements for the opening of a mobile money account are essentially the same as for opening a bank account. In a country with 2,000 bank branches and 8,000 M-PESA agents, the value proposition to customers of opening an M-PESA account over a bank account is not immediately clear, at least in terms of access and availability. This problem of an unclear value proposition was compounded by the presence right across the country of well-established non-bank domestic remittance providers, such as chains of grocery stores.

Pick N Pay, for example, is a chain of popular retail outlets across the country that offers domestic remittances without the need to open an account. Customers simply present their I.D., money and receive a PIN, which the recipient can then use to receive their cash at a different Pick N Pay store. (Though it is of note that Pick N Pay has partnered with MTN to offer "Pick N Pay MTN SIM cards", that include "Pick N Pay Mobile Money", which can, in addition to sending and receiving money, be used to pay at the till in Pick N Pay and Boxer stores.)

Critical Mass

This gets to the heart of the problems of mobile money in South Africa: it is not clear who would use it and why. Vodacom CEO Shameel Joosub stated in his press release announcing the closure of M-PESA that he believed that opportunities still existed for the company in the financial services market in South Africa. He may be right, and MTN is clearly sticking it out, for now at least. But there are many who disagree. South Africa has the most extensive banking system of any country in Africa, and 70% of South Africans have a bank account, up from 54% just a few years ago, according to World Bank data. And although that still leaves 30% of the population unbanked (and an unknown proportion of the 70% underbanked or irrelevantly banked), mobile money appears to offer little in the way of convenience or security that cannot easily be accessed elsewhere.

The one market segment that was most likely to benefit from a mobile money service was also excluded from it. Foreign migrant workers in South Africa, from neighboring countries such as Lesotho, Swaziland and Zimbabwe, would have benefited enormously from a lower cost alternative to their current options, which are among the most expensive remittance corridors in the world. However, e-money foreign remittances were not permitted in South Africa until recently, when EcoCash in Zimbabwe last summer finally received approval from SARB to begin operations in cross-border mobile money transfers. The approval might have been meaningless, however, had the government not eased the regulatory requirements for small-value cross-border remittances, which it thankfully did. Nevertheless, the move may have come too late for M-PESA. In the announcement of the shuttering of M-PESA, Joosub noted that it had failed to generate a "critical mass of users" upon which the business model rested. It is ironic, then, that the one "killer" use case that could have helped generate that critical mass, and reversed M-PESA's fortunes in South Africa, was strangled by red tape and banking regulation.

Only time will tell whether mobile money in South Africa has any future at all, and all eyes will be watching MTN with interest. However, the signs are not good. The turbulent history of mobile money in South Africa teaches us that it is difficult for mobile money to take root in a bank-led regulatory environment (a point borne out by the experience of Ghana). However, even if this were to be changed, it is still unclear whether the market conditions in South Africa are right for mobile money. M-PESA did not appear to do anything that South Africans were not already able to do elsewhere, and nor did it offer greater convenience, reliability or value (though it was competitive on price). Vodacom was unable to provide South Africans with a compelling reason to trust them with their money. And without that, mobile money is dead in the water.

Click here to subscribe and receive a weekly Mondato Insight direct to your inbox.