East African Interoperability: Dispatches from the Home of M-Pesa

“Is it live yet? Have you checked?”

For weeks, these were the refrains around Nairobi as customers checked their mobile phones to see whether mobile money interoperability had finally arrived. Fortunately, the wait is over. As of 10 April, customers in Kenya can send money between the country’s two largest wallet services, Airtel and M-Pesa, without the intervention of an agent. Telkom’s T-Kash is planned to connect very soon. Ironically, Kenyans will still not see the option to send to another network on their phone menus. The service is available via USSD, but not as part of the SIM Toolkit. This means that users will need to remember a series of codes to perform the transaction (if they are aware of the service at all).

Kenya is not alone in establishing digital payments interoperability. Domestic interoperability for real-time payments is steadily gaining traction across East Africa. In 2014, mobile money providers in Tanzania came together to develop a scheme for multilateral interoperability. In 2017, banks in Kenya launched the real-time money transfer service PesaLink, and mobile network operators (MNOs) in Uganda launched a scheme in their own market. Kenya’s new MNO-led scheme adds to a growing list.

Each of these models is slightly different in terms of the paths followed, the rules agreed, the business models chosen and the technologies that connect participants. The one thing they all have in common is that they represent a multilateral approach to interoperability. Rather than a third-party aggregator facilitating transactions or an uncoordinated array of bilateral contracts setting terms between pairs of providers, providers chose to balance competition and collaboration to make digital payments more widely accepted, convenient and useful for customers.

The success of any interoperability model hinges on whether it expands the value of digital payments for customers. In terms of transaction volumes, MNO interoperability in Tanzania remains East Africa’s most successful scheme, with around 30 percent of all person-to-person (P2P) mobile money transactions moving between wallets of different providers. Tanzania’s MNOs achieved this success in part by adopting an economic model that balanced participants’ incentives and prevented additional surcharging to customers. As noted in a previous post, “Balancing the Economics of Interoperability,” aligning incentives is important to ensure participants remain committed to interoperability.

Yet the single top-line number of 30 percent tells us relatively little about interoperability’s impact on customers — who uses these services, why they send across networks and how the services affect their lives.

A nationally representative survey of Tanzanian mobile phone owners, conducted by CGAP and BFA in 2018 (publication forthcoming), offers some answers. It found that over 60 percent of the country’s mobile money customers had used interoperable transfers. Interestingly, interoperable customers looked almost exactly like their noninteroperable counterparts by gender, age, education and the reasons for using mobile money. However, they differed in a few important ways. Users of interoperable wallet services tended to transact more frequently, send smaller values and keep higher balances on their wallets.

In short, it is fair to say that Tanzanians who use wallet-to-wallet interoperability are doing more to integrate payments into their daily lives. They also appear to be expanding the economic pie for providers. It is impossible to know what market performance might have looked like without interoperability, but based on historical trends, it seems safe to assume that interoperability has grown the overall P2P market around 10 percent by volume and value (comparing market performance before and after the 2014 launch).

Each of the schemes launched in East Africa focused initially on the P2P use case. P2P is a low-hanging fruit because it often makes up a large segment of transactions, does not face the same technical compatibility challenges of merchant payments (for example, when one provider relies on QR codes and another uses SMS) and is far less contentious among industry players than cash deposit or withdrawal.

As the East Africa market evolves, however, the conversation has started expanding to other use cases. Some merchant transactions are already flowing through the interoperable P2P rails in Tanzania, and providers have incentive to formalize this service because it could help them to kick-start the merchant payments space more broadly. As CGAP pointed out in another blog post, IFC estimates that Africa sees $1.5 trillion in merchant payments per year, and digitizing even a fraction of this amount is a “much bigger prize” for providers than P2P transfers.

Interoperability across borders is also emerging as a potentially valuable use case. Another forthcoming study by CGAP and BFA found that around 8 percent of people in East Africa actively send or receive payments across borders. Anecdotally, this percentage actually sounds about the same as the domestic/international breakdown in the developed markets cards space in Africa. Notably, almost half of these cross-border payments are already being made using mobile money through bilateral agreements. However, these bilateral corridors can be expensive, lack transparency for consumers and regulators and provide inconsistent user experiences.

Despite these opportunities for increasing value to East African customers, P2P transfers may not even be the largest value proposition related to cross-border transactions. The same study found that the largest unmet customer demand for cross-border payments was not P2P, but cash withdrawal. Around 75 percent of travelers take cash with them, and most would use their mobile money account abroad if they could. Intuitively, this makes sense. When wealthier banked customers travel abroad, stopping at an ATM to withdraw local currency is often one of their first priorities.

Scheme conversations will continue to progress domestically and regionally to address customers’ pain points. Progress is being made, and that is good. Yet there is also a larger objective to keep in mind as more and more payment schemes reach the market: not repeating the mistakes of the past.

Today, the East Africa retail banking sector serves less than half the value of mobile money, yet it relies on more than three times the amount of switching infrastructure. Costs are high. Penetration is low. The East Africa mobile money sector should avoid replicating an ecosystem that includes separate switches for each use case, channel, provider type or even just each country. The result would be an industry locked into subscale arrangements with customers bearing the burden in the form of high costs and poor user experiences.

A more holistic strategy means looking beyond individual use cases to consider what makes sense, on an economic basis, to drive scale and customer value. This is why the latest developments in places like Kenya, Tanzania and Uganda are encouraging. Industry is focusing on schemes over switches, and keeping its options open for the future. 

A supplementary slide deck can be found here.