M-Pesa and the African Fintech Revolution

A Mobile Operator’s Foray into Financial Services

This is a free version of Net Interest, my newsletter on financial sector themes. For additional content and supplementary features, please consider signing up as a paid subscriber.

A recurring theme here at Net Interest is that inside every company there’s often a financial services business dying to get out. Sometimes, management buries the financial services business deep – usually the case when it provides credit for the rest of the company to grow. But sometimes, led by market demand, management is keen to set its financial services business free.

That’s very much the case in the African telecom market right now. 

Over the past few months, Airtel Africa has lined up a number of investors to take minority stakes in its mobile money business. Airtel Africa is the second largest telecom operator in Africa, active in 14 markets across the continent. It offers the usual range of telecom services: mobile voice, mobile data, fixed line, broadband and data centre services. It also operates a mobile money business, through which it offers payments, microloans, savings and international transfers. The mobile money business is relatively small – only 10% of overall revenues – but it’s hot. While stock in the overall company trades in London at a 3.9x multiple to EBITDA, investors such as the Qatar Investment Authority and Mastercard value the mobile money business at a multiple of 13.6x EBITDA. Mobile money may only be 10% of Airtel Africa’s business, but it's a third of its enterprise value. No wonder management wants to set it free.

MTN is shaping up to do something similar. MTN is the largest telecommunications operator in Africa by subscribers, with 278 million customers across Africa and the Middle East. It too has a mobile money business, accounting for around 8% of group revenues. The company’s target is to get that to 20% by 2025. MTN is in the process of structurally separating out this business. Its CEO said recently:

“With similar valuations to that of Airtel, our valuation would sit at 75 billion rand, or about $5 billion. No decision has been made as yet, but listing will be an option considered if that will be the best approach to unlock value.”

In Africa today, mobile money is bigger business than mobile voice or mobile data. And it all started in Kenya fifteen years ago. 

The Story of M-Pesa

Across huge swathes of the globe, banking penetration is low: people don’t live near banks and they don’t have bank accounts. For many, not having a bank account means not being able to send money, save money or get loans; it also means having to navigate the risks associated with handling cash. 

Back in the mid 2000s, not being able to get a loan was seen as a particular obstacle to economic development. The United Nations declared 2005 the International Year of Microcredit and many agencies analysed ways to provide credit to would-be entrepreneurs in developing markets. Among them was the UK’s Department for International Development. They noticed that in many of these markets, more people had mobile phones than bank accounts, so they partnered with Vodafone to explore ways for mobile phone credits to be used to disburse micro loans. 

Vodafone brought the idea to Kenya, where it managed the country’s leading telecom operator, Safaricom. Only 15% of adults in Kenya had a bank account, but 54% had access to a mobile phone. The company trialled a mobile money scheme in the town of Thika. The trial proved very popular, yet rather than using it for loans, people realised that they could use the system to send money between themselves. One woman on the pilot texted some money to her husband after he was robbed so he could get a bus home; businesses deposited money overnight rather than keep it in a safe; city dwellers used the system to send money back home. Seeing these alternative use cases, Safaricom marketed the project around just that: sending money back home. 

Named M-Pesa (pesa is Swahili for money, m is for mobile), the scheme came to market in March 2007. It immediately took off. Safaricom’s business plan projected it would have 350,000 customers by the end of the year; within nine months it had hit 1.2 million.

M-Pesa allowed customers to deposit, transfer and withdraw money or pay for goods using a virtual account on their mobile phone. The system operated as an application on subscribers’ SIM cards – these were the days before smartphones. To pay money in, subscribers would hand cash to one of Safaricom’s agents who then credited it to their account in the same way they would add airtime. Cashing out could similarly be done through the network of agents, who were incentivised using commissions. 

Today, M-Pesa is ubiquitous in Kenya. It has 28.3 million active users – in an adult population of 30.3 million – and 248,000 agents. In the year to March 2021, $200 billion of transactions flowed through M-Pesa, equivalent to over 2x GDP.

The impact that M-Pesa has had on the Kenyan economy has been significant. One study estimates that M-Pesa has increased per capita consumption levels and lifted 2% of Kenyan households out of poverty. By increasing financial resilience and savings and creating greater occupational choice (especially for women, who moved out of agriculture and into business) the impact may have been greater than via the microloans the UK Department for International Development originally had in mind.

From Mobile Money to ‘The Number One Provider in Financial Services’

At the start, M-Pesa wasn’t really viewed as a profit centre. Safaricom’s first CEO, Michael Joseph, recalled a few years later: 

“A mobile phone operator doesn’t need to make money from M-Pesa. It’s nice if they do, but they don’t need to, because the benefit for them is the stickiness and loyalty of the customer.” 

Yet as it gained scale, M-Pesa did begin to make money. In the latest financial year ended March 2021, M-Pesa revenues outstripped voice revenues as the largest single contributor to Safaricom’s overall revenues. From 21% five years ago, M-Pesa revenues now contribute 31% to total group revenues.

The biggest source of revenue is payment fees. Safaricom assesses a fee for each transaction, based on the amount of money being transferred and whether the payee is a user of M-Pesa. Although fees are waived on very small transfers less than KShs 100 (equivalent to around $0.90), fees are levied on peer-to-peer payments even within the M-Pesa closed system – not always the case in mobile money systems elsewhere in the world. Consequently, M-Pesa’s ‘take rate’ is higher than global peers. In the year before Covid, its take rate was 0.61%, which compares with less than 0.05% at Alipay in China.

M-Pesa profits did take a knock during the pandemic. Guided by the Central Bank, the company exempted certain transactions from fees in order to encourage wider use of electronic payments over cash. Activity levels rose, with active user numbers growing at their fastest rate in four years and the value of transactions growing at a multiple of that. But revenues declined modestly, as around a fifth of transactions by value attracted no fees. At the beginning of this year, the company reinstated fees albeit at a reduced rate, so the take rate is likely to come down. 

In addition to fees from payments, M-Pesa earns revenues from other financial products it has rolled out over the years:

  • In 2012, it launched a loan product, M-Shwari, in conjunction with the Commercial Bank of Africa (now NCBA Bank). The average value of a loan is around $50, on which it charges a facilitation fee of 7.5%.

  • In 2013, Lipa Na M-Pesa was introduced for merchants to accept M-Pesa payments. There are currently 302,000 merchants in Kenya active using the system.

  • More recently, at the start of 2019, the company introduced an overdraft facility, Fuliza, enabling customers to complete transactions at the point of sale when they have insufficient funds, subject to predetermined limits. There are now 1.4 million active users of this product. Fuliza loans are underwritten by partner banks NCBA and KCB Bank. M-Pesa charges an access fee plus daily maintenance fees. 

Lending revenues increased over Covid and in the latest financial year contributed 8% to M-Pesa revenues. 

M-Pesa’s biggest direct cost is the commission it pays to agents. As it has scaled, the share of revenues it pays out to agents has declined. Having paid out 40% of revenues to agents in the form of commissions in 2014, that share has declined to 28%. After commissions, contribution per user has increased from KSh1,300 in 2013 ($16) to KSh2,500 in the full year before COVID ($25). 

Note: Full financial data on M-Pesa is available to paying subscribers. To become one, click here.

Given these strong economics, it’s no surprise M-Pesa has become a central plank in Safaricom’s strategy. In its latest annual report, the company exclaims:

“We plan to expand and diversify our financial service suite of products to make us the number one provider in financial services in the country”.

To achieve this, M-Pesa plans to offer new financial products across wealth management, savings, insurance and credit. In addition, in June this year, it launched a super app. The company’s CFO previewed the move on his interim earnings call last year:

“If you're seeing what Alipay had done, it's the same concept we are looking at and putting M-Pesa as a fully – as a super app for both merchants as well as for the consumer side of it.”

Traditional mobile phones are still dominant in Kenya, but Safaricom has around 8.5 million smartphone users in Kenya for whom an app-based interface makes more sense. Whether the company can get the same penetration in this market, though, remains to be seen. A large part of the success of M-Pesa stems from Safaricom’s control of both the GSM network and the SIM card gateway. Without control of the gateway, growth may be more difficult. 

Data from Apptopia suggests that there have been 1.06 million downloads of the M-Pesa app in Kenya since the beginning of June. In August, the rate of downloads dropped off. However, engagement from these early adopters remains high, with monthly average users tracking at around 860,000.

M-Pesa Outside Kenya

If new financial products and the super app are the first two legs of M-Pesa’s growth strategy, the third is international expansion.

Historically, Safaricom (40% owned by Vodafone) ran M-Pesa in Kenya and Vodacom (60.5% owned by Vodafone) ran M-Pesa elsewhere. The M-Pesa brand was retained by Vodafone. Although Vodacom never had that much success with M-Pesa in its home market of South Africa, it has had varying degrees of success exporting it to some of its international markets: Tanzania, DR Congo, Mozambique, Ghana and Lesotho. 

In March 2020, Vodacom and Safaricom acquired the M-Pesa brand, product development and support services functions from Vodafone in a joint venture. Last year, they invested heavily in the joint venture “so that we can create the future M-Pesa from a platform perspective, but also start to accelerate and share best practice in the way we develop products” (as per the Safaricom CEO on last year’s interim earnings call). 

M-Pesa has 16.1 million customers across these other countries, fewer than in Kenya even though the combined population of these countries is four times larger. It is also less profitable in these other markets. Average revenue per customer in these markets is around $20, which compares with around $32 in Kenya (based on Vodacom disclosures). 

The reasons reflect the difficulties getting new payments architecture off the ground anywhere, and are instructive of the challenges that financial technology faces all over the world. Vodacom tried to launch M-Pesa in South Africa twice, but was unsuccessful each time. Today, it operates a financial services business – Vodacom Financial Services – with 13.3 million customers, operating independently of M-Pesa. But revenue per customer is lower than at M-Pesa ($13 per customer).

One of the reasons for M-Pesa’s inability to get off the ground in South Africa is the relative strength of the banks there: over 70% of adults have a transaction account. But there are many other reasons that explain M-Pesa’s success in Kenya, not all of them replicable in other markets. 

They include:

↣ Permissive regulation. When M-Pesa launched in Kenya, it did so with the financial regulator on side. The Central Bank requires 100% of balances to be held in a trust account, segregated from the balance sheet of Safaricom, but otherwise it adopted a fairly relaxed position. Faced with pressure from banks (who were able to witness what was going on in Kenya), regulators in other markets have not all been as hospitable. In Ghana, for example, the central bank required mobile operators to partner with banks in guidelines released in 2008; it wasn’t until 2015 that the regulatory framework was reformed to allow mobile operators direct access to the market. M-Pesa itself was confronted with similar obstacles when it tried to enter India ten years ago.

↣ Strong distribution. M-Pesa nailed its distribution strategy in Kenya, but it wasn’t easy. As long as cash remains dominant in an economy, a network of agents facilitating ‘cash in’ and ‘cash out’ is paramount. But investing in a network of agents before widespread adoption of the system is expensive. A paper published last year concludes that mobile money struggled to take off in Niger, in spite of latent demand, because of low agent density. 

↣ Mobile dominance. Safaricom has the benefit of a massive market share in Kenya in its core mobile business. When it launched M-Pesa, its share of the mobile market was around 60%; its share today is over 64%. Customer conversion is a lot cheaper than customer acquisition.

↣ Extraneous stimuli. In early 2008, violence erupted across Kenya after a disputed election result. M-Pesa was a convenient way to transfer funds to people who were trapped and, among many, it was seen as a safer harbour of cash than many banks. Electronic money solutions had a similar fillip in India after demonetisation in 2016. More recently, mobile money providers all over the world have experienced something similar after Covid. 

These days, there’s an extra obstacle for mobile money operators to navigate. The growing penetration of smartphones is a headwind that fosters a much more competitive marketplace. Banks as well as startup fintechs are muscling in. Just this week, Standard Bank of South Africa launched a mobile payments platform called Unayo to compete with M-Pesa and others across multiple African markets. Mobile operators still have control of the core infrastructure, but they no longer have a monopoly on the gateway.

Valuing M-Pesa

In the past, Safaricom has bounced around the idea of spinning off M-Pesa. After it took control of the infrastructure from Vodafone, its former CFO said on a call to investors, “never say never… we will continuously evaluate it.” M-Pesa generates higher revenue per customer than Airtel Africa, so even though its business mix is somewhat different, it could attract a similarly high multiple. And at 31% of group earnings, that could mean something.

The challenge is how to deal with the Government of Kenya, which owns 35% of Safaricom. The current structure reflects a happy mix of private and public cooperation. The Government’s stake gives it sufficient incentive to see M-Pesa flourish but, with Safaricom as a buffer, direct intervention is held at bay. A spin-off could disturb this balance. 

An alternative access point is Vodacom, which has joint control of the M-Pesa infrastructure with Safaricom and operates the franchise outside of Kenya. It is listed in South Africa and although the Government Employees Pension Fund has a large shareholding, it is free of government intervention. It also directly owns 35% of Safaricom, so has exposure to the Kenyan M-Pesa business via that shareholding. 

Vodacom bundles its financial services business into a single segment, comprising its proportional share in M-Pesa Kenya, the M-Pesa businesses it operates in other African countries, and Vodacom Financial Services in South Africa. Across them all, it generated $290 million of pre-tax profit last year, 17% of the group total.

According to Vodacom’s CEO: “What's clear is that the financial service part is growing at different multiples, okay? And the multiples are hugely attractive. We’re seeing multiples of 26x-plus.” 

It’s an exciting time for African fintech right now, and M-Pesa sits at the heart of it. As mobile money outgrows mobile everything else, we may yet get to see it as a stand-alone entity.

Full financial data on M-Pesa is available to paying subscribers. To become one, click here.

For more on M-Pesa and the entire African fintech ecosystem, I would recommend Samora Kariuki’s newsletter, Frontier Fintech. Thanks also to former colleague Justin Funnell, a long-time top-rated European telecoms analyst, for his insights on the market. 

Subscribe for More Net Interest