M-PESA, its 12.2 million active customers, and 81,000 agent outlets have fundamentally altered the landscape of financial services in Kenya. More than half the adult population now uses M-PESA, which has contributed to a dramatic increase in the percentage of adults with access to formal financial services since 2009, when just 41% of adults had access. Now, this figure is at 67%.
Back in 2010, CGAP debunked some common misconceptions about the mobile money service. Overall, there is now a much better understanding of how M-PESA works, but there are still some misconceptions.
1. Myth: M-PESA customer funds are held by Safaricom.
Many policymakers we speak with around the world are still concerned about the safety of M-PESA customer funds. The truth is that M-PESA funds are not held (nor used) by Safaricom. The funds are held by a trust which is owned by Vodafone, deposited in several commercial banks, and cannot be accessed by Safaricom. In the event of Safaricom going bankrupt, the creditors of Safaricom would not have access to the M-PESA funds. Safaricom is not even allowed to use the interest earned on these accounts. Instead, it has established the M-PESA Foundation to spend the interest. The Foundation aims to promote education, health and environmental conservation in Kenya.
2. Myth: since 43% of Kenya's GDP is sent through M-PESA, it must pose a systemic risk.
Many publications from the Economist to the Financial Times have quoted the large percent of Kenya’s GDP that flows through M-PESA. However, this statistic is misleading. First, the value of mobile money transactions quoted includes all aggregate flows both into and out of the mobile money system – so at the very least this statistic is double counting. More importantly, GDP is a measure of the value of goods and services an economy produces and does not represent the amount of money that had to flow through it to pay for these goods and services. In actuality, M-PESA flows are roughly equal to the transaction flows of one of Kenya’s larger commercial banks. According to the Central Bank, mobile money contributes to 6.59% of the total national payments throughput value (including both gross and retail) but a staggering 66.56% of the total NPS throughput volume. This means that M-PESA is important, but does not necessarily pose a systemic risk. However, M-PESA’s reach across all economic segments of Kenyans, coupled with its high usage for financial transactions, means that the product’s security can influence public perceptions of safety and security of financial transactions.
3. Myth: Since M-PESA is so popular, most transactions in Kenya must be going through it.
Actually, cash is still king in Kenya. Although 62% of Kenyans are active mobile money users, among the 300 low-income households that took part in the recently released Kenya Financial Diaries, just 1% of expenditures (mostly airtime top-ups) and 3% of all transactions were made electronically. Although a broader ecosytem is growing to facilitate usage of M-PESA at merchants and to pay school fees, for example, it is still primarily used to send and receive money, buy airtime top-ups and as a transactional storage account.
4. Myth: M-PESA-related fraud and crime mostly impacts customers.
Fraud remains a concern, but maybe not in the way you think. Safaricom has aggressively tackled the fraud issue, which has helped to contribute to reduced incidences of fraud for customers. However, fraud is the #1 issue most burdensome to agents, not only in Kenya but throughout East Africa. This is followed closely by the threat of armed robbery. Providers have done little to protect agents against these risks, especially in providing safety against armed robbery.
5. Myth: M-PESA costs virtually nothing to send money to a friend.
M-PESA is certainly a low-cost option for sending money from one part of Kenya to another, especially compared with alternatives like sending money on a bus. However, M-PESA’s price structure for person-to-person transfers is non-linear, even though the costs to Safaricom are the same regardless of the amount sent (as long as cash-in or cash-out is not involved). More interesting still, Safaricom recently slashed P2P prices on low-value transfers, while raising them at the high end. This was likely a response to Equity’s planned launch of the FinServe MVNO, and may be based in part on evidence that consumers sending large values will be willing to pay more for the service than those sending low values. It also raises the question of how high Safaricom’s margins are, if they were able to reduce prices by as much as 67% overnight.
6. Myth: M-PESA agents are the most profitable in the world.
The Helix Institute of Digital Finance's Agent Network Accelerator Survey reports that the world’s largest mobile money agent network is well-managed and generates healthy transaction levels. Kenyan agents do a median of 46 transactions a day. However, they make just $70 profit per month which is less than Uganda’s $78 per month and significantly lower than Tanzania’s $95 per month. This is most likely due to the fact that until very recently, Kenyan agents were restricted to just transacting for Safaricom and not its competitors while in neighboring countries (especially Tanzania), agents can transact across multiple providers.
7. Myth: Poor people don’t use M-PESA.
It’s true that low-income and unbanked households were not early adopters of M-PESA. But, they caught on quickly. In 2008, fewer than 20% of the population living outside Nairobi on less than $1.25 per day used M-PESA. By 2011, this share had expanded to 72 percent. More recent data suggests that active mobile money users are no more likely to be literate or numerate than the national average, further indicating that the service is accessible and used by lower income households. Unfortunately, rural residents, those below the poverty line, and women are less likely to use the newer value-added services such as Lipa Na M-PESA (merchant payments) compared with their counterparts who are urban, above the poverty line and male.
8. Myth: M-Shwari, the saving and loan product, is an M-PESA product.
M-Shwari is a savings and loan product that has proven enormously popular since its launch in November 2012. Commercial Bank of Africa (CBA), who offers the product via M-PESA, now has 897,000 loan accounts, more than any other bank in the country, and 5.6 million deposit accounts, 26% of the country’s total. Each individual customer has a separate bank account within CBA that is subject to all the normal regulatory requirements of any other bank account. This means that these are customers of CBA, deposits are held by CBA and loans are in the books of CBA. M-PESA merely provides access to this simple account through its menu.
9. Myth: M-PESA is only being used to offer financial services.
An increasing number of organizations (at least 55 by CGAP’s latest count) are leveraging M-PESA’s infrastructure to make basic, essential services and utilities – in energy, health, education and water, for example – more accessible to people at the base of the economic pyramid (what we’re calling Digital Finance Plus). Angaza Solar and Grundfos Lifelink are two such examples. These services leverage M-PESA’s “rails” to make payments easier for customers and less expensive for providers.
10. Myth: Safaricom is so dominant that no competitor can threaten its quasi-monopoly status.
M-PESA has faced limited competition but that may be changing fast. The Kenya Communications Authority recently authorized a series of new MVNO licenses, including to Equity Bank (through its subsidiary Finserve Africa). Equity Bank will use SIM overlay technology to give it reliable access to the mobile channel through which it will serve customers – without relying on Safaricom. This could put Equity in a prime position to challenge Safaricom’s dominance. Safaricom is feeling the heat, vigorously challenging the proposed move and defending its turf. Perhaps partly in response to the recognition of changing times, Safaricom recently announced that it will allow its agents to work for other mobile money services, a move that its competitors have been fighting for, for years.
All these new developments may bring headaches to the Kenyan regulators but are fascinating for those of us interested in the potential of digital financial services for financial inclusion, and carry significant potential benefits for consumers and market competition.
Great blog. thanks for demytifying M-pesa mechanism.
According to the Financial Inclusion Research program by InterMedia (linked in #7 bullet point in the post), even the users of more established mobile money products like M-Shwari, not just Lipan a M-Pesa, still exhibit all the characteristics of early adopters. The majority are urban, males, those living above the poverty line, mostly 15 and 34 years old. They are well educated and are likely to report professional or service jobs (teacher, doctor, salesperson, or policeman). They also have access to financial services: two thirds are banked. Overall, M-Shwari users appear financially comfortable but also young and advantageous; in combination, these two factors allow them experiment with M-Shwari without fearing potential financial risks. Currently, only about 1 in 5 users of M-Shwari are those below the poverty line. Yet, among M-Pesa users, 44 percent are below the poverty line. This means that while Kenyans living below the poverty line trust M-Pesa and recognize its importance for their financial lives, this is not the case for M-Shwari and other VAS. It seems that a good question would be, what type of assurance in terms of customer protection, would help boost confidence in VAS among M-Pesa users below the poverty line and encourage them to at least try the services? And following, what type of informational support this group will require to ensure they understand how to use VAS to their advantage?
Thank you for pushing a bit further on the VAS', which are only going to become more important on these types of delivery channels. I would wonder if part of the issue why lower income use M-Shwari less is that the loans are very small if you save only a small amount, and the savings side of this product is emphasized as much in consumers' minds as the borrowing side. I think the marketing of M-Pawa in Tanzania has moreso emphasized the savings element, so I wonder if there may be greater uptake by consumers who don't have enough money to save to borrow in significant amounts, or just want a secure savings vehicle more than they do a loan product. Or, another idea is it could be as simple as the less money you have, the less money you have available to save, hence lower use of M-Shwari to store value by lower-income populations even if it is attractive in theory. What we really need here is to know how much M-Shwari is displacing other savings or credit products used by these households.
I don't feel the uptake of M-Shwari is really about confidence (or lack of) in the product within the different demographics. Because of the ease of accessing and spending money that's been brought about my MPesa, those who want to save are aware that they cannot really trust themselves to save money "in a mobile phone" and retain it as savings without getting the urge to want to spend it. This is a common negative effect experienced by those who tried to use MPesa to save money (before MShwari was launched) and ended up spending the money in Airtime and other mostly unnecessary expenditures because of the ease with which that money could be accessed. (There's practically an MPesa agent everywhere you turn!)
So many would rather have a bank account (not linked to an ATM Card) as a savings account. This way, before they can decide to act on the urge to spend their savings:
1) It must be really really important.
2) There's no other option but to access these savings.
3) They are ready to queue in line at the Banking Halls.
It's mostly about what these customers who use MPesa would logically prefer, and their decision is affected by amongst other considerations, their financial responsibilities ex. do they have families to plan the future for?)
Myth #1 - Missing the point - the whole point is that the Telco's, not the banks, own the MPesa customers' money. The local banks are bypassed and used merely as custodial banks for the Vodafone trust fund. Does Kenya Central Bank have protection for Mpesa customers' money in case Vodafone goes down?
You are not correct. The telcos do not own customers money, the customers do. As the article states, customers M-Pesa funds are held in Trust accounts by the CBK. So to answer your question, the funds are sheltered from Vodafone / Safaricom and would be entirely protected in the event of bankruptcy. In fact compared to a normal bank account, M-Pesa funds are backed 100% by deposits which cannot be said for banks who operate a lower level of liquidity.
Correction regarding point10. Due to the e-payments regulations that came into effect on the 1st of August 2014, Safaricom was no longer allowed to claim exclusivity. Their pr department successfully managed to 're-frame' this as that Safaricom had opened up their agency network. So this is myth nr 11 that needed to be demystified....
Hi Frederik, thanks for the comment. You are right that Safaricom managed to make their announcement just a few weeks before they would have been forced to open their agent network anyhow. Perhaps we should have included this nuance in the blog post. In any case, the general point was that there are positive signs towards some increasing competition in the market.
But are the commercial banks allowed to intermediate on the funds? I assume they are because a nominal interest is paid to the Trust Foundation. Therefore the customers cannot own the money in the trust account because if they did the interest would accrue to them.
Thanks for an informative blog. Myth 4, is actually a fact. Kenya has set a good precedence in adoption of M-PESA generally, and in accordance, your justification on associated crime and fraud should also give assurance to emerging users, on management and/or mitigation thereof. The lessons emanating from this are diverse and very useful to new markets. This is one of the causes for liquidity challenges across the agents, as people now fear holding large balances due to increasing crime. I think the response here should also indicate the safety and security measures taken, what role does insurance companies play? what is the role of the network in ensuring safety of deposits made/loss of cash holdings with agents, you mention that they have done nothing to protect the agents, isn't this discouraging new entrants, especially low income users?
Currently, what is the maximum daily transaction in M-Pesa and M-Shwari? What is the maximum balance in an M-Pesa and M-Shwari account?
Does Safaricom or the Central Bank of Kenya set the maximum daily transaction and the maximum balance in the M-Pesa and M-Shwari account. To what extent are these maximums related to the maximums in accounts in a commercial bank?
Maximum daily transaction limit is KES 140,000 for personal accounts; KES 70,000 per transaction; However there are business accounts that allow businesses to have MPesa "accounts" allowing them higher transaction limits. This is especially convenient for cases of #9 with MPesa increasingly being used for Bill Payments via it's Lipa Na Mpesa product.
"To what extent are these maximums related to the maximums in accounts in a commercial bank?" No relation whatsoever.
On Myth #1: How protected are the Trust Funds?
Granted the customer funds are not held by Safaricom and in the event of insolvency on the part of Safaricom; its creditors would not touch the trust funds. But what about in the event of insolvency of the bank(s) holding the trust account? Are the funds ring-fenced? If so is this double ring fencing? If they are allowed to intermediate on the funds (and that is the only way they can pay interest on it) can this not lead to the economic phenomenon called 'moral hazard'? I say this because the interest payable on the trust account is nominal so ideally this presents itself as a cheap source of funds which if not carefully managed can be leveraged rather 'carelessly'
Great article. Thanks
Amazing stuff! It would be great if this could be serialized across the markets, with country-specific debunks!
is there anyway to adopt or finance an m-pesa user this would be a great way to directly help some of the poorest of the poor