NextBillion - An Initiative of the William Davidson Institute at the University of Michigan

A funny thing happened to me the other day. I was giving a presentation on the potential of branchless and in particular mobile banking at an international forum, and apparently I upset the Kenyans in attendance for not having lavished enough praise on M-PESA. (Incidentally, they were not with parent company Safaricom). When I talk about mobile banking I generally use the example of M-PESA because there are few others that are really worth talking about at this point (sadly), and I mostly get “enough with that already” kind of looks. So this was new.

Mind you, I’d began my remaks by saying that M-PESA is “fantastic” and rolled off the usual impressive stats: from 0 to 80 percent of the adult population in just 5 years; cash in/out at more than 30,000 stores across the land; more than 70 percent of electronic transactions are now done by M-PESA; median transaction sizes of around $15, much smaller than normal banking. So what upset them?

I suspect it was three points I made. First, most M-PESA wallets contain very little or no value. I can’t prove it because Safaricom does not disclose these stats, but this statement is consistent with all the evidence I’ve seen. (Safaricom, please disprove me if I am wrong.) I know that Safaricom can’t actively push the savings benefits of M-PESA for regulatory reasons, but my point was that if people do not have significant balances in their accounts they are very unlikely to flash their mobiles for payments in day-to-day life. Instead, they count on M-PESA mostly for situations where cash is truly an urgent need, such as an immediate bill that has come or is past due.

That’s why most people conduct one to three transactions a month. (Again, this is estimated, as Safaricom reports total transactions by value, but not the number of transactions). Stories abound about how you can pay for taxis and late night bars with M-PESA, but in a small survey of everyday stores that I was associated with, less than 1 percent of actual transactions during a four-day study period were carried out using M-PESA. As I have argued elsewhere, you cannot go to cash-lite on empty accounts.

Second, I argued that formal businesses are not using M-PESA much. This is direct observation from a qualitative study I did with my former colleague Amolo Ng’weno for the Financial Sector Deepening Trust of Kenya. (For a summary of findings, click here on why it’s not used more and here for how that could be changed). Most formal businesses are surprisingly wedded to checks to pay suppliers, and they feel entirely comfortable paying salaries into bank accounts. Paradoxically, M-PESA is in much more demand by informal businesses – those without access to M-PESA’s corporate payment services for paying bills or bulk payments.

The previous two points are of course related: Both suggest that there isn’t a whole lot of recycling of M-PESA value going on. That is, most transactions start and end in cash. I often say that M-PESA, rather than displacing cash, has made cash so much more efficient because it acts as a bridge between local cash ecosystems.

Third, I made the point that M-PESA has filled a market gap – remote money transfers — in a stunningly fast and ubiquitous way, but there are many precedents for that happening in other countries, and we don’t celebrate those much, if at all. In the Philippines, it was not the telcos but the pawnshops that spotted the opportunity. Many customers selling goods to pawnshops had the ultimate goal of sending that money to help a friend or family member. Pawnshop owners realized their business could be much improved if they helped people send the money home and especially if they could help their clients get their money back to reclaim their pawned assets. In Colombia, it was the courier companies that noticed people were stuffing cash into envelopes, which became a temptation for thievery for highway robbers and the mail carriers themselves. In both countries, pawnshops and courier companies have now created an electronic cash distribution system through thousands of their own retail outlets. M-PESA is more impressive than any of these independent networks – it’s cheaper, more widely available — but there may be a little phone fetishism when we talk up sub-scale mobile money systems in other countries and fail to mention these very real and very useful alternative payment networks.

I am enormously impressed with M-PESA, and any criticism stems from wanting it to be still much more than it already is. That might be my own intellectual hangup. But equally, we must recognize M-PESA for what it is today: mostly a cash-to-cash money transfer system. If M-PESA accounts were actually used as store-of-value accounts – well, that would definitely set it apart from these other schemes.

Whatever its actual usage, with my recent episode at this conference I now see what is most impressive of all about M-PESA: it is now a source of national pride. What higher aspiration could there be for a business brand? Through financial services; who would have thought that was possible?! Let the record state: Well done, Safaricom.

Ignacio Mas is a freelance consultant focused on mobile money technology and policy.