Mad Dog 21/21: Kenya Hear Me Now
March 12, 2012 Hesh Wiener
How do you win a war no matter how it turns out? You sell weapons to all sides. That is what IBM is doing in Kenya, where Vodafone’s Safaricom runs the fabulous M-Pesa mobile payment system. M-Pesa’s transaction volume dwarfs that of Kenya’s banks, five of which have now hired IBM to help them get a piece of the action. Meanwhile, Bharti Airtel, also powered by IBM, has similar ambitions in mobile telephony and micropayment.
M-Pesa is an offspring of Islamic hawala funds transfer; this worries Interpol as it fascinates the Gates Foundation and cheers IBM.
Everyone in Kenya knows about M-Pesa and more than 10 million people or close to half the adults in the country use what amounts to branchless banking system. Even the brand name M-Pesa is sort of familiar. The M stands for mobile and the hyphen makes it all look high tech but the rest of the word, pesa, is Swahili for money.
The whole thing began about five years ago. Two people at Vodafone in London, Nick Hughes and Susie Lonie, dreamed up a scheme for mobile payments. They farmed out the development to a geeky outfit called Sagentia, which orbited Cambridge University. (The British government paid for much of the development as part of its imaginative foreign aid program.) The Sagentia implementation of a secure system for moving funds only needed a GSM cellphone and an ordinary one at that. After the system was up and running, loved by nearly everyone in Kenya and envied by Safaricom’s mobile phone rivals to say nothing of Kenya’s banks, it was time for Sagentia to move on. Vodafone brought in IBM to run the system and most likely to also provide support services wherever the M-Pesa story is retold, a process that seems to be underway in Tanzania, in South Africa, in Afghanistan and, perhaps soon, in India, where the user base could reach hundreds of millions.
When funds are moved with the help of M-Pesa, transaction data is transmitted as SMS text and the actual code that makes M-Pesa work can be baked into the SIM card that gives a GSM phone its identity. In practice this means that users of M-Pesa do not even need their own phone to send or receive funds; they only need their SIM card and access to a phone.
What M-Pesa lets people (and institutions) do is send money from one part of the country to another swiftly and cheaply without all the rigmarole banks require. A typical transaction might involve the movement of about $10 value (in Kenyan Shillings) sent by a breadwinner in a city to his family in the sticks at a cost of $0.30 to $0.50 (but again in Kenyan Shillings). The sender gives the money to an agent. The recipient gets the money (less fees) from another agent. The agents ultimately settle out between themselves. Along the way, many user put funds into their accounts and their accounts are then handed to the agency network, which ultimately puts the money into trust accounts at major banks. The end users don’t get any interest, which keeps their activities more or less in compliance with the no-interest aspects of Islamic banking law. However, the aggregated funds held by the M-Pesa network, Sharia law or not, do earn interest and that interest is apparently used to support and improve the system, not just to feather the banks’ nests.
Most of Kenya is rural and life is lived in villages where most people don’t get a round much. But that’s fine. M-Pesa will always come to town, even if town is only a few buildings with just enough electricity to power a mobile phone mast. In larger villages and in cities, there are outlets in every neighborhood, and more all the time, perhaps one for every 2,000 people. As a result, there are now M-Pesa agents all over Kenya; their green signs are as common as Coca-Cola ads in Mexico or pub signs in London. The network went from nothing to total national coverage in something like three years. It’s probably fair to say no technology ever caught on so fast in Kenya, possibly due to a combination of circumstances that might never occur elsewhere, nor again even in Kenya.
First of all, there was the near-monopoly on mobile phone service held by Safaricom, which probably has 75 to 80 percent of the market. This means that if Safaricom offers something, it’s available everywhere in Kenya at once. When the M-Pesa concept was proposed to the various government authorities, including the telecoms and banking regulators, the government decided to just plain lay off and see what would happen. By the time the banks realized they had missed a terrific opportunity, they could only gripe and learn to live in cooperation with M-Pesa. The basic idea, the concept that made it all so attractive to Kenyans, boils down to three words: send money home.
Basically, Kenyan breadwinners migrate to cities looking for work. They find the work, but they often could never move their families from the sticks to the towns, not without putting their wives and kids into slums. So, to support their families, many Kenyans would send funds home by one or another means. None of the available methods, such as sending a parcel containing cash by bus, was very safe or inexpensive. For the hard-working, low-paid Kenyans, then, a fast, cheap M-Pesa alternative was just about a miracle.
About two years ago, the Bill and Melinda Gates Foundation took a long look at M-Pesa and said that it had established outlets in nearly 17,000 stores across the country. Something like 10 percent of Kenyan GGP was moving through the network. Businesses were joining using the system to make payments as well as to receive payments. Even people traveling home to visit their families were using M-Pesa instead of cash because in Kenya going anywhere with cash in your pocket can be very dangerous, while a SIM card is very easy to hide. In general, the Gates report suggested that M-Pesa was here to stay and that it had brought many benefits to little people and big shots across Kenya.
By the end of last year, IBM was busy signing up Kenyan banks to implement mobile and Internet services with the help of Big Blue’s knowhow, experience, and networking backbone.
What worked in Kenya seemed to work elsewhere, although no country has yet reached the state of M-Pesa ubiquity that Kenya boasts. Still, the power of the scheme is considerable. In Afghanistan, M-Pesa payments enable the government to deliver salaries to policemen (particularly in the boondocks), a process that was leaky and corrupt, to say nothing of physically dangerous, when it used large shipments of cash instead of many tiny mobile messages.
The reason M-Pesa can catch on in Afghanistan, where hardly any other systems from elsewhere are accepted, is in part cultural. M-Pesa is basically a high tech implementation of hawala. Hawala is an Islamic payment system that has roots and law tied to the Koran and the Sharia law that evolved from Muslim preaching and teaching. In a hawala system a party can move funds to another party by bringing the money to a local hawaladar. Next thing you know the recipient will be able to get the money from another hawaladar, also close at hand. The receiver might get the funds in another country and another currency. Still, the transaction will occur, the time lag will be minimal, and the fees will be reasonable. There is no interest payment in the hawala system; interest is prohibited by Sharia law; there are fees for services rendered. The integrity of hawala depends on the honor of hawaladars, and that honor is considerable. Hawala works.
The funds move through hawala systems out of sight of banks and authorities, and that makes police agencies nervous. Like hawala, M-Pesa is also largely invisible, and the records kept by agents are not detailed the way, for example, American bank records of wire transfers, are recorded. But one big difference between M-Pesa and hawala helps M-Pesa remain in the good graces of Kenyan and other authorities: There is a value cap of $500 on M-Pesa transactions. The small value makes it unlikely if not impossible for M-Pesa to be used for money laundering by, for example, drug kingpins or arms dealers. That is not true of hawala, which is a system with no intrinsic boundaries on the size of a transaction and no network boundaries that confine activities to a single country. (The restrictions on M-Pesa may soon be relaxed, however, as mobile carriers, banks, and various authorities are trying to find ways to enable M-Pesa transfers to cross national boundaries and, perhaps, to have a higher ceiling, too.)
As the various players, including in a prominent way IBM, try to extend and expand M-Pesa, along the way possibly adapting it for use in other nations, cultures, and legal systems, the failings of Western law may stand in the way of progress. The legal framework of Europe, based on Roman law, makes agency, an aspect of M-Pesa and hawala, unnecessarily difficult and in some ways impossible, and provides a terrible foundation for efficient small value funds transfer systems. Even English law, which does include a considerable body of thought governing trusts and agency, is complicated and tortuous in ways that militate against M-Pesa and hawala . . . even before the jealousies of banks come into play.
It might take a force more powerful than history, tradition or government to bring about the kind of reform that would permit an M-Pesa funds transfer system to take root in the USA or EU. That might mean Vodafone, but it might take a more powerful force, such as IBM. And if IBM can’t bring about reform, the west might have to wait for Amazon, Google, or Apple to decide it wants to take over the banking system, or maybe only a large portion of it.