At the launch of Google Wallet, Osama Bedier, the company’s
VP of payments, shows how to pay for a Coke with a phone 
                    

Perhaps the world’s most famous mobile payments solution is Kenya’s M-PESA, provided by Safaricom, which is part-owned by Vodafone and also has management responsibility for the company.
The UK operator owns the M-Pesa solution, now run by IBM, and licenses Safaricom to use it. In August 2011, the Wall Street Journal reported that Safaricom’s latest annual report shows Vodafone earned $21 million through its Kenyan subsidiary, with $15.6 million coming from M-Pesa in licence fees.
As of November 2011 M-Pesa had over 14 million subscribers — out of Kenya’s population of about 40.5 million — and more than 28,000 agents across the country versus around 600 ATMs.
About a third of Kenya’s gross domestic product passes through M-Pesa and Safaricom earns more money from M- Pesa than it does from SMS — although this is in part because SMS tends to be bundled, free of charge, in the payments system.
M-Pesa is operator-centric, working through an application that sits on all Safaricom SIM cards. To put money on your phone, you walk into an authorised agent, hand over your money, then receive an SMS saying that the money has arrived on your phone.
To send money to someone, you go to the pay menu on the phone, look for the person in your phonebook, or add their details, then send them the amount. They get a message saying, in effect: “If you have an M-Pesa account, you now have 50 shillings [say] on your phone. If you are not an M-Pesa account holder, go to any agent and they will give you the money.” 
                     
                   
Who lost out? 
                    
Arguably the biggest losers from M-Pesa’s success were bus drivers, who previously were often paid by city dwellers to take money back to their home villages each week.
The other network service providers were losers too, as lots of people moved to Safaricom because none of the others offered anything like it — before M-Pesa, Safaricom had something like 45% market share, now that’s at about 70%.
Other service providers have since launched mobile payments services, but none is anywhere near reaching the scope and scale of M-Pesa. It benefited hugely from being highly innovative, the first to market and from enabling — though not coming up with — lots of applications beyond simple person-to-person payments.
It’s possible that if Vodafone and its partners were building the service from the ground up today, one thing they would include is open application program interfaces so that applications, including those from banks, could be plugged into the platform easily and quickly.
A good example of where integrating M-Pesa with banking applications would provide savings and efficiencies is bars and shops. Subscribers can use M-Pesa to pay for goods in supermarkets and buy drinks in a bar — indeed there are some bars in Nairobi that will only accept payments from M-Pesa, because it massively cuts the cost and time involved in handling and banking cash, as well as greatly reducing the risk of robbery.
It would be hugely convenient for those running small businesses if they could plug M-Pesa into a simple payroll application linked to a bank account to pay staff, without having to enter the information manually into software.
Including open APIs would not only enable unforeseen new services, allowing banks and individuals to innovate, but would also work in concert with other M-Pesa-type schemes, which at the moment generally operate in silos.
While this isn’t yet an issue, the wider the spread of these services, the more it is likely to become a problem and potentially inhibit the sector’s growth. 
               
               
NTT DoCoMo and other lessons 
               
In Japan, the mobile money market started more than a decade ago, when NTT DoCoMo introduced i-mode phones and their supporting payments ecosystem.
Although there are six or seven contactless card payment systems in the market place, there are relatively few debit and credit cards in circulation and Japan is still a heavily cash-based economy. The proximity payments sector is very competitive. Transport companies have a contactless card whose use has been extended to enable people to pay in convenience stores, at newspaper stands, other stores and venues such as swimming pools.
The biggest point about the Japanese market is how NTT DoCoMo has navigated the situation. It has probably 80 million handsets in circulation with near proximity payment interfaces.
To achieve this, NTT DoCoMo decided that the portfolio of services that it would run over the phones would be very wide.
DoCoMo has been successful in introducing a credit product into Japan, which is not a credit-oriented society. This is good news from the operator’s point of view as credit is a more profitable product than simple payments.
Contrary to common perception, it appears that the number of mobile proximity payments made in Japan fell during 2010. About 10% of all mobile subscribers, or about 9.8 million users, made a mobile proximity payment during December 2010.
It seems that consumers found that ordinary mobile payments involved pushing too many buttons on their phone and having to wait for software to load — poor customer experience in other words. This appears to explain why e-payments continue to be dominated by proximity prepaid cards, except that these payments too appear to be falling. Why?
Prepaid cards are generally only used for public transport and convenience stores, as all the other shops also accept payments from the credit facility embedded in the card’s integrated circuit. Most of these other stores also make a 3% charge for credit-based payments, instead of just 1% that is usual for prepaid transactions. Also, these stores often don’t bother to ask for a PIN entry for credit-based payments, making them painless and very fast.
The rest of us can learn much from this: poor customer experience has a dramatic effect on usage and so do incentives.
But the Japanese aren’t flocking to buy proximity-enabled phones because they want to make payments, but rather because they enable people to do all sorts of other interesting things, which are proving very popular indeed. So while payments might be the essential ingredient in the e-payments and mobile cake, it’s the icing on top that’s drawing the customers.
The most striking example is McDonald’s Club: one out of every eight Japanese are members. Every week their mobile phone receives a voucher for extra fries or a free Coke, say. Club members go to McDonald’s, place their order including the goods covered by the voucher, and tap their phone on the
terminal.
Within 300 ms, the coupon is transferred from the phone to the point of sale terminal, where it is instantly accounted for. Then the terminal requests the balance, which is paid from the phone. 
               
               
Google Wallet and Isis 
               
When Google first looked at providing payments services, it discovered that it’s hard to make money from payments if you aren’t already part of the established processes: something that DoCoMo realised and hence the reason it went about introducing credit, through well-established channels.
Google also understood that while payments are an essential element of the m-commerce process, the way to make money from them is the other services that can be offered around them, such as vouchers, loyalty programmes and promotions — again as appears to be borne out by the Japanese experience.
Google considered the issue of how a consumer finds and engages with a merchant. What does the merchant do to entice the consumer to buy from them? Then how is the payment handled?
Google is highly experienced and efficient at doing all the front-end processes — providing search, delivering relevant advertising and location based services — so it is logical that adding payments would be adding value, getting merchants and consumers to engage more quickly and easily.
The Google Wallet, launched in September 2011, uses near field communications standard and the economics are essentially the same as if it were a card-based payment. The value chain is not changed by the fact that the transaction is taking place via a mobile phone, but it runs in parallel with a new marketing and promotion value chain, with the payments people capturing the added value.
Google is focusing on the consumer engagement, distributing the coupons, enabling consumers to redeem those coupons in retailers through a single tap of their phones on an NFC-enabled PoS terminal and completing the payment in one go. The launch partners for Google Wallet were Citi, MasterCard, First Data and Sprint. If a customer does not have a Citi MasterCard, then there’s a Google prepaid card. The subscriber can link any card they have, regardless of the brand or issuer, to fund the prepaid account.
Google is not sharing revenue with its launch partner, Sprint, which arguably is doing little more than acting as an NFC-enabled phone distribution channel for Google Wallet.
Isis is a joint venture by AT&T Mobility, T-Mobile USA and Verizon Wireless, with partners MasterCard, Visa, Discover and American Express. It will launch in 2012, and has the clear intention of making money from mobile payments by owning the mobile wallet, albeit relying on established payment mechanisms, and the economics and scale they bring.
Whether one model prevails over the other remains to be seen, it seems likely there is room for both models, although one might work better than the other in certain markets. 
               
               
Billing systems 
               
One of the big questions that has been asked around mobile money is could operators use their charging and billing systems for payment of goods — regulations allowing? It would seem that the answer is no: being able to take 30% of the cost of a ringtone, for instance, is one thing, but — as many have discovered to their cost — disrupting the existing payment chain is all but impossible.
While customers don’t even realise the operators are taking 30% to deliver a ringtone, neither they nor the merchant are prepared to pay an extra 30% for being able to make or accept payment for a candy bar through a mobile phone.