The m-banking industry (for the unbanked) has become accustomed to describing m-banking businesses as bank-led, telco-led, and third-party. The three categories are particularly useful in explaining who is driving the business operationally and who might control the revenues. The categories have particular currency in policy circles where the top-line question is what to do with non-banks.

I am particularly intrigued by the different arrangements we see today and would ideally prefer a typology of sorts that helps us make sense of a more complex market. For example, it is useful perhaps to ask where is Safaricom likely to take M-PESA next and situating M-PESA as something other than telco-led may help us shed a light on that future.

Anyone taking a quick scan of the market today will in fact start seeing a range of business arrangements. We have identified four in particular that stand out and, for now, let us call them:

• Money service provider
• M-wallet aggregator
• Mobile bank
• Agent acquirer

Most of the businesses we are looking at are customer facing businesses – so no technology companies that provide platform services or service a part of the value chain. And we are not considering mobile banking businesses that provide information services.

What follows are brief high-level descriptions of each of the arrangements and we’d appreciate your thoughts.

Money service provider: A company that moves money electronically for the customer for a fee per transaction. Remittance providers fall in this category. Any mobile network operator (MNO) doing what the industry understands as straitlaced mobile money – a transfer service or payment service offered, no bank account offered to the customer, a tie-up by the provider with one or more banks to hold the float. The provider earns on transactions alone and there is no explicit intent (or regulatory permission) to offer deposit services or to intermediate funds. Examples are M-PESA in Kenya, M-paisa in Afghanistan and G-CASH in the Philippines. A money service provider could be licensed as an e-money issuer that allows people to maintain balances (without interest) and either runs its own network of locations where customers can cash-in or cash-out of their account or ties up with a company that provides that service. Banks hold float in this instance on behalf of the provider. Banks can issue their own e-money, of course.  You see businesses where the issuer of the e-money float is the bank and a MNO brands and ‘distributes’ the e-money as a money service provider. Banks get a share of the revenue.

M-wallet aggregator:  Here the customer accesses services from multiple financial services providers through a single interface with an intermediary company. The customer opens a service account with this aggregator company and links multiple bank accounts or e-money accounts to it. Even though the company is customer facing, the company needn’t charge customers and sees most of its revenue from servicing multiple financial service providers. An IT solutions company could make it possible for customers to link m-wallets issued by multiple MNOs who are in a money service business. The m-wallet aggregation may be particularly appealing to merchants. You could see a revenue model where the IT company charges the merchants and the banks and pays the MNO for making the m-wallet aggregation solution available on its network. Are both M-chek and SMART Money m-wallet aggregators?

Mobile bank: A mobile operator owns all or part of the bank or MF provider. For example, EasyPaisa in Pakistan which is a tie-up between Telenor and Tameer Microfinance Bank or BanKO in Philippines which is owned in part by GXI, a money service provider owned by Globe Telecom . The business model innovation here, if any, is that transfer pricing between the MNO and bank means a virtual P&L with mostly variable costs on that P&L. So, in other words, both the bank and MNO leverage already incurred costs on distribution and banking-related risk management. Of course, leveraging each other’s “natural” advantages is easier said than done for banks and MNOs. Take distribution, for instance. MNOs have that advantage but it is not clear cut: they may not have strong relationships with their wholesalers of airtime especially if they work in extremely saturated telecom markets where even distributors work with multiple operators. On risk management, for example, we are currently looking at a business where the bank does not think that their risk management policies apply to the agent network. The biggest advantage of this model of course is the ability to bundle services and do additional configurations of products in-house. This could mean more choices on both savings and loan products for the poor.

Agent acquirer: There a number of businesses that provide cash conversion (cash to electronic and vice-versa) for money service providers or m-wallet aggregators. For example, ATM network operators. But we are particularly interested in businesses that operate similar to a white label ATM company or merchant acquirer business. The company primarily handles what could be called banking distribution – identifying, training, branding and setting-up retail points to be a transaction point on behalf of multiple financial services providers. The company could also host accounts and switch transactions. The company could be licensed as an e-money issuer. The revenue model can be both transactional and float based. The provider also gets service fees for opening and maintaining accounts. The company could have control over its brand. We see these businesses as closing the gap significantly for banks to compete with MNOs. We are seeing examples in Colombia, VISANet with Banco do Brasil, Eko in India, Nationlink in the Philippines and possibly GXI for  BanKO.

-Kabir Kumar