Focus of Agent Based Banking Should be Disbursal of Government Benefits, Not Provision of Savings

A couple of years ago, the RBI generated a lot of optimism in financial inclusion circles by passing regulation allowing banks to outsource functions to third party agents. It was hoped that the RBI’s new model for agent based banking, called the “business correspondent” model, would reduce the cost to banks of providing financial services, especially savings products, to un-served populations while simultaneously reducing the cost to potential clients of accessing banking services.

Unfortunately, only two days after releasing the initial circular on the business correspondent model the RBI tacked on an additional restriction which effectively killed the new model: it prevented Non-Banking Finance Companies, a legal category which includes all of the largest MFIs, from acting as business correspondents. Despite recommendations from both the Rangarajan and Rajan committees as well as numerous independent observers to drop this restriction, the RBI has remained steadfast in its refusal to allow NBFCs to operate as business correspondents. The result is that two years on adoption of the business correspondent remains extremely limited.

The RBI’s motives for barring NBFCs from handling savings under the business correspondent model are not without merit. Up until the mid-90s ordinary NBFCs were allowed to accept savings deposits (certain types of NBFCs are still allowed to accept deposits but only if they have a very large capital base and are subject to numerous restrictions on how they handle funds) and the results were disastrous: by the end of the decade the acronym “NBFC” became virtually synonymous with “Ponzi scheme”. Worse still from the RBI’s perspective, NBFCs proved very adept at finding legal loopholes to continue to accept savings long after the RBI realized that it was a very bad idea for them to do so.

In my own opinion, this regulatory standoff over whether NBFCs should be allowed to handle savings misses the point: the real potential for agent based banking in India lies in delivery of government benefits rather than provision of savings products. Between NREGA, the national old age pension scheme, and Indira Awas Yojana, the government of India will disburse nearly 40,000 crore rs worth of benefits directly to individuals this fiscal year and that figure is set to grow substantially next year. Despite efforts by the central government to force these payments to be made via the formal banking system (for example, the Ministry of Rural Development recently mandated that all NREGA wages be made via a bank account), most of these payments are still made via GP officials or post offices due to the limited number of bank branches in rural areas. There is a huge need for an effective agent based model to allow these payments to be made through the formal banking system. Further, for potential third party agents such as MFIs, delivery of government benefits is a more attractive business proposition than handling savings. Unlike management of savings accounts, delivering government benefits is relatively straightforward and predictable business with relatively easy cash management requirements. In the case of a typical MFI, delivering government benefits might even reduce their cash handling cost as loan officers could collect cash from borrowers making repayments and then distribute the same cash to beneficiaries.

In fact, some companies are already using the business correspondent model to deliver government benefits. Technology companies FINO and ALW (among others) are currently delivering NREGA wages and other benefits on behalf of banks in AP through the use of biometric smartcards and a network of agents equipped with mobile smartcard readers. (In AP, the state government has tasked banks with the responsibility of delivering these government benefits. The banks in turn have outsourced the actual delivery to FINO and ALW, paying them 1.75% of the total amount disbursed as service fee.) An independent case study of the FINO payment system conducted by CMF found that the smartcard system led to greater convenience for the end beneficiary and reduced leakage from corruption all for a nominal cost.

If companies are already using the business correspondent to deliver government benefits then what’s the problem you may ask. The problem is that the business correspondent model doesn’t exactly allow FINO and ALW to do what they are doing. The business correspondent regulation excludes for profit companies, which both FINO and ALW are, from acting as business correspondents. To get around this restriction, FINO and ALW have set up semi-independent section 25 companies (non-profit companies) to interface with banks and clients. These section 25 sister organisations funnel money back to FINO and ALW through “technology license fees”.

Forcing companies to play this shell accounting game to act as third party agents and deliver government benefits has two main drawbacks. First, it leaves them in a state of regulatory limbo subject to the whims and fancies of the RBI. FINO and ALW are both doing great work and they obviously have the state governments in their corner in the case of a showdown with the RBI. Still, it would be better to normalize this practice and save FINO and ALW the worry of wondering whether the RBI will crack down on this practice in the future.

The second, and more significant, drawback is that it effectively forces the organisation acting as the business correspondent to serve as both the technology provider and the deliverer of payments. This means that only large technology companies like FINO and ALW which have the capacity to both develop suitable technology and also set up and monitor a team of agents disbursing payments can really enter the market. (This is also why you probably don’t hear FINO or ALW complaining too much about the current regulatory framework: it effectively works to their favour by limiting the entrance of other players.)

Here’s what I propose: the RBI should create a new type of banking agent authorized to deliver government benefits (but not handle savings) and include NBFCs in the list of organisations approved to act as this type of agent. The benefit of this move would be that many more organisations – large MFIs such as SKS and Spandana, deposit taking NBFCs such as Sahara and Peerless, and others – could then serve as agents to deliver government benefits. For organisations such as MFIs, the business case of delivering government payments would be very attractive: after visiting a village to collect money from borrowers they could simply turn around and distribute the same money to beneficiaries. It might even reduce the overall cash handling costs.

From a regulatory perspective, allowing an organisation to disburse government benefits is a much less scary proposition than allowing it to handle savings. Down the road, if the RBI finds that NBFCs acting as agents to deliver government benefits are behaving prudently and responsibly they could gradually expand the number of activities these agents could perform on behalf of banks (for example, first they could add on remittances, then savings, and finally insurance).