Chapter 3.1
Introduction and Strategic Direction

Introduction:

The vision statements for payments and savings for small businesses and low-income household are:

By January 1, 2016 each Indian resident, above the age of 18 years, would have an individual, full-service , safe , and secure electronic bank account.

By January 1, 2016, the number and distribution of electronic payment access points would be such that every single resident would be within a fifteen minute walking distance from such a point anywhere in the country. Each such point would allow residents to deposit and withdraw cash to and from their bank accounts and transfer balances from one bank account to another, in a secure environment, for both very small and very large amounts, and pay “reasonable” charges for all of these services. At least one of the deposit products accessible to every resident through the payment access points would offer a positive real rate of return over the consumer price index.

India has a rich and diverse payments infrastructure in place already. A number of measures have been taken by RBI to “activate” this infrastructure⁶⁶ including the creation of:

1. The National Electronic Clearing Service (NECS) to “facilitate centralised processing for repetitive and bulk payment instructions”⁶⁷.

2. The National Electronic Funds Transfer (NEFT) payment system to facilitate (in a batch-mode with netting), “one-to-one funds transfer” from an account in “any bank branch” to an account in “any other bank branch in the country”⁶⁸.

3. The Real Time Gross Settlement (RTGS) System which is similar to NEFT but operates in real-time on a transaction to transaction basis and is “primarily meant for large value transactions”⁶⁹.

4. The Immediate Mobile Payment Service (IMPS), “an instant interbank channel agnostic electronic fund transfer service which includes mobile phones”⁷⁰ between customers of different banks.

5. White Label ATMs (WL-ATMs) by “non-bank entities” recognising that, “investments in ATMs have been leveraged for delivery of a wide variety to customers” and “expanded the scope of banking to anytime, anywhere banking through interoperable platforms”⁷¹. Although there has been nearly 23-25 per cent year-on-year growth in the number of ATMs (1,26,950 as of 30th August 2013⁷²), their deployment has been predominantly in Tier I & II centres. There is a need to expand the reach of ATMs in Tier III to VI centres (classification of centres as prescribed under the Census of India 2011). In spite of the banks' pioneering efforts in this direction, much needs to be done.

6. A Payments system for the issuance and operations of Pre-paid Instruments that are issued by banks and non-banks on an “Open System” and a “Semi-Closed System” basis respectively.⁷³

7. A Mobile Banking system permitting banks to offer services to their customer over mobile phones⁷⁴.

8. A network of Business Correspondents across the length and breadth of the country by permitted Banks to engage individuals, and for-profit and non-profit organisations as their agents for the performance of some of the core banking functions⁷⁵.

However, despite all this effort and the number of initiatives that have been taken, as can be seen from the data and discussion in Chapter 2.2, there is still a vast gap in the availability of even basic payments and savings services for small businesses and low-income households both in rural and urban India. However, the large presence of Business Correspondents, advances in mobile telephony, rural and urban broad-band connectivity, and the large scale roll-out of Unique-ID (Aadhaar) numbers (more than 50 crore generated by the end of November 2013 with more than 3 crore generated in November alone)⁷⁶ now offer an unprecedented opportunity to realise this vision and for India to leap-frog over several other far more developed countries in this area just as it has done in mobile telephony.

Strategic Direction:

As evidenced by the July 2002 date of its “Report of the Working Group on Electronic Money” India was one of the early movers amongst both developed and developing countries in formally exploring the potential of electronic money⁷⁷. However, driven by concerns relating to the loss of seigniorage income of the Reserve Bank of India (“…lesser private sector demand for central bank money and the consequent threat to the existence of the central bank”⁷⁸), and the lower perceived “credibility of non-bank financial institutions in India”⁷⁹, the Working Group recommended a model which restricted the full use electronic money only to scheduled commercial banks with very limited participation being permitted to non-banks⁸⁰. In the last ten years since the report was published, this appears to have been the guiding regulatory philosophy in India, even if not entirely for just these two reasons.

Given the sheer size of the country (33 lakh sq. km., 120 crore population) and the absolute poverty of a majority of its citizens (more than 60 per cent with an income below USD 2 per day)⁸¹, a large proportion of them do not have the financial capacity to absorb the costs associated with the management of physical currency notes and traditional branches. This implies that while some progress can indeed be made using a cross-subsidisation route or the use of more efficient non-branch but cash-based channels such as ATMs and Business Correspondents, the financial inclusion and efficiency gains associated with the wide-spread, even ubiquitous use of electronic money are likely to be very high. For these reasons making access to formal electronic payments infrastructure universal is a key component of the overall vision of financial inclusion and the RBI vision document on payments⁸² correctly aims towards an economy that is eventually entirely cash-less. In this document, RBI lays out the over-arching vision to “proactively encourage electronic payment systems for ushering in a less-cash society in India”. In addition to the obvious benefits to individuals and businesses, moving away from cash-based transactions to electronic transactions also has important efficiency benefits for the Government⁸³. A 2010 McKinsey study⁸⁴ while assessing the costs and benefits of digitising government payments in India estimates that the Government would save more than Rs.1 lakh crore per annum if it were to digitise all payments to and from the government.

In order for this state of the future to be realised, while clearly bank-led strategies will have a key role to play, it will become necessary to:

1. Address the concerns relating to the loss of seigniorage income by either concluding that it is not a serious problem or by finding alternate solutions for it.

2. Add to the work being done by the bank-led channels and develop new ways of harnessing the capabilities of non-bank institutions as independent participants in this strategy while ensuring that concerns of Stability, Transparency, Neutrality, and Responsibility are not violated in the process and in particular a level-playing field is ensured for the existing scheduled commercial banks.

It is estimated that in order to achieve the vision, more than 30 lakh payment points will be required. The number of 30 lakh points itself is not daunting given the number of transaction points that already exist in urban India. What is required is a greater proliferation in rural India so that there is a more even distribution of points which have the ability to handle very small value transactions efficiently and in a secure manner. While, through the dedicated Business Correspondent model, banks have indeed been able to provide this functionality quite effectively in more densely populated urban areas, it is in the very low-revenue rural areas that they have faced a particular challenge and is perhaps the place where independent non-bank players such as telecommunications companies which have far bigger customer bases and a significantly deeper presence, may be able to add the most incremental value for the following reasons:

1. Banks are the most effective and efficient when credit and savings products are offered together. However, while ubiquity and the presence of 30 lakh access points is a desirable vision for payments and savings services, credit by its very nature is a significantly more specialised activity with important customer protection, and risk considerations for the financial system requiring the presence of better trained and higher cost personnel whom the customer needs to access relatively less frequently. Far fewer such points would be needed or for that matter would be supportable from a feasibility and viability point of view. And, while they would be more than adequate in number for the provision of credit, they may not be able to meet the requirements of a ubiquitous payments system.

2. While provision of credit is an important “adjacency” that is available when payment and savings services are integrated with it, there are a number of other businesses with adjacencies to payments which have even higher risk-adjusted margins than credit does, such as retailing and mobile telephony that, under the appropriate regulatory architecture, can increase the penetration of very low cost payment points, particularly in rural and low-income urban neighbourhoods where the revenue flows from purely “bank-like” businesses are not sufficient to ensure channel viability.

3. This approach also brings these non-banks firmly into the fold of financial services and facilitates system wide access to the very valuable data that they hold on their customers which otherwise would be accessible only with great difficulty. There would of course in any case be a need to put in place some carefully designed complementary infrastructure in partnership with these institutions that would facilitate such access. This issue is discussed in great detail in the chapter on Complementary Infrastructure later in the report.

4. When banks seek to partner with such businesses even for purely payments purposes, a problem of “pancaking” also occurs where both the bank and the front-end “Business Correspondent” partner such as a telecommunications company need to maintain account information for the client. Client revenues now have to support the cost of two accounts when they are barely sufficient to support one.

The above discussion does not, by any means, intend to conclude that the existing banks, if they so wished, using a variety of partnerships would not be able to offer payment and savings services to small businesses and low-income households, but it does suggest that the inclusion of additional participants could significantly accelerate progress towards universal access as long as it is done in a way that does not disadvantage existing banks. The experience of both developed and developing countries has been consistent with this view.

In addition to exploring the potential of new channels, given the sheer size of the country and its unbanked population, there are also some critical enablers that will need to be in place to enhance the potential for impact of all the channels and to achieve faster and more orderly progress towards the goal of universal access to payments and savings services. The solutions to each of these problems will necessarily need to be very different from those taken by other smaller developing countries or the developed countries which have followed a more traditional pathway. These are related to the existence of a universal bank account, resolution of Know-Your-Customer (KYC) issues, suitable authentication strategies, and existence of smooth inter-operability across channels. The rest of this chapter will discuss each of these issues.

Universal Electronic Bank Account (UEBA)

There is considerable value in the existence of an electronic bank account that is automatically assigned to every citizen at the point that she gets an acceptable identity number such as Aadhaar. Opening bank accounts in field settings for individuals is both difficult and expensive. However, not having an account becomes a barrier even for the government to operationalise its various DBT schemes to low-income households. Additionally, the bank account has become an essential gateway to all financial services, including those supervised by other regulators. Not having one is equivalent to being shut out of the entire financial system.

One of the vision statements is that by January 1, 2016 each Indian resident, above the age of eighteen years, would have an individual, full-service, safe, and secure electronic bank account. The automatic bank account has to be provided by a high-quality, national, full-service bank. It is recommended that the following approach that leverages the Aadhaar number issuance process be pursued:

1. An instruction to open the bank account should be initiated by UIDAI upon the issuance of an Aadhaar number to an individual over the age of 18. The bank would need to be designated by the customer from amongst the list of banks that have indicated to UIDAI that they would be willing to open such an account with the understanding that it would attract no account opening fee but that the bank would be free to charge for all transactions, including balance enquiry with the understanding that such transactions’ charges would provide the host banks with adequate compensation. In its discussions with members of the Committee, State Bank of India, Bank of Baroda and Axis Bank have indicated their willingness to participate. The Committee recommends that RBI issue a circular that permits banks to open such an account upon the receipt of such an electronic instruction. The Committee also requests all the other banks that have an interest in this to formally write to UIDAI.

2. UIDAI has already indicated that there are over 24 crore individuals that have an Aadhaar number and wish to have a bank account. UIDAI has agreed that it would facilitate an automatic opening of these accounts with banks that show an interest. The Committee also recommends that the RBI issue a circular indicating that no bank can refuse to open an account for a customer who has adequate KYC which specifically includes Aadhaar (as this data sent by UIDAI is same as e-KYC already recognised by RBI). The UIDAI would therefore be able to automatically send a request for those individuals who have not designated a specific bank but have indicated an interest in having a bank account opened to a bank which has a branch closest to the customer. Such a bank would be required to open this account under extant RBI regulations and send a letter to the customer indicating the details of this account.

3. The resident is of course free to open an additional account with any financial institution and map their Aadhaar number to it but the UEBA would remain active as a perpetual account as long as the Aadhaar number remains active. In order to ensure that the account does not get misused without the knowledge of the account-holder, the account could remain inactive until, by using it at least once in a properly authenticated manner she indicates her acceptance of the account, and its terms of use.

If a UEBA is not possible to be issued automatically there is a real risk that banks may directly or indirectly refuse to open bank accounts for under-banked customers which are full service or, as has been seen recently in the context of Basic Savings Bank Deposit accounts⁸⁵, if they indeed open them, refuse to make full service functionality operational on account of the costs involved. To address this issue, in Sweden, for example, banks cannot refuse to open a saving or deposit account under Section 2 of the Banking Business Act of 1987 while in France, Article 58 of the Banking Act, 1984 recognises the principle of the right to a bank account⁸⁶. A similar guideline would need to be issued to banks in India and aggrieved individuals would need to have the right to seek redress from RBI in case they have not received this account. Upon receiving such a complaint the RBI would need to ensure that such an account is opened within 30 days of receipt of the complaint.

An important issue behind the reluctance of banks has been the requirement to bundle a number of free transactions along with a basic bank account⁸⁷. This represents a real cost to the bank⁸⁸ and it is not clear that it represents an equivalent source of value to the customer or if indeed what is offered for free is the best route for the customer to access these services. In order to ensure that each customer is offered a UEBA in a manner that does not become a substantial drain on the banking system and that multiple channels evolve to serve the customer, it would be important for policy to move away from the notion of forced bundling (which is equivalent to hidden subsidies) to a strategy in which the customer needs to pay for every transaction but has a choice of competitive channels through which she can access her account at an affordable cost.

RBI and UEBA

Concerns have been raised around the digitisation of cash and its effect on monetary policy and in particular seigniorage revenue of the RBI⁸⁹. Given the fact that at Rs.12 lakhcrore of notes in circulation⁹⁰ this amounts to over 50 per cent of the RBI’s balance sheet and at an average T-Bill rate of about 7 per cent represents an interest savings of about Rs.84,000 crore (about 1 per cent of Indian GDP) to the RBI (more than 100 per cent of its Gross Income during the same period), this is an important concern. From a public policy perspective seigniorage represents an effective tax on the financial assets of households and firms that are being held using currency notes. This is a global phenomenon and applicable in every country of the world and is a key component of the revenues of the Central Bank and the national exchequer. However, in India (and other developing countries) since the poor are likely to hold a larger proportion of their financial assets in such a form and, given poor progress on financial inclusion, are the least likely to substitute away from it, this tax, levied in its current form, acquires a highly regressive character.

Reports of the BIS Committee on Payment and Settlement Systems suggest that while, “so far no Central Bank has indicated an adverse impact on the size of its balance sheet due to a decline in the value of the banknotes in circulation as a consequence of widespread adoption of e-money. The ECB is of the view that the national Central Banks can maintain the size of their balance sheet if necessary by imposing minimum reserves on e-money issuers or by issuing e-money themselves”⁹¹. In India the switch to electronic money is already on its way but is likely to happen much more slowly but result in the steady reduction in the seigniorage income of the RBI and impair its ability to finance its assets such as the foreign exchange reserves of the country.

All of this builds a case for the RBI or an Agent appointed by it to become the sole issuer of e-money and to invest in its rapid adoption as a medium of exchange with the goal of rapid phase-out of currency notes in circulation on a nation-wide basis. If indeed the RBI is able to become the sole issuer of e-money it could issue it directly to the other purveyors of e-money (as it does with notes in circulation now) and thus maintain only wholesale accounts on its books.

KYC Issues

Current KYC guidelines from the RBI are very different from those of other regulators and require that in addition to a proof of identity the individual provide documented proof of her current address. Proof of identity can be obtained by the individual from local authorities at the place at which they were born but proof of local address is much harder to obtain and is perhaps now the most significant barrier to opening a bank account for many individuals in urban and rural environments.

This requirement appears to be unique to the RBI and is neither required by the global Financial Action Task Force (FATF) nor even by other equally conservative financial regulators. FATF Guidance on Financial Inclusion (2013)⁹² states that, “…in a normal Customer Due Diligence (CDD) scenario, the FATF Recommendations do not require information to be gathered on matters such as occupation, income, or address, which some national AML/CFT regimes mandate, although it may be reasonable in many circumstances to seek some of this information so that effective monitoring for unusual transactions can occur.” The FATF Guidance also expresses a concern that⁹³: “There are potentially negative consequences if the controls designed for standard risks and higher risks are also applied to situations where the risks are lower. This “over-compliance” approach by regulators and financial institutions could exacerbate financial exclusion risk, thereby increasing overall money laundering/terrorism financing risk. Regulators/supervisors should play a role and provide further guidance when institutions overestimate money laundering/terrorism financing risks or adopt overly-conservative control measures.” The US Code of Federal Regulations governing this aspect of banking (31 CFR 103.121) not only does not require banks to verify the current address but defines address extremely broadly to mean: “For an individual: a residential or business street address, or if the individual does not have such an address, an Army Post Office (APO) or Fleet Post Office (FPO) box number, the residential or business street address of next of kin or of another contact individual, or a description of the customer’s physical location.”

The current approach being followed by the RBI is to allow customer accounts to be opened without any documentary proof of identity or current address if the amounts involved are less than Rs.50,000 and to insist on both once that limit is crossed⁹⁴. The Committee believes that a superior approach would be to insist on a strong proof of identity like Aadhaar in all cases and to require financial institutions to internally develop their own risk based processes which are linked to transactions monitoring and usage patterns to identify and address high risk cases once identity has been clearly established. This approach, in the Committee’s view, particularly with the advent of Aadhaar, will make it much easier for multiple channels of financial access to become fully engaged in the task of ensuring universal access to payments and savings services. It will also serve to converge the KYC requirements of multiple financial and non-financial regulators making the process of customer acquisition and customer migration from one service provider to another much easier and more importantly, at lower costs. If this issue is clarified, the recently announced e-KYC guidelines⁹⁵ for example, can significantly accelerate the process of account opening.

Customer Authentication

As the payments network is sought to be scaled up, the manner in which a customer is sought to be identified and authenticated so that repudiation and fraud risks are minimised, becomes very important. Aadhaar is the crucial piece of infrastructure in this regard. If each of the payments points is enabled with an acquiring device with biometric capability, identification and authentication of the customer become very secure. However, there are three concerns on this issue, one that the authentication and transaction platform represents an additional investment of Rs. 15,000 for the transaction point and whether the merchant will be willing to invest in this; two, given that Aadhaar has not been activated as of this date for about 50 per cent of the population and that there are entire states such as Chhattisgarh and Madhya Pradesh which have low penetration, there may be a need for some intermediate solutions; and three, given the slow roll-out of broadband and GPRS, biometric authentication may not be feasible in the most remote parts of the country. In order to build ubiquity each of these issues needs to be addressed carefully:

1. The cost of biometric authentication devices is expected to fall sharply over a period of time due to the multiple-vendor architecture as well as developments in integrating biometrics with phones. Even in the current context if the merchant in-charge of the payment access point has a smart phone already then the add-on biometric device costs only Rs.3,000. This cost is sufficiently low that either merchants would invest in it themselves or the government could launch a scheme through the UIDAI to give these devices away.

2. Currently, India has an optical fibre network at the block level. Bharat Broadband Network Limited (BBNL) has initiated National optical fibre network (NOFN) project to provide connectivity in 2.5 lakh gram Panchayats in India. The Recent TRAI report: “Recommendations on Improving Telecom Services in the North-Eastern states: An Investment Plan”, makes valuable suggestions on improving telecommunications and optic fibre outreach in the context of the North-Eastern states.

3. State Governments need to coordinate more closely with UIDAI, NPR, and BBNL to ensure rapid coverage of their states for both Aadhaar as well as broadband. The Committee recommends that authentication for the purpose of transactions happen in either of three ways:

   1. Fingerprint in combination with the Aadhaar number or the bank account number (Token-less authentication)

   2. One-time Password (OTP) in combination with the Aadhaar number or the bank account number (Token-less authentication)

   3. PIN in combination with the Aadhaar number or the bank account number (Token authentication)

In the context of a UEBA, at the time that the account is created or an Aadhaar number is generated, the customer also provides her mobile number which is then mapped to her Aadhaar number and UEBA. Using this information where the customer does not have access to a payment access point, assuming that she has not changed her mobile number, she could request that a One-Time-Password (OTP) be sent to her registered mobile phone number so that she can then use it in combination with her Aadhaar number to authenticate herself. Developing additional alternative authentication strategies, in the Committee’s view, would only delay the entire process.

Rural Branching Mandate

As discussed in Chapter 2.2, there still is a large gap between the number of full-service access points required for financial inclusion particularly as it concerns payment services and deposit products. While it is possible that the strong emphasis on achieving priority sector outcomes by building specialisations and the inclusion of newer business models and participants would have the effect of automatically and quickly filling in this gap, the Committee feels that just as in the case of PSL requirements, the time is not yet opportune for the complete removal of the 25 per cent rural branching requirement for unbanked rural (Tier 5 and Tier 6) centres⁹⁶. It welcomed the fact that in order to suit customer requirements RBI has already changed its understanding of what constitutes a branch⁹⁷. However, it felt that there is need for additional clarification on issues such as the nature of services provided, hours of operation, and the number of days/week that it should be open. The Committee proposes that under the rural branching mandate, a qualifying branch be understood to have the following features:

1. Minimum services available: account opening with e-KYC on an off-line mode, cash in, cash out, transfer initiation, balance enquiry.

2. Minimum hours of operation: 1,000 working hours per annum.

3. Minimum days of operation: 100 days.

4. Nature of employment of staff: not applicable/any.

5. Minimum infrastructure configuration: ability to directly transact on bank’s CBS, ability to print an account statement.

6. Nature of ownership of infrastructure and premises: not applicable/any.

7. Minimum customer protection: bank takes full responsibility for all banking related grievances of the customer, customers covered by branch audit process of the bank, secured authentication as per bank’s internal guidelines and extant RBI regulations.

The Committee also recommends that the policy of mandatory rural branching be reviewed regularly and be phased out once the goals specified in the vision statement for payments services and deposit products have been achieved.

Designs
The next few chapters in this section of the report will examine each relevant bank and non-bank channel carefully for its potential to serve small businesses and low-income households and, where necessary, make recommendations on how the channel could be strengthened and its potential better realised. The institutional designs that are relevant for the provision of payments and savings services in the Indian context are given below. Each of these designs and their variants has been described earlier in Section 4 in some detail:

1. National Banks operating through their branches or through agents;

2. Regional Banks;

3. Payment Network Operators (PNO); and

4. Payments Banks.

Recommendations:

3.1 Every resident should be issued a Universal Electronic Bank Account (UEBA) automatically at the time of receiving their Aadhaar number by a high quality, national, full-service bank. An instruction to open the bank account should be initiated by UIDAI upon the issuance of an Aadhaar number to an individual over the age of 18. The bank would need to be designated by the customer from amongst the list of banks that have indicated to UIDAI that they would be willing to open such an account with the understanding that it would attract no account opening fee but that the bank would be free to charge for all transactions, including balance enquiry with the understanding that such transactions’ charges would provide the host banks with adequate compensation. The Bank would be required to send the customer a letter communicating details of the account thus opened. The Committee recommends that the RBI issue a circular indicating that no bank can refuse to open an account for a customer who has adequate KYC which specifically includes Aadhaar. [Identical to Recommendation 5.1]

3.2 RBI should require a strong Proof of Identity (POI) for each and every customer and a documentary proof of one national address but waive the requirement of documentary proof for the current address, for the purpose of opening a full-service bank account. It should instead enjoin upon banks to carry out careful tracking of usage and transactions patterns to ascertain the risk levels of the customer and take necessary action based upon risk-based surveillance processes developed internally by each bank. [Identical to Recommendation 5.2]

3.3 Under the existing rural branching mandate, a qualifying branch may be understood to have specified features regarding minimum services available, minimum hours of operation, nature of employment of staff, minimum infrastructure configuration, nature of ownership of infrastructure and premises, and minimum customer protection. In addition, this mandate is to be reviewed regularly and be phased out once the goals specified in the vision statement for payments services and deposit products have been achieved.

3.4 Aadhaar is the key piece of infrastructure to enable a customer to be identified and authenticated so that repudiation and fraud risks are minimised and therefore should become the universal basis for authentication. However, with slow enrolment in some areas and low penetration of biometric devices and internet network connectivity in many areas, intermediate authentication methods such as PIN numbers and OTP could be used. State Governments need to coordinate more closely with UIDAI, NPR, and BBNL to ensure rapid coverage of their states for both Aadhaar and broadband.

Chapter 3.2
National Bank with Branches

Design Description

This is a design that comprises large scheduled commercial banks that operate on a nation-wide basis or a supra-regional basis using traditional branch banking and have the capacity to provide the full suite of payment and savings services including electronic bank accounts, payments services, and a full-range of savings products.

High Cost to Serve

In the context of bank accounts with small-value deposits or nil balances, the branch becomes a very high cost-to-serve channel for banks. Currently, RBI requires all banks to open a certain number of accounts in un-banked areas. The first generation of these accounts were referred to as “No-Frills Accounts” (NFAs) and had very little functionality beyond local deposits and withdrawals. The absence of integration with the Core Banking platforms (CBS) of the banks had created this situation. More recently, RBI has mandated that all these NFAs need to be converted to “Basic Savings Bank Deposit Accounts” (BSBDAs) accounts with a minimum set of features including inter-operability between different banks. Progress on this has been slow because from the bank’s perspective, the principal barrier is the cost of opening these BSBDAs without clear visibility on account balances and usage levels.

The traditional bank branch is neither an effective nor an efficient channel for directly serving low-income households and small businesses for their basic payments and savings product needs. There is also the added factor that branches have limited working hours and are fundamentally not suited for the “anytime access” that customers need. There is sometimes a view that is expressed that income from credit products that are offered to low-income households and small businesses will be able to compensate for losses from savings products. However, as is discussed in detail later in the section on credit products, for loans less than Rs.5 lakh in value, the bank-branches have shown themselves to be an extremely high cost and high-risk credit channel.

Valuable National Resource

Despite these limitations, from a financial inclusion perspective, the branch infrastructure of the banks continues to represent an invaluable national resource that can be more effectively utilised to provide a number of important services on the payments and savings front:

1. Branch as a Gateway: For larger businesses and upper income households who would necessarily seek to build integrated relationships with their financial services provider, as the proportion of such households grows even in rural and semi-urban areas, branches would become an essential component of the delivery strategy of any financial institution for both payments and credit. For small businesses and low-income households that seek to graduate from their relationships with local providers, the presence of such “gateways” to National Banks would be invaluable. In order to free-up capacity to service such customers and to thus maximise their own profitability, branches of National Banks could be strongly encouraged to use agent based models to address routine requests from all customers and all requests from small businesses and low-income households. They would also need to slow down the growth of such branches to match the requirements of larger businesses and upper income households otherwise they risk adversely impacting their own profitability and thus have a negative impact on the stability of the financial system as a whole.

2. Branch as a Cash and Cheque Management Hub: With strong support from the RBI, even if e-money becomes the dominant medium of exchange it is likely to take a number of years and notes in circulation are likely to have an important role. In this context branches, with their strong rooms, trusted and highly trained personnel, and a close relationship with the RBI could become operators of a much larger number of RBI Currency Chests so that both dematerialisation and re-materialisation of currency notes to and from e-money can happen smoothly at a number of locations spread-out nationwide. For the provision of these services these branches could be paid out a service charge.
   
   In terms of its core business, a framework could be developed that allows each bank branch to operate in a fully inter-operable manner vis-à-vis other banks. For example while cash-in, cash-out, and remittances can be handled at agent locations in an interoperable manner, only bank branches currently have the ability to handle cheques. Cheque clearing, can be carried out by branches in an inter-operable manner as the Cheque Truncation process does not require the physical cheque issued by the drawer to be presented for clearing. This could become yet another source of fee-income for bank branches.

3. Branch as an Enrolment Hub: Similarly there will be a continuing need to ensure that customers are properly identified and their electronic identities established. Once the current “camp-mode” of providing Indian residents with Aadhaar numbers concludes, given their extensive presence, bank branches could become a very good system-wide resource for the issuance of Aadhaar numbers and for the careful verification of proof of identity and proof of address documents submitted by residents.

4. Branch as a Risk Management Hub: Experiences both globally as well as in India clearly point to fact that while from a cost perspective the branch is an expensive resource, from a risk management perspective, in an environment where multiple channels will be deployed by the Bank to serve the low-income household and the small business customer, the branch does have an important role to play, particularly if the branch is organised in a manner that allocates substantial decision making powers to it and tracks its overall performance using metrics such as RAROC – thus creating a full “bank-in-a-bank” model.

Chapter 3.3
National Bank with Agents

Design Description

This design refers to banks that seek to extend the reach of their branches through the use agents. It is one of the ways employed by them for reducing costs and improving outreach. Under current regulation an agent used by the bank for these purposes is referred to as the Business Correspondent (BC).

Low Financial Viability

This is a very powerful channel and countries such as Brazil have ensured that vast segments of their populations have been covered by the Business Correspondents of existing commercial banks. Every municipality in Brazil is reached by banks with 1 in 4 municipalities being served only by business correspondents. In 2008, bill payments comprised 75 per cent of volume and 70 per cent of value transacted through Brazilian correspondents⁹⁸. Various studies reveal that in absolute terms, the number of BC points in rural India is very large and showing very encouraging signs of growth both in rural and urban areas. As of March 2013, there were 2,21,341 business correspondents engaged by banks, having grown by 548 per cent since 2010.⁹⁹ However, the fraction of them actively transacting is very low, particularly in rural areas and low-income urban neighbourhoods where they are most needed and alternate channels such as ATMs are not available¹⁰⁰. This has resulted in an absence of financial viability and several have now started to shut down. Field research by College of Agricultural Banking, CGAP, and MicroSave suggest that over 75 per cent of accounts opened and over 25 per cent of BCs are dormant¹⁰¹.

This suggests that urgent action needs to be taken if all the investments that have been made in building these networks are to result in the financial inclusion benefits that were anticipated when the channel was originally conceived. The financial viability of the Business Correspondent, particularly in rural areas, is a key challenge but a number of steps can be taken to address it. The core of the strategy for doing this has three components – improve revenues by exploiting existing or building new “adjacent” revenue streams, ensure fair compensation for services provided by the agent, and reduce costs.

Building Adjacencies

“Adjacencies” can arise from three sources: transactions, credit, and the real-sector.

1. Transactions Adjacencies: Instead of treating the BC as an independent third-party operating on a stand-alone basis, some banks have successfully integrated them seamlessly into their day-to-day operations. Routine transactions of all customers such as balance enquiry, balance statement, and small-value cash-withdrawals are all handled by the Business Correspondent thus allowing the branch itself to focus on high value customers and transactions. The savings in costs that have accrued to the branch have more than paid for the fixed costs associated with operating this channel and have allowed the bank to extend the reach and profitability of its entire operations in its command area. Dena bank illustrates such an arrangement with its BC wherein it pays a monthly remuneration of Rs. 2,000 plus commissions for carrying out operations such as account opening, organising financial literacy camps, sourcing loans resulting in sanction and disbursal, maintaining records, and post-sanction monitoring and recovery of loan accounts.

2. Credit Adjacencies: The services of the BC can also be employed to provide access to a broader range of financial services, including credit and insurance so that he now has multiple revenue streams. While, as discussed later in Chapter 6.1, the existing BC model does not lend itself very well to such an approach being adopted, there are ways in which the BC can be made into a stronger credit originator for the Bank and thus acquire this adjacency effectively but this may well come at the cost of ubiquity since not all BCs would be able to benefit from this adjacency.
   
   The other alternative is for the bank to work with ND-NBFCs as their BCs. These entities currently offer credit on their own account and are regulated like banks in their discharge of that function but, under extant regulations, are not permitted to act as BCs. This stems in part from the concern around the isolation of the cash from deposits collected by the BC on behalf of the bank from the lending activities of the NBFC (comingling risk), and in part from the desire to ensure that banks do not inadvertently end-up strengthening a potential competitor on the assets side (competition risk). The issue of “competition risk” is very much a commercial one and is best left to the bank to resolve as it builds out its partnerships. Even the former issue is related to how the bank is able to ensure that the funds of its customers are not comingled with that of the funds of the ND-NBFC. With technology enabling intra-day clearing of funds, it is not obvious that this is anymore a risk factor that cannot be managed by a bank. Regulation should permit ND-NBFCs to be BCs for National Banks where they can take deposits on behalf of the National Bank and offer credit from their own balance sheet or in risk-sharing partnerships with that bank or any other bank. ND-NBFCs have a vast reach on the ground and those that are classified as NBFC-MFIs in particular are able to access customers well beyond the reach of the bank branches. BC partnerships between banks and such NBFCs could prove to be very valuable.

3. Real Sector Adjacencies: Real sector adjacencies obtain when BCs provide access to payments as well as a broader range of non-financial services such as mobile phone top-ups and retailer revenues. This approach extends what existing merchants with POS terminals do in urban areas into the rural context. Globally large non-financial companies offer payment services to increase the potential to garner new revenues. For instance, Diconsa, an operator of 22,000 grocery stores in rural Mexico, began a program to deliver cash payments from government benefit programs to people in its stores. Since they began doing this, they have seen customer visits and foot traffic increase by 20 per cent¹⁰².This model of ‘adjacent’ non-financial revenue possibilities seems highly scalable without creating additional risks to the financial system. Similar partnerships are being explored in India between banks and Community Service Centres (CSC) as well as mobile phone companies and have demonstrated a great deal of potential for growth. Many CSCs already facilitate the opening of savings accounts, the distribution of government payments, and the distribution and processing of loans and Kisan Credit Cards¹⁰³. Axis Bank, for instance, has partnered with Airtel and Idea to leverage their wide outreach in rural areas.

Adequacy of Fees

These are principally related to when BCs intermediate G2P payments and when they support other transactions.

1. G2P Payments: The Report of the Taskforce on the Aadhaar-enabled Unified Payments Infrastructure recommended that a fee of 3.14 per cent (subject to a cap of Rs. 15.71 per transaction) be paid for Direct Benefit Transfer (DBT) payments originating from governments. However government departments have been reluctant to pay these fees to the banks. This situation needs to be urgently corrected if the governments are to realise the benefits of DBTs actually reaching the end-beneficiary. The Madhya Pradesh State Government has followed an entirely different path and has ensured, through a heroic coordination and database creation effort, that all the government schemes pay-out electronically and do so through a single bank / BC location in every 5 kilometre radius (implying an area of about 80 sq. km.) which also includes the bank account of the Panchayat. This has allowed them to successfully make the DBT payments in some of the poorest parts of the State¹⁰⁴ while ensuring that the branches and BCs that provide these services remain viable despite very low levels of fees being paid to them by the government on account of the sheer volume of money flowing through them. It is hard to see however, how such an effort, while highly admirable, can be replicated elsewhere or even extended deeper into those very regions to provide comprehensive financial services beyond DBT, in the absence of an adequate amount of fees being paid out by the State Government.

2. Charging for services: A rural BC carries out between 1 and 10 transactions per day while an urban agent does 10 to 60 transactions a day. However, despite this, banks are much more willing to charge urban customers than they are rural customers. For instance, on a deposit of Rs. 5,000 an urban customer is charged 2 per cent (or Rs. 100) while the same transaction for a rural customer yields a total charge of Rs. 6 to be shared between the Bank and the BC¹⁰⁵. As discussed earlier, while there is no regulatory barrier, banks would need to allow the BC channel to charge a remunerative fee to the customer if the BC channel is to thrive.

Reduction of costs

While the BC provides an additional channel for Account Opening that is proximate to low income households, the account is hosted with the Sponsor Bank. As a result, the costs associated with account opening and maintenance discussed in the earlier chapter, are applicable here as well. These costs, particularly those associated with Information Technology and the high level of branch-personnel engagement in the process of account opening would need to be resolved to ensure the viability of this channel as a whole.

Interoperability of BCs

Another key limitation that has severely restricted the number of transactions that are being carried out by each BC is the fact that it is restricted to operating only those accounts directly opened by it within the parent bank. Given multiple banks and multiple schemes, it is entirely possible that the members of the community hold bank accounts with multiple banks. A BC with full inter-operability would be able to easily give them access to their own bank accounts despite being appointed by another bank. There are no regulatory barriers to accomplishing this¹⁰⁶. A number of banks have activated this and an inter-change price has been agreed to as a result of the efforts of the IBA. The only barrier now is for each bank to migrate their legacy No-Frills Accounts to an inter-operable switch and follow the RBI guidelines on BSBDAs. There is slow progress on this front.

Specific Regulatory Barriers

There is also value in eliminating regulatory barriers that are preventing the effective utilisation of this channel. In particular historically there are several approvals available to merchants that operate POS Terminals in urban India which, for some reason, have been taken away from rural BCs who offer an identical functionality. These need to be made available to the entire BC network as well. These include:

1. NBFCs are not permitted to own and operate BC networks while they are permitted the same for ATMs and POS terminals. The concern that has been expressed here is that there may be potential for conflict of interest because the NBFC is a lending channel and may co-mingle deposits raised through its BC activity. The Committee is of the view that this may be left to the judgment of banks. There are adequate technology solutions available to do intra-day reconciliations between the Bank and its BC.

2. There are distance criteria (30 km for rural and 5 km for urban) which are specified in terms of the distance between the BC and a branch of the Sponsor Bank, while there is no such restriction for ATMs. The concern that has been expressed here is that some branch oversight of BCs is required from a cash management and operational perspective. The Committee is of the view that this may be left to the judgment of banks to decide.

BC Networks

There are two types of BC networks emerging in the country – one established directly by the bank through their branches and others created by various government and private sector entities. There are a number of benefits and challenges associated with each of them.

1. Pay Parity: Demands for pay and job parity from individual BCs directly engaged by bank branches. This concern arises from a similar experience that banks had when the “pygmy banking” scheme was deployed by them several years ago. Ensuring that each BC has adequate work and is able to earn more on the basis of increased effort and engagement would be the best way to address this concern.

2. Technical Support: Poor technical support provided by branches to individual BCs particularly in case of breakdown of machinery. Ensuring that the contracts with equipment providers include a provision for on-going maintenance will be key to addressing this issue.

3. Performance of CSPs: Low performance by the Customer Service Points (CSPs) of corporate BC networks. The concern here is that while corporate BC networks are able to successfully address the human resources and technical support issues, the quality of performance by their CSPs has a highly variable nature. Compensating corporate BC networks adequately, building in strong penalties for non-performance, and getting branches engaged in the supervision of the CSPs would all be key aspects of addressing the performance issue of CSPs.

Recommendations:

3.5 Restore the permission of ND-NBFCs to act as BCs of a bank. Concerns around commingling can be effectively handled through technology-based solutions such that all settlements happen on an intra-day basis. In addition, eliminate the distance criteria between the BC and the nearest branch of the sponsor bank. Allow Banks to decide operational criteria.

3.6 The Taskforce on Aadhaar Enabled Unified Payment Infrastructure recommended that State Governments pay a fee of 3.14 per cent (subject to a cap of Rs. 15.71 per transaction) for Direct Benefit Transfer (DBT) payments originating from governments. RBI should enjoin upon State Governments to implement the same.

Chapter 3.4
Regional Bank

Design Description

This is a design in which there are several regionally focussed full-service banks, each relatively small in size, which offers credit, deposits, and payments services. This design is also sometimes referred to as “Small Banks” or “Community Banks”. These banks are supervised by the RBI through NABARD and are governed on a day-to-day basis by their local boards. The Cooperative banks though licensed by RBI are registered under the respective State Cooperative banks. A Regional Bank does not use capital markets for its resource raising. While cooperative banks with their network of PACS do not use agents in any form to reach their customers, the RRBs in addition to their branches, have begun to deploy BCs and CSPs to reach out to their clients. As on 30th September 2013, RRBs have deployed 30,352 BCs and 24,279 CSPs¹⁰⁷.

Poor Technology Infrastructure

Regional Banks have lower human resources and infrastructure costs and are closer to the customer and are therefore, better equipped to originate deposits and offer payment services vis-à-vis National Banks that operate through branches. They also have an extensive branch network and very large existing customer bases with the potential to add new ones relatively easily¹⁰⁸. They have experienced personnel who have had many years of experience and training in handling cash and have well developed cash management and fraud-control processes. They are also well equipped to handle all the traditional banking instruments such as cash, cheque, demand drafts, and can relatively easily be trained in the use of new devices that work with biometric authentication. The environment of a branch offers the customer safety and privacy that compares very favourably with that offered by a local agent.

While there has been some progress, more effort is urgently needed to complete the task of upgrading the technological infrastructure of these banks. As on 31 October 2013, 23 of the 31 StCBs and 257 of the 370 DCCBs were on the CBS platform and will be offering technology based services such as NEFT and ATM by 31st March, 2014. While PACS are yet to fully embrace technology, a few State Governments such as those of Odisha and Gujarat have initiated the process of computerising them¹⁰⁹. Such measures will make them very well placed to offer a much broader range of savings products to their customers as well as connect smoothly with national platforms such as RTGS.

Significant Challenges on Credit

However, a significant challenge that they face is in regard to deployment of their resources since their local nature also makes them more prone to “capture”. This has led to persistent governance problems and owing to the higher exposure that they have to local systematic risk (weather, crop prices, and regional economic performance), they are likely to have to pay a higher rate to their depositors which in turn, might create the need to make “riskier” loans resulting in a vicious cycle of rising non-performing assets and eventual losses to their depositors. Addressing these issues is likely to be difficult but feasible for those institutions and State Governments that are willing to resolve them effectively. Possible directions that this effort could take are discussed in detail in Chapter 4.6.

Payments Bank

Regional Banks that are demonstrably unable to perform the credit function could move towards an environment in which they are able to function purely as Payments Banks. While this would be an option available to all banks, it could be mandated for those Regional Banks that have a non-investment grade rating from an independent rating agency or a low internal supervisory rating from the RBI.

Branch as a Cash & Cheque Management Hub

As in the case of National Bank with branches, the branch network created by RRBs and DCCBs represents a useful infrastructure that should be leveraged effectively. Therefore existing rural branches of Regional Banks should be made inter-operable and act as acquirers for all transactions in that region and the regulator must allow Regional Banks to charge a fee for providing inter-operability. Unlike National Banks with branches, not all Regional Banks have a Core Banking solution at present, though this process is on-going. In order to even move in the direction of interoperability, this will be a fundamental prerequisite.

Chapter 3.5
Payment Network Operator

Design Description

A payment network operator is responsible for interoperability, interconnectivity, clearing, and settlement between various payment ecosystem providers such as NPCI, Visa, MasterCard, RBI, and clearing houses such as NEFT.

These are very important channels through which a very wide variety of customers, both low-income and middle-to-high income are able to access the payments systems and their savings and credit accounts with various banks and financial institutions. RuPay cards have been recently issued to farmers that have KCC accounts and represent a promising low-cost innovation. VISA, Mastercard, and RuPay are PNOs that have direct access to the settlement system without the need to process their payments through a sponsor bank even though all their cards are issued only in collaboration with a bank.

Contagion Risk Concerns

Historically ATMs in India have been an integral part of a bank network. As a recent innovation White Label ATMs (WL-ATMs) have been permitted to emerge as independent PNOs. There are currently 2 White Label ATM operators authorised by the RBI, namely Prizm and Tata Communications¹¹⁰. However, currently White Label ATMs (WL-ATMs) require a sponsor Bank at the back-end that is linked to the payment and settlement system. The WL-ATM directly does not have access to the settlements system. As discussed in Section 2, this ‘nested’ design where the PNO speaks to the settlements system through the Sponsor Bank is fraught with contagion risk. If the WL-ATM grows very large compared to the Sponsor Bank, it could create undesirable levels of contagion risk from the PNO to the Bank. In order to prevent this build-up of risk, the WL-ATM could be given direct access to the settlement system.

White-Label BC

A challenge faced by Corporate BCs today is the extent of dependence on the Sponsor Bank, oftentimes for simple operational decisions such as the hiring and location of agents. BCs also tend to be constrained for liquidity on account of high levels of Account Receivables from Banks¹¹¹ and a high level of variability in the interest that the bank and its branch managers have in supporting this network. This level of dependence and the business uncertainty that arises as a consequence is not conducive to long-term planning and growth of the BC channel and nor is it desirable for each bank to try and put in place parallel networks making it difficult for even one of them to be viable, on account of fragmentation of the business. The National Bank with BC design must be enabled to grow, and one way to enable this is by allowing high-quality, independent White Label BC Network Operators to emerge as PNOs which can also increase penetration of payment points in the country. This has already happened in the context of ATMs and the objective here would be to loosen the tight coupling between the BC and a Sponsor Bank by allowing the BC greater operational flexibility as well as more degrees of freedom in determining charges to customers. The White Label BC should be fully inter-operable and will have the ability to work with multiple banks at the back-end. Potential candidates for such a license could include NBFCs, existing corporate BCs, mobile phone companies, consumer goods companies, the post office system, and real sector cooperatives

Emergence of Money Services Businesses
With the emergence of NPCI’s Immediate Payments Service which can allow individuals to transmit money to each other and to and from merchants using mobile phones, even with different banks hosting their bank accounts, there is the very real and very welcome possibility that there will be rapid and spontaneous proliferation in the network of individuals and merchants that are willing to participate in “money services businesses” allowing purchase and sale of goods as well as cash-in-cash-out and remittance services using mobile based transactions to grow rapidly. In the Committee’s view the RBI should be supportive of this trend but require those individuals, merchants, or large companies that wish to become formal Money Services Businesses (MSBs) and advertise themselves as such, to register themselves with local State level authorities and place a deposit with them of an appropriate amount and sign an undertaking to follow a suitable “Fair Practices” code. This has very much been the practice followed in the USA for example to govern the behaviour of the various Check-Cashing Businesses that have proliferated.

RBI should provide an enabling environment for such MSBs to emerge and once it appears that significant players could emerge in this space, issue clear guidelines on licensing and authorisation of MSBs. MSBs in the US, for instance, are governed by State level statutes. The Vermont statutes on banking and insurance covering MSBs for example, provides detailed guidance on eligibility for becoming an MSB and the process for obtaining an MSB licence. There is a one-time application and licence fee of USD 1,000 and the issuance of a licence is based upon a thorough investigation of the applicant’s financial condition and responsibility, financial and business experience, character, and general fitness¹¹².

Recommendations:

3.7 In order to address contagion risk concerns, instead of requiring White Label ATMs to access the settlement systems in a “nested” manner through a sponsor bank, provide them direct access to the settlement system subject to certain prudential conditions, to mitigate operation risk.

3.8 In order to ensure that the BC infrastructure that is established is utilised in an optimal manner and shared by multiple banks, which may each have account holders in a specific geography, allow high-quality White Label BCs to emerge with direct access to settlement systems subject to certain prudential conditions. This would be similar to Recommendation 3.7 vis-à-vis mitigating operations risk in the White Label ATM network.

Chapter 3.6
Payments Bank

Design Description

This design is functionally equivalent to a recently introduced class of companies called pre-paid instrument providers (PPIs) that are permitted to receive cash deposits from customers, store them in a digital “wallet”, and allow customers to pay for goods and services from their digital “wallet”. These companies are currently permitted to accept a maximum amount of Rs.50,000 in their “wallet” from their customers and are required to maintain an escrow account with a scheduled commercial bank where these aggregate amounts received from customers are credited immediately upon receipt. There are currently 27 PPIs authorised by the RBI under the Payment and Settlements Act, 2007¹¹³. These players have enabled significant expansion of low-value payments services among individuals who hitherto have never used banking services.

Even globally, similar innovations have taken root and gone further than they have in India. Most jurisdictions have created room for the participation of non-bank institutions in enabling payments. In 2007, the EU adopted the Payment Services Directive (PSD) for a harmonised legal framework for retail payment services. The PSD contains both prudential requirements and civil law provisions pertaining to the various payment service providers and the payment services they provide. To promote competition, a new group of payment service providers, the so-called “payment institutions”, has been created. They can offer payment services without being a bank and do not have to cover the entire range of services provided by a bank. In addition, the rules pertaining to the execution of transactions have been clearly defined¹¹⁴.A BIS report on payments states that, in Japan, non-banks are allowed to provide funds transfers. In South Africa, non-banks can become designated clearing system participants and have full access to the clearing system provided that they meet the Central Bank’s requirements¹¹⁵.

Given all these developments, any financial inclusion strategy would not be credible if it did not envisage a clear role for independent non-bank participation in the provision of payment and deposit services.

Pre-Paid Instrument Providers

1. KYC Issues

PPIs have been provided relaxed KYC requirements for their customers¹¹⁶ in exchange for limiting the value of transacted amounts on a wallet to Rs. 50,000 and restricting cash-out on wallets to banking outlets alone. Recognising that not allowing cash-out represents a key limitation of the product, a limited pilot has recently been permitted to the PPIs for cash-out. While the restriction of transaction amounts may be justified given that the focus of PPIs is to enable payment services for unbanked individuals, these two measures do not provide adequate protection against AML/CFT risks. Given the growing spread of e-KYC, it may be feasible for PPIs to benefit from this and have KYC standards at par with banks, particularly if the stipulation to obtain documentary evidence for current local address is removed for all providers, including banks.

2. Customer Authentication

As the PPI network is sought to be scaled up, the manner in which a customer is sought to be identified and authenticated so that repudiation and fraud risks are minimised, becomes very important. As in the case of KYC, Aadhaar is the crucial piece of infrastructure in this regard. If each of the payments points is enabled with an acquiring device with biometric capability, identification and authentication of the customer become very secure and concerns regarding AML/CFT are also addressed satisfactorily. However, there are three concerns on this issue, one that the authentication and transaction platform represents an additional investment of Rs. 15,000 for the transaction point and whether the merchant will be willing to invest in this; two, given that Aadhaar has not been activated as of this date for about 50 per cent of the population and that there are entire states such as Chhattisgarh and Madhya Pradesh which have low penetration, there may be a need for some intermediate solutions; and three, given the slow roll-out of broadband and GPRS, biometric authentication may not be feasible in the most remote parts of the country. As discussed earlier, these concerns are either not serious ones or are expected to get resolved in the near future. While Aadhaar is essential for account opening, it is not essential for account operations. The Committee recommends that authentication for the purpose of transactions happen in either of three ways:

1. Fingerprint in combination with the Aadhaar number or the bank account number (Token-less authentication)

2. One-time Password in combination with the Aadhaar number or the bank account number (Token-less authentication)

3. PIN in combination with the Aadhaar number or the bank account number (Token authentication)

Developing additional alternative authentication strategies, in the Committee’s view, would only delay the entire process.

3. Savings Product Designs and Payment of Interest

Currently PPIs are prohibited from paying interest on the balances held by them in the customer’s digital “wallet”. Given that balances in wallets may represent substantial amounts relative to the savings of low-income households in particular, it seems very important to find a mechanism to pay interest on these balances. However, in the current environment, PPIs do not earn interest on their escrow balances with the sponsor bank and hence their ability in turn to pay interest to their wallet customers also would be constrained even if they were permitted to do so.

4. Contagion Risk

There is also the concern about the safety of funds being held by the PPIs that arises from contagion risk. If the sponsor bank fails for some reason then since the amounts held by the PPI with the sponsor bank are at risk, the amounts held by individuals with the PPI are also at risk and do not enjoy the benefit of deposit protection unlike the direct depositors of the sponsor bank itself. Such a nested approach creates opacity and screens the build-up of risk in the system. The PPI has to take a view on the riskiness of its Sponsor Bank that holds its deposit balances and the Sponsor Bank has to worry about the operating quality and the likelihood of a “run” on its partner PPIs¹¹⁷. All nested structures have this feature and there may be greater stability obtained from independent designs where the PPI deals directly with the RBI rather than through a Sponsor Bank.

There is global evidence of a shift in this direction. Brazil’s law 12865 creates a new legal entity known as a 'payments institution,' which will be regulated by the Brazilian Central Bank. Under the law, electronic money is not considered a deposit and is issued by a third party under a licence from the financial sector authority. Digital wallet providers, card issuers, and mobile network operators are eligible to apply for payments institution licences. Initial minimum capital requirements have been established by Circular 3683 (Brazilian Real 2 million), and Circular 3681 includes several risk management requirements (including a minimum net equity). Licensed payments institutions will be granted access to the clearing and settlement facilities operated by the Central Bank, but they are not mandated to use it. Upcoming circulars/resolutions will likely elaborate on this point. The balance of payment accounts must be fully allocated to a specific account at the Central Bank, or else in federal government bonds, as established under Circular 3680. Moreover, this circular allows for simplified KYC procedures for low-value pre-paid payment accounts¹¹⁸.

South Africa is another country that has ventured in this direction. In 2007, the South African Reserve Bank acknowledged that non-bank Payment Service Providers (PSPs) have an important role to play in the payments system. The Reserve Bank published directives in 2007 regulating their participation in the payments system. By the end of 2008, amendments were made to the National Payment System Act to provide the Reserve Bank with the mandate to designate non-banks as designated clearing system participants, thereby formalising their participation in the payments system¹¹⁹.

Payments Bank

Given these significant concerns with the current PPI model with respect to KYC, inability to pay interest on balances, and contagion risk; and taking into account the need to urgently provide access to payment services and deposit products to millions of individuals, the Committee recommends that a set of banks may be licensed under the Banking Regulation Act, which may be referred to as Payments Banks. Potential candidates for such a license could include separately capitalised subsidiaries of NBFCs, existing corporate BCs, mobile phone companies, consumer goods companies, the post office system, and real sector cooperatives.

1. Regulation Neutrality

Payments Banks would need to have the following characteristics to balance proportionate regulation with competitive neutrality:

1. Given that their primary role is to provide payment services and deposit products to small businesses and low-income households, they will be restricted to holding a maximum balance of Rs. 50,000 per customer.

2. They will be required to meet the CRR requirements applicable to all the Scheduled Commercial Banks.

3. They will be required to deposit the balance proceeds in approved SLR securities with a duration of no more than three months and will not be permitted to assume any kind of credit risks.

4. As in the case of a full service bank, it would be required to satisfy capital adequacy requirements for both market risk and operations risk¹²⁰.

5. In view of the fact that they will therefore have a near-zero risk of default, the minimum entry capital requirement for them will be Rs. 50 crore compared to the Rs. 500 crore required for full-service SCBs.

6. They will be required to comply with all other RBI guidelines relevant for SCBs and will be granted all the other rights and privileges that come with that licence.

7. Existing SCBs should be permitted to create a Payments Bank as a subsidiary¹²¹.

2. Competitive Concerns

There is also a concern regarding Payments Banks that are subsidiaries of mobile phone companies, which could potentially face a conflict of interest while serving their partner banks and operating as independent Payments Banks at the same time. In order to pre-empt these anti-competitive concerns, the Committee recommends that all mobile phone companies must be mandated to provide USSD connectivity as per recent TRAI regulations with the price cap of Rs 1.5 per 5 interactive sessions. In addition, they must be mandated to categorise all SMSs related to banking and financial transactions as Priority SMS services with reasonable rates and to be made available to the banking system.

3. Efficacy of Financial Inclusion

There is a concern that since customers need payments, savings, and credit services an approach that offers these to clients separately will not end up serving them adequately. No doubt customers require access to payments and credit on a convenient, continuous basis. However, there is demonstrated customer preference for a combination of differentiated channels to access payments and credit. A customer would often access payments through a 24X7 ATM whether or not owned by her bank branch or via a credit card POS terminal at a merchant location while for credit, they would be happy to go to have their application evaluated at a more central location.

NBFCs, including MFIs and HFCs, have successfully demonstrated that with no access to payments information, they have been able to build strong credit portfolios using a number of credit risk proxies as well as soft information. There is also evidence that for first time borrowers, information from mobile phone records and utility bill payments provides very strong credit intelligence. Furthermore, it is true that when a customer operates her payments account, she also generates information that would be useful for credit decision making. However, it is not clear that bundling any one of these services with credit is the best way to identify credit-worthy customers. The Committee feels that the best way to do this would be to make sure that adequate complementary infrastructure is created (as discussed in chapter 4.9) that the customer herself is able to aggregate data from all these sources and authorise usage by designated providers who assume credit risk on her, such as post-paid mobile phone accounts, utility providers, and formal lenders.

Nothing in the design limits the ability of the customer or the full service Bank to build a relationship on payments and lending. In fact, the current approach uniquely advantages a full service bank to use its own private information on the customer’s payment behaviour in combination with information from credit bureaus, mobile phone records, and utility bill payments to offer competitively priced products to high-quality customers. No other provider brings this competitive advantage to the table. Even in the current landscape, customers often will seek payment and deposit services from one provider while borrowing from another provider. In any case, for equally important services such as life insurance, the underlying provider is distinct from a full service bank. Banks have reached out to providers with adjacencies such as retailers and mobile phone operators in order that they may benefit from the lower marginal cost at which they may be able to offer payment services to their customers. Finally, there is no compelling logic to suggest that customers must be forced to seek payment services only from the specific provider that offers them credit.

4. Viability of Payments Banks

There are concerns about the viability of the Payments Bank model but discussions of the Committee with existing providers suggest that with adequate regulation, the market will be extremely competitive with participation from big and small players alike¹²², particularly if neutral but proportionate regulation is applied to them.

Recommendations:

3.9 Given the difficulties being faced by PPIs and the underlying prudential concerns associated with this model, the existing and new PPI applicants should instead be required to apply for a Payments Bank licence or become Business Correspondents. No additional PPI licences should be granted.

3.10 Under the Banking Regulation Act, a set of banks may be licensed which may be referred to as Payments Banks with the following characteristics:

1. Given that their primary role is to provide payment services and deposit products to small businesses and low-income households, they will be restricted to holding a maximum balance of Rs. 50,000 per customer.

2. They will be required to meet the CRR requirements applicable to all the Scheduled Commercial Banks.

3. They will be required to deposit the balance proceeds in approved SLR securities with a duration of no more than three months and will not be permitted to assume any kind of credit risks.

4. In view of the fact that they will therefore have a near-zero risk of default, the minimum entry capital requirement for them will be Rs. 50 crore compared to the Rs. 500 crore required for full-service SCBs.

5. They will be required to comply with all other RBI guidelines relevant for SCBs and will be granted all the other rights and privileges that come with that licence.

6. Existing SCBs should be permitted to create a Payments Bank as a subsidiary.

3.11 RBI to work with TRAI to ensure that all mobile phone companies, including those with Payments Bank subsidiaries, be mandated to provide USSD connectivity as per recent TRAI regulations with the price cap of Rs. 1.5 per 5 interactive sessions and to categorise all SMSs related to banking and financial transactions as Priority SMS services with reasonable rates and to be made available to the banking system.

Chapter 3.7
Recommendations Regarding Payments and Savings

3.1 Every resident should be issued a Universal Electronic Bank Account (UEBA) automatically at the time of receiving their Aadhaar number by a high quality, national, full-service bank. An instruction to open the bank account should be initiated by UIDAI upon the issuance of an Aadhaar number to an individual over the age of 18. The bank would need to be designated by the customer from amongst the list of banks that have indicated to UIDAI that they would be willing to open such an account with the understanding that it would attract no account opening fee but that the bank would be free to charge for all transactions, including balance enquiry with the understanding that such transactions’ charges would provide the host banks with adequate compensation. The Bank would be required to send the customer a letter communicating details of the account thus opened. The Committee recommends that the RBI issue a circular indicating that no bank can refuse to open an account for a customer who has adequate KYC which specifically includes Aadhaar. [Identical to Recommendation 5.1]

3.2 RBI should require a strong Proof of Identity (POI) for each and every customer and a documentary proof of one national address but waive the requirement of documentary proof for the current address, for the purpose of opening a full-service bank account. It should instead enjoin upon banks to carry out careful tracking of usage and transactions patterns to ascertain the risk levels of the customer and take necessary action based upon risk-based surveillance processes developed internally by each bank. [Identical to Recommendation 5.2]

3.3 Under the existing rural branching mandate, a qualifying branch may be understood to have specified features regarding minimum services available, minimum hours of operation, nature of employment of staff, minimum infrastructure configuration, nature of ownership of infrastructure and premises, and minimum customer protection. In addition, this mandate is to be reviewed regularly and be phased out once the goals specified in the vision statement for payments services and deposit products have been achieved.

3.4 Aadhaar is the key piece of infrastructure to enable a customer to be identified and authenticated so that repudiation and fraud risks are minimised and therefore should become the universal basis for authentication. However, with slow enrolment in some areas and low penetration of biometric devices and internet network connectivity in many areas, intermediate authentication methods such as PIN numbers and OTP could be used. State Governments need to coordinate more closely with UIDAI, NPR, and BBNL to ensure rapid coverage of their states for both Aadhaar and broadband.

3.5 Restore the permission of ND-NBFCs to act as BCs of a bank. Concerns around commingling can be effectively handled through technology-based solutions such that all settlements happen on an intra-day basis. In addition, eliminate the distance criteria between the BC and the nearest branch of the sponsor bank. Allow Banks to decide operational criteria.

3.6 The Taskforce on Aadhaar Enabled Unified Payment Infrastructure recommended that State Governments pay a fee of 3.14 per cent (subject to a cap of Rs. 15.71 per transaction) for Direct Benefit Transfer (DBT) payments originating from governments. RBI should enjoin upon State Governments to implement the same.

3.7 In order to address contagion risk concerns, instead of requiring White Label ATMs to access the settlement systems in a “nested” manner through a sponsor bank, provide them direct access to the settlement system subject to certain prudential conditions, to mitigate operation risk.

3.8 In order to ensure that the BC infrastructure that is established is utilised in an optimal manner and shared by multiple banks, which may each have account holders in a specific geography, allow high-quality White Label BCs to emerge with direct access to settlement systems subject to certain prudential conditions. This would be similar to Recommendation 3.7 vis-à-vis mitigating operations risk in the White Label ATM network.

3.9 Given the difficulties being faced by PPIs and the underlying prudential concerns associated with this model, the existing and new PPI applicants should instead be required to apply for a Payments Bank licence or become Business Correspondents. No additional PPI licences should be granted.

3.10 Under the Banking Regulation Act, a set of banks may be licensed which may be referred to as Payments Banks with the following characteristics:

1. Given that their primary role is to provide payment services and deposit products to small businesses and low-income households, they will be restricted to holding a maximum balance of Rs. 50,000 per customer.

2. They will be required to meet the CRR requirements applicable to all the Scheduled Commercial Banks.

3. They will be required to deposit the balance proceeds in approved SLR securities with a duration of no more than three months and will not be permitted to assume any kind of credit risks.

4. In view of the fact that they will therefore have a near-zero risk of default, the minimum entry capital requirement for them will be Rs. 50 crore compared to the Rs. 500 crore required for full-service SCBs.

5. They will be required to comply with all other RBI guidelines relevant for SCBs and will be granted all the other rights and privileges that come with that licence.

6. Existing SCBs should be permitted to create a Payments Bank as a subsidiary.

3.11 RBI to work with TRAI to ensure that all mobile phone companies, including those with Payments Bank subsidiaries, be mandated to provide USSD connectivity as per recent TRAI regulations with the price cap of Rs. 1.5 per 5 interactive sessions and to categorise all SMSs related to banking and financial transactions as Priority SMS services with reasonable rates and to be made available to the banking system.

⁶⁶RBI (2012) provides a detailed discussion, see: http://rbi.org.in/scripts/PublicationVisionDocuments.aspx?ID=664

⁶⁷Source: National Electronic Clearing Service, Procedural Guidelines, Reserve Bank of India, October 2008,
see: http://rbidocs.rbi.org.in/rdocs/Content/PDFs/87706.pdf

⁶⁸Source: RBI FAQs on NEFT, November 12, 2012, see: http://www.rbi.org.in/scripts/FAQView.aspx?Id=60

⁶⁹Source: RBI FAQs on RTGS, August 30, 2013, see: http://www.rbi.org.in/scripts/FAQView.aspx?Id=65

⁷⁰Source: NPCI FAQs on IMPS, see: http://www.npci.org.in/impsFaqCust.aspx

⁷¹Source: White Label ATMs in India – Guidelines, June 20, 2012, see: http://rbidocs.rbi.org.in/rdocs/notification/PDFs/CWLA200612_F.pdf

⁷²Source: RBI Bankwise ATM/POS/Card statistics, August 2013, see: http://www.rbi.org.in/scripts/ATMView.aspx?atmid=30

⁷³Source: Issuance and Operation of Pre-paid Payment Instruments in India (Reserve Bank) Directions, 2009,
see: http://rbidocs.rbi.org.in/rdocs/notification/PDFs/RFGF280409.pdf

⁷⁴Source: RBI Master Circular – Mobile Banking Transactions in India – Operative Guidelines for Banks, July 1, 2013,
see: http://rbidocs.rbi.org.in/rdocs/notification/PDFs/116MCMB030713.pdf

⁷⁵Source: RBI Guidelines for engaging of Business Correspondents (BCs), September 28, 2010,
see: http://rbidocs.rbi.org.in/rdocs/notification/PDFs/CPC28092010.pdf

⁷⁶Source: UIDAI website, see: https://portal.uidai.gov.in/uidwebportal/dashboard.do

⁷⁷Cama (2002).

⁷⁸Ibid., Page 9.

⁷⁹Ibid., Page 10.

⁸⁰Ibid., Page 11 – 13.

⁸¹Source: World Development Indicators, World Bank, see: http://data.worldbank.org/indicator/SI.POV.2DAY

⁸²RBI (2012).

⁸³Radcliffe & Voorhies (2012).

⁸⁴McKinsey & Company (2010).

⁸⁵Damodaran (2011).

⁸⁶Source: Submission by Bank of Baroda to the Committee.

⁸⁷Source: RBI Circular on BSBDA, August, 2012), see: http://rbi.org.in/scripts/NotificationUser.aspx?Id=7519&Mode=0

⁸⁸RBI (2010), see: http://www.rbi.org.in/Scripts/bs_viewcontent.aspx?Id=2344

⁸⁹“…relative shrinkage in balance sheet [of the RBI] may also have implications regarding loss of seigniorage revenue for the central bank. This is because currency notes and coins are the interest-free liability of the central bank towards public (i.e., public have in effect lent to the central bank such amount free of interest) which are then used by the central bank to purchase interest bearing assets. Thus, interest earned on these assets constitutes its seigniorage revenue. Therefore, if the central bank's operating costs are high which are more likely in an environment where central bank is required to intervene more, it may even incur losses on account of seigniorage revenue foregone. This may also have adverse implications for governments with chronic budget deficits as it would be deprived of the transfer of surplus to its treasury from the profitable central bank”, Cama (2002)

⁹⁰Source: RBI Annual Report 2012-13, see: http://rbidocs.rbi.org.in/rdocs/AnnualReport/PDFs/P2_11RBA220813.pdf

⁹¹Bank of International Settlements Committee on Payment and Settlement Systems (2004), Page 6, March 2004.

⁹²Financial Action Task Force (2013), Paragraph 67, page 28.

⁹³Ibid., Paragraph 39, page 18.

⁹⁴Source: RBI KYC Guidelines Master Circular, July, 2013, Section 2.5, see: http://www.rbi.org.in/scripts/BS_ViewMasCirculardetails.aspx?id=8179#f9

⁹⁵RBI guidelines on e-KYC, see: http://www.rbi.org.in/scripts/BS_CircularIndexDisplay.aspx?Id=8397

⁹⁶RBI Master Circular on Branch Authorisation, July, 2013, see: http://www.rbi.org.in/scripts/BS_ViewMasCirculardetails.aspx?id=8136

⁹⁷RBI Master Circular on Branch Licensing – RRBs, July, 2013, Sections 4 and 8,
see: http://www.rbi.org.in/scripts/BS_ViewMasCirculardetails.aspx?id=8184#8

⁹⁸CGAP - Branchless Banking Agents in Brazil: Building viable networks 2010,
see: http://www.slideshare.net/CGAP/branchless-banking-agents-in-brazil-building-viable-networks-2010

⁹⁹RBI Trend and Progress in Banking in India 2012-13, Table IV.33, page 84,
see: http://rbidocs.rbi.org.in/rdocs/Publications/PDFs/0RTP21112013_F.pdf

¹⁰⁰In its interaction with the regional/zonal heads of PSU banks and RRBs in Uttarakhand, the Committee found that there is a saturation point for BCs' income after the initial opening of deposit accounts. The transactions based model does not generate sufficient revenue on account of various factors such as low population density, hilly terrain, and lack of awareness of banking products and government schemes.

¹⁰¹Wright et al (2013).

¹⁰²Bill & Melinda Gates Foundation (2013).

¹⁰³International Telecommunication Union (2013).

¹⁰⁴Submission by Dr. Aruna Sharma, Additional Chief Secretary, Rural Development, Government of Madhya Pradesh to the Committee.

¹⁰⁵Workshop with BC Organisations, Committee on Comprehensive Financial Services for Small Businesses and Low Income Households, New Delhi, 8th November 2013.

¹⁰⁶Financial Inclusion by Extension of Banking Services - Use of Business Correspondents (BCs), March 2012, RBI,
see: http://www.rbi.org.in/scripts/NotificationUser.aspx?Id=7038&Mode=0

¹⁰⁷Submission from NABARD to the Committee.

¹⁰⁸As on 31 March 2013, 64 Regional Rural Banks (RRBs) function in 635 districts with loans outstanding of Rs.1.37 lakh crore and deposits of Rs.2.11 lakh crore. Rural Cooperative Credit Institutions comprise 31 StCBs and 370 DCCBs that function through 92,432 PACS at the village level (of which 66813 are categorised as Viable). The credit outstanding for StCBs and DCCBs stood at Rs.77,644 crore and Rs. 1.57 lakh crore respectively, and they had deposits of Rs.86,430 crore and Rs.1.88 lakh crore as on March 31 2012. There are also 1606 UCBs across 330 districts with advances and deposits of Rs.1.8 lakh crore and Rs. 2.77 lakh crore respectively, as on March 31, 2013. Source: Data from NABARD, NAFSCOB, RBI.

¹⁰⁹Submission from NABARD to the Committee.

¹¹⁰Source: Certificates of Authorisation issued by the Reserve Bank of India under the Payment and Settlement Systems Act, 2007,
see: http://www.rbi.org.in/scripts/PublicationsView.aspx?id=12043

¹¹¹From discussions at the Workshop with BC Organisations, Committee on Comprehensive Financial Services for Small Businesses and Low Income Households, New Delhi, 8th November 2013.

¹¹²Source: Vermont status on banking and insurance, see: http://www.leg.state.vt.us/statutes/fullchapter.cfm?Title=08&Chapter=079

¹¹³Source: Certificates of Authorisation issued by the Reserve Bank of India under the Payment and Settlement Systems Act, 2007,
see: http://www.rbi.org.in/scripts/PublicationsView.aspx?id=12043

¹¹⁴Bank of International Settlements Committee on Payment and Retail Systems (2012), pg. 82.

¹¹⁵Ibid.

¹¹⁶Source: Issuance and Operation of Pre-paid Payment Instruments in India (Reserve Bank) Directions, 2009,
see: http://rbidocs.rbi.org.in/rdocs/notification/PDFs/RFGF280409.pdf

¹¹⁷Tarazi and Breloff (2010).

¹¹⁸“New Legal and Regulatory Framework for Payment Arrangements and Payment Agents,
see: http://www.pinheironeto.com.br/publicacao/4072

¹¹⁹Bank of International Settlements Committee on Payment and Retail Systems (2012), pg. 84.

¹²⁰Based on the Committee’s interaction with Mrs. Chanda Kochhar, MD & CEO.

¹²¹ There are examples of BRAC Bank in Bangladesh setting up a subsidiary Bkash, which is focussed on remittance services through mobile banking.

¹²² Submission from Oxigen to the Committee.