Chapter 4.1
Introduction and Strategic Direction

The vision statement for credit is:

By January 1, 2016, each low-income household and small-business would have “convenient” access to formally regulated lenders that have the ability to assess and meet their credit needs, and offer them a full-range of “suitable” credit products, at an “affordable” price.

If this vision is to be achieved, it will require a significant expansion in the quantum and range of credit products to be provided by the banking sector. In this context, there are some significant overall trends impacting the banking sector that are worth noting.

Risk Based Supervision Regimes as a Driver of Specialisation

These include the new Basel III guidelines and the RBI’s new Risk Based Supervision (RBS) framework which will be driven by periodic stress tests and an internal Supervisory Program for Assessment of Risk and Capital (SPARC). In addition, there will be heightened regulatory and supervisory scrutiny of Systemically Important Financial Institutions (SIFIs)¹²³. A number of market observers believe that these new developments, particularly the imposition of Basel III, will result in more banks seeking to move away from “purely holding deposits and making its own loans”¹²⁴. Banks will need to manage their balance sheets actively to ensure that they are using their capital resources efficiently in order to preserve their Return on Equity. Even prior to the implementation of the Basel III Accord, the Banking System, at under 13 per cent per annum showed a low and declining Return on Equity¹²⁵, and Stress Tests carried out by the Reserve Bank of India show that, “…while the present level of [NPA] provisions is adequate, a gap may arise under severe stress scenario”¹²⁶. Increasingly, each bank will need to become inherently stronger, focus more sharply on their core capabilities, and have the flexibility and the regulatory mandate to collaborate actively with other market participants who have complementary capabilities instead of being forced to follow identical strategies as every other participant. Policy and regulatory action will need to facilitate this process in order to ensure that the vision is achieved while simultaneously enhancing the stability of the financial system as a whole. All of these will have a bearing on the strategies that Banks pursue for credit delivery.

Non-Performing Assets in the Banking System and Financial Inclusion

In India PSL guidelines require banks to allocate 40 per cent of their lending book to it, making it the most important part of the bank‘s lending book and that has a significant impact on the performance of the bank as a whole. However, whenever financial inclusion goals are generally specified and strategies articulated, there is little acknowledgement of risk and cost-to-serve considerations. Banks therefore approach this sector with a complete lack of enthusiasm. Where they have a choice, they choose to under-serve this segment, and where they do not, they end up bearing enormous losses. As one of the direct consequences of this, government owned banks have ended up carrying a disproportionate share of the burden of meeting the PSL obligations, particularly as it concerns rural branching and serving the agricultural sector. This has had a severe impact on their performance overall and, as will be discussed in the next chapter, despite the loan waivers that took place a few years ago, more than half the total NPAs on their books are attributable to this sector with an NPA ratio that is close to double that of the rest of the asset book. While this would be a matter of serious concern for any bank, it is even more so for these banks because, given their ownership, as group they represent a direct contingent liability of the Government of India for their entire balance sheet and there is a grave risk of a large systemic shock if there is a large build-up of non-performing assets on their books.

A larger expectation has gradually been created in the country that this entire business has to be seen as a part of a series of hand-outs to low-income families, farmers, and small-businesses and a viable business proposition is simply impossible to construct and that rates of interest for providing an unsecured loan of Rs.25,000 to small farmers have to necessarily be lower than the rates charged to secured home-loan borrowers of Rs.10 lakh. This is seriously hurting both the banking system as well as the client on whose behalf it is being decided that a 10% rate of interest is a more important dimension than access itself.

Even at the regulator level this approach has unfortunately led to a number of gaps in both perception and implementation. For example, while for the much smaller NBFC-MFIs, linkages with credit bureaus have been made mandatory, for the rest of the banking system, particularly for the rural lending portfolio of banks, no such requirements have been imposed; despite the skew in NPA ratios the standard asset provisioning norms for PSL loans have been pegged at 0.25 per cent versus 0.40 per cent for the rest of the loan book; and most damagingly, interest rates for small loans to farmers have been permitted to be below the base rate when rates to large corporates with a far lower cost-to-serve, are restricted to be at a minimum of base rates. On the other hand, save for some work on the weather insurance side, risk management issues for these sectors have received no attention at all. When Regional Banks have failed, poor governance is usually held responsible and not also the fact that these banks were taking on concentrated risks that they could not manage in times of crisis. Development of important risk management tools such as commodity options, catastrophic portfolio insurance, and active warehouse receipt trading is moving at a very slow pace. Unless these issues are integrated in discussions on financial inclusion, banks will always be reluctant participants.

Given this important background, this Report overall and this section in particular, makes a conscious effort to redress this balance and issues of risks and costs have been kept at the very centre in the discussions of each of the strategies for providing better access to financial services to small businesses and low-income households.

Potential for Differentiated Banks in Deepening Credit Delivery

The RBI Discussion Paper on Banking Structure in India discusses the need for differentiated banks as a way to increase the size and strength of the banking sector relative to the needs of the real economy. In terms of new banks, the minimum capital requirements to license a National Bank which can perform both retail deposit taking and lending would need to be kept very high, with the result that only very few new entrants would qualify. However, for differentiated banks such as Wholesale Consumer Banks and Wholesale Investment Banks who are permitted to accept only wholesale deposits, these requirements could potentially be significantly lowered, thus permitting many more new entrants.

While National Banks is the dominant design and will remain one, it is clear that there is a need to proceed with extreme caution in creating new ones. Wholesale Consumer and Wholesale Investment Banks can significantly add strengths to the task of financial deepening without the risks of creating several new retail deposit-taking National Banks. Wholesale Consumer Banks are well placed to build deep specialisations in categories such as small business loans, student loans, and commercial vehicle finance, while Wholesale Investment Banks can effectively channel market liquidity as well as make it possible for an active market to emerge in the trading of such loans.

Role of Non-Banking Financial Companies and their Potential Transition to Banks

In addition to Banks – existing and new, there is also a continuing role for Non-Banking Financial Companies (NBFCs). The bulk of the NBFC sector in India remains very small, does not have the ability to garner public deposits, and in aggregate has performed at a very high level of quality. The sector as a whole therefore does not constitute a source of systemic instability. It has instead been playing the role of extending the reach of the banking system to the more difficult parts of the economy. While NBFCs in their current format play a useful role in credit delivery and must continue to do so with active regulatory support, they will have limits to how large they can become. This will be the case on account of differentiated capital adequacy requirements, absence of access to payment systems, constraints imposed by dual regulation, and other restrictions on the nature of business they are allowed to undertake and the nature of risks they are allowed to assume. From a systems design perspective this would be required because as players outside the banking system they would not enjoy the same degree of protection as banks would, and nor would the three principles of Stability, Transparency, and Neutrality apply to them in the same degree. The differentiated banking design offers multiple end points towards which NBFCs could head, particularly if issues such as SLR requirements, uniform applicability of CRR requirements and the current definition of PSL which is rooted in historical perspectives, are satisfactorily dealt with.

The table below provides an illustration of how various norms may be harmonised in a differentiated bank regime that also provides room for NBFCs.

The rest of the section discusses issues in credit delivery relevant to various banking designs. The two banking designs that are relevant for the provision of credit in the Indian context are given below. Each of these design choices and their variants have been described in some detail in Chapter 2.4:

1. National Banks operating as: Branch-based lenders; or working through Agents; Wholesale Banks; Consumer Banks using credit scores and analytics.

2. Regional Banks operating through their branches.

Although not a banking design, issues around NBFCs are also examined in detail given their relevance to credit delivery.

Within each of these designs, the following chapters examine their performance and potential generally and specifically in terms of: a) small loans to low-income households for various life-cycle needs; and b) lending to small businesses.

Chapter 4.2
National Bank with Branches

This is a design that covers large scheduled commercial banks that operate on a nation-wide or supra-regional basis using traditional branch banking.

This has been the dominant model of Indian banking for several decades. In aggregate, there are 105,753 branches across all scheduled commercial banks in India¹²⁷. Of these, about 39,336 branches are in rural India. Nearly 10,000 rural bank branches have been added in the last three years (2010-2013) alone – a fourth of the total number of rural branches. While this branch network is inadequate relative to the vision of one credit access point for every 10,000 population, even this level of penetration has been achieved as a result of single-minded regulatory and policy focus. For instance, Banks are required to maintain 25 per cent of their branch network in rural India.

In the context of rural credit to low-income households, this bank branch network has been focussed on advancing loans to farmers through the Kisan Credit Card (KCC) and the General Credit Card (GCC) products and advances to non-farming households through the Self-Help Group (SHG) product. KCCs advanced through bank branches account for the largest share of credit activity under financial inclusion with a portfolio outstanding of over Rs. 2.62 lakh crore as on March 31, 2013.

Despite this sustained effort to improve credit delivery through bank branches, rural outreach indicators remain poor in absolute terms with significant regional and segmental inequities. In 2002, non-institutional agencies accounted for 42.9 per cent of rural credit¹²⁸. A study showed that an increasingly large share of agricultural credit is going to farm sizes of more than 5 acres¹²⁹. As Chapter 2.2 points out, there is also significant regional skew in financial depth. During 2007-2012, 37.55 per cent of agricultural credit was accounted for by the Southern states despite them constituting less than 20 per cent of India’s Gross Cropped Area while the Eastern and North-Eastern states accounted for only 7.71 per cent, despite having comparable Gross Cropped Area¹³⁰.

Under the SHG-Bank linkage program, as on March 2012, 43.5 lakh SHGs were provided with bank loans through the commercial banks, cooperative banks and RRBs, of which 54 per cent belonged to the Southern region. With an average amount of Rs.144,086 disbursed as loan per group, the total outstanding amount to SHGs stood at Rs.36,341 crore as on March 2012. NABARD refinance of Rs.3,073 crore was provided to banks covering their lending to SHGs in the same year, amounting to 19.9 per cent of NABARD’s long-term refinance disbursements¹³¹. The NPAs of banks against loans to SHGs stood at 6.38 per cent as on March 2012¹³².

Similarly for SMEs, the Economic Census of 2005-06 revealed that 90 per cent of enterprises are self-financed. Of the balance 10 per cent, Banks account for 3.25 per cent of funding. An estimate done for Bangalore Urban District which has an overall credit to GDP ratio greater than 100 per cent and ranked among the top ten financial inclusion districts by CRISIL Inclusix, revealed a presence of five lakh SMEs with a projected credit need of Rs. 17,000 crore. Against this, the total lending across all segments by all commercial banks is approximately Rs. 12,000 crore¹³³.

As far as risk and quality of this credit is concerned, Public Sector Banks are the largest participants in priority sector lending, accounting for more than 75 per cent of the over Rs.15 lakh crore of the total. The Gross Priority Sector NPA ratios for these Banks are about 50 per cent higher than their overall Gross NPA Ratios, with small loans (below Rs.25,000) having a ratio that is 4.5 times higher¹³⁴. As seen in Table 4.2.4¹³⁵ below, for private sector banks and public sector banks, non-performing loans are far higher in the priority sector than in the non-priority sector.

On the cost front the picture is equally grim for both Private Sector and Public Sector Banks with the cost of operations (not including interest cost) alone approximating 30 per cent to make a one year loan of Rs.10,000¹³⁷. For a Private Bank the cost of an ultra-small brick and mortar branch (USB) is estimated to require an investment of Rs.8-10 lakh and an annual recurring cost of Rs.20 lakh¹³⁸ - this high a level of expenditure may not be consistent with the level of business activity that this USB will be able to sustain.

While there may be notable exceptions from the above analysis it is very clear that the current branch-based approach is not an effective credit delivery channel. As a rule very high-cost and very high-risk strategies are being pursued by the National Banks in the context of financial deepening. If National Banks wish to act as originators of credit risk using their branches, they will have to develop far more effective, lower cost, and far lower-risk models of outreach.

How can the efficacy of branch-based lending for National Banks be increased? While the empirical evidence using data even from developed nations suggests that large banks are disadvantaged in lending to low-income households and small-businesses¹³⁹, a number of banks within India, as well as globally, have found ways of doing this by placing dedicated staff inside their branches with the full power to make local decisions and by effectively treating branches focussed on this business as stand-alone profit centres and carefully measuring their performance using Risk Adjusted Return On Capital (RAROC) type integrated performance measures. Banks in India have centralised loan processing “factories” and rely on the separation of sales, credit, and collections to ensure focus and high-quality in their consumer finance operations. However, the lessons from successful examples of branch-based lending appear to suggest the opposite direction. Banks would have to de-centralise loan decision making and integrate sales, credit, and collections. Some of the notable examples are given below and all carry the message that if this design has to work, decision making needs to be moved very close to the customer so that “soft information” can be acted upon.

1. Bank Rakyat Indonesia is the oldest bank in Indonesia and has created a separate structure for its micro-banking business which is addressing its objectives of financial inclusion and penetrating into the mass market segment very well. This is done by creating a structure of small dedicated units (referred to as a BRI Unit) under a Branch office which caters to its mass micro business activities. The BRI unit is a small micro Branch structure established in interior places with a small compact staff team comprising 3 staff members (1 Account officer, 1 teller for Cash transactions, and 1 customer service executive for non-cash transactions). Under each BRI unit, there are separate “Teras BRI” units in the style of extension counters to expand the reach of the BRI units where warranted and “Teras mobile” units which travel to the villages for carrying out banking activities. This dedicated micro banking vertical is 100 per cent self-funded i.e., the savings/deposits from micro banking clients fund the loans to the micro banking clients. All loans below USD 10,000 are handled by micro banking branches and thus there is no overlap in customer base between different verticals of the bank. Each branch is a profit centre with a defined break-even period and incentive structure is linked to the performance of the branch¹⁴⁰. High interest rates on loans has been noted as a concern with the BRI model.

2. Lloyds Banking Group is one of the large banks in the UK which has placed dedicated relationship managers in 500 of its locations. They specialise in certain sectors such as agriculture, franchising, property, healthcare, legal, and manufacturing. They can decide on loans of up to GBP 500,000 and cover 90 per cent of lending applications at Lloyds¹⁴¹.

3. A Large US Bank: This bank (which is amongst the top five commercial banks in the US) operates more than 1500 branches and uses standardised credit scoring methodology to assess its small business loans. However, “… credit decisions ultimately reside with branches because local managers can alter credit scores on the basis of a standard set of subjective criteria [based on non-verifiable “soft” information that the branch manager has] that the final score reflects [through this process “hardening” their “soft” information]. Similarly, they can adapt loan terms including pricing to the specific circumstances of the application. However, branch managers’ career prospects and remuneration depend on the overall success of their credit decisions, and local overrides are closely monitored by the bank’s risk management”. Using this approach the bank has been able to successfully build a client base that is low risk but pays a higher margin on account of the banks’ proximity to them¹⁴².

4. HDFC Bank: They have over 1,000 branches in rural and semi-urban areas and a dedicated field force of 3,500 to serve these markets using products such as Self-Help Group loans and Joint Liability Group loans. As of this date, they have disbursed Rs. 1,600 crore during the current financial year, at lending rates ranging from 21 – 26 per cent and have reached out to over 23 lakh unique rural borrowers spread across 24 states. This dedicated field force is another example of a “bank-in-a-bank” model which, while using a shared branch infrastructure, is managed as an entirely separate unit within the bank. In terms of products, the loan product is combined with a recurring deposit product and a savings account. There has been a reasonable build-up of balances in these recurring deposit accounts. HDFC Bank expects this dedicated unit to form the foundation of its financial inclusion efforts and over a five year period, build a commercially sustainable business. The key success drivers of this business appear to be the ability to hire local staff, maintain a low-cost structure, and control operational and credit risks in a tight manner.

It is clear that there are ways in which large branch based models can be made more effective in the context of credit delivery to low-income households and small-businesses. However, this goes well beyond the mere physical presence of a branch. Executing this strategy is going to be challenging given all the human resources and other challenges that are likely to be faced by banks in the process. Several aspects of the functioning of the Bank will have to change.

Banks will also have to develop strong risk management capability and the power to use various tools to manage their exposures to these segments. These include:

1. Systematic Risk: Irrespective of how origination exactly happens, it is entirely possible that National Banks may end up building portfolio concentrations (to a particular sector or region) that are unhealthy and attract a higher level of penalty in their Stress Tests than they have comfort with. They will need to continually and actively rebalance their portfolio using sales and purchases of loans, bonds, securitised portfolios, and credit derivatives. They will need to move away from an exclusive originate-and-hold-to-maturity strategy and among other things, will need the ability to be able to hold bonds in their banking book without having to mark them to market and will gradually need to start to document all their loans using debenture / bond documentation so that the liquidity of their balance sheet improves.

2. Rainfall Risk: In rural and agricultural lending, rainfall is the most important source of exogenous risks. Banks have thus far tried to get their clients to hedge this risk using crop insurance that is bundled along with a farm loan. Unfortunately however, the entire rural book is exposed to this risk even in the non-farm portion of it and the most efficient way for banks to hedge this risk is to seek to protect the entire book against these shocks. Such a strategy also allows the bank to carry out large scale waivers in case of a region-wide or a nation-wide shock to rainfall without having to request for government funded bail-outs which have created a serious distortion in the farm level credit culture.

3. Commodity Price Risk: Commodity price risk is the other big risk that needs to be hedged using either insurance or commodity put options that the bank purchases on a wholesale basis in global options markets. These are not available in India. These could also be used to offer large scale loan-waivers in case of extremely negative price events.

4. Moral Hazard: Since there is no uniform reporting of credit data mandated for all National Bank branches, there are weak controls against moral hazard. Universal reporting to credit bureaus of all loans, both individual and SME, but in particular SHG loans, Kisan Credit Card, and General Credit Card is important to mitigate this risk.

Irrespective of how well National Banks execute their origination and their balance sheet management strategies, given their systemic importance and the need to ensure that any signs of incipient weakness are quickly identified and dealt with, it is very important to ensure that there is complete transparency and accuracy in the financial statements of these financial institutions, including weaknesses stemming from priority sector lending. Such transparency also ensures that the competition between different types of institutions is on a level playing field and that the apparently stronger player is indeed inherently stronger and not simply being “protected” by its lower levels of disclosure. In order to ensure complete transparency and accuracy in the financial statements of these institutions, several steps would need to be taken:

1. While NPA recognition norms in India are comparable to those currently prevailing in most countries, Indian NPA coverage ratios are far lower. Comparison with developing countries such as China, Brazil, and Mexico, who are likely to have comparable levels of risk and Loss Given Default statistics, reveals that India has both lower provisioning and higher observed NPAs¹⁴³. However, this approach to NPA recognition assumes that all banks follow similar strategies and have the same asset mix. In view of the fact that banks may choose to focus their strategies on different customer segments and asset classes, it is recommended that the regulator provide specific guidance on provisioning norms at the level of each asset class.

For instance, given the fact that historical NPAs on Micro Finance Joint Liability Group (JLG) loans are in the region of 0.7 per cent¹⁴⁴ and assuming loss given defaults to be 100 per cent, the risk based estimate of the NPA Coverage Ratio for this asset class would be 100 per cent of historical NPAs or 0.7 per cent. For KCC, currently banks formulate their own internal policies in terms of fixing the repayment period and determining the scale of finance. RBI may require banks to formulate broad guidelines so that there is some degree of standardisation in the way the KCC product is administered and it becomes possible to properly ascertain if the asset is performing as expected or not. The guidelines should prescribe various factors which banks should consider before fixing repayment period and scale of finance. The repayment period should be linked to the income generation pattern of farmers which in turn depends on numerous factors including cropping pattern, cropping season (Kharif and Rabi), harvesting season (short duration or long duration), acreage, yield, and economies of scale. For instance, a 9-12 month repayment period may be fixed for short duration crops (rice, wheat or pulses) whereas a longer repayment period of one year may be fixed for long-duration crops like sugarcane or banana. Tracking of actual repayments against these periods on a 90 past-due basis can then be used to carry out asset classification. And, over the over the last decade, since the gross NPAs in the agriculture segment have been higher than banks loans to the non-agriculture segment on an average by 1.21 times, the standard asset provisioning, for all agriculture exposures including KCC should be set at 0.5 per cent (~1.21 times of 0.4 per cent)¹⁴⁵.

Similarly, based on the data available for different asset classes, the regulator should recommend NPA Coverage Ratios for each of them. A bank’s overall NPA Coverage Ratio would be a function of its portfolio asset mix.

2. Even on standard assets, provisioning levels would need to reflect the underlying level of riskiness of each asset class (combination of customer segment, product design, and collateral). For example, for a bank pursuing a broader priority sector strategy, a long-term analysis of priority sector asset quality indicates that there needs to be an increase in standard assets provisioning from the current level of 0.25 per cent to the level of 0.40 per cent.

3. Different customer-asset combinations behave very differently from each other and mechanical rules (such as a 90 day past-due rule) applied at the level of each loan may not accurately capture the riskiness of each sub-group. It is recommended that the regulator mandate NPA recognition rules at the level of each asset-class and require that all banks conform to these mandates. For instance, a 52-week, weekly repayment JLG loan to a low-income household and a 15-year, monthly repayment mortgage to a high-income household are fundamentally different type of customer-asset sub-groups that require differential NPA recognition treatment. In view of these differences, the regulator might specify that the NPA recognition norm for the weekly JLG loan be 30 days, while that for the mortgage be 120 days.

4. Require all financial institutions to publicly disclose the results of their stress tests both at an overall balance sheet level as well as at a segmental level on an annual basis.

For a variety of reasons governments may decide to offer interest subventions to farmers and on occasion loan waivers as well. It is strongly recommended that such subventions and waivers be offered directly to beneficiaries by the governments using DBT into their bank accounts and allow banks and financial institutions to act strictly according to market principles so as not to introduce strong moral hazard and make it difficult for financial institutions to enforce contracts. This is the standard manner in which such subsidies are offered in the EU ¹⁴⁶ and the US ¹⁴⁷ to ensure that there are minimal distortions in the market prices and production decisions of farmers while providing them some minimum income protection.

Banks are also required to price farm loans below Rs. 3,00,000 at 7 per cent for which they receive an interest subvention of 2 per cent in the consequent period. This pricing does not appear consistent with the risk parameters discussed previously. From the perspective of Stability that entails sustainable pricing, Banks must be allowed to freely price these loans based on their risk models. The current permissions to price below their base rate must be withdrawn.

Recommendations:

4.1 In order to encourage banks to actively manage their exposures to various sectors, including priority sectors, a number of steps would have to be taken:

1. Banks must be required to disclose their concentration levels to each segment in their financial statements.

2. Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation. [Identical to Recommendations 4.10 and 4.30]

3. RBI must represent to the MoF to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment pointing out the role it would play in ensuring efficient risk transmission. [Identical to Recommendations 4.11 and 4.38]

4. Banks must be permitted to purchase portfolio level protection against all forms of rainfall and commodity price risks, including through the use of financial futures and options bought either within India or globally.

4.2 Universal reporting to credit bureaus should be mandated for all loans, both individual and SME, but in particular SHG loans, Kisan Credit Card, and General Credit Card. [Identical to Recommendation 4.43]

4.3 In view of the fact that banks may choose to focus their priority sector strategies on different customer segments and asset classes, it is recommended that the regulator provide specific guidance on differential provisioning norms at the level of each asset class. A bank’s overall NPA Coverage Ratio would therefore be a function of its overall portfolio asset mix. On standard assets, provisioning levels as well as asset classification guidelines specified by RBI would need to reflect the underlying level of riskiness of each asset class (combination of customer segment, product design, and collateral) and not be uniform across all the asset classes. Additionally, different customer-asset combinations behave very differently from each other and it is recommended that the regulator mandate NPA recognition rules at the level of each asset-class and require that all banks conform to these mandates. [Identical to Recommendation 4.21 for NBFCs]

4.4 All banks should be required to publicly disclose the results of their stress tests both at an overall balance sheet level as well as at a segmental level at least annually. [Identical to Recommendation 4.22 for NBFCs]

4.5 From the perspective of Stability that entails sustainable pricing, banks must be required to freely price farm loans based on their risk models and any subventions and waivers deemed necessary by the government should be transferred directly to the farmers and not through interest subsidies or loan waivers. The permission to price farm loans below the base rate should be withdrawn. [Also see Recommendation 4.34]

Chapter 4.3
National Bank with Agents

This design refers to Banks that have a limited branch network but rely on sourcing credit through agents. It is one of the ways of reducing costs, improving outreach, improving Suitability, and simultaneously reducing risks associated with lending.

Agents with no Capital Commitment

Indian banks have thus far used two types of agents to carry out their lending activities – the Direct Sales Agent (DSA) and the Business Correspondent / Business Facilitator (BC/BF). The BC/BF can either be a corporate or an individual. A corporate BC/BF can either be a dedicated entity or an existing real-sector intermediary such a sugarcane company that has farmer relationships. The BC/BF currently accounts for a small share of priority sector lending, approximately Rs. 2,500 crore of KCC and GCC loans are outstanding through these channels. Not much is known about the quality of these loans. It is clear from global experiences in consumer credit¹⁴⁸ that credit-score based lending with no capital commitment by the Agent (in this case the BC/BF) is prone to significant adverse selection and moral hazard. The urban DSA model for short-term personal loans and credit cards attracted significant negative attention on account of customer protection concerns and had to be discontinued by all the National Banks.

Agents with Capital Commitment

In the rural context, ICICI Bank in 2005-06 successfully experimented with the concept of a Credit Franchisee who was similar to a Business Correspondent / Business Facilitator but was required to place a fixed deposit with the Bank of 10 per cent of the amounts of loans that he would sanction. Upon default, the deposit would function as a First Loss Deficiency Guarantee¹⁴⁹.

Globally, many banks have used agents to effectively increase penetration and reach. The factor crucial to the success of this model is the careful selection of agents. For example, shop keepers who are successful in their core business, enjoy the trust of the community, have a large cash inflow in their business and are already familiar with providing credit to the customers for their purchases, were chosen to be agents by an Indonesian bank. These selection criteria ensured that the shop keeper had the requisite skill and understanding to connect with the customers and also see the banking opportunity to do effective cash management for him. His cash inflows are used for customer withdrawals / loan disbursements / remittances and he gets paid for that, as also for getting additional footfall. In the Philippines, pawn shops have been used as agents by banks as they already have a natural interest and connect with the customers¹⁵⁰.

Dedicated Subsidiaries as Agents

For National Banks, fully owned subsidiaries can also potentially act as dedicated agents of a National Bank subject to the guidelines on investments in subsidiaries¹⁵¹. They can be a cost effective alternative which can address some of the challenges around having wage flexibility, different compensation structures, local hiring practices, contractual arrangements / outsourcing, etc. Also, a subsidiary route enables a focussed approach to economise on the cost of lending operations. This route will also be able to connect with the local community better and thus enable better quality of assets. For example, in the case of Bank Danamon in Indonesia, the vehicle finance and equipment finance business is conducted through Adira Finance, a fully owned subsidiary that sells the loans that it originates to Bank Danamon through an arm’s length arrangement for funding purposes. In India, HDFC Limited has promoted a company focussed on education loans called Credila. The motivation of these banks to set-up a subsidiary has been to get the right focus and appropriate cost structure for the specialised business and also attract other experts as investor partners in the company.

In the Indian context, however, while considering the concept of a subsidiary route, the case of Regional Rural Banks (RRBs) needs to be kept in view, which were also started by Public Sector Banks on this premise but subsequently their salary structure had to be made at par with public sector banks. So the very purpose of setting up the RRBs as low cost structures could not be served. Permitting strong National Banks to float specialised subsidiaries with an ownership structure that can address this concern may allow them to benefit from the inherent strengths of these institutions.

Real Sector Intermediaries as Agents

Real sector intermediaries such as sugarcane companies are used by Banks, particularly Private Sector Banks with limited branch networks, as agents. While this may work as an interim solution for outreach, given that credit sourcing and servicing are peripheral activities for these agents, it is unlikely that these agents will be able to provide the full range of credit products as defined in the credit vision statement. There also appear to be greater customer protection concerns here given the ability of the agent to tie credit in with its procurement activities and offer lower procurement prices to the farmer. Agents must create dedicated operations for lending so that there is separation of credit and non-credit activities.

Outsourcing Guidelines

It is clear that there are ways in which Agent based models can be made to work for providing comprehensive financial services to low-income households and small-businesses. However, while this strategy has the potential to address the costs of intermediation, it has significant risks vis-à-vis customer protection in the case of individual Agents and credit risk in the case of Agents who commit no capital to the lending business. As a consequence all of the safeguards specified earlier for National Banks with Branches would need to be adhered to here as well. Given the inherent moral hazard entailed in agent-based lending models, risk-sharing between the Bank and its agents is an important enabler to consider. Current RBI outsourcing guidelines do not permit outsourcing of credit sanctions of loans (including retail loans) by Banks¹⁵². This will need to be suitably amended so that Agents can sanction loans on behalf of Banks where risk is then managed through risk-sharing arrangements and statistical quality control processes.

Recommendations:

4.6 Banks are already permitted to set up specialised subsidiaries upon getting specific approvals from the RBI. However, no approvals have been granted; potentially due to concerns around circumvention of branch licensing guidelines. In light of the recent relaxation of branch licensing guidelines and the capability to carry out consolidated supervision, the requirement of prior approvals may be removed for the purpose of creating dedicated subsidiaries for financial inclusion.

4.7 The decision on the manner in which risk sharing and credit approval arrangements need to be structured between banks and their agents can be left to the judgment of banks. Outsourcing guidelines should be amended to permit this.

Chapter 4.4
National Consumer Bank

These banks lend directly to certain consumer segments using credit scores¹⁵³ and other analytical models, either through agents or product-market partnerships with entities such as department stores, vehicle manufacturers, mobile phone companies, and utility services providers.

At the heart of the branch and agent-based models is the assumption that proximity matters. While “soft information” and the associated need for proximity would continue to be important for many segments of customers, advances around “big data” that promise to deliver a far more accurate picture of the credit-worthiness of the individual and business than traditional branch-based under-writing or credit scores could completely transform the nature of credit decision making in the years to come. Even currently these new approaches seem particularly well-suited to specialised segments such as student loans and early stage financing for small businesses. Globally, while concerns have been expressed about the over-reliance on credit scores for credit-decision making, particularly in the context of the US subprime-mortgage crisis¹⁵⁴, longer-term trends, such as those documented for the first time by Petersen and Rajan (2002)¹⁵⁵, suggest that the distance between lenders and small business borrowers is growing and that they are communicating less and less with each other. They find that the principal factor accounting for this is the fact that technology of information gathering, aggregation, and reporting has improved dramatically and allows firms to address the traditional problems of adverse selection and moral hazard in a completely different manner.

Even in India such approaches have been found to be successful by the Credit Card industry – they permitted banks such as Citibank¹⁵⁶ which had very few branches to become one of the largest credit card issuers and ICICI Bank to reach a leadership position in mortgages. In the priority sector lending space, NBFCs have successfully used such approaches combined with some “soft” information to build large equipment and vehicle finance businesses. In the microfinance industry which deals with the poorest of clients, credit bureau records have become near universal. According to data from MFIN, in about 18 months' time, credit bureaus have been able to gather information about 10 crore loan accounts of about 2.5 crore individual customers from about 42 MFIs.¹⁵⁷

These are very welcome trends and better ability to store, analyse, and share data collected by financial institutions, mobile phone companies, and electricity utilities would allow high quality borrowers located even in remote rural locations to signal to low cost lenders such as a National Consumer Bank, their credit worthiness can go a long way towards ensuring that sufficient access to adequate levels of credit is made available to them at low prices. For several National Banks, this may even be a better strategy to grow their client and asset base rather than through a branch network as discussed previously. The value of real-sector companies like utilities and mobile phone companies may also be in the context of their information assets and not just their outreach.

In its chapter on Protection of Personal Information of Consumers¹⁵⁸, the draft Indian Financial Code recommends that the regulator may make regulations to:

a. Provide additional requirements for the collection, storage, modificationand protection of personal information by financial services providers, including:

1. The manner of maintenance of records of personal information and the time-periods for which the records are to be maintained; and

2. The manner in which records of personal information should be dealt with after the expiry of the specified period;

b. Exempt a class of financial services providers from the application of all orany portion of this Chapter or modify the manner in, or extent to which, all or any portion of the Chapter applies to them, subject to any specified conditions; or

c. Establish mechanisms to ensure that consumers have access to, and are given an effective opportunity to seek modifications to, their personal information.

Recommendations:

4.8 There is a need to develop a robust legal and regulatory framework around customer data generated in various transactions (credit and payments, digital and off-line), with the objective of customer ownership of their own transactions data and its use, among others, for signalling credit-worthiness. RBI should constitute a Working Group comprising TRAI, CERC, and Credit Information Companies to develop a framework for sharing of data between telecom companies, electrical utilities, and credit bureaus. This framework should be in keeping with the FSLRC’s draft Indian Financial Code which recommends the creation of regulations on the collection, storage, modification and protection of personal information by financial services providers; and establishment of mechanisms to ensure that consumers have access to, and are given an effective opportunity to seek modifications to, their personal information. [Identical to Recommendation 4.42]

Chapter 4.5
National Wholesale Bank

Some National Banks without the branch network or expertise of the underlying priority sector asset classes, have sought to fulfil their priority sector obligations by purchasing these assets from other banks and financial institutions, while continuing to remain wholesale in character. The strategy of international banks in India and some of the mid-sized private sector banks have been along these lines.

The RBI guidelines on securitisation have provided a strong framework for the development of this market. The performance of securitised priority sector assets has been quite robust. As an illustration, 89 per cent of securitised micro-loans rated by CRISIL were rated ‘A’ and above.

Given the enormous cost and informational disadvantages that National Banks face in India it is possible that this may be an entirely acceptable and even a preferred strategy for a large, systemically important bank to follow, relative to all the others that have been discussed thus far, so that it is able to maintain an extremely high level of safety in its credit ratings and can therefore act as a high quality aggregator of both deposits and loans allowing smaller and more specialised banks and financial institutions to transfer their own systematic exposures to such a Wholesale Bank.

In order to make several markets such as those relating to credit derivatives, corporate bonds, warehouse receipts, and take-out financing function well, the presence of such Wholesale Banks as aggregators and market makers is essential. The presence of such banks can have an important and beneficial impact on financial inclusion.

In order for a National Bank to successfully follow the strategy of a Wholesale Bank and act as an aggregator and market maker, significant regulatory anomalies need to be addressed:

1. Removing price caps on priority sector assets originated indirectly. Current PSL guidelines state that “investments by banks in securitised assets, representing loans to various categories of priority sector, except 'others' category, are eligible for classification under respective categories of priority sector (direct or indirect) depending on the underlying assets provided….the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the investing bank plus 8 per cent per annum”¹⁵⁹. Since there are no such price caps when the Bank originates a loan directly, this regulation appears to violate the principle of Neutrality.

2. Securitisation markets had been growing steadily in India since around 2005 owing to a strong and conducive regulatory environment. Securitised debt instruments were listed for the first time in 2013, thus improving standards of transparency and reporting and widening the potential investor base. However, post facto claims by income tax authorities in October 2011, stating that the gross income of such SPVs was liable to tax, have effectively hampered the growth of the market. The matter is presently sub-judice at the Bombay High Court. The Finance Bill, 2013, has sought to clarify the tax position by stating that a securitisation SPV is not liable to pay income tax. However, the Bill also states that trustees of such SPVs must pay tax on distributed income calling into doubt the pass-through status of these vehicles. Restoring the pass-through status of securitisation SPVs will help to develop this market and therefore, create pathways for Wholesale Banks to provide liquidity to other Banks and Financial Institutions directly originating assets in priority sectors¹⁶⁰.

3. Addressing the loan-bond arbitrage in terms of opacity of the former by allowing banks to classify (and reclassify) bonds into a held-to-maturity (HTM) or available-for-sale (AFS) bucket based on their declared intention rather than automatically based on legal documentation.

4. This design would be vastly enabled if the role of apex agencies such as NABARD, CGTMSE, SIDBI, and NHB are aligned towards the provision of risk-based credit enhancements, market-making and facilitation of listing of debt securities rather than provision of direct finance, automatic refinance or automatic credit guarantees. The professional capabilities of these apex organisations need to be built so that they can transition to these roles. This transition would considerably enhance the impact that these institutions have with their limited refinance resources and simultaneously serve to strengthen both originators as well as investors. This is also in line with the recommendations of the Rajan Committee (2009).

5. This strategy seems particularly well-suited to those Public Sector Banks that are unable to develop low-cost, low-risk, and effective direct origination strategies.

Recommendations:

4.9 The stipulation that the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the purchasing bank plus 8 per cent per annum should be removed. [Identical to Recommendation 4.37]

4.10 Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation. [Identical to Recommendations 4.1(b) and 4.30]

4.11 RBI should represent to the Government of India to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment so as to create pathways for Wholesale Banks to provide liquidity to other Banks and Financial Institutions directly originating assets in priority sectors. [Identical to Recommendations 4.1(c) and 4.38]

4.12 Reorient the focus of NABARD, CGTMSE, SIDBI, and NHB to be market makers and providers of risk-based credit enhancements rather than providers of direct finance, automatic refinance, or automatic credit guarantees for National Banks.

Chapter 4.6
Regional Bank

This is a design in which there are several Regional Banks, each relatively small in size, that are full-service Banks (offering credit, deposits, and payments services). This design is also sometimes referred to as “Small Banks” or “Community Banks”.

These are supervised directly by the national regulator (and depending upon their specific structure and licence also by a State Regulator) and governed on a day-to-day basis by their local boards. While these banks may borrow some amounts in wholesale markets their principal source of funds is their local deposit base. A Regional Bank does not use Capital Markets extensively for its resource raising nor does it use Agents in any form to reach its customers – its only means of serving customers is typically through its branches.

History and Global Perspective

A country where this model of banking continues to be very important, though less so each year, is the United States. It has 8,100 commercial banks, 1,200 Savings and Loans Associations, and 12,000 credit unions (cooperative banks). Its top ten banks have only 60 per cent of total assets compared to Canada or UK where 4 or 5 banks dominate the industry and have close to 90 per cent of the assets¹⁶¹. However, despite its resilience, the model is in retreat in the USA because National Banks and Non-Bank Financial Institutions have been able to bring both a wider range of products as well as lower cost of funds into the home markets of the Regional Banks. And, as Petersen and Rajan (2002) suggest, technology of information gathering, aggregation, and reporting has improved dramatically and allows firms to address the traditional problems of adverse selection and moral hazard in a completely different manner even when the distances between the bank-branch and the firm are larger. The imposition of Basel III requirements on all these banks is likely to compound the problems of survival faced by the local banks. Amongst the few countries where this model of banking continues to thrive are Germany and Switzerland and important lessons can be learnt from their experiences.

Developing countries that mandated community banks with low minimum capital requirements produced hundreds of institutions that have had solvency problems while not solving the inclusion problem (Nigeria, Tanzania) or have required intensive, subsidised upgrading programs (Philippines, Ghana). The jury is still out on Mexico, but risks are apparent. Supervisory capacity has been overwhelmed by many small institutions, leaving weaknesses uncorrected.

This is not a new idea even in India. As of March 2012, there were 1618 Urban Cooperative Banks in India, 82 Regional Rural Banks, 31 State Cooperative Banks (StCBs), and 370 District Central Cooperative Banks (DCCB) – all of which broadly meet the definition of Regional Banks. In addition there are 92,432 Primary Agricultural Cooperative Societies (PACS) which are single branch operations, but full-service. Within the rural cooperative institutions, credit disbursement for agriculture is being done by PACS, while deposit mobilisation is done by the PACS, DCCBs and StCBs. These organisations have enjoyed a great deal of support from the financial system and from the government and have achieved considerable outreach. However, with very few notable exceptions, they have performed very poorly in terms of financial performance and will need a great deal of on-going support to reach their full potential¹⁶².

In the private sector as well, while once again there are indeed notable exceptions such as the highly profitable 110 year old City Union Bank which is focussed on the Thanjavur-Kumbakonam region of Tamil Nadu or the Jammu and Kashmir Bank, over the years many Regional Banks such as the Bank of Rajasthan, Bank of Madura, United Western Bank, and Ratnakar Bank have either failed and have had to be merged with stronger banks and substantially recapitalised, or have sought to change their character to that of a National Bank.

Benefits and Challenges

Regional Banks are likely to be lower cost and closer to the customer and therefore, better equipped to originate vis-à-vis National Banks, but 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. A design that seeks to significantly grow this category of institutions will place significantly higher regulatory and supervision demands. And, even if the local regulatory and governance issues are taken care of, while there are clear advantages to having a regional /local banking structure, there are also long-term structural issues such as concentrated risks and the gradual “hardening of soft information” that has been discussed earlier which will erode the very raison d’être of these institutions unless they are dealt with effectively.

There is clearly one benefit from a development and financial inclusion perspective that Regional Banks have that neither the large National Banks nor the non-bank financial institutions demonstrate – their ability to collect local deposits ends up permanently anchoring them to the local community in a way that the other institutions are not able to match. This anchoring gives them both the desire and the requirement to stay connected to the local community during both good times and bad ones. A National Bank could allow a local branch to simply stay dormant because it is experiencing some difficulties and an NBFI can simply pull-out but neither option is available to the Regional Bank and it is forced to adapt itself to the local environment.

Improving Governance

The Regional Rural Bank which was conceived of as a subsidiary of larger National Banks was potentially a stronger design because it successfully dealt with the challenge of “capture” by local political interests owing to its relationship with its parent, but eventually did not perform as expected because gradually the culture and the cost structure of the parent National Bank permeated into the Regional Bank as well and overwhelmed the attempts at building a truly regionally focussed institution. Permitting strong National Banks to float such subsidiaries with a different ownership structure may allow them to overcome some of these difficulties and benefit from the inherent strengths of these institutions.

A study on best practices followed by community banks in the USA that were able to maintain the highest rating from their supervisors from 2006 to 2011 highlighted the following¹⁶⁴:

1. Commitment to conservative lending practices with emphasis on detailed underwriting and credit policies even if this meant losing business to some competitors who were increasing LTV ratios to gain more business

2. Presence of experienced senior management, coupled with a supportive, engaged board of directors

3. Balance between growth objectives and risk level

4. Rigourous follow-up of delinquent loans – in some cases as early as 5 days past due

5. Emphasis on relationship banking based on detailed knowledge of their markets and customers and avoiding market or products that they did not understand.

Board and management oversight is the fundamental element of ensuring a safe and sound bank and director oversight is the primary driver that keeps a bank moving in a positive direction, and is a critical component of a bank's success¹⁶⁵. Therefore there is a strong need to strengthen the board governance at the Regional Banks by ensuring that there are high standards for board composition and operations as well as by proactively grooming competent board members. Board governance for Regional Banks can be strengthened by taking the following steps¹⁶⁶:

1. Mandating high standards for Board Composition and Operations by specifying clearer “fit and proper” standards for board members that ensure some relevant experience. (e.g. mix of skills and stakeholders; independent directors; higher qualifications for board chairs; board chair and CEO not the same person; no family members of CEO on board). A few operational items could be mandated (existence of audit committee and risk committee; reporting of internal auditor directly to the board; conflict of interest standards; rotation of members). The German approach of a dual board structure: a supervisory board and an executive board that runs the bank could also be explored.

2. Grooming competent board members. Governance training could be required of all board members, or at least Chairs. Retired senior bankers may be appointed to Board roles in these institutions and institutions such as College of Agricultural Banking (CAB), Bankers Institute of Rural Development (BIRD), and Centre for Advancement in Financial Research and Learning (CAFRAL) could be requested to develop such training programmes along the lines of The Bank Director’s Desktop developed by US Fed, the Global Corporate Governance Forum developed by the IFC, and the Bank Negara Malaysia sponsored Institute, the ICLIF, which offers a Financial Institution Directors Education course to board directors across Asia.

Structural Issues

While improving the regulation and supervision of these institutions will address some of the problems that they face there will be an active need to actually strengthen the capabilities of these institutions so that they are able to overcome some of the structural challenges that they face. Some of the steps, such as the adoption of Core Banking Systems by these banks is on-going¹⁶⁷ and will increase the capacity to do off-site supervision but there are several more that will need to be taken, particularly with regard to their capabilities to manage the variety of risks that they face:

1. Systematic Risk: Given their regional focus it is inevitable that in the process of originating both larger and smaller loans they will end up building portfolio concentrations that are unhealthy and attract a higher level of capital-penalty in their Stress Tests than they have comfort with. They will need to continually and actively rebalance their portfolio using sales and purchases of loans, bonds, securitised portfolios, and credit derivatives. They will need to move away from an exclusive originate-and-hold-to-maturity strategy and among other things, will need the ability to be able to hold bonds in their banking book without having to mark them to market and will gradually need to start to document all their loans using debenture / bond documentation so that the liquidity of their balance sheet improves.

2. Rainfall Risk: In rural and agricultural lending, rainfall is the most important source of exogenous risks which can have a region-wide impact. Banks have thus far tried to get their clients to hedge this risk using crop insurance that is bundled along with a farm loan. Unfortunately however, the entire rural book is exposed to this risk, even in the non-farm portion of it, and the most efficient way for banks to hedge this risk is to seek to protect the entire book against these shocks. Such a strategy also allows the bank to carry out large scale waivers in case of a region-wide or a nation-wide shock to rainfall without having to request for government funded bail-outs which have created such a massive distortion in the farm level credit culture.

3. Commodity Price: Commodity price risk is the other big risk that needs to be hedged using either insurance or commodity put options that the bank purchases on a wholesale basis in global options markets. These are not options that are permitted to be traded locally and in any case, from a national point of view there is value in laying off the risks overseas. These could also be used to offer large scale loan-waivers in case of extremely negative price events.

There are differences on this count even in the global experience. While the Swiss and the German Regional Banks have explicitly tried to address this issue of risk management, the US design has not done so and this perhaps is the main reason why the model is thriving in these two countries while it is shrinking in the US. For example, the German Sparkassen Banks are part of a national system that spreads risk across a system of Regional Banks and national institutions which include the “Joint Liability Scheme” of national and regional guarantee schemes coordinated by the Deutscher Sparkassen und Giroverband (DSGV: the German Savings Bank Association). Germany also has credit guarantee banks that lend within each federal region or Land organised through the Verband Deutscher Bürgschaftsbanken (VDB: the Association of German Guarantee Banks). They are non-profit associations of lenders that historically provided sureties worth 80 per cent of the loan value. Each guarantee bank would take on up to 35 per cent of the risk, while the federal government took 40 per cent and the Land 25 per cent. The borrower pays a fee of 1-1.5 per cent of the loan plus an annual commission of 1-1.5 per cent on the amount outstanding each year. Historically borrowers were at risk for 20 per cent of the loan value, but as a result of the recent financial crash the German Government has encouraged guarantee banks to cover 90 per cent of the risk and to take up to 50 per cent themselves¹⁶⁸. Similarly, the Swiss cantonal banks benefit from being part of the Association of Cantonal Banks. The Association facilitates cooperation between the banks, this allows the banks to benefit from economies of scale in providing products such as pensions, investment advice and asset management. These services are provided by 20 network providers overseen by the ASCB, in an effort to ‘produce centrally, provide locally’. This local provision of services allows cantonal banks to create support schemes or products specifically tailored for firms in their local market.

In India, SIDBI has been developing a pilot programme of lending to SMEs using a specialised lending tool that they have developed in which they have partnered with Urban Cooperative Banks and Regional Rural Banks. A broader set of similar partnerships would encourage the stronger institutions to grow faster and build more product depth relative to the poorer performing ones thus incentivising the weaker institutions to also improve their performance.

A review of these experiences suggests some directions for improving performance of Regional Banks in India.

Regulation and Supervision

Currently, there is considerable opacity about the true health of Regional Banks. Cooperative institutions, given their unique structure, have very little interaction with debt and equity markets – both public and private. In the context of the Cooperative Banks the Rajan Committee (2009) recommended that: “Indeed it [the Committee] would suggest rethinking the entire cooperative bank structure and moving more to the model practiced elsewhere in the world, where members have their funds at stake and exercise control, debtors do not have disproportionate power, and government refinance gives way to refinancing by the market”¹⁶⁹. This suggests that integrating Regional Banks into the more mainstream financial markets may offer one way out in which refinance by NABARD or Credit Guarantee support by CGTMSE is offered not on automatic basis but as a fairly priced second-loss deficiency guarantee allowing the Regional Bank to sell its originated loans either directly to other institutional investors, including National Wholesale Banks. All of this will serve to reveal more information about the financial health of these institutions and enable the stronger ones to forge more partnerships and grow.

While the Risk Based Supervision process has been designed for the larger and more complex institutions, a similar effort could be conceived of for the Regional Banks allowing supervisors to direct scarce on-site supervision resources to higher-risk institutions. The help of commercial ratings agencies such as CRISIL could also be taken to formally rate these Regional Banks and provide the ratings to the depositors on a regular basis – thus including them more directly into the risk-containment process. Since DICGC offers deposit insurance to these banks, the Agreement between DICGC and the bank provides an additional opportunity to both ensure that the bank is run well on an on-going basis and resolved quickly in the event that it turns insolvent. Such an Agreement would provide for risk-based pricing of deposit insurance; the right to carry out both on-site and off-site inspections; and to take possession of the bank in the event that it goes into liquidation. FDIC provides a good model for such an engagement between a deposit insurer and a Regional Bank.

Given the sheer size of the country and each state, it is clear that there will be a need to build much stronger regulatory capacity at the state level. There is a considerable amount of financial regulation that happens at the state level even today in India. For instance, the Registrar of Chits regulates the chit fund industry and the Registrar of Cooperatives regulates cooperatives in each state. In addition the RBI has recommended that regulation of NGO-MFIs also be done at the state level. However, this state level financial regulatory framework is fragmented and there is merit in the creation of a State Finance Regulatory Commission (SFRC) into which all the existing State Government-level regulators could be merged and functions like the regulation of NGO-MFIs and local Money Services Business (referred to in Chapter 3.1) could be added on. In some states, the SFRC could be created by upgrading existing Institutional Finance Cells. There is also value in bringing regulatory function close to the enforcement function under the Economic Offences Wing (EOW), so as to ensure that they are working closely together. The RBI must be closely involved over a longer time frame in training the commissioners and licensing and accrediting the Commission itself. The local regional directors of the RBI could, for example, be ex-officio Chairs of the Commission’s Board of Governors while the Commissioner could be a senior State level appointee drawn from the local Banking community. The District Magistrates would also play an important role in their respective districts. In the US which has the most varied and deep financial system relative to any other country, each state has a similar commission¹⁷⁰. This is the process through which, jointly with the FDIC, the US ends up effectively regulating the variety of institutions that make up the financial services landscape in that country¹⁷¹.

Recommendations:

4.13 Regional Banks continue to have a strong appeal for inclusion but low demonstrated stability in the Indian context. Robust solutions are required vis-à-vis regulation, supervision, risk management, and governance of the existing Regional Banks before any new ones are created.

4.14 In a manner similar to National Banks, for Regional Banks as well, refinance by NABARD or credit guarantee support by CGTMSE should be designed as risk-based guarantees and not available automatically. [Similar to Recommendation 4.12]

4.15 While the Risk Based Supervision process has been designed for the larger and more complex institutions, a similar effort could be conceived of for the Regional Banks allowing supervisors to direct scarce on-site supervision resources to higher-risk institutions. The help of commercial ratings agencies could also be taken to formally rate these Regional Banks and provide the ratings to the depositors on a regular basis – thus including them more directly into the risk-containment process.

4.16 Since DICGC offers deposit insurance to these banks, the Agreement between DICGC and the bank provides an additional opportunity to both ensure that the bank is run well on an on-going basis and resolved quickly in the event that it turns insolvent. Such an Agreement would provide for risk-based pricing of deposit insurance; the right to carry out both on-site and off-site inspections; and to take possession of the bank in the event that it goes into liquidation.

4.17 A State Finance Regulatory Commission (SFRC) could be created into which all the existing State Government-level regulators could be merged and functions like the regulation of NGO-MFIs and local Money Services Business could be added on. In some states, the SFRC could be created by upgrading existing Institutional Finance Cells. There is also value in bringing the regulatory function close to the enforcement function under the Economic Offences Wing (EOW), so as to ensure that they are working closely together. The RBI must be closely involved over a longer time frame in training the commissioners and licensing and accrediting the Commission itself. The local regional directors of the RBI could, for example, be ex-officio Chairs of the Commission’s Board of Governors while the Commissioner could be a senior State level appointee drawn from the local Banking community. The District Magistrates would also play an important role in their respective districts.

Chapter 4.7
Non-Banking Financial Company

The total number of registered NBFCs was 12,225 as on March 31, 2013 comprising 254 deposit taking NBFCs (D-NBFCs), 418 systemically important non-deposit taking NBFCs (NBFCs-ND-SI) whose asset size exceeded Rs. 100 crore¹⁷². The ratio of the deposits of all NBFCs relative to the liabilities of the banking system is less than 1 per cent¹⁷³. NBFCs are classified on the basis of their activity into six categories: Loan Companies, Investment Companies, Asset Finance Companies, Infrastructure Finance Companies, Systemically Important Core Investment Companies, and Micro Finance Institution NBFCs. Retail NBFCs have better origination and collection capabilities and are able to reach out to the customer to perform door-step services. For instance, comparing the portfolio of housing loans originated by Banks and NBFCs, it emerges that Housing Finance Companies outperform Banks in terms of Gross NPAs with defaults of only 0.68 per cent on 2012, when compared to 2.63 per cent for Banks¹⁷⁴. Eighty per cent of AUM of all retail NBFCs in the CRISIL-rated portfolio have an outstanding rating in ‘AA’ or above. These institutions have anywhere between 10 to 20 per cent of capital adequacy.

One of the most important concerns expressed about the NBFCs is that they are shadow banks since they operate outside the regular banking sector but perform many of the same functions. The European Commission in its most recent paper lists entities and activities that it considers as a part of Shadow Banking¹⁷⁵. These include: Special purpose entities which perform liquidity and/or maturity transformation; for example, securitisation vehicles such as ABCP conduits, Special Investment Vehicles (SIV) and other Special Purpose Vehicles (SPV); Money Market Funds (MMFs) and other types of investment funds or products with deposit-like characteristics, which make them vulnerable to massive redemptions ('runs'); Investment funds, including Exchange Traded Funds (ETFs), that provide credit or are leveraged; Finance companies and securities entities providing credit or credit guarantees, or performing liquidity and/or maturity transformation without being regulated like a bank; Insurance and reinsurance undertakings which issue or guarantee credit products; Securitisation; and Securities lending and repo.

Based on the above definition, NBFCs fall in the category of finance companies that provide credit. However, as can be seen from the Table given below, NBFCs in India, for the most part are regulated proportionately to a Bank. Therefore in accordance with the European Commission definition, with some changes, they could be considered as an integral part of the larger banking system and not as Shadow Banks. In fact in “French banking legislation the definition of a bank arises from only the asset side of the balance sheet, that is, from lending. Consequently, in France regardless of how credit institutions fund themselves, they are considered banks, and, as such, subject to all banking regulation”¹⁷⁶.

Rationale for Differentiated Treatment of Banks and NBFCs

The gaps in the manner in which Banks and NBFCs are currently treated in India are mentioned in Table 8. It is not clear however, if full convergence is immediately possible and in some cases even desirable. For example, while it is clear that SLR as a prudential tool has outlived its utility for both Banks and NBFCs and eventually needs to be removed, it is not as obvious that even niche NBFCs which are engaged in promoter funding or specialised infrastructure lenders should be required to adhere to Priority Sector Lending norms or be given access to the Lender of Last Resort (LOLR) facilities.

While NBFCs in their current format play a useful role and will continue to do so, and every effort needs to be made to ensure that they are able to perform that role effectively, they will, of necessity, have limits to how large they can become. This will be the case on account of differentiated capital adequacy requirements, absence of access to payment systems, constraints imposed by dual regulation, and other restrictions on the nature of business they are allowed to undertake and the nature of risks they are allowed to take. From a systems design perspective this would be required because as players outside the banking system they would not enjoy the same degree of protection as banks would, and nor would the three principles of Stability, Transparency, and Neutrality apply to them in the same degree. The differentiated banking design offers multiple end points towards which NBFCs could head, particularly if issues such as SLR requirements, uniform applicability of CRR requirements, and the current definition of PSL which is rooted in historical perspectives, are satisfactorily dealt with. It needs to be borne in mind that other than a small number of entities, the bulk of the NBFC sector remains very small, does not have the ability to garner public deposits, and in aggregate has performed at a very high level of quality. The sector as a whole therefore does not constitute a source of systemic instability. It has instead been playing the role of extending the reach of the banking system to the more difficult parts of the economy.

1. Risk Weights: While it is true that an NBFC may have assets on its books that have a lower expected loss rate and therefore should see the benefit in terms of Standard Asset Provisioning, it is also the presumption that an NBFC is likely to have a sectorally or regionally concentrated asset profile relative to a bank and therefore the aggregate portfolio level risk of default would not benefit from the kind of diversification that a bank would have. It would therefore be prudent to not give a lower risk weight to an asset on the books of an NBFC to compensate for the higher correlation risk between assets.

2. Capital Adequacy: The NBFC is presumed to be a niche participant getting into harder to reach categories, and therefore, relative to a full service National Bank which in contrast is presumed to operate in less risky categories, while permitted to enter with lower thresholds so that it can serve even very small niches, it will need to maintain a higher level of capital adequacy.

3. Priority Sector: While several NBFCs are deeply present in priority sectors already, for others, given their relatively small size and niche role, they may not be in a position to fulfill priority sector obligations and it may be desirable to have them focus on other sectors.

4. Entry Capital Requirement: Given the non-deposit taking nature of these entities, a high barrier in terms of entry capital does not seem necessary. As noted above, several of them have assets under management of less than Rs. 100 crore and may be quite regionally or sectorally specialised.

As noted previously, there are several categories of non-deposit taking NBFCs based on their activity and new ones are being proposed such as the Small Business NBFC. The Committee believes that a proliferation of categories is unwieldy, creates room for regulatory arbitrage, and hinders the evolution of NBFCs which have the ability to provide the broad range of credit products envisaged in the vision statement. The Committee recommends a consolidation of multiple NBFC definitions into a two categories: a distinct category for Core Investment Companies (CIC) and another category for all other NBFCs. Benefits that were previously available to specific NBFC types, such as tax benefits, bank limits, and priority sector status may continue to be available even after consolidation on a pro-rata asset basis. As discussed previously, the supervision strategy is recommended to be risk-based. The Rajan Committee (2009), in its report also notes¹⁷⁷ that balkanisation forced by regulation even between areas of the financial sector that naturally belong together can result in financial institutions not being able to realise economies of scope in these areas, leading to inefficiency and slower growth. The asset class differences in behaviour can be accommodated through differential provisioning on the basis of asset class rather than by creating new NBFC categories.

Funding Issues:

Given that the large majority of NBFCs are not deposit-taking, they rely on wholesale funding and this has a direct bearing on their ability to grow. Other than equity, wholesale funding includes the following sources: borrowings from banks, refinance from apex institutions, borrowings through External Commercial Borrowing (ECB), sale of securitised assets, and issuance of bonds to capital markets investors. There are constraints within each of these. These constraints need to be addressed in a systematic manner instead of creating new deposit-taking NBFCs or following an accelerated process of full-service bank licensing.

1. Banks have been an important source of funding for NBFCs and will continue to be so. While NBFCs have historically not been a source of systemic risks for the banks, it has indeed been the case that shocks to the banking system (as witnessed in 2008) have had a significant negative impact on the NBFCs¹⁷⁸. Therefore, from the perspective of diversifying the liabilities of NBFCs and minimising liquidity risks, it is important to deepen capital markets access for NBFCs. Investors such as mutual funds, insurance companies, provident and pension funds and private accredited investors could complement bank funding to this sector.

2. There has been concern around NBFCs raising public deposits via the Non-Convertible Debentures route. It is recommended that a clear framework be developed for Qualified Institutional Buyers (QIBs) and Accredited Individual Investors (AIIs) who may participate in these issuances and clarifications issued on NBFC eligibility for the shelf prospectus facility. Section 60A of Companies Act, 1956 had explicitly provided the eligibility list for availing the “Shelf Prospectus” facility only to public sector banks, scheduled banks and public financial institutions. Even though SEBI had amended the SEBI (Disclosure and Investor Protection {DIP} guidelines)¹⁷⁹ to this effect to give effect to the same and then through SEBI (Issue and Listing of Debt Securities) (Amendment) Regulations, 2012 issued on 12th October 2012 had permitted the availability of ‘shelf prospectus’ for a period of 180 days for any issuer going for private placement of debt. Now that Section 31 of the new Companies Act, 2013 does not prescribe the eligibility list and enables the entities as approved by SEBI to avail the said facility, it would ensure level playing field if all issuers are brought into this ambit so that the benefit of ‘shelf prospectus’ is available for one year to all issuers instead of 180 days.

3. NBFCs have been important issuers in the securitisation market, particularly in the context of priority sector assets. As discussed previously, this is a promising alternative to forcing all Banks to originate assets directly. RBI guidelines on securitisation have been very enabling. However, the tax-exempt status of pass-through securitisation vehicles needs to be restored.

4. Norms on External Commercial Borrowing are rigidly defined and eligibility varies across different categories of NBFCs. It is recommended that there be more flexibility within this category. Specifically, ECB in Rupees needs to be permitted for all institutions. For ECB not in Rupees, eligibility to be linked to size and capacity to absorb foreign exchange risk rather than specific NBFC categories.

5. Access to various refinance schemes is restricted by institution type rather than activity, thus violating the Neutrality principle as far as NBFCs are concerned. For instance, most of NABARD’s existing refinance schemes (includes schemes which amounted to Rs. 66,095 crore in 2012-13 for crop loans to farmers by commercial banks, RRBs, and cooperative banks)¹⁸⁰ and guarantees under SIDBI’s Credit Guarantee Trust for Micro and Small Enterprises (CGTMSE)¹⁸¹ are not available for NBFCs even if they are engaged in eligible lending activities under these schemes.

6. As far as equity funding is concerned, the minimum capitalisation norms are provided for NBFCs based on foreign ownership. The capitalisation slabs are USD 0.5 million, USD 5 million and USD 50 million for foreign capital below 51 per cent, below 76 per cent, and above 75 per cent respectively. These limits are quite steep and, in the context of scarce domestic equity capital, make it far too difficult to access foreign sources of equity, including patient capital from DFIs and this may have a link to higher pricing of loans from NBFCs, in particular NBFC-MFIs. If the root cause of these restrictions is related to money laundering concerns, that may be mitigated by instituting additional reporting requirements on Banks/Authorised Dealers (AD) that enable these transactions.

Risk Management:

While it is indeed the case that the non-bank character of NBFCs makes them less of a systemic risk concern on a national basis, on a regional or a product market basis, failure of an NBFC can have severe consequences for clients and local economies where they may have been the sole or dominant formal provider. Research by IGIDR that studied the impact of the Microfinance Ordinance on households in Andhra Pradesh, found that consumption dropped by 19.5 per cent over the one year after the crisis across all income classes, and the income class with the highest use of micro-credit, experienced a negative impact on even food consumption¹⁸². It can therefore be concluded that failures in the NBFC sector can have negative consequences for the real sector. It is important therefore to ensure that while NBFCs may continue to enjoy some exemptions, such as that on minimum entry capital, adequate attention is paid to the manner in which they manage their risks.

1. Liquidity risk: Due to the added significance of the underserved populations to the politics of both State and Central governments, as well as the existence of more than one regulatory authority for certain types of credit institutions and the delays in attaining regulatory clarity wherever loopholes exist, extending financial services to underserved segments is fraught with risk. This in turn has real consequences for credit delivery. It drives investors in such institutions to demand a higher risk premium to mitigate this perceived risk, which translates to higher prices of offering credit to the end customer. This risk is best exemplified by the Andhra Pradesh Micro Finance crisis of 2010, where an ordinance passed by the State Government resulted in large losses to portfolios of several MFIs; impairment in the credit behaviour of borrowers and a withdrawal of credit facilities by all banks to all MFIs, even those outside Andhra Pradesh¹⁸³. There is a need to adequately protect the sector against correlated behaviour by commercial banks and regulators. If this risk can be mitigated through catastrophic insurance or a guarantee fund, it can benefit lenders and improve ratings of these entities.

2. Provisioning: Different customer asset sub-groups behave very differently from each other and it is recommended that the regulator specify NPA recognition and provisioning rules, including for standard assets, at the level of each asset-class and require that all NBFCs conform to these mandates.

3. Risk Measurement and Disclosure: Require all NBFCs to have better on-going risk measures. These include disclosure of their stress test results both at an overall balance sheet level as well as at a segmental level on an annual basis. All NBFCs must adopt core banking systems so that they are able to do the sophisticated reporting required for effective off-site supervision.

Micro Finance Institution-NBFCs

NBFC-MFIs are a specific type of NBFC created by RBI in December 2011 as the seventh category within NBFCs. NBFC-MFIs typically serve low-income women clients in rural and urban environments through a group loan product. 22 per cent of all small borrower accounts are with MFIs. This is higher than the number of such accounts with Commercial Banks or RRBs. An NCAER study found that most MFI borrowers belonged to the 31-40 years group, in their productive years, who needed credit most in the absence of employment opportunities and a vast majority of the borrowers were either illiterate or had studied only up to primary level¹⁸⁴. These loans are used for a variety of purposes including repayment of informal debt, house repair and renovation, purchase of livestock, agriculture, education and household consumption including health expenditures.

The NBFC-MFI guidelines of 2011, subsequently amended in 2012, laid down specific requirements in terms of: minimum net owned funds requirement which at Rs. 5 crore is higher than the Rs. 2 crore requirement for regular NBFCs; margin caps of 10 per cent and 12 per cent for large and small MFIs respectively; caps on number of loans to a client; caps on loan ticket size; restrictions on loan tenure; defining qualifying assets as comprising pre-dominantly income-generating loans to clients of a certain income; and mandatory credit bureau reporting. Many of the criteria that determine whether an NBFC can be an NBFC-MFI, such as the borrower income criteria¹⁸⁵, loan size and tenure¹⁸⁶, number of borrowings¹⁸⁷, purpose of loan¹⁸⁸, have the potential to create undesirable consequences, both for the end borrowers and the lending institution.

There has been a lot of concern about the rates of interest charged by MFIs and is the reason why the NBFC-MFI regulations included a price cap, in what was viewed as a significant reversal of policy stance on interest rate deregulation being followed by RBI. This was also in contrast to recommendations of the Rajan Committee (2009), which advocated for liberalising the interest rate charged, but putting in place requirements of transparency as well as eligibility for accessing priority sector funding¹⁸⁹. Despite all this, it was a welcome first step because it assuaged the anxieties around “extreme profiteering” by MFIs and also ensured that there was a public endorsement of the critical role that these institutions play; a reaffirmation of the fact that RBI was actively involved in their oversight; and that indeed in its judgment these rates were justified on account of the higher cost to serve this very vulnerable client group.

MFI interest rates are currently around 26 per cent with cost of funds accounting for about 13-14 per cent of this. Competitive forces have not produced price competition in this sector and most MFIs have similar levels of pricing. There is a concern that the current NBFC-MFI guidelines that restrict the number of MFIs serving a client are effectively acting as a barrier to entry thus limiting the competitive forces operating in the sector. Another reason why sharp reductions in interest rate, proportionate to costs are not being witnessed is that many MFIs finance growth through revenues rather than equity capital. Increasing the sources of equity funding is therefore an important aspect of bringing down interest rates in the MFI sector even as it continues to expand its outreach. There are wide variations in the cost structures of MFIs and the price cap has indeed created conditions in which the low cost players are no longer required to offer the benefit of these costs to their customers. Some of the ways to address this issue include the following:

1. Enable better benchmarking by requiring all NBFC-MFIs to disclose their operating costs (direct and indirect) of a mature branch to the RBI or MFIN once it becomes an SRO.

2. Total borrowing limit for the small borrower segment may be increased immediately to Rs. 100,000 across all lenders. If total indebtedness is being tracked adequately, the stipulation of a maximum number of lenders appears redundant and can be gradually removed as this would also help in creating intensified price competition. This is consistent with focus on total indebtedness of the small borrower in relation to their debt-servicing capacity and not just indebtedness per se or merely from NBFC-MFIs.

3. Gradually phase out the restriction on NBFC-MFI from making larger value loans (> Rs. 50,000). This constrains their diversification into categories such as home loans and small business loans.

Regulation such as restricting the proportion of consumption finance by NBFC-MFI has constrained lending for consumption purposes for low-income households and has overwhelmingly laid policy emphasis on credit for income-generation. That this is a misplaced notion is well-articulated by the RBI Governor Raghuram Rajan’s statement of the importance of credit to smoothen consumption needs and to tide over emergencies for low-income households¹⁹⁰. A mechanical emphasis on income-generating loans also produces new risks linked to success of the project and may trigger a shift to more expensive informal loans for the purpose of consumption smoothing. There is also no empirical evidence that at the current levels increases in the quantum of formal credit has led to unsustainable levels of consumption producing higher levels of credit risk. All policy biases against consumption finance need to be removed, particularly since total formal sector indebtedness can now be tracked through credit bureaus.

Additionally, meeting the complete financial services needs through household assessments is a more promising approach to serving low-income households than narrowly assessing product requirements of individuals.

Transition of NBFCs to Differentiated and National banks

Entities that start out as a niche NBFC may over time acquire the profile of a Bank, and would need to be able to transition smoothly into either a Full Service Bank or a Wholesale Consumer Bank or a Wholesale Investment Bank.

Currently there are specific barriers that prevent several of those entities that already have the character of Banks from becoming one. These include:

1. Current definitions of PSL are restricted only to a few sectors. Specialised NBFCs operating in sectors such as infrastructure are not able to include their loans as PSL and will have to originate in sectors where they have no expertise, for example, agriculture.

2. Absence of markets to trade in PSL assets. The definitional issue in PSL is compounded by inability to buy and sell PSL assets. This forces every player to create their own origination capability and this is challenging for players who may have the liquidity to purchase PSL assets but not the expertise to build origination engines.

3. SLR is applied on all banks despite no prudential reason to do so in the presence of capital adequacy guidelines.

4. CRR is uniformly applied on demand and time liabilities, instead of just on demand liabilities.

The Committee recommends that these barriers be expeditiously addressed because they will both enhance the effectiveness of existing Banks as well as allow strong NBFCs to enter as Differentiated Banks or National Banks, without abandoning their core capabilities. As noted previously, this has the potential to significantly strengthen financial deepening in India.

This transition will have a number of benefits to the NBFCs:

1. The concerns of dual regulation will be properly addressed.

2. Being accorded the status of a bank, even if only a Wholesale Bank, will give them access to the payments system and protection under LOLR facilities.

3. There will be full convergence on all other fronts, including capital adequacy.

However, even without these changes, for a group of NBFCs, such as those that are already engaged substantially in originating loans that qualify under current PSL guidelines, may benefit from a transition even under the current regime.

Accordingly, the Committee recommends that under the Banking Regulation Act, a set of banks may be licensed which may be referred to as Wholesale Banks with the following characteristics:

1. Given that their primary role is lending and not the provision of retail deposit services, they will only be permitted to accept deposits larger than Rs. 5 crore.

2. Since they could expect to borrow large amounts from other banks, net liabilities from the banking system will be permitted to be deducted from their NDTL computation for the purposes of ascertaining their SLR obligations on par with the treatment currently given for CRR.

3. Since other banks are expected to lend large amounts to Wholesale Banks, those other banks will be permitted to deduct their net assets to the banking system from the computation of their ANBC (the amounts on which PSL requirements are to be applied).

4. In view of the fact that they will not take retail deposits, the minimum entry capital requirement for them will be Rs. 50 crore compared to the Rs. 500 crore required for full-service SCBs.

5. If the institution has fewer than twenty branches through which it operates, it will not be required to meet the 25 per cent branching requirement. Institutions with twenty or fewer branches could be referred to as Wholesale Investment Banks while those with a larger branch network could be referred to as Wholesale Consumer Banks.

6. Wholesale Consumer Banks could be permitted to act as BCs for other full service banks.

7. 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.

Recommendations:

4.18 The Committee recognises that a partial convergence of NBFC and Bank regulations may be desirable. It recommends the following:

4.19 Multiple NBFC definitions should be consolidated into two categories: a distinct category for Core Investment Companies (CIC) and another category for all other NBFCs. Benefits that were previously available to specific NBFC types, such as tax benefits, bank limits, and priority sector benefits should continue to be available even after consolidation, on a pro-rata asset basis.

4.20 The Committee recommends addressing wholesale funding constraints faced by NBFCs in a systematic manner. The following are the specific recommendations in this regard:

1. A clear framework to be developed by RBI and SEBI for Qualified Institutional Buyers and Accredited Individual Investors who may participate in debt market issuances of NBFCs.

2. Benefit of ‘shelf prospectus’ should be available for one year to all issuers including NBFCs.

3. Permit ECB in Rupees for all institutions.

4. For ECB not in Rupees, eligibility should be linked to size and capacity to absorb foreign exchange risk rather than specific NBFC categories.

5. The nature of activity, rather than institution type, must be made the criterion for availing refinance from NABARD, NHB, SIDBI and credit guarantee facilities.

6. Current capitalisation slabs on foreign equity funding should be relaxed and money laundering concerns should be mitigated by instituting additional reporting requirements on Banks/Authorised Dealers (AD).

4.21 In a manner similar to banks, different customer-asset combinations behave very differently from each other and it is recommended that the regulator specify NPA recognition and provisioning rules, including for standard assets, at the level of each asset-class and require that all NBFCs conform to these mandates. [Identical to Recommendation 4.3 for National Banks]

4.22 Require all NBFCs to have better on-going risk measures. These include disclosure of their stress test results both at an overall balance sheet level as well as at a segmental level at least annually. All NBFCs must adopt core banking systems so that this can enable better off-site supervision. [Identical to Recommendation 4.4 for National Banks]

4.23 Enable better benchmarking by requiring all NBFC-MFIs to disclose their operating costs (direct and indirect) of a mature branch to the RBI or MFIN once it becomes an SRO.

4.24 The regulatory focus must be on total indebtedness of the small borrower in relation to their debt-servicing capacity and not just indebtedness per se or merely from NBFC-MFIs. Keeping this in mind, the total borrowing limit for the small borrower segment may be increased immediately to Rs. 100,000 across all lenders, including bank-lending to this segment. In order to implement this, all lenders to this segment will need to be mandated to report to the credit bureau as has been the case with NBFC-MFIs. If total indebtedness is being tracked adequately, the stipulation of a maximum number of lenders appears redundant and can be gradually removed as this would also help in creating intensified price competition.

4.25 All policy biases against consumption finance need to be removed. An example of this is restricting the proportion of consumption finance that is permitted for NBFC-MFIs.

4.26 In order to enable the gradual transition of eligible and interested NBFCs to Wholesale Consumer Banks or Wholesale Investment Banks or National Banks, the Committee recommends a re-examination of PSL definitions [also see Recommendation 4.41], creating an active market for PSL assets, assessment of the relevance of SLR in light of capital adequacy norms, and application of CRR on time liabilities.

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

1. Given that their primary role is lending and not the provision of retail deposit services, they will only be permitted to accept deposits larger than Rs. 5 crore.

2. Since they could expect to borrow large amounts from other banks, net liabilities from the banking system will be permitted to be deducted from their NDTL computation for the purposes of ascertaining their SLR obligations on par with the treatment currently given for CRR.

3. Since other banks are expected to lend large amounts to Wholesale Banks, those other banks will be permitted to deduct their net assets to the banking system from the computation of their ANBC (the amounts on which PSL requirements are to be applied).

4. In view of the fact that they will not take retail deposits, the minimum entry capital requirement for them will be Rs. 50 crore compared to the Rs. 500 crore required for full-service SCBs.

5. If the institution has fewer than twenty branches through which it operates, it will not be required to meet the 25 per cent branching requirement. Institutions with twenty or fewer branches could be referred to as Wholesale Investment Banks while those with a larger branch network could be referred to as Wholesale Consumer Banks.

6. Wholesale Consumer Banks should be permitted to act as BCs for other full service Banks.

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.

Chapter 4.8
Priority Sector Lending

Introduction

Directed lending through efforts such as the Priority Sector Lending (PSL) program of the RBI have had a well-established history across many countries at different points in their development and have served a different purpose in each country¹⁹¹. In India while priority sector focus has existed from the 1950s in some form or the other, the Priority sector Lending (PSL) program in its current form has been implemented by the RBI since 1974, when banks were advised to raise credit to specified priority sectors of the economy to the level of 33.3 per cent by March 1979¹⁹². Currently this number stands at 40 per cent with the sectoral allocation specified in Table 4.8.1¹⁹³.

Starting with this sectoral allocation the two important questions that need to be answered are: how best to achieve PSL targets? and how to set these targets? This chapter discusses both of these questions and makes some recommendations on how the current approaches may be improved upon.

Definitional Issues

While a later section in this chapter discusses whether the current sectoral approach towards setting PSL obligations is the best one and whether, over time, there would be a case to move to an entirely different approach, there are aspects of the current PSL guidelines that need to be reaffirmed and certain anomalies that need to be addressed.

1. There have been several suggestions made to relax the eligibility criteria within current sectors of PSL. However, given the vast unmet need for credit for sectors under even current definitions, this does not appear desirable. For example, while exact data is not available, both national level and field level data indicate that agriculture continues to be heavily under-banked. At the national level, from Table 2.3 in Chapter 2.2 it can be seen that against a minimal aspiration of 50 per cent Credit to GDP for each sector, while Industry is at 87 per cent, agriculture is only at approximately 36 per cent - offering perhaps one explanation for the inflationary and supply side shocks emerging from that sector of the economy. At the field level, based on the data from a rural financial institution in Tamil Nadu and NABARD’s scale of finance estimates, the demand for credit by small and marginal farmers at a Gram Panchayat level comprising fewer than 2,000 small and marginal farmer households, is about Rs. 1.95 crore¹⁹⁴. If this is assumed to be the level of demand for all the 2.46 lakh Gram Panchayats in India, it produces an estimate of Rs. 4.8 lakh crore as the credit demand from small and marginal farmers alone. This is approximately equal to the amount of total credit that currently flows to the entire agricultural sector in India – both direct and indirect. Similarly, another field level estimate¹⁹⁵ shows credit demand for micro-enterprises just in Bangalore Urban District alone to be approximately Rs. 17,000 crore. If merely 25 urban centres have a similar level of credit demand from micro-enterprises, just this sector alone would require more than Rs.4 lakh crore of credit. From these admittedly crude approximations, it appears that there is sufficient room for growth of credit even within current definitions and that there is a fairly severe incidence of under-penetration for these very important sectors of the economy. Unless more carefully done research produces numbers that are very different from these approximations, there appears to be no case to relax and broaden the PSL definitions in the context of the current PSL guidelines.

In this context it is interesting to note that despite the many years of focus on agriculture overall as a sector it has not crossed even the very low financial depth threshold of 50 per cent Credit to GDP. Agricultural GDP in India is about Rs. 13 lakh crore. Achieving the 18 per cent agricultural targets under PSL at the current Net Bank Credit level would imply a credit number of Rs. 9 lakh crore giving a relatively healthy sectoral credit-to-GDP of 70 per cent. However, in reality, as mentioned earlier, the sectoral credit-to-GDP of agriculture is much lower. In addition, over the years, there is also a disproportionate increase in credit per capita to farmers with large land-holdings even though their share in agricultural production is lower.

This is also corroborated by the continued prevalence of informal indebtedness among farmers - AIDIS (2002) reveals that non-institutional sources account for almost 40 per cent of borrowing by cultivator households. In fact only 14 per cent of the marginal farmers (with land holdings less than 1 hectare) were taking institutional credit in 2009, with the remaining largely relying on informal sources of credit for their credit needs¹⁹⁷. It could well be the case that low credit penetration is one of the important factors responsible for the rapid decline in the share of agriculture, despite the involvement of such a large part of the rural populace in this activity.

2. Some PSL delivery models depend on lending through real sector intermediaries such as fertiliser dealers and sugarcane companies. While these channels do offer a measure of convenience to the bankers, they also run the risk of tying the hands of the farmer and leading to malpractices such as over-consumption of inputs like fertilisers and pesticides, over-pricing of inputs, and under-pricing of produce. Moreover these non-financial channels are not bound to follow the customer protection guidelines applicable to the financial sector. These channels therefore represent at best a stop-gap arrangement and there is a need for formal financial sector channels to emerge that are capable of dealing with this customer base directly. Accordingly, it is consistent for PSL guidelines to encourage this shift by not making lending through real-sector intermediaries eligible for PSL. Similarly, lending to real-sector intermediaries has also been removed from the eligible list in the last change of PSL guidelines. This is consistent with the objectives of directing credit towards segments that experience the most credit rationing. Several real-sector intermediaries are highly rated companies and should have no constraints in accessing credit from Banks outside the PSL framework.

3. Direct Agriculture lending targets, in the manner in which they are currently defined, appear however, to be biased against landless agricultural labourers as well as marginal farmers who self-supply labour. The sole factor of production available to the landless labourer is her own labour. In order to maintain this factor of production (self-supplied labour), she has to spend on health, food, life insurance and disability insurance premiums, and other critical consumption items throughout the year. In order to meet these needs, due to the seasonality of farm incomes, she will necessarily have to borrow to manage expenses particularly during the agricultural off-season. However such borrowing by a landless labourer does not qualify under Direct Agriculture and is viewed as financing consumption expenditure instead of production expenditure. In direct contrast to this, a medium-farmer who relies on farm equipment as the main factor of production can borrow to purchase and maintain such equipment even when it is not being used, and his borrowing is treated as a Direct Agriculture. This distinction between labour and other factors of production results in landless labourers systematically facing higher costs of borrowing relative to land-owning farmers. This difference in costs of borrowing could be between 20-60 per cent depending on the source of credit. The issue faced by the landless labourer also persists in the context of marginal farmers who self-supply labour as well as large farmers using labour-intensive techniques.

The Committee therefore recommends that all loans given to landless labourers and small and marginal farmers should be counted as a part of Direct Agriculture and not merely the wages component of a loan given to a farmer for financing her agricultural production.

4. PSL is currently defined with respect to loans made to various sectors and segments identified in the guidelines. It also includes investment by banks in securitised assets and Inter-Bank Participation Certificates where the underlying assets qualify under the PSL criteria. The Committee recommends the following with respect to PSL investments:

1. Investment by banks in bonds of institutions must qualify for PSL where wholesale lending to the same institutions already qualifies under PSL. The fact that loans to an NBFC-MFI qualify for PSL while bonds do not for the same institution appears to be an anomaly. Separately there is also a need to ensure that bond and other investments of banks are permitted to be held in the “banking book” of the bank based on declared intent and not merely based on source or legal documentation.

2. Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation.

3. Investment by banks in RIDF must not qualify for PSL. RIDF investments are intended only to address non-compliance and also because RIDF proceeds are made available to State Governments, who, while pursuing important development goals, are not credit constrained and could raise money from market sources and should be encouraged to do so instead of crowding out bank lending. Instead, efforts need to be directed towards ensuring that Banks are able to effectively lend to the under-banked sectors of the economy.

4. Investment by banks in the form of non-fund based limits (such as guarantees) should also qualify for PSL to the extent of the credit equivalent amount of the off-balance sheet facility where loans to these categories qualify for PSL. Since a bank is taking on the risk of this facility and has offered it to the client based on assessed need, this treatment would appear to be more internally consistent than excluding such facilities and therefore denying such clients access to non-fund based facilities. ANBC should also be adjusted to include such PSL-linked, non-fund based limits.

5. As an added category, within the overall equity investment limits of banks, RBI could permit equity investments by banks in complementary infrastructure within the purview of PSL guidelines, such as rural warehouses, market yards, godowns, silos, and NBFCs in low financial depth districts. Unlike investments in downstream enterprises, these investments would have the effect of directly contributing to increased lending to the current PSL sectors and face a severe constraint in the amount of equity available to them. These equity investments may be included as a part of the overall priority sector lending targets. They should only be permitted where debt already qualifies for PSL but with a multiplier of four, to reflect the higher risk and the illiquid character of these investments. For example, a Rs. 100 crore equity investment in a rural warehousing company by a bank would be equivalent to Rs. 400 crore of priority sector lending. The benefit must accrue as long as the equity investment is held by the Bank. This list of eligible equity investments may be varied from time-to-time and marked to market as per existing investment guidelines applicable to banks.

Strong Year End Bias in Agricultural Flows

Credit flows to agriculture under PSL display considerable seasonality not explained by cropping patterns. The Task Force on Credit Related Issues of Farmers, in its report submitted to the Ministry of Agriculture, Government of India, in 2010, observed that while month-wise credit disbursement patterns should have been in line with ground level requirements of Kharif (June, July, September) and Rabi (December, January) seasons, one-fourth of the disbursements instead happen in March, a month that is not critical to agriculture production. This is explained by the fact that PSL is a financial year-end target for banks.

Considering the fact that a bulk of the PSL lending is focussed in March every year because there is only a year-end target, the Committee recommends that the current PSL targets be applicable on the last reporting Friday of the last month of each quarter in exactly the same manner as it is currently applicable in the month of March, so as to ensure more timely and continuous credit flow into priority sectors. In order to ensure administrative ease, requirements such as investment into RIDF can continue to be levied on an annual basis and computed on the basis of the average of the quarterly requirements. This may at first glance appear to make it more difficult for banks to achieve their priority sector targets but on careful examination it is clear that this will in fact make it easier for them to build a strong and vibrant priority sector business:

1. From a demand side there is nothing particularly significant about the March quarter relative to the other quarters. Historically this was chosen as the date because in an environment where banking data was all managed without the use of computers, it was convenient since it was also the financial year-end. In a Core Banking System enabled banking environment it is now possible to track these targets, if necessary, on a daily basis. From a supply side, therefore, spreading out the achievement of these targets will allow lenders to more closely match demand throughout the year and build a lending division that is equally active throughout the year and not overly so only at the year-end.

2. Currently those banks that lend for short-term crops during the Kharif and Rabi seasons do not get any benefit of that lending for the priority sector target computation since the crop loans are, for the most part, already repaid by the end of March. This also in part accounts for the lack of an adequate supply response from the agricultural sector since credit is not made available in a timely manner. A more continuous lending programme would address this issue satisfactorily as well as ensure that credit risk is managed much better.

3. The need of the entire banking system to lend large amounts of credit during a single quarter end results in undue pressure on interest rates that can be charged and since the borrowers do not really need the money at that time it also increases the risk of diversion of funds and therefore of default.

4. The need to maintain a near continuous PSL portfolio outstanding rather than just at one point during the year will increase the focus of banks on investment credit, since, unlike crop loans, it has the property that it remains outstanding over a number of years.

Restrictive interest rate policies, subventions, and waivers on agricultural loans

Banks are required to price farm loans below Rs. 3,00,000 at 7 per cent for which they receive an interest subvention of 2 per cent in the subsequent period. This pricing is not consistent with the risk levels associated with loans to agriculture. The principle of Stability requires banks to price all loans, including agricultural loans, on a sustainable basis. This stipulation of 7 per cent also violates the RBI guideline that requires banks to price all loans above their Base Rate. However, the RBI has made a special exemption for crop loans up to Rs. 3,00,000 and said that banks should charge farmers the interest rates as stipulated by the Government of India. In case the yield to the bank (after including subvention) is lower than the Base Rate, such lending will not be construed as a violation of the Base Rate guidelines¹⁹⁹.

In addition, the implementation of the Debt Relief and Debt Waiver Scheme for Small and Marginal Farmers in 2008 by the Central Government has had serious negative implications for the banking system since it created the expectation in the minds of farmers that more such schemes would follow and therefore farm-loans did not have to be serviced regularly.

The Committee is of the view that if the government does desire to provide relief in any form to the small farmer, it would be best carried out as a direct benefit transfer to the bank account of the farmer and not through the mechanism of either interest subvention or debt waiver. This would ensure that the banking system is able to price loans in a sustainable manner and also protect credit discipline amongst its borrowers. Adding a universal requirement to report all defaults to credit bureaus would ensure that the borrower also builds a strong interest in protecting his credit history. Canara Bank for example recently reported that using CIBIL data there were able to halve the NPA levels on their retail asset portfolio²⁰⁰.

The Committee also recommends that in order to guard against large scale defaults resulting from catastrophic events, banks work closely with insurance companies to purchase bank-wide portfolio level insurance against events such as large scale rainfall failure on a regional or national basis, instead of having an expectation that relief would be provided from national or state budgets.

No Market for Sale and Purchase of Agricultural Credit

While a number of other financial markets have evolved in India, including for the trading of agricultural futures contracts, there has not been any impetus for the evolution of a market for the sale and purchase of agricultural credit. The Narasimham Committee II (1998) had recommended in this context that “As a measure of improving the efficiency and imparting a measure of flexibility, the Committee recommends consideration of the debt securitisation concept within the priority sector. This could enable Banks which are not able to meet their priority sector targets to purchase debt from institutions which are able to lend beyond their mandated percentage”.

The absence of this market has meant that each financial institution has had to limit its origination of PSL loans exactly to the extent of its own deposits and market borrowings and not originate in excess of their requirements to sell to those institutions that may be flush with funds but do not have the capability or the confidence to generate an adequate quantum of PSL loans. This has also inhibited the growth of institutions that may have origination capability in low financial depth regions but not the liquidity to support it. This has not only prevented specialisations from building up but also has contributed to the imbalances in lending to those regions that have not been able to build up their own banks, as cash rich banks have chosen to focus their lending efforts in regions where they are already strong such as the Southern and Western regions of the country. As the discussion on National Banks in Chapter 4.2 reveals, in the absence of such a market, many banks are being forced to pay large penalties or to originate PSL assets directly even though in doing so, they are incurring unsustainably high costs and risk-related charges.

The creation of such a market would also enable differentiated banks with different foci and strategies such as Wholesale Investment Banks and Wholesale Consumer Banks to contribute actively to supporting the priority sectors. Such markets would also help naturally concentrated lenders such as Regional Banks (including RRBs, LABs, and Cooperative Banks) to transfer local systematic risks away from their balance sheets and thus become structurally stronger. The absence of this market has meant that the only way to redress concentration risks arising from focussed lending to specific sectors or regions has been through a broadening of origination strategy forcing them to assume risks that are inconsistent with their core capabilities.

While currently, RBI guidelines on securitisation are quite enabling vis-à-vis trading of PSL assets, one specific regulatory barrier is that the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the purchasing bank plus 8 per cent per annum. Since such a restriction does not exist on an asset directly originated by a bank and is also inconsistent with the cap on interest rates permitted to NBFC-MFIs, this should be removed for securitised assets as well.

Creation of such a market will also enable a broader group of participants in the market both as suppliers and users of credit. For example, the Food Corporation of India (FCI) and State Governments can originate warehouse receipts and raise low-cost funds against these receipts instead of being reliant only on bank credit and crowding out other borrowers that do not have the ability to access the market directly.

The Committee recommends that the RBI take steps to encourage the creation of such a market for Agricultural Credit by permitting banks to hold purchased assets within their banking book on par with assets originated directly by them, based entirely on the declared intent of the bank and not based on the legal format of the documentation.

The Committee recommends that the RBI remove the stipulation that the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the purchasing bank plus 8 per cent per annum. Further, as mentioned in Chapter 4.5, the RBI should also represent to the Government of India to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment so as to create pathways for Wholesale Banks to provide liquidity to other Banks and Financial Institutions directly originating assets in priority sectors.

While a market that trades PSL assets will be of critical importance, the Committee also recommends that regulation should additionally enable the use of risk-free PSL Certificates as a means to achieving PSL compliance amongst banks that wish to do so.

Specialisation through Differential Sectoral and Regional Weights

Current PSL strategies require each bank to originate assets in exactly the same manner as every other bank. While such an approach could be entirely feasible for banks with universal banking aspirations, it is not well-suited to benefit from the unique capabilities of more regionally or sectorally specialised banks. For example, rural cooperative banks or National Banks with a large rural branch network may be better suited to focus their entire attention on direct agricultural lending while their urban counterpart would do well to concentrate only on lending to small businesses. This has meant that not only are the banks as a whole not achieving their PSL targets, particularly to key sectors such as agriculture but are also experiencing a very high incidence of loan losses.

An alternate way of achieving the same overall goals would be to not impose a uniform sectoral allocation on each and every bank but instead to reflect the importance of a particular sector through an adjustment factor and replace the current 40 per cent PSL requirement with a 50 per cent Adjusted PSL (APSL) requirement. Banks could achieve their APSL requirement either by concentrated origination in a sector of their choice or through market purchases of qualified assets. The sector weightages can be derived every three years on the basis of relative shortfalls in achievements by banks. This is illustrated below:

According to the relative shortfalls seen in the above, sectors like Agriculture and Weaker Sections may receive higher weightages than export. Even within agriculture, Direct Agriculture would receive a higher weightage, while Indirect Agriculture would not, based on the difficulty banks face in specifically achieving their Direct lending targets. There is evidence²⁰³ to show that credit deepening in the indirect finance category has been much more pronounced than in the direct finance category. Based on the data in Table 4.8.5, the weightages for Direct Agriculture and Weaker Sections are calculated depending on the extent of underachievement relative to the target; all other sectors do not have sub-targets and it is not obvious that banks find it difficult to lend to these sectors. The following multipliers for investments in different PSL sectors are proposed for the FY 2014-15 to FY 2016-17 period:

As the scheme in Table 4.8.6 indicates, Rs.1 lent by a bank for direct agriculture (or 1 per cent of Net Bank Credit) would be multiplied by a factor of 1.25, and the Adjusted PSL achievement would be equal to Rs. 1.25 (or 1.25 per cent of Net Bank Credit). The objective of this design is to incentivise banks and others to lend to sectors that are severely starved for credit and to build specialisations in those sectors.

In addition to the shortfalls associated with the credit flows to specific sectors, there also exist large regional disparities both in the total quantum of credit disbursed by the banking system as well as for each sector. During the 11th Five Year Plan²⁰⁴, 37.6 per cent of agricultural credit was disbursed in Southern India, which accounts only for 18.7 per cent of India’s gross cropped area. On the other hand, Eastern India received only 7.3 per cent of agricultural credit with 14.7 per cent of gross cropped area and Central India received only 13.2 per cent of agricultural credit with 27.2 per cent of gross cropped area. To address this regional disparity, there is a need for differential regional multipliers that will value Rs. 1 invested in a credit starved district more than in a district well-penetrated by credit. These differential multipliers for districts are constructed based on an objectively measured index of financial inclusion - Inclusix published by CRISIL²⁰⁵. The index provides a score out of 100, with 100 indicating a fully financially included district. Scores in the index range from 96.2 for Pattanamthitta district in Kerala to a low of 5.5 for Kurung Kumey district in Arunachal Pradesh. A raw district weightage for each district is calculated based on the distance of the score for that district from that of the best performing district, i.e., the worst performing district would have the highest weightage. However, in order to ensure that the district weightages do not become too high, they need to be damped down by a factor which ensures that the district with a CRISIL Inclusix score of 50 gets an APSL of 50 per cent for achieving 40 per cent ANBC in PSL. Based on the data, Faridkot (Inclusix Rank: 141, Score: 50) in Punjab is the district for which achieving 40 per cent ANBC should yield a APSL of 50 per cent. By adjusting for this, the damping factor is obtained, which is applied to the raw district weightages. Table 4.8.7 illustrates the damped district weightages and the PSL sector weightages, which together yield the district-sector combination weightages that will need to be applied to the PSL portfolios to calculate APSL.

Based on this analysis, it emerges that Rs. 1 invested in Direct Agriculture in Pattanamthitta district (Rank: 1) yields an APSL of Rs. 1.25, while if it were invested in Kurung Kumey district (Rank: 632), it would yield an APSL of Rs. 1.56. Depending on the sector and the district that different PSL investments are made in, the overall APSL for the portfolio is calculated.

It must however be noted that the APSL is designed to be an incentive for institutions to go into more difficult-to-do regions and sectors, as well as to promote greater specialisation; it is not meant to be a disincentive for banks from lending in more better covered regions and sectors. As is clear from the design, each Rupee invested into PSL anywhere in India, in any sector, yields an adjusted value of at least 1, there is no penalty for operating in “easy” regions and sectors. What the APSL design instead offers is the choice of going to more excluded regions and sectors to get the benefits of increased weightages.

As an illustration, three different banks could choose three completely different strategies to fulfil PSL obligations based on their specialisations and strengths:

1. Bank 1 follows a strategy to lend 40 per cent of ANBC to the sectors and sub-sectors in proportions defined in the PSL guidelines and lends to a number of districts above and below the middle score district, which yields an APSL of 50 per cent.

2. Bank 2 focuses all of its lending in Direct Agriculture to be done equally among all districts, so as to get an overall weightage benefit (sectoral and regional combined) of 1.45. Such a strategy will need the bank to lend only 34.5 per cent of its ANBC to reach its 50 per cent adjusted PSL requirement. If, however, the bank were to focus all its Direct Agriculture lending to top ranked district, it would need to lend 40 per cent of its ANBC to get to an APSL of 50 per cent, while, focusing only on Direct Agriculture to the bottom ranked district would require it to lend 32 per cent of ANBC.

3. Bank 3 focuses all of its lending on SMEs in the major Indian cities, thereby foregoing any sectoral weightage benefit, but gaining some regional benefit, yielding an overall weightage of 1.16. In this case, so as to reach the APSL target of 50 per cent, the bank will need to lend 43.2 per cent of its ANBC to SMEs.

The Committee recommends that the RBI revise the PSL targets and require banks to meet an Adjusted PSL target of 50 per cent against the current requirement of 40 per cent of ANBC. The adjusted sectoral and regional weights to be used in computing the APSL for the three year period from FY 2014-15 to FY 2016-17 are specified in Annexure A.

A Fresh Examination of the PSL Framework

While the banking sector in India has grown substantially since the 1970s, the size of the banking sector in India is still inadequate to meet all the needs of the real economy. This is evident in the low financial depth ratio and the concentrated nature of balance sheets of banks. These data point to the real possibility that in a completely deregulated environment, there might not be adequate flows to sectors of the economy that have a particular importance, even if they are credit-worthy and can be served profitably. In order to ensure that India exhibits balanced growth and uses its scarce banking system resources in a manner best suited to the needs of the economy it therefore becomes necessary to “nudge” the system towards these sectors. The definition provided in the Narasimhan Committee Report I (1991) suggests that directed credit policies in India have fundamentally aimed to extend bank credit to under-banked regions as well as sectors that are deemed important to national growth and otherwise neglected sectors. All of this builds the rationale for a continuation of a priority sector lending policy until the overall financial system acquires enough depth so that no critical sector of the economy or region of the country is so starved of access to finance that it eventually starts to exert a deleterious influence on the growth of the entire economy and weakens its ability to address the challenges of deep seated underdevelopment in those parts.

However, the basis of the total quantum of PSL obligations and how that is allocated across various sectors is not clear and appears to be lacking any dynamic features. As a consequence, even though the structure of the economy has changed quite dramatically, the priority sector requirements have remained essentially unaltered for four decades, since they were first specified in 1974.

Currently, priority sectors include agriculture, which has historically contributed a major portion to India’s GDP; weaker-sections, which represent the most marginalised groups in society; and housing and education, which are deemed to require investment for creating long-term social benefits. However, since its inception, the scope of priority sector has been widened to include more and more sectors. For instance, since the 1970s there have been several additions to the sectors covered under directed bank credit – infrastructure, retail traders, small businesses, and education (1970) and the addition of a sub category of weaker sections (1980). These additions have been frequently modified and even repealed in the subsequent years. This apparent lack of any particular pattern strongly suggests that there is a need to closely examine the fundamental design of the PSL program and its designated sub-targets for different sectors in order to ensure that indeed the sequestered resources are being put to the best possible use and serving the purpose for which they were originally intended. For example, the logic for higher education loans being reckoned as PSL and primary education loans not being eligible is not clear. There are also inadvertent biases that have gotten created in PSL definitions. The issue of labourers and marginal farmers versus mechanised farmers was discussed earlier. Another concern is the bias towards ownership housing and not rental housing by making home loans up to a specific amount eligible for PSL.

In order for PSL targets to automatically reflect the needs of the underlying economy, the Committee recommends that since in accordance with its vision, each “significant” sector or sub-sector (with more than a 1 per cent contribution to the GDP) of the economy should achieve at least a 50 per cent credit to GDP ratio (financial depth) in order to ensure that the absence of finance is not retarding its growth, the difference between actual financial depth and this 50 per cent goal should determine the weight assigned to it.

The Committee also recommends that since, in accordance with its vision, each district of the country should achieve at least a 50 per cent credit to GDP ratio (financial depth) in order to ensure that the absence of finance is not retarding its growth, the difference between actual financial depth and this 50 per cent goal should determine the weight assigned to it. The table below gives the weights that are assignable to each sector of the economy using this approach.

For the illustration below, only the three main sectors of the economy have been used (agriculture, manufacturing and services), and two customer segments (sub-sectors) under each of the sectors. A deeper exercise with accurate district GDP data²⁰⁶, more granular sectors and detailed customer sub-segments would be required to design a more precise mechanism for PSL.

To reiterate, a comprehensive measure to assess the sectors and regions eligible for PSL would be built on two broad parameters:

1. District level Credit Depth

2. Sector and sub-sector level Credit Depth

The Committee recommends that RBI seriously examine moving to a new framework in which two parameters: District level credit depth, and sector and sub-sector level credit depth be used to determine the sector, sub-sector, and regional weights which are published every three years. Using these weights banks would be required to reach an adjusted PSL target of 150 per cent of ANBC.

Recommendations

4.28 All loans given to landless labourers and small and marginal farmers should be counted as a part of Direct Agriculture and not merely the wages component of a loan given to a farmer for financing her agricultural production.

4.29 Investment by banks in bonds of institutions must qualify for PSL where wholesale lending to the same institutions already qualifies under PSL.

4.30 Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation. [Identical to Recommendations 4.1(b) and 4.10]

4.31 Investment by banks in the form of non-fund based limits (such as guarantees) should qualify for PSL to the extent of the credit equivalent amount of the off-balance sheet facility where loans to these categories qualify for PSL. ANBC should also be adjusted to include such PSL-linked, non-fund based limits.

4.32 Equity investments by banks in complementary infrastructure within the purview of PSL guidelines, such as rural warehouses, market yards, godowns, silos, and NBFCs in low financial depth districts. These equity investments should be eligible for contribution to the overall priority sector lending targets. They should be permitted where debt already qualifies for PSL but with a multiplier of four, to reflect the higher risk and the illiquid character of these investments. The benefit must accrue as long as the equity investment is held by the Bank. This list of eligible equity investments may be varied from time-to-time. [Identical to Recommendation 4.44]

4.33 PSL targets should be applicable on the last reporting Friday during the last month of each quarter in exactly the same manner as it is currently applicable in the month of March, so as to ensure more timely and continuous credit flow into priority sectors. In order to ensure administrative ease, requirements such as investment into RIDF can continue to be levied on an annual basis and computed on the basis of the average of the quarterly requirements.

4.34 If the government does desire to provide relief in any form to the small farmer, it would be best carried out as a direct benefit transfer (DBT) to the bank account of the farmer and not through the mechanism of either interest subvention or debt waiver. This would ensure that the banking system is able to price loans in a sustainable manner and also protect credit discipline amongst its borrowers. Adding a universal requirement to report all defaults to credit bureaus would ensure that the borrower also builds a strong interest in protecting his credit history, even if he is a recipient of DBTs. [Similar to Recommendations 4.2 and 4.43]

4.35 In order to guard against large scale defaults resulting from catastrophic events, banks should be permitted to work closely with insurance companies to purchase bank-wide portfolio level insurance against events such as large scale rainfall failure on a regional or national basis, instead of having an expectation that relief would be provided from national or state budgets.

4.36 For the provision of food-credit, Food Corporation of India (FCI) and State Governments should be required to originate warehouse receipts and raise low-cost funds in the market against these receipts instead of being reliant only on bank credit. [Identical to Recommendation 4.46]

4.37 The stipulation that the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the purchasing bank plus 8 per cent per annum should be removed. [Identical to Recommendation 4.9]

4.38 The RBI should represent to the Government of India to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment so as to create pathways for Wholesale Banks to provide liquidity to other Banks and Financial Institutions directly originating assets in priority sectors. [Identical to Recommendation 4.11]

4.39 While a market that trades PSL assets will be of critical importance, regulation should additionally enable the use of risk-free PSL Certificates as a means to achieving PSL compliance amongst banks that wish to do so.

4.40 In order to enable greater regional and sectoral specialisation among Banks, the Committee recommends that the RBI revise the PSL targets and require banks to meet an Adjusted PSL target of 50 per cent against the current requirement of 40 per cent. Districts and sectors are weighted based on the difficulty in lending to them, and a Bank lending to a difficult sector in a difficult to reach district can benefit from a multiplier value based on the specific sector and district. Every sector-district combination has a weight associated with it and the Bank will have to reach an adjusted PSL value of 50% taking these weightages into account.

4.41 The Committee recommends that RBI seriously examine moving to a new framework in which two parameters: District level credit depth, and sector and sub-sector level credit depth be used to determine the sector, sub-sector, and regional weights which are published every three years. Using these weights banks would be required to reach an Adjusted PSL target of 150 per cent of ANBC.

Chapter 4.9
Complementary Infrastructure

In order for credit to be delivered in a sustainable manner and for it to be distributed in an equitable manner across regions, there are several elements of complementary infrastructure that need to be encouraged. This Chapter discusses some of the important elements of complementary infrastructure including:

1. Customer Data Architecture

2. Warehousing Infrastructure

3. Land Registries

4. Weather Stations

5. Registries for Movable Collateral

6. Development of second-hand asset markets

Customer Data Architecture

Robust customer data architecture is critical infrastructure for financial deepening in the country. In particular, it enables Banks to extend their outreach beyond physical branch networks. A large number of borrowers lack access to formal credit since they do not have acceptable forms of collateral or verifiability of income. These borrowers consequently don’t have formal credit histories or payment histories, which further limits their ability to access formal credit. Such borrowers are called ‘thin-file’ borrowers in the credit reporting world. The following steps can be taken to ensure that such customers can build credit histories over a period of time:

1. Use of Alternative or Non-Traditional Data: Most borrowers, the poor included, have alternate payment history track records that do not include making repayments on formal loans or credit. This could be the payment of utility (water, gas, telephone, cable) bills, rentals, and instalment payments for consumer goods and appliances, which occur on a recurring basis. These recurring payments provide an alternative method of tracking payment behaviour for potential borrowers. For instance, the Italian credit bureau, CRIF, set up a credit scoring model, the “water score,” which took up to 3 years of payment of water bills into consideration. More than 83 per cent of water customers who previously had no credit history now have a positive one thanks to paying their water bills, which made it easier for them to obtain credit.

The most significant example of a service with far-reaching impact is that of mobile service providers. Mobile service providers operate in virtually every economy today, and serve large segments of the population in these economies. Mobile phones have deep outreach and penetrate the poorest segments of the population – those without access to basic amenities, or regular phones/landlines. Mobile service providers are, therefore, natural collectors of information on the entire range of financial clients (including the unbanked), and are likely participants in credit reporting systems.

Such non-traditional or alternative data allows lenders to make more reliable assessments of applicants that have sparse “formal” credit histories, and increase access to credit to those that are typically left out of formal lending channels. For example, a 2006 study by the Policy and Economic Research Council (PERC) in the United States, that looked at 80 lakh thin-file consumer records held by a mainstream credit bureau in the United States, found that including non-traditional data increased acceptance rate (by 10 per cent in the case of energy utility data and 9 per cent in the case of telecom data). With the introduction of alternative data, a larger per cent of the thin-file population became score-able (using traditional lending models). In a one-year observation period, the study found that 16 per cent of thin-file borrowers whose credit report included alternative data had opened a new credit account as compared with 4.6 per cent of thin-file borrowers whose credit report did not include alternative data.

2. Use of Big Data: The use of transactional data such as mobile minutes, top-ups, and Voice and SMS usage to create scoring models that predict consumer behaviour presents the next frontier in credit scoring. This new set of data can be a lot richer and more voluminous. For example, a mobile account may generate hundreds or even thousands of calls and text messages per month, each carrying a rich data set that includes when the call was made, the location of the caller at the time of the call, who received the call, the type of information accessed via text messaging, and the types and number of payment transactions made through the device.

However, big data is more than financial data or mobile data. It includes all types of digital data that can be studied and analysed to construct the profile of an individual and provide tailored products and services catering to this individual. Several start-up companies have emerged around the globe that use social data feeds (like Twitter, Facebook, LinkedIn) to develop credit profiles and scores of people that are not traditionally banked, and offer credit to them.

One example is Cignifi, a start-up that partnered with Oi Telecom, Brazil’s largest telecom operator, to develop a pilot scorecard using consumer call data records that would predict creditworthiness of low-income borrowers in Brazil’s poorest region. Specifically Cignifi worked with data of Oi Telecom’s ‘Oi Paggo’ product which provided pre-pay customers with a credit card delivered entirely through their mobile phone. Oi Paggo was looking for ways to improving its risk assessment of pre-pay customers, the majority of whom did not have formal credit histories, to reduce loan loss rates and improve uptake of the product. Cignifi used call data records (CDR) of 27 lakh pre-pay mobile consumers in 21 cities in north-east Brazil specifically looking at detailed calls, SMS, and top-up activity. The scoring model they built was back tested against loan and payment data from a portfolio of 40,000 virtual credit cards issued to a set of those same customers by Oi Paggo. Cignifi was able to develop a scorecard that provided Oi Paggo a comprehensive underwriting strategy for targeting low-income consumers in a profitable manner while limiting risk exposure. Cignifi analysis showed that if Oi Paggo had used their scorecard during preapproval and underwriting, they would have approved 24 per cent net new customers and reduced default rate in its portfolio from 12 per cent to 9 per cent. Even though big-data related developments are at a nascent stage in India, creating robust customer data architecture will be a critical complementary infrastructure for financial deepening in the future.

3. Credit Bureaus to cover all credit institutions: With the integration of microfinance borrowers into main stream credit reporting, a larger segment of the borrowing population (around 2.5 crore microfinance loan records are uploaded monthly on the credit bureaus) is being included in the coverage of bureaus. However, to play a more effective role in inclusion, credit bureaus will eventually have to cover all credit institutions including banks, cooperatives, and NBFCs. There is an urgent need for credit bureaus to tap into such sources of credit information in order to build credit histories of people. The regulator should require that all loans by regulated institutions should be reported to at least one credit bureau.

4. Expand list of specified users for Credit Bureau data: The Credit Information Companies Act (CICA), 2005 provides for specified users, notified by the RBI, who can get access to bureau data. The list of specified users, as well as non-credit contributors needs to be expanded as India gains experience with credit information. Extant regulations denote cellular and phone service providers (amongst others) as specified users. However, while these entities are not ‘banking companies’ and are therefore not mandated to share data with the bureaus, they are allowed to access bureau data in making decisions relevant to their activity. Good practice credit reporting systems generally operate on the principle of reciprocity²⁰⁷, whereby a data provider would only be allowed to access credit bureau data if it had provided its own data to the bureau.

In this context, customer data protection assumes great importance. India needs to move to a system where any person having a written authorisation of the borrower or entity on whom information is being sought, ought to be able to have access to the credit records in question, with only the truly necessary safeguards. Among potential users should include a prospective employer, landlord, or creditor, whether bank or non-bank. One of the best ways to limit unauthorised use of credit information is to develop systems, which record all queries for an individual’s report. Consumers can review this information if they think their data has been used in an inappropriate manner. This simple reporting tool can greatly help to detect misuse of the data by lenders and others who may request this information, as well as by staff of the credit reporting firm.

5. Incentives to “Clean Up”: Procedures should be in place to facilitate challenges to erroneous data. Consumers should be able to review their reports and identify reporting errors via the Internet and by phone. It is particularly important that consumers have access to reports when an adverse action has been taken. Clear procedures should be established in regulations specifying the steps in the dispute resolution process and the time that credit reporting firms have to verify and respond to complaints. International best practice is to establish time limits on the length of the credit history available to a potential lender. Economic research shows that recent credit payment record is of most relevance for predicting future default. Moreover, the fact that after a certain period of time, information, especially regarding defaults, will not be distributed to lenders creates additional incentives for the borrower to improve credit repayment behaviour and to ‘clean up’ the record. For example, records are available only for 5 years in Australia, Brazil, Germany, Ireland, Peru, and Spain, and for 7 years in the US, and Mexico.

6. Ownership of Customer Data: There is a need to develop a robust legal and regulatory framework around customer data generated in various transactions (payments and credit, digital and off-line), with the objective of customer ownership of their own transactions data and its use for signalling credit-worthiness. One approach to this is the architecture developed by companies like Idcubed.org who have an open source software that can be built as standard into every smartphone, tablet and wearable digital device that allow users to own their data through a highly secure ‘core identity’, which makes it harder for the state and companies such as Google and Facebook to scrape private data for their own ends. The availability of such a 'lock box' of data, authentication and 'priming' by the phone vendor, and employer requirement to 'log in' such authentication, provide alternative approaches to developing an architecture that enables use of data while respecting individual rights. In countries where a separate data protection authority or privacy commissioner exists, all data providers/users of a credit reporting system would fall under the purview of this authority, since the authority’s reach extends across sectors and thus overcomes some of challenges associated with overlapping regulatory boundaries. In Australia and New Zealand for instance, the respective Data Privacy Commissioner is responsible for regulating all entities participating in the credit reporting system and ensuring their compliance with the respective credit reporting legislations. Data protection is still largely in its infancy in India, although privacy is afforded through several acts, like the Information Technology Act of 2000, Information Technology (Reasonable Security Practices and Procedures and Sensitive Personal Information) Rules 2011 (IT RSPPSPI Rules), the Indian Constitution, to name a few. As the data landscape evolves in India, it becomes critical that a cogent framework for data protection is developed.

Warehousing Infrastructure:

Warehouse receipts finance can play an important role in smoothing income for farmers by providing liquidity against the harvested commodity. When delivering the product to an accredited warehouse, the farmer obtains a Warehouse Receipt that can be used as collateral for short-term borrowing to obtain working capital. The Rajan Committee (2009) estimated that this had the potential to inject over Rs. 1 lakh crore of agricultural credit, based on a projection that at any time, about 15–20 per cent of the annual agricultural produce is stored in warehouses.

With the promulgation of the Warehousing (Development and Regulation) Act, 2007, warehouse receipts have become negotiable instruments. They offer reliable form of collateral in the agricultural sector. Warehouse receipts can be in physical or electronic form and must be issued by registered warehouses, which will be accredited by the Warehousing Development and Regulatory Authority. The warehousing capacity available in India, in public, cooperative and private sector is about 11.24 crore MTs (Table 4.9.1²⁰⁸).

The storage space available in the country is not sufficient to cater to the procured stocks. As a result, substantial quantity of food grains (wheat) is stored in Covered and Plinth (CAP), an open storage system. For instance, in the Rabi Marketing Season (RMS) 2012-13, the Food Corporation of India (FCI) had procured 3.8 crore tonnes of wheat which was around 1 crore tonnes higher than the earlier record procurement.

Land Registries:

A critical barrier to financial access is the frequent inability of small and informal borrowers to secure loans with collateral, often a necessary condition for participation in formal credit markets. One frequently cited contributing factor is the fact that in much of the developing world a large percentage of both rural and urban property is untitled²⁰⁹. In Peru, introduction of land titling was associated with a 9-10 percentage point increase in loan approval rates from the public sector bank for housing construction materials, while there appears to be no effect on the loan approval rate of private sector lenders. However, conditional on receiving a loan, private sector interest rates are an average of 9 percentage points lower²¹⁰ for clients with a clear title to their land.

As per Indian law, while the compulsory registration of sale of land and property is mandated, the Registration Authority does not need to verify the history or ownership of the property from the seller. The deeds registry simply gives public notice of a transaction and does not imply any inference about the legal validity of the transaction or that the parties were legally entitled to carry out that transaction. Therefore, land registration does not mean registration of title of land; it is only a deed of transaction. This considerably weakens the protection offered by the system to property holders.

Considering the multiple gaps in the present system of land registration, the Government of India needs to lead an urgent initiative to establish the legal and institutional framework aimed at establishing a transparent and streamlined process for surveying, recognising, validating, registering, issuing and guaranteeing land titles. Several State Governments including Rajasthan and Andhra Pradesh have taken the lead in moving towards this approach.

It has been recommended by the Working Group on Partnerships for Land Title Implementation in Urban Management (PLATINUM)²¹¹, that the long-term solution to the issue of land titling will require State Governments to constitute Land Titling Authorities (LTA), which will undertake surveys of all lands and issue title certificates after undertaking the required enquiries and maintain registers of titles issued. It may also administer a system of indemnification of the titles issued against errors. For this purpose, the Authority could constitute specialised divisions working under its control and supervision.

In a guaranteed land title registration system, the title register shows the accurate legal situation of the land and therefore removes the need for retrospective examination of title and maintenance of deeds. This system is based on three basic principles:

1. Curtain: There is no requirement of further proof of ownership beyond the Title Certificate and the entry in the title registry of the LTA. The LTA will maintain an accurate and updated Title Register of all land through a system of Unique Property Identification Numbers (UPIN). Any conveyance registered with the Stamps and Registration Department is updated in the LTA’s Title Registry, and a fresh title is issued by the authority indicating all the changes in the title.

2. Mirror: Under this principle, the LTA ensures that what is shown on the title certificate reflects the ground reality. The LTA will survey, geo-locate and allocate UPINs to land parcels before the first title is issued.

3. Indemnity: The title is guaranteed by the State Government to the title holder and any legitimate counter claimant is indemnified against loss by the government. The land Tribunal will hear appeals against orders of the Authority and adjudicate on them.

This approach has also been endorsed by the Rajan Committee (2009) and offers a systematic long-term path to land titling. However, considering the fact that land is a subject in the State List of the Constitution, the establishment of a nation-wide system that guarantees land titles is bound to be a long process. As interim measures that pave the way for creation of guaranteed land titling system, the immediate steps outlined by the Rajan Committee (2009) were:

1. Full Computerisation and Integration of land records: Government measures to encourage computerisation could include:

1. Clarifying the policy and establishing clear criteria and accountability mechanisms for allocation of central funds on this;

2. Identifying and publicising best practices on technical and legal issues and promoting exchange and communication among technical staff across states; and

3. Prioritising full functional integration between records and registry.

2. Full cadastral mapping of land: An important problem is that existing cadastral survey records are largely limited only to agricultural land. The inhabited portions of villages, as well as towns and cities, have largely remained un-surveyed. Identification of urban property is, therefore, only by means of description of boundaries. This has to be remedied. A relatively low cost method to implement basic cadastral mapping is to combine satellite imagery with existing village maps and other readily available spatial products.

3. Reduction of stamp duty: All states receiving funds under JNNURM are required to provide a roadmap for reducing stamp duties to no more than 5 per cent in a definite timeframe.

4. Compulsory registration of all transactions: A large number of land transactions, especially in case of succession, do not need to be registered, partly because it is deemed unreasonable to charge stamp duty on these. Requiring that any change in the revenue records as a result of succession triggers a corresponding change in the land registry, without any payment in stamp duty, will go some way in ensuring registries are complete.

5. Elimination of restrictions on land markets: Widespread prohibition of land leasing is not consistent with efficient resource allocation. It raises the cost to rural–urban migration as villagers are unable to lease their land, and often have to leave a family member (typically the wife) behind to work the land. Lifting these restrictions can help the landless (or more efficient large land owners) get land from those who migrate, and allow those who currently lease land informally to formalise their transactions and thus obtain institutional credit and other benefits.

Weather stations:

One of the ways in which the risk of loss of crop yield on account of adverse weather has been mitigated in lending to farmers is through the use of a crop insurance protection linked to the crop loan. Weather based insurance products are designed to provide insurance protection against losses in crop yield resulting from adverse weather incidences. For example, the Weather Based Crop Insurance Scheme (WBCIS) provides payout against adverse rainfall incidence (both deficit and excess) during Kharif and adverse incidence in weather parameters like frost, heat, relative humidity, and un-seasonal rains during Rabi season. The weather index insurance market in India is the world’s largest, having transitioned from small-scale and scattered pilots to a large-scale weather based crop insurance program²¹². In the 2010-11 agricultural year over 90 lakh Indian farmers held WBCIS policies with premium volume of over Rs. 1,300 crore and total sum insured over Rs. 16,000 crore²¹³. For the purposes of compensation, the State Government notifies a Reference Unit Area (RUA), which is deemed to be a homogeneous unit. All the insured cultivators of a particular insured crop in that area are deemed at on par in the assessment of claims. Each RUA is linked to a Reference Weather Station (RWS), on the basis of which current weather data and the claims would be processed. Claims arise when there is a certain adverse deviation in actual weather parameter incidence in RUA as per the weather data measured at RWS. According to the India Meteorological Department (IMD), there are 701 departmental hydro-meteorological observatories in the country. Further, the IMD operates 125 Automatic Weather Stations and 500 automatic rain gauge stations²¹⁴. In addition, there are 8,579 non-departmental rain gauge stations²¹⁵.

In weather based insurance products, the insured holds the ‘basis risk’. Basis risk is defined as the variability in the relationship between the value of losses as measured by a weather index and the value of losses experienced on the farm by the farmer²¹⁶. Basis risk could occur due to spatial variation in weather variables (particularly where there are local micro-climates) as well as differences in management practices, soil quality or crop varieties. Because no individualised loss adjustment occurs with weather index insurance, the policy-holder must always carry the basis risk. Geographic basis risk, which is likely to arise from the distance between a farmer’s plot of land and the contractual weather station, can be reduced by an increase in granularity of weather stations²¹⁷. Paucity of weather stations in the country limits demand for weather insurance since the farmer knows that she holds the risk of receiving no claim payment despite having experienced a severe crop loss.

Further, “suggestive evidence of the role of liquidity constraints implies that policies should be designed to provide payouts as quickly as possible, especially during the monsoon when measured discounting of future cash flow is high. For example, payouts from a policy covering the first phase of the monsoon, if paid immediately, could be used to help fund crop replanting later in the monsoon. In practice to date, payouts are not made until after the end of the monsoon, in part because of delays in receiving certified rainfall data from government rainfall stations.”²¹⁸ Private entities like ICICI Lombard, for instance has begun using automated rain gauges that report rainfall immediately. This allows payouts to be made more quickly, and reduces basis risk by increasing the density of rainfall stations.

More geographically dispersed weather stations therefore reduce basis risk and increase the speed and accuracy of weather data. This could lead to a higher take-up of weather indexed insurance and higher rate of acceptance of weather-indexed contracts by reinsurers, translating into improved reinsurance rates. According to the Draft Guidelines for setting up Automatic Weather Stations (AWSs) and Automatic Rain Gauge (ARGs), crops within a radius of 5 km from a weather station can be insured with a reasonable reduction in basis risk, while anything beyond 5 km tend to increase the uncertainty. Assuming the 5 km radius, about 40,000 weather stations would be required at the national level to service weather insurance²¹⁹. Private entities like ICICI Lombard have started to form partnerships with National Collateral Management Services Limited (NCMSL) to install automated weather stations throughout India. Such initiatives need to be strongly encouraged by the Government. A wide network of weather stations also enables the growth of the domestic weather risk market and the placement ability of domestically underwritten weather contracts in the international market.

Registries for movable collateral and development of second-hand markets:

According to World Bank Enterprise Surveys performed in over 100 countries, collateral was required in over 75 per cent of all loans. Studies have noted that the availability of collateral remains a binding constraint on financing, and that this constraint binds harder in underdeveloped financial markets, where insufficient collateral remains one of the main reasons firms are rejected when they apply for bank credit²²⁰.

Movable assets, as opposed to fixed assets such as land or buildings, often account for most of the capital stock of private firms and comprise an especially large share for micro, small and medium-size enterprises. For example, in the developing world, it is estimated that 78 per cent of the capital stock of businesses is typically in movable assets such as machinery, equipment or receivables, and only 22 per cent is in immovable property²²¹. Hence, movable assets are the main type of collateral that firms, especially those in developing countries, can pledge to obtain bank financing. However, banks in developing countries are usually reluctant to accept movable assets as collateral due to the inadequate legal and regulatory environment in which banks and firms co-exist. In this context, movable assets become “dead capital”. For instance, studies have estimated that nearly 90 per cent of movable property that could serve as collateral for a loan in the United States would likely be unacceptable to a lender in an underdeveloped financial market²²².

While a sound legal and regulatory framework is essential to allow movable assets to be used as collateral, registries for movable assets fulfil two key functions:

1. to notify parties about the existence of a security interest in movable property (of existing liens) and

2. to establish the priority of creditors vis-a-vis third parties

Therefore, without a well-functioning registry for movable assets, even the best secured transactions laws could be ineffective. A recent study that explored the impact of introducing collateral registries for movable assets on firms' access to bank finance across 73 countries found that the introduction of registries for movable assets is associated with an increase in the likelihood that a firm has a bank loan, line of credit, or overdraft; a rise in the share of the firm’s working capital and fixed assets financed by banks; a reduction in the interest rates paid on loans; and an increase in the maturity of bank loans²²³. In 2007, China enacted the new Property Law which adopted important principles of modern secured transactions system and it opened up the scope for movables lending (including receivables), set up a clearer priority rule, and provided a better basis for enforcement. A modern security interest registry for account receivables was created in October 2007. It was China’s first nation-wide, central and internet-based filing system for secured transactions. The new receivables registry incorporated all the key features of a modern movable collateral registry such as on-line accessibility to the users and public, user accounts, notice based registry in which information limited to the creditor, debtor, loan amount and the description of assets, centralised information, and reasonable fee. As a result, the total number of commercial loans involving movable assets grew by 21 per cent per year in 2008-2010. Further, 84 per cent of loans secured with movables (receivables) went to SMEs. Almost none of the SMEs surveyed had any loans secured with receivables before the reform.

The Rajan Committee (2009) notes the need for establishing a comprehensive registration regime. India has a “system for registration of security interests created by companies incorporated under the Companies Act, 1956, but there is no registration process mandated for certain types of security interests created by individuals, partnership firms, cooperative societies and trusts. Additionally, for certain categories of movable assets, there are asset specific registration systems in operation, and registration is required in respect of charges created on such assets irrespective of who holds the asset.²²⁴” The Rajan Committee (2009) pointed to three possible paths to establishing a comprehensive registration regime:

1. Using the network of Registrar of Companies and the Ministry of Corporate Affairs’ “MCA 21” e-governance initiative.

2. Using credit information companies (or alternatively, the depository infrastructure—which serves the capital markets)

3. Using the SARFAESI Act provisions dealing with the registry of security interests: The Central Registry of Securitisation Asset Reconstruction and Security Interest of India (CERSAI) was set up under Section 20 of the SARFAESI Act. The object of setting up the Registration System under Chapter IV of the SARFAESI Act was to create a public data base about encumbrances created on properties to secure loans and advances given by the banks and financial institutions, as also transactions of securitisation or asset reconstruction undertaken pursuant to the provisions of the SARFAESI Act.

A liquid market for second-hand assets increases the ability of banks to do asset backed lending to small businesses as it serves to minimise loss given default in segments such as commercial vehicles finance, tractor finance and finance of other farm equipment. The second-hand asset market is often dominated by brokers and other intermediaries who have constrained access to finance. Initiatives that increase the transparency and liquidity of these markets are significant.

Recommendations:

4.42 There is a need to develop a robust legal and regulatory framework around customer data generated in various transactions (payments and credit, digital and off-line), with the objective of customer ownership of their own transactions data and its use, among others, for signalling credit-worthiness. RBI should constitute a Working Group comprising TRAI, CERC, and Credit Information Companies to develop a framework for sharing of data between telecom companies, electrical utilities, and credit bureaus. This framework should be in keeping with the FSLRC’s draft Indian Financial Code which recommends the creation of regulations on the collection, storage, modification and protection of personal information by financial services providers; and establishment of mechanisms to ensure that consumers have access to, and are given an effective opportunity to seek modifications to, their personal information. [Identical to Recommendation 4.8]

4.43 Universal reporting to credit bureaus should be mandated for all loans, both individual and SME, but in particular SHG loans, Kisan Credit Card, and General Credit Card. [Identical to Recommendation 4.2]

4.44 Equity investments by banks in complementary infrastructure within the purview of PSL guidelines, such as rural warehouses, market yards, godowns, silos, and NBFCs in low financial depth districts. These equity investments should be eligible for contribution to the overall priority sector lending targets. They should be permitted where debt already qualifies for PSL but with a multiplier of four, to reflect the higher risk and the illiquid character of these investments. The benefit must accrue as long as the equity investment is held by the Bank. This list of eligible equity investments may be varied from time-to-time. [Identical to Recommendation 4.32]

4.45 For the provision of food-credit, Food Corporation of India (FCI) and State Governments should be required to originate warehouse receipts and raise low-cost funds in the market against these receipts instead of being reliant only on bank credit. [Identical to Recommendation 4.36]

4.46 RBI needs to write to each of the State Governments expressing its support for the recommendations of both the PLATINUM Group and the Rajan Committee (2009) and urge them to implement those ideas by pointing out the potential benefits to the expansion of banking and financial activity in their respective states.

4.47 Banks and Financial Institutions should be required to verify the land records of their clients at the time of making loans and in those states where this is possible, to insist that transfers take place before a loan can be renewed for a second time.

4.48 Equity investments by banks in private companies engaged in the task of installing and operating weather stations, or in creating markets for second-hand assets should be eligible for PSL treatment. These investments should also get a multiplier of four, to reflect the higher risk and the illiquid character of these investments. [Also see Recommendation 4.44]

Chapter 4.10
Recommendations Regarding Credit

4.1 In order to encourage banks to actively manage their exposures to various sectors, including priority sectors, a number of steps would have to be taken:

1. Banks must be required to disclose their concentration levels to each segment in their financial statements.

2. Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation. [Identical to Recommendations 4.10 and 4.30]

3. RBI must represent to the MoF to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment pointing out the role it would play in ensuring efficient risk transmission. [Identical to Recommendations 4.11 and 4.38]

4. Banks must be permitted to purchase portfolio level protection against all forms of rainfall and commodity price risks, including through the use of financial futures and options bought either within India or globally.

4.2 Universal reporting to credit bureaus should be mandated for all loans, both individual and SME, but in particular SHG loans, Kisan Credit Card, and General Credit Card. [Identical to Recommendation 4.43]

4.3 In view of the fact that banks may choose to focus their priority sector strategies on different customer segments and asset classes, it is recommended that the regulator provide specific guidance on differential provisioning norms at the level of each asset class. A bank’s overall NPA Coverage Ratio would therefore be a function of its overall portfolio asset mix. On standard assets, provisioning levels as well as asset classification guidelines specified by RBI would need to reflect the underlying level of riskiness of each asset class (combination of customer segment, product design, and collateral) and not be uniform across all the asset classes. Additionally, different customer-asset combinations behave very differently from each other and it is recommended that the regulator mandate NPA recognition rules at the level of each asset-class and require that all banks conform to these mandates. [Identical to Recommendation 4.21 for NBFCs]

4.4 All banks should be required to publicly disclose the results of their stress tests both at an overall balance sheet level as well as at a segmental level at least annually. [Identical to Recommendation 4.22 for NBFCs]

4.5 From the perspective of Stability that entails sustainable pricing, banks must be required to freely price farm loans based on their risk models and any subventions and waivers deemed necessary by the government should be transferred directly to the farmers and not through interest subsidies or loan waivers. The permission to price farm loans below the base rate should be withdrawn. [Also see Recommendation 4.34]

4.6 Banks are already permitted to set up specialised subsidiaries upon getting specific approvals from the RBI. However, no approvals have been granted; potentially due to concerns around circumvention of branch licensing guidelines. In light of the recent relaxation of branch licensing guidelines and the capability to carry out consolidated supervision, the requirement of prior approvals may be removed for the purpose of creating dedicated subsidiaries for financial inclusion.

4.7 The decision on the manner in which risk sharing and credit approval arrangements need to be structured between banks and their agents can be left to the judgment of banks. Outsourcing guidelines should be amended to permit this.

4.8 There is a need to develop a robust legal and regulatory framework around customer data generated in various transactions (credit and payments, digital and off-line), with the objective of customer ownership of their own transactions data and its use, among others, for signalling credit-worthiness. RBI should constitute a Working Group comprising TRAI, CERC, and Credit Information Companies to develop a framework for sharing of data between telecom companies, electrical utilities, and credit bureaus. This framework should be in keeping with the FSLRC’s draft Indian Financial Code which recommends the creation of regulations on the collection, storage, modification and protection of personal information by financial services providers; and establishment of mechanisms to ensure that consumers have access to, and are given an effective opportunity to seek modifications to, their personal information. [Identical to Recommendation 4.42]

4.9 The stipulation that the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the purchasing bank plus 8 per cent per annum should be removed. [Identical to Recommendation 4.37]

4.10 Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation. [Identical to Recommendations 4.1(b) and 4.30]

4.11 RBI should represent to the Government of India to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment so as to create pathways for Wholesale Banks to provide liquidity to other Banks and Financial Institutions directly originating assets in priority sectors. [Identical to Recommendations 4.1(c) and 4.38]

4.12 Reorient the focus of NABARD, CGTMSE, SIDBI, and NHB to be market makers and providers of risk-based credit enhancements rather than providers of direct finance, automatic refinance, or automatic credit guarantees for National Banks.

4.13 Regional Banks continue to have a strong appeal for inclusion but low demonstrated stability in the Indian context. Robust solutions are required vis-à-vis regulation, supervision, risk management, and governance of the existing Regional Banks before any new ones are created.

4.14 In a manner similar to National Banks, for Regional Banks as well, refinance by NABARD or credit guarantee support by CGTMSE should be designed as risk-based guarantees and not available automatically. [Similar to Recommendation 4.12]

4.15 While the Risk Based Supervision process has been designed for the larger and more complex institutions, a similar effort could be conceived of for the Regional Banks allowing supervisors to direct scarce on-site supervision resources to higher-risk institutions. The help of commercial ratings agencies could also be taken to formally rate these Regional Banks and provide the ratings to the depositors on a regular basis – thus including them more directly into the risk-containment process.

4.16 Since DICGC offers deposit insurance to these banks, the Agreement between DICGC and the bank provides an additional opportunity to both ensure that the bank is run well on an on-going basis and resolved quickly in the event that it turns insolvent. Such an Agreement would provide for risk-based pricing of deposit insurance; the right to carry out both on-site and off-site inspections; and to take possession of the bank in the event that it goes into liquidation.

4.17 A State Finance Regulatory Commission (SFRC) could be created into which all the existing State Government-level regulators could be merged and functions like the regulation of NGO-MFIs and local Money Services Business could be added on. In some states, the SFRC could be created by upgrading existing Institutional Finance Cells. There is also value in bringing the regulatory function close to the enforcement function under the Economic Offences Wing (EOW), so as to ensure that they are working closely together. The RBI must be closely involved over a longer time frame in training the commissioners and licensing and accrediting the Commission itself. The local regional directors of the RBI could, for example, be ex-officio Chairs of the Commission’s Board of Governors while the Commissioner could be a senior State level appointee drawn from the local Banking community. The District Magistrates would also play an important role in their respective districts.

4.18 The Committee recognises that a partial convergence of NBFC and Bank regulations may be desirable. It recommends the following:

4.19 Multiple NBFC definitions should be consolidated into two categories: a distinct category for Core Investment Companies (CIC) and another category for all other NBFCs. Benefits that were previously available to specific NBFC types, such as tax benefits, bank limits, and priority sector benefits should continue to be available even after consolidation, on a pro-rata asset basis.

4.20 The Committee recommends addressing wholesale funding constraints faced by NBFCs in a systematic manner. The following are the specific recommendations in this regard:

1. A clear framework to be developed by RBI and SEBI for Qualified Institutional Buyers and Accredited Individual Investors who may participate in debt market issuances of NBFCs.

2. Benefit of ‘shelf prospectus’ should be available for one year to all issuers including NBFCs.

3. Permit ECB in Rupees for all institutions.

4. For ECB not in Rupees, eligibility should be linked to size and capacity to absorb foreign exchange risk rather than specific NBFC categories.

5. The nature of activity, rather than institution type, must be made the criterion for availing refinance from NABARD, NHB, SIDBI and credit guarantee facilities.

6. Current capitalisation slabs on foreign equity funding should be relaxed and money laundering concerns should be mitigated by instituting additional reporting requirements on Banks/Authorised Dealers (AD).

4.21 In a manner similar to banks, different customer-asset combinations behave very differently from each other and it is recommended that the regulator specify NPA recognition and provisioning rules, including for standard assets, at the level of each asset-class and require that all NBFCs conform to these mandates. [Identical to Recommendation 4.3 for National Banks]

4.22 Require all NBFCs to have better on-going risk measures. These include disclosure of their stress test results both at an overall balance sheet level as well as at a segmental level at least annually. All NBFCs must adopt core banking systems so that this can enable better off-site supervision. [Identical to Recommendation 4.4 for National Banks]

4.23 Enable better benchmarking by requiring all NBFC-MFIs to disclose their operating costs (direct and indirect) of a mature branch to the RBI or MFIN once it becomes an SRO.

4.24 The regulatory focus must be on total indebtedness of the small borrower in relation to their debt-servicing capacity and not just indebtedness per se or merely from NBFC-MFIs. Keeping this in mind, the total borrowing limit for the small borrower segment may be increased immediately to Rs. 100,000 across all lenders, including bank-lending to this segment. In order to implement this, all lenders to this segment will need to be mandated to report to the credit bureau as has been the case with NBFC-MFIs. If total indebtedness is being tracked adequately, the stipulation of a maximum number of lenders appears redundant and can be gradually removed as this would also help in creating intensified price competition.

4.25 All policy biases against consumption finance need to be removed. An example of this is restricting the proportion of consumption finance that is permitted for NBFC-MFIs.

4.26 In order to enable the gradual transition of eligible and interested NBFCs to Wholesale Consumer Banks or Wholesale Investment Banks or National Banks, the Committee recommends a re-examination of PSL definitions [also see Recommendation 4.41], creating an active market for PSL assets, assessment of the relevance of SLR in light of capital adequacy norms, and application of CRR on time liabilities.

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

1. Given that their primary role is lending and not the provision of retail deposit services, they will only be permitted to accept deposits larger than Rs. 5 crore.

2. Since they could expect to borrow large amounts from other banks, net liabilities from the banking system will be permitted to be deducted from their NDTL computation for the purposes of ascertaining their SLR obligations on par with the treatment currently given for CRR.

3. Since other banks are expected to lend large amounts to Wholesale Banks, those other banks will be permitted to deduct their net assets to the banking system from the computation of their ANBC (the amounts on which PSL requirements are to be applied).

4. In view of the fact that they will not take retail deposits, the minimum entry capital requirement for them will be Rs. 50 crore compared to the Rs. 500 crore required for full-service SCBs.

5. If the institution has fewer than twenty branches through which it operates, it will not be required to meet the 25 per cent branching requirement. Institutions with twenty or fewer branches could be referred to as Wholesale Investment Banks while those with a larger branch network could be referred to as Wholesale Consumer Banks.

6. Wholesale Consumer Banks should be permitted to act as BCs for other full service Banks.

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.

4.28 All loans given to landless labourers and small and marginal farmers should be counted as a part of Direct Agriculture and not merely the wages component of a loan given to a farmer for financing her agricultural production.

4.29 Investment by banks in bonds of institutions must qualify for PSL where wholesale lending to the same institutions already qualifies under PSL.

4.30 Credit facilities documented as bonds or Pass-Through Certificates (PTC), whether originated directly or purchased in the secondary markets should be permitted to be held in the “banking book” of a bank based on declared intent and not merely based on source or legal documentation. [Identical to Recommendations 4.1(b) and 4.10]

4.31 Investment by banks in the form of non-fund based limits (such as guarantees) should qualify for PSL to the extent of the credit equivalent amount of the off-balance sheet facility where loans to these categories qualify for PSL. ANBC should also be adjusted to include such PSL-linked, non-fund based limits.

4.32 Equity investments by banks in complementary infrastructure within the purview of PSL guidelines, such as rural warehouses, market yards, godowns, silos, and NBFCs in low financial depth districts. These equity investments should be eligible for contribution to the overall priority sector lending targets. They should be permitted where debt already qualifies for PSL but with a multiplier of four, to reflect the higher risk and the illiquid character of these investments. The benefit must accrue as long as the equity investment is held by the Bank. This list of eligible equity investments may be varied from time-to-time. [Identical to Recommendation 4.44]

4.33 PSL targets should be applicable on the last reporting Friday during the last month of each quarter in exactly the same manner as it is currently applicable in the month of March, so as to ensure more timely and continuous credit flow into priority sectors. In order to ensure administrative ease, requirements such as investment into RIDF can continue to be levied on an annual basis and computed on the basis of the average of the quarterly requirements.

4.34 If the government does desire to provide relief in any form to the small farmer, it would be best carried out as a direct benefit transfer (DBT) to the bank account of the farmer and not through the mechanism of either interest subvention or debt waiver. This would ensure that the banking system is able to price loans in a sustainable manner and also protect credit discipline amongst its borrowers. Adding a universal requirement to report all defaults to credit bureaus would ensure that the borrower also builds a strong interest in protecting his credit history, even if he is a recipient of DBTs. [Similar to Recommendations 4.2 and 4.43]

4.35 In order to guard against large scale defaults resulting from catastrophic events, banks should be permitted to work closely with insurance companies to purchase bank-wide portfolio level insurance against events such as large scale rainfall failure on a regional or national basis, instead of having an expectation that relief would be provided from national or state budgets.

4.36 For the provision of food-credit, Food Corporation of India (FCI) and State Governments should be required to originate warehouse receipts and raise low-cost funds in the market against these receipts instead of being reliant only on bank credit. [Identical to Recommendation 4.46]

4.37 The stipulation that the all-inclusive interest charged to the ultimate borrower by the originating entity should not exceed the Base Rate of the purchasing bank plus 8 per cent per annum should be removed. [Identical to Recommendation 4.9]

4.38 The RBI should represent to the Government of India to restore the tax-free status of securitisation SPVs as pass-through vehicles for tax treatment so as to create pathways for Wholesale Banks to provide liquidity to other Banks and Financial Institutions directly originating assets in priority sectors. [Identical to Recommendation 4.11]

4.39 While a market that trades PSL assets will be of critical importance, regulation should additionally enable the use of risk-free PSL Certificates as a means to achieving PSL compliance amongst banks that wish to do so.

4.40 In order to enable greater regional and sectoral specialisation among Banks, the Committee recommends that the RBI revise the PSL targets and require banks to meet an Adjusted PSL target of 50 per cent against the current requirement of 40 per cent. Districts and sectors are weighted based on the difficulty in lending to them, and a Bank lending to a difficult sector in a difficult to reach district can benefit from a multiplier value based on the specific sector and district. Every sector-district combination has a weight associated with it and the Bank will have to reach an adjusted PSL value of 50% taking these weightages into account.

4.41 The Committee recommends that RBI seriously examine moving to a new framework in which two parameters: District level credit depth, and sector and sub-sector level credit depth be used to determine the sector, sub-sector, and regional weights which are published every three years. Using these weights banks would be required to reach an Adjusted PSL target of 150 per cent of ANBC.

4.42 There is a need to develop a robust legal and regulatory framework around customer data generated in various transactions (payments and credit, digital and off-line), with the objective of customer ownership of their own transactions data and its use, among others, for signalling credit-worthiness. RBI should constitute a Working Group comprising TRAI, CERC, and Credit Information Companies to develop a framework for sharing of data between telecom companies, electrical utilities, and credit bureaus. This framework should be in keeping with the FSLRC’s draft Indian Financial Code which recommends the creation of regulations on the collection, storage, modification and protection of personal information by financial services providers; and establishment of mechanisms to ensure that consumers have access to, and are given an effective opportunity to seek modifications to, their personal information. [Identical to Recommendation 4.8]

4.43 Universal reporting to credit bureaus should be mandated for all loans, both individual and SME, but in particular SHG loans, Kisan Credit Card, and General Credit Card. [Identical to Recommendation 4.2]

4.44 Equity investments by banks in complementary infrastructure within the purview of PSL guidelines, such as rural warehouses, market yards, godowns, silos, and NBFCs in low financial depth districts. These equity investments should be eligible for contribution to the overall priority sector lending targets. They should be permitted where debt already qualifies for PSL but with a multiplier of four, to reflect the higher risk and the illiquid character of these investments. The benefit must accrue as long as the equity investment is held by the Bank. This list of eligible equity investments may be varied from time-to-time. [Identical to Recommendation 4.32]

4.45 For the provision of food-credit, Food Corporation of India (FCI) and State Governments should be required to originate warehouse receipts and raise low-cost funds in the market against these receipts instead of being reliant only on bank credit. [Identical to Recommendation 4.36]

4.46 RBI needs to write to each of the State Governments expressing its support for the recommendations of both the PLATINUM Group and the Rajan Committee (2009) and urge them to implement those ideas by pointing out the potential benefits to the expansion of banking and financial activity in their respective states.

4.47 Banks and Financial Institutions should be required to verify the land records of their clients at the time of making loans and in those states where this is possible, to insist that transfers take place before a loan can be renewed for a second time.

4.48 Equity investments by banks in private companies engaged in the task of installing and operating weather stations, or in creating markets for second-hand assets should be eligible for PSL treatment. These investments should also get a multiplier of four, to reflect the higher risk and the illiquid character of these investments. [Also see Recommendation 4.44]

¹²³ The FSLRC has proposed a separate regulatory framework for SIFIs. This is likely to have implications for all Government-owned Banks and the larger Private Sector Banks.

¹²⁴ Citigroup (2013).

¹²⁵ Source: Financial Stability Report, Issue Number 7, Table 2.4, Page 32, Reserve Bank of India, June 2013.

¹²⁶ Source: Page 2, Financial Stability Report, Issue Number 7, Reserve Bank of India, June 2013.

¹²⁷ Source: Handbook of Statistics of the Indian Economy, 2012-13.

¹²⁸ Pradhan (2013).

¹²⁹ Radhakrishna (2007).

¹³⁰ Thorat (2011).

¹³¹ Used on a second-loss basis, Rs. 3,073 crore could have covered over Rs. 30,000 crore of SHG portfolio.

¹³² Puhazhendhi (2013).

¹³³ Submission from Jana Foundation to the Committee.

¹³⁴ Submission from Bank of Baroda to the Committee.

¹³⁵ Source: Submission from CRISIL to the Committee.

¹³⁶ Adjusted GNPA equals GNPAs in non-priority sector including 30% of restructured standard assets (excluding state power utilities)/advances outstanding in the non-priority sectors.

¹³⁷ Based on submission from Bank of Baroda to the Committee and on the findings of the Committee on Financial Inclusion, 2008, headed by C.Rangarajan (Annexure IV, page 129). The submission by Bank of Baroda concludes that rural Bank branches become cost-competitive at loans above Rs.100,000.

¹³⁸ Submission from Axis Bank to the Committee.

¹³⁹ Berger et al (2004), Page 11.

¹⁴⁰ Based on submissions from Bank of Baroda and CAFRAL to the Committee.

¹⁴¹ The Banker, June 2013 issue, Pages 16 - 21.

¹⁴² Agarwal & Hauswald (2010).

¹⁴³ Provisioning for NPAs is lower in India compared to many countries including France, US, Russia, China, Thailand, Brazil, and Mexico. This coupled with higher gross NPA levels, means that the Indian banking system’s net NPA levels as a percentage of capital are weaker than that in most developing economies.

¹⁴⁴ Source: MFIN’s Micrometer data for all non-Andhra Pradesh MFIs FY 2011-12 and 2012-13. The operation of AP based MFIs were severely curtailed following the AP State Government action in 2010.

¹⁴⁵ Submission from CAFRAL to the Committee.

¹⁴⁶ Of the 57 billion Euros in subsidy offered to farmers in 2010, there was no interest subsidy or loan waiver component and about 70% was handed out as direct cash subsidy.

¹⁴⁷ USA paid out direct subsidies of about $10 billion in 2010. It was careful to ensure that these subsidies were decoupled from actual production related activities so as to ensure that there is minimum distortion in market prices and production decisions of the farmers, see: http://www.ers.usda.gov/topics/farm-economy/farm-commodity-policy/government-payments-the-farm-sector.aspx#.UnPAEnBmim4

¹⁴⁸ Pendley et al (2007) find that residential mortgages originated in 2006 and 2007 with high FICO scores under-performed substantially and that the credit scoring model was susceptible to fraud, not picking up on obviously suspicious items on credit reports. Rhyne (2001) finds that complete dependence on scoring for credit risk assessment can lead to adverse outcomes such as in the case of a finance company in Bolivia which went bankrupt after introducing a credit appraisal system for microcredit borrowers that was based only on a credit scoring model that did not take subjective risk factors into account.

¹⁴⁹ Wade & Magleby-Lambert (2008).

¹⁵⁰ Submission from CAFRAL to the Committee.

¹⁵¹ Source: Master Circular on Para-banking Activities, July 2013, see:
http://www.rbi.org.in/scripts/BS_ViewMasCirculardetails.aspx?id=8196#2

¹⁵² Guidelines on Managing Risks and Code of Conduct in Outsourcing of Financial Services by banks, November 2006,
see:http://www.rbi.org.in/commonman/English/Scripts/Notification.aspx?Id=40

¹⁵³ Mester (1997) provides a good overview of the benefits of credit scoring.

¹⁵⁴ Demyanyk (2008) found that during the crisis credit scores did not predict either the true risk of default of a subprime mortgage loan or the crisis as a whole.

¹⁵⁵ Peterson & Rajan (2002).

¹⁵⁶ “Citigroup to Expand Credit-Card Business in India”, Wall Street Journal, May 25, 2012,
see: http://online.wsj.com/news/articles/SB10001424052702304707604577426011248201708

¹⁵⁷ “Credit bureau of MFIs brings details of small borrowers to the fore”, Business Standard, October 15, 2013, see:
http://www.business-standard.com/article/finance/credit-bureau-of-mfis-brings-details-of-small-borrowers-to-the-fore-113101500111_1.html

¹⁵⁸ Source: Chapter 19, Draft Indian Financial Code, Report of the FSLRC

¹⁵⁹ RBI Master Circular on Priority Sector Lending: Targets and Classification, July 2013,
see: http://rbidocs.rbi.org.in/rdocs/notification/PDFs/107010713PSLFL.pdf

¹⁶⁰ Mukherjee (2013).

¹⁶¹ Mishkin (2013).

¹⁶² In Maharashtra there are another set of cooperative financial institutions which are called 'Pata Sansthas' or 'Pata Pheris'. There are entirely unregulated, very small, completely self-funded, but highly effective and well managed. There are about 40,000 of them in Maharashtra, have deposits of over Rs 70,000 crore and give loans to small businesses in rural and urban areas. Almost all of the 50,000 autorickshaws plying in Mumbai have been financed by them.

¹⁶³ Source: Handbook of Statistics on the Indian Economy, RBI, 2012-13.

¹⁶⁴ Gilbert et al (2013).

¹⁶⁵ Source: “The Importance of Effective Corporate Governance”, by Kevin Moore, Federal Reserve Bank of Kansas City, Community Banking Connections, Fourth Quarter 2012., see:
http://www.communitybankingconnections.org/articles/2012/Q4/Importance-of-Effective-Corporate-Governance.cfm

¹⁶⁶ Based on the submission from ACCION to the Committee.

¹⁶⁷ Source: RBI has directed all UCBs to complete CBS implementation by December 31, 2013. NABARD has also initiated a CBS project for all cooperatives.

¹⁶⁸ Clarke (2010).

¹⁶⁹ Rajan (2009), Page 8.

¹⁷⁰ The Pennsylvania Department of Banking and Securities for example regulates more than 225 state-chartered banks, savings associations, trusts and credit unions; licenses and registers about 14,000 non-bank lenders, including mortgage brokers and lenders, auto sales finance companies, debt management companies, check cashers, pawnbrokers and money transmitters; enforces compliance with state and federal laws for non-bank lenders so that consumer and business borrowers will have confidence in the fairness of their transactions; and licenses and registers more than 190,000 securities companies and professionals doing business with Pennsylvania residents, including agents, broker-dealers, investment advisors and notice filers, and investment advisor representatives.

¹⁷¹ Based on discussions with Professor Robert DeYoung of the University of Kansas School Of Business and the lead author of the 2004 Chicago-Fed study on “The Past, Present, and Probable Future for Community Banks”.

¹⁷² Source: Report on Trend and Progress of Banking in India 2012-13, RBI.

¹⁷³ Acharya (2013).

¹⁷⁴ Submission from CRISIL to the Committee.

¹⁷⁵ European Commission (2012).

¹⁷⁶ Kumar et al (1997), Page 63.

¹⁷⁷ Rajan (2009), Page 13.

¹⁷⁸ Acharya and Oncu (2013).

¹⁷⁹ Vide their circular SEBI/CFD/DIL/DIP/12/2004/8/4 dated 8th April 2004.

¹⁸⁰ Source: Highlights 2012-13, NABARD, see: www.nabard.org

¹⁸¹ Member Lending Institutions under CGTMSE are all scheduled commercial banks, specific RRBs, NSIC, NEDFi and SIDBI,
see: http://www.cgtsi.org.in/Eligibility_criteria.aspx

¹⁸² Sane & Thomas (2013).

¹⁸³ The reported overdues of 10 large MFIs in the State amounted to Rs.3174 crore by the end of March 2011. This would have adversely affected the 2.7 crore borrowers in Andhra Pradesh who relied on microfinance institutions operating in the State. Studies find that the crisis resulted in a drop in average household expenditure in Andhra Pradesh by 19% (Sane and Thomas (2013)). The crisis also contaminated the SHG portfolio of banks in the State, where at the end of July 2011, about 17% of linked groups had defaulted on their repayments, constituting a portfolio at risk of 16.7% (Puhazhendhi (2013)). A study by MicroSave that looked at the effect on AP clients from the cessation of MFI activity found that more than 70% of respondents had to borrow from other sources and more than 10% reported distress sale of assets.

¹⁸⁴ Submission from MFIN to the Committee.

¹⁸⁵ Only loans given to households falling within specific income levels can qualify – For rural households, the annual income must not exceed Rs.60,000, and for urban households, the income must not exceed Rs.120,000. This criterion completely disregards the limitations of traditional lending mechanisms in overcoming challenges that the borrower segment has, namely, the difficulty in estimating their incomes, the lack of reliable credit history records, and the difficulty in collateralising their assets.

¹⁸⁶ Loan amount cannot exceed Rs. 35,000 in the first cycle and Rs. 50,000 in subsequent cycles, and the total indebtedness of the borrower must not exceed Rs. 50,000. Tenure of the loan cannot be less than 24 months for loans in excess of Rs.15,000.

¹⁸⁷ A borrower cannot be a member of more than one SHG/JLG. No more than two MFIs can lend to a borrower. This essentially places a restriction on choice of the borrower, who if having availed a JLG loan, will now be forced to build a savings history for 6 months before he or she can decide to carry out the activity for which she needs financing support. The decision on whether a borrower has the capacity to repay the loan she is seeking or not, is best left to the borrower who is taking the loan, and the institution that is willing to underwrite such a loan with the help of available credit bureau information.

¹⁸⁸ Loans given for income generation should constitute at least 70% of the total loans of the MFI and the remaining 30% can be for other purposes such as housing repairs, education, medical and other emergencies. The distinction between income-generation and consumption is a superficial one and questions the credibility of the borrower in her ability to make rational choices for her household. Improving one’s living conditions, investing in education or health, or even purchasing a two-wheeler can have welfare-enhancing prospects for the household and the decision on how to prioritise on expenditures is best left to the borrower household.

¹⁸⁹ Rajan (2009), Proposal 6.

¹⁹⁰ Statement by Dr. Raghuram Rajan on taking office on September 4, 2013.

¹⁹¹ A number of countries like Japan, South Korea, Columbia, Costa Rica, Brazil and China have used directed lending to achieve a variety of goals. For instance, studies like Calomiris et al (1994) and Horiuchi & Sui (1992) note the influence of directed lending in the growth of Japanese industries. Similarly, Amsden (1989) argues that subsidy components of directed lending were used in South Korea as a disciplining mechanism to enforce strict performance standards on firms.

¹⁹² Source: Master Circular on Priority Sector Lending, July 2013. RBI, see:
http://rbidocs.rbi.org.in/rdocs/notification/PDFs/107010713PSLFL.pdf

¹⁹³ Ibid.

¹⁹⁴ Estimation Methodology: Data from the IFMR Trust shows that there are 1713 small and marginal farmer households (owning land for cultivation up to 5 acres) in the village of Karambayam, Tamil Nadu. The total acreage amounts to 2926 acres. Assuming that all farmers in the village cultivate paddy on their lands for 2 crop seasons in a year, and given that the scale of finance for paddy for one crop season is about Rs.10000/acre (which is a conservative reflection of the costs of cultivation) in the region, and assuming that 100% of the cultivation cost is financed with credit, there is a demand for Rs.1.95 crore by small and marginal farmers at a Gram Panchayat level (since the Karambayam area covers a population that is spread across approximately 3 Panchayats). If this were applied to all of the 2.46 lakh Panchayats in India, the total credit demand by small and marginal farmers can reasonably be estimated at Rs.4.8 lakh crore. Further, an analysis of the per acre cost of cultivation for some of the principal crops in 5 major producing states for each crop, indicate that it ranges from about Rs.8000 for arhar, Rs.13000 for paddy, to Rs.29000 for sugarcane (www.data.gov.in).

¹⁹⁵ Submission from Jana Foundation to the Committee.

¹⁹⁶ Source: RBI Handbook of Statistics on Indian Economy 2012-13.

¹⁹⁷ Source: All India Debt and Investment Survey (2002) and Sarangi (2010).

¹⁹⁸ Sarangi (2010).

¹⁹⁹ RBI Master Circular on Interest Rates on Advances, July 2013, see:
 http://rbidocs.rbi.org.in/rdocs/notification/PDFs/73MCIRA01072013F.pdf

²⁰⁰ Speaking at the Fourth Annual Credit Information Conference in Chennai, organised in December 2013, R K Dubey, chairman and managing director of Canara Bank said, 'The way our credit on retail is working, we are hopeful that the gross NPA rate to be below three per cent'. He said that all the 4,500 branches of Canara Bank are using the data of CIBIL while taking a decision on the loan applications and this has helped the Bank to avoid lending to customers with higher risk. Source: http://www.business-standard.com/article/finance/canara-bank-reduced-retail-gross-npa-using-cibil-information-says-dubey-113120400940_1.html.

²⁰¹ Source: RBI Reports on Trends and Progress of Banking in India, 2011-12.

²⁰² Adjusted GNPA equals GNPAs in non-priority sector including 30% of restructured standard assets (excluding state power utilities)/advances outstanding in the non-priority sectors.

²⁰³ Thorat (2011) found that there has “.. been a distinct shift in the preference of commercial banks towards indirect finance (IF) vis-à-vis direct finance (DF) to agriculture. Of the total credit outstanding to agriculture, the share of indirect finance was 16% in 2000 which increased to 28% in 2006 and stood at 24% in 2010.” The report also finds that credit deepening in the indirect finance category has been much more pronounced than the direct finance category.

²⁰⁴ Ibid.

²⁰⁵ CRISIL Inclusix is a comprehensive index launched by CRISIL for measuring the progress of financial inclusion in the country, down to the district-level. CRISIL Inclusix, whose methodology is similar to other global indices, such as UNDP's Human Development Index, measures financial inclusion on the three critical parameters of basic banking services - branch penetration, deposit penetration, and credit penetration. The index uses parameters that focus only on the 'number of people' whose lives have been touched by various financial services, rather than on the 'amounts' deposited or loaned. The As of now, the CRISIL Inclusix index covers 632 districts over a three-year timeframe (2009-2011).

²⁰⁶ Estimation Methodology: GDDP numbers have been calculated as explained in End Note 30. In this absence of district level GDP data, as is the case for all districts of Gujarat and some other districts in the country, district level weights are calculated by using state level GDP numbers and adjusting them using the CRISIL Inclusix to estimate district GDP. In the case of some Union Territories, when there is no GDP data available, those districts have been left empty.

²⁰⁷ Most credit bureaus including Credit Information Bureau (India) Limited (CIBIL) work on the principle of reciprocity, which means that only members who had submitted all their credit data could access credit information reports.

²⁰⁸ Source: Annual Report 2011-12, Warehousing Development and Regulatory Authority.

²⁰⁹ Holden (1997).

²¹⁰ Field & Toreno (1994).

²¹¹ India Urban Space Foundation (2011).

²¹² Clarke et al (2012).

²¹³ Collier et al (2009).

²¹⁴ Source: IMD Pune, see: http://www.imdpune.gov.in

²¹⁵ Source: IMD, see: http://www.imd.gov.in/doc/obs_surface.htm

²¹⁶ Collier et al (2009).

²¹⁷ Clarke et al (2012).

²¹⁸ Cole et al (2009).

²¹⁹ Source: Draft Report/Guidelines for setting up Automatic Weather Stations (AWSs) and Automatic Rain Gauge (ARGs) & their accreditation, standardisation, validation and quality management of weather data for Implementation of Weather Based Crop Insurance Scheme (WBCIS)

²²⁰ Liberti & Mian (2010) and Fleisig (2006).

²²¹ Alvarez de la Campa (2011).

²²² Safavian at al. (2006).

²²³ Love et al (2013).

²²⁴ Rajan (2009), Page 164.