I had an opportunity to participate in the excellent conference organised by the Stanford Centre for International Development (SCID) on Indian Economic Policy as a discussant for a presentation by Dr. Rakesh Mohan. My comments drew heavily from a forthcoming paper on “Modernising Indian Banking” by my colleague Deepti George. The transcript is below, would welcome thoughts/feedback:
- I want to start by thanking SCID for inviting me to this conference and for an opportunity to discuss this excellent, data-rich presentation on India’s financial sector reforms by Dr. Rakesh Mohan and Prof. Partha Ray. I will focus my comments on the banking sector and financial inclusion aspects.
- From all accounts, it appears that the traditional banking business model will face some threat going forward starting with increased competition on the liabilities side. Low cost deposits are a significant factor contributing to the risk absorption capacity of banks’ balance sheets, a hidden capital buffer of sorts. Raghuram Rajan in his Committee’s Report of 2009 calls this the ‘grand bargain’[i]: where cheap deposits were available to banks in an environment marked by low competitive intensity, in exchange for financing Governments and priority sectors of the Government. This is unravelling with significantly more competition expected for the CASA business. This competition is particularly likely to be sharp from the newly licensed category of Payments Banks whose sole focus will be on deposits and payments. Many of these new banks are expected to exploit the adjacencies with their telecom businesses and significantly increase the outreach of the banking sector and ease of depositing small amounts, frequently. Even over the past few years, data shows that there has been a “flight of CASA” to a few banks that are perceived as being strong. This “sorting” is happening even within Public Sector Banks and represents an important nuance to the authors’ observation on reversal of convergence.
- As the presentation notes, a number of measures have been launched by the RBI and the Government including the “Indradhanush” package announced in August 2015[ii]. However, most of these measures stop short of addressing the root cause of troubles in the banking sector. NPAs reflect the outcome of decisions made several years ago by banks and while the current debates on provisioning levels are important from the perspective of assessing the fair value of banks, it does not examine the conditions under which these assets were originated and monitored and indeed, how we ensure that these issues are not recurrent themes in the Indian banking sector. A lot is attributed to ownership issues of banks. However, ownership notwithstanding, there are several pragmatic measures that can improve management of banks and oversight by the Board – these have not received much attention. Specifically, adoption of three building blocks: Risk-based pricing of loans, Activity-based costing and Matched Funds Transfer Pricing will ensure more rational pricing of assets by banks even as the ownership issues get resolved.
- In addition to actions that can be taken by bank management, the supervisory regime has an important role to play to enable banks to reveal more information on an on-going basis about the true performance of the bank as well as create an environment where more capabilities are getting built within banks. Over the past few years, there has been a definitive shift towards risk-based supervisory approaches. This has been driven by Basel II requirements on banking regulators to undertake the Supervisory Review and Evaluation Process (SREP) of supervised banks, which includes the review and evaluation of the bank’s Internal Capital Adequacy Assessment Plan (ICAAP)[iii], conducting an independent assessment of the bank’s risk profile, and if necessary, taking appropriate prudential measures and other supervisory actions such as based on the severity of risks, requiring banks to follow through on a prescribed Monitorable Action Plan (MAP)[iv]. However, the RBS framework is still historic and partial in its approach because, for example, in looking at credit risk while it does, for the first time, go to the performing book, it only examines the rating migrations that have already taken place, and uses as a measure of concentration risk, only the top twenty assets.
- The Ministry of Corporate Affairs and RBI have recognised the urgent need to converge Indian Accounting Standards with International Financial Reporting Standards (IFRS) – the Ind AS has been recommended by RBI to be commenced for scheduled commercial banks from April 1, 2018 onwards. IFRS 9 represents a significant move in that it will require the computation of provisioning based on a forward looking Expected Credit Loss (ECL) impairment model, even for the performing book. This is likely to result in significantly higher impairment provisions and therefore more capital requirements. This will be a watershed moment for Indian banking.
- RBI through its Framework for Revitalising Distressed Assets in the Economy[v], the Guidelines on the Joint Lenders Forum (JLF) and Corrective Action Plan (CAP)[vi], the Strategic Debt Restructuring Scheme[vii], has put in place machinery for the rectification and restructuring of stressed assets on banks’ books.[viii]. However, just as with standard assets, there is a need for recognizing and incorporating expected losses into the loss recognition in restructured assets.
- Also, the RBI has indicated its interest in moving to a dynamic provisioning framework in which banks will need to make dynamic provisions which would be the difference between the long run average expected loss of the portfolio for one year and specific provisions[ix] made during the year. Thus, this will ensure less P&L volatility[x].
- With the accounting for financial assets moving towards better capturing the effects of potential impairment for the remaining life of the asset either through mark-to-market and expected loss approaches, there is broad consensus that these measures will ensure adequate cover for expected losses in the form of better provisioning, while unexpected losses are to be covered by capitalization[xi] and by more efficient use of banks’ capital. Overall, a move from high regulatory forbearance-low bank competencies equilibrium to low forbearance-high competencies would be essential.
- I want to spend a few minutes on financial inclusion. Specifically, the differentiated banking design that was discussed, has great potential for combining sharp increases in financial inclusion while preserving stability. This category includes Payment Banks and Wholesale Banks (the discussion paper on this is awaited). After the last round of universal bank licensing in which 2 new banks were licensed, it was clear that it is not possible to find sufficient number of qualified candidates that satisfy “fit and proper” requirements to significantly accelerate the number of banks in the system in the near-term. An integrated banking regulation framework that permits differentiated banking business models appears desirable for a number of reasons:
- There would be flexibility to approach payments, savings, and credit both independently (in a Vertically Differentiated Banking Design) and to bring them together (in a Horizontally Differentiated Banking Design) when the efficiency gains are high and the other costs are low. Concerns relating to finding fit-and-proper candidates in the case of vertically differentiated institutions would be far fewer and licensing a relatively large number of them would, consequently, be far easier. These, over time, could also provide a pipeline for future universal banks.
- The current fragmented regulatory structure creates far too many arbitrage and lobbying opportunities, and in the absence of a single unifying framework, measures are continually being taken to respond to them in a somewhat ad-hoc manner (such as higher capital adequacy norms for NBFCs combined with an easier Non Performing Assets recognition norms and 100% risk weights)
- Even as new differentiated banking business models take root, there is a need to reimagine the role of universal banks as one that is no longer engaged as risk originators, particularly in high-risk segments, but rather as being risk aggregators, with freedoms to rebalance their portfolios based on risk-profiles and diversification outcomes that each bank decides for itself.
- Given the progress on the JAM trinity and the emergence of differentiated banks, it may well be the case that the credit and payments strategy evolve differentially within the broader financial inclusion strategy. While progress on credit would necessarily have to be much more measured and prudent no matter what strategies are adopted given the inherent risks and customer protection concerns, there is an urgent need to make access to payments ubiquitous & this seems within striking distance. In addition, there is also a pressing need to create an architecture that allows information relating to customer behaviour, in particular, transactions histories with financial institutions, telecommunications companies, and utilities, to be captured and transmitted with high integrity, while simultaneously maintaining the highest standards of customer privacy. The development of this payments and information architecture will not only have enormous inherent value but could also be thought of as “highways” on which a more diverse credit intermediation system can be built. Fortunately, India already possesses the necessary tools to bring about this rapid change. We have a one billion plus strong Unique ID database; a rapidly growing telecommunication network in rural areas with over a billion mobile phone users; expanding broadband connectivity which is expected to cover every village in the next 12 to 24 months; and multiple credit bureaus which are all very active.
- Getting the financial inclusion strategy on the credit side right has important consequences for the economy and the country. Despite the authors’ observation that the financial depth at the national level has plateaued off at roughly 60%, we continue to have fairly low levels of financial depth (credit-to-GDP) in several pockets of the country. States such as Bihar have an overall credit to GDP ratio of less than 16% despite the fact that it has one of the lowest levels of GDP in the country. It is arguable whether the binding constraint is the availability of credit or the opportunities available in the regional economy – but at very low absolute levels of credit availability, this may be a self-fulfilling prophecy. We need a banking sector that is capable of meeting the growth needs of all sectors and regions of the country and for this; the sector has to exhibit a high degree of resilience and profitability.
[i] Chapter 4 from A Hundred Small Steps: Report of the Committee on Financial Sector Reforms, Government of India, 2009
[ii] Indradhanush Plan for Revamp of Public Sector Banks, Department of Financial Services, Ministry of Finance, August 2015
[iii] The ICAAP is a forward-looking risk-based process that is approved by banks’ boards and submitted to the RBI annually. It sets risk tolerance levels and lays out processes for managing and monitoring risks, stress testing and scenario analysis, and links back to a strategic plan for meeting current and future needs for capital and reserve funds given the risk tolerance levels.
[iv] RBI can require banks to modify or enhance risk management and internal control, reduce risk exposure to specific risk levels, achieve minimum CRAR levels above the minimum regulatory capital requirements, and so on
[ix] These are the provisions made for NPAs in accordance with existing RBI circulars
[x] B Mahapatra: Underlying concepts and principles of dynamic provisioning. Keynote address by B Mahapatra, ED, RBI at the Conference on “Introduction of dynamic provisioning framework for banks in India”, organised by CAFRAL, 21 September 2012