Thursday, August 15, 2013

The case for differentiated bank licenses

by Harsh Vardhan.

The much hyped applications for new bank licenses are in – all 26 of them. Now, RBI will evaluate these applications which will require scrutinising mountains of documents that have been presented.

A casual look at the applicants reveals remarkable diversity in their character. The applications include non-banking finance companies, micro-finance companies and brokerage firms. There are established and new firms. There are focused (i.e. single business oriented) and diversified firms. There are public and private firms. A number of applicants have well established niche businesses: gold loans, vehicle finance, micro lending, etc. They clearly have established capabilities and credibility in these businesses over years of operating them. Now they want to morph into a generic universal banking model, large parts of which they have no experience in. In effect, all these players transform from a specialised business to a more generic, undifferentiated business.

This is surprising. As any business evolves, we expect specialisation to develop. It is a natural response to innovation and deepening of capabilities. In response to competition, firms become specialists in some things and develop a competitive edge there. Many of the specialist players in Indian finance have done so over the last few decades – they have focused on serving specific needs (e.g. loans against gold), specific customer segments (e.g. semi-urban, self-employed), and specific geographies. Many of the aspirants of banking license today are very successful specialists in one sub-component of finance.

Then why is it that they want to jettison a specialisation built over years of effort, and merge into this undifferentiated, monolithic group called commercial banks? More importantly, will the financial system become weaker when these specialists become generalists?

Ask the applicants why they want to become a bank and the most common answer is access to low cost deposits. This is, at best, a half-truth. While, it is true that only licensed banks can access putative low cost deposits (CASA), the belief that they are actually `low cost' is not supported, once we account for all the costs of getting them. New banks will have a hard battle on their hand getting a foothold into this intensely competitive business of CASA. It is not a coincidence that the only banks to have hiked savings account interest rates, post deregulation, are the newest and the smallest ones: Kotak Mahindra, Yes, and IndusInd.

But there are other important parts of financial services that are reserved for banks – access to payment systems is the most important one. Payments have seen enormous innovation across the world primarily driven by non-banking players. In India, we have lagged in this area. Non-banks cannot enter the payments business, and banks are laden with legacy technology and processes that make them slow to innovate and reluctant to cannibalise their existing cashflows.

RBI recognises only one type of bank. So, the moment an entity becomes a bank, it is subject to the same rules and regulations as every other bank. All large Indian banks look like each other, and they all look like mere enlarged versions of what they were 10 years ago. Bank regulation is a giant homogenising force that kicks in when a banking license is granted. If any of the currently specialist players get a license, they will undergo this transformation and begin to look like the 90-odd scheduled commercial banks.

As India grows and becomes more sophisticated, banks must keep pace with the sophistication of the real economy, where most firms do not look much like they were 10 years ago. An important feature of a sophisticated banking system is specialisation. How do we achieve specialisation and diversity in our banking system?

First, in the current round of licensing, I suggest that RBI not only recognises diversity but encourages it. It can do so by giving licenses to applicants with diverse and specialist business models that focus on specific customer segments, products, and technology and process platforms. More importantly, it needs to show flexibility in regulation, and not ask all banks to look the same in the short run. Financial regulation is about consumer protection and micro-prudential regulation, and not homogenisation.

Second, RBI needs to start thinking about differentiated banking licenses. Why should we have only one type of license? Why can’t we have a utility bank – one that provides only transaction services (e.g. security related services such as custodian, trade related services, etc) and does not lend or borrow (significantly)? Why can’t we have a payment specialist bank or a home loan banks (one that can take deposits but lend only in form of home loans, by far the safest lending for Indian banks)? Everyone talks about financial inclusion, so why can’t we have an inclusion bank – one that serves only the bottom of pyramid customers?

Such specialist banks will be expected to focus on their own individual niches and strengthen their capabilities in serving these niches as a bank - through access to payment systems, ability to raise deposits, regulatory oversight, etc. The presence of such banks will make the system less monolithic and hence better placed to face economic cycles. With differentiated banking licenses, we will have banks that do not face boom and bust at the same time. Reduced correlations between banks will give lower systemic risk.

Issuing differentiated licenses will require that RBI has to become more sophisticated in how regulations are drafted. RBI's regulatory staff will need to understand each business model, and write regulations about consumer protection and micro-prudential regulation that cater to the unique features of that business model. These regulations will have to differentially use the main tools of micro-prudential regulation – capital adequacy and reserve requirements, regulatory audits, reporting and compliance -- so as to achieve the desired failure probability, while recognising differences in business models. RBI will also have to ensure that there is no regulatory arbitrage that is created across different types of banks.


  1. Harsh,

    This got me thinking: How would this work under the draft Indian Financial Code? I think there are two parts to the story.

    The first is that the IFC clarifies the classes of financial firms that can exist. At present, in India, we have terms like NBFC and `Merchant Banker' and so on. Under the IFC, there are exactly two groups of firms. Either a firm is doing banking or payments, in which case it's regulated by RBI, or it is not, in which case it's regulated by the second regulatory agency (the UFA). The definition of finance is very general. This will open up many new and innovative lines of business that are presently infeasible. Some of this is what you might think of as a special purpose bank.

    The second line of thought is about the drafting of regulations. Under the IFC, the two main elements of financial regulation are consumer protection and micro-prudential regulation. The regulator would be obliged to make documentation, and address public comments, about the soundness of both these.

    The objectives of consumer protection would have to be achieved for all kinds of banks and payments companies, and this may well require drafting differentiated regulations. Complaints at the FRA would generate feedback about places where there are problems in consumer protection, and would encourage revisiting the drafting of regulations.

    Similarly, prudential regulation would need to be tailored to the risk profile of specialised forms of banks, such as banks that only give home loans, so as to achieve the same failure probability for all.

    All these good things are in the IFC. The real problem is one of accountability. If a regulator is lazy or ignorant, how to push towards better performance? Suppose a regulator merely writes simplistic flat regulations that are blindly applied to all banks; suppose it is tone deaf and ignores the principles of prudential regulation in the IFC. What would then happen? The IFC embeds a great deal by way of accountability mechanisms (which is what is generating some of the hostility from existing regulatory staff), which should reduce the extent to which such misbehaviour happens. Over the years, jurisprudence would develop around the key question of challenging regulations at the FSAT on the grounds that they are incompatible with the principles written in the IFC.

  2. Ajay;:

    The IFS, correctly, anchors the definition of banking around deposit taking. If we take this definition to its logical end, if an entity does not take deposits - its not a bank. A payment company, whose model does not entail deposits, can thus be a non-bank. Unfortunately, today such a company is not possible as a non bank cannot participate in the payment system at all.

    To make it easy, let me use an analogy. Consider the demat accounts for securities. All the securities are held by the depository which has the legal ownership of them and the Depository Participants (DPs) carry the customer accounts and facilitate transactions - buying and selling of securities. Why can't such a model work for a payment company? Can we have a situation where money is held by a few players but a larger set of players facilitate instructions that create flows between the accounts - which is effectively payments. Let's separate the money flows and the information flows. All the innovation is around how information flows around payments happen.

  3. Get rid of this monster RBI instead of all that regulatory chik-chik

  4. A lot of RBI Regulations are dilapidated and does not make sense. First step RBI can take is to conduct the feasibility of some age old regulatory clauses and remove the ones which doest not render any useful ness in current competitive world.


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language. I delete any comment which does not contribute to the intellectual discussion about the blog article in question.

Please note: LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.