## Tuesday, September 01, 2015

### Payment bank entry process considered inconsistent with the rule of law

by Shubho Roy and Ajay Shah.

### First best

Banking is the business of accepting deposits from the public with the promise of repaying such deposits on demand at an agreed rate of return. Payments is a distinct business from banking. A payment system is one which enables payment of funds to be effected between payer and a beneficiary. In the olden days, payments was primarily (though not entirely) the business of banks. In the modern world, banks should face competition from technology companies who will use modern telephony to do payments. These companies are bringing in a new business strategy of leveraging their client base and using analytical tools to provide a superior consumer experience.

The first best approach is to establish a regulatory strategy for payments which is not deferential to the business interests of banks,i.e. which has a level playing field between banks and non-banks.

### Second best

The second best strategy is to invent the moniker `payments banks',which suggests some kind of entity which does payments and sounds like bank but isn't a bank. In this case, a mechanism is required through which new entrants will establish payments banks. Anything that we do in public policy in the Republic of India should be grounded in the rule of law. How would a rule-of-law entry mechanism work?

Under version1.1 of the draft Indian Financial Code, the legal process for considering an application for carrying out a new business is governed by Chapter 20. This induces the following procedure:

1. The regulator has to make clear regulations on the criteria on which an application will be judged.
2. Section 73(6) of the Code provides that before rejecting an application, the Regulator must provide a show cause notice to the applicant.
3. Section 91 provides the details of how a show cause notice is issued and its contents.
4. The Regulator must give an hearing, and provide all the material the Regulator relied on to come to its decision.
5. The hearing will be before a staff member of the regulator, who is not involved in the actual evaluation of the proposal (separation between executive and and quasi-judicial functions)
6. There will be detailed regulations on how the hearings will be carried out; the procedure will not be made up on the fly.
7. If the applicant is unhappy with the decision of the regulator, the applicant may appeal to the Financial Sector Appellate Tribunal (FSAT).

This procedure yields the rule of law in entry of new firms.

### Third best

The RBI recently announced that it has given "in principle" approval to 11 applicants for payments banks licenses. These were chosen from 41 applicants who wanted to start payment systems. The manner in which 11 payments banks were short-listed is inconsistent with the rule of law. It is not clear how the RBI came to the number of 11 payments banks, or why licenses were denied to the others.

Right at the outset of this process, it is not clear why RBI chose to use the Banking Regulation Act, 1949 to license "payments banks" when Parliament has made a separate law for governing payment systems under the Payments and Settlement Systems Act, 2007. Since paragraph iv of the payment bank guidelines do not allow such entities to undertake any lending activities, it is not even clear why RBI is classifying them as banks.

The RBI states that an external evaluation committee chose the 11. The RBI claims that the committee determined its own procedure and analysed the applications. Without casting any aspersions on the members of the committee, this is, sadly, not a due process of law.

The RBI notification claims that the committee screened the applicants for financial soundness, fit and proper criteria, physical outreach, business model innovation, capability of volumes of transactions and money and their proposed business plan. Additional data was also requested from the applicants. While this may satisfy a casual reader, these phrases are vague and empty. There is no explanation on how the committee evaluated each criterion mentioned in the paragraph. Were marks given for each criteria? Was there a pre-decided cut off marks below which licenses would be denied? What was the basis/system of providing marks or comparative ranking? How was innovation adjudged and comparatively scored/ranked between the 41 applicants?

In addition to these process failures described above, the RBI guidelines provide no review or appeal provision for rejected applicants. This is also a violation of the rule of law.

Compare and contrast this with examples for the existing legal process for selection/short-listing in the government. If you apply for a driving license and it gets rejected, you get a specific reason for the rejection. The Motor Vehicles Rules specify what the department tests. The Motor Vehicles department does not state "the driving instructor will formulate his/her own procedure for determining the competence of the driver". Giving out licenses to payments banks is more important than giving out driving licenses; the quality of procedure at RBI should be even more thorough than what is done by Motor Vehicles departments all across India.

Look at the UPSC exams which select the senior most officials for all of Government of India and State Governments. There is a set syllabus. The rules about how marking is done are stated before the exam. There is a clear system of assigning ranks. After the examination, the answers are also published for examinees to determine their scores independently and more importantly to understand where they went wrong. The process has a subjective part of interviews; but that forms a smaller component of the overall evaluation which is predominantly objective.

The process used for the entry of payments banks is strikingly different from the quality of governance seen in main line government. It is, however, disturbingly similar to the process of allocating coal blocks through a Screening Committee, which the Supreme Court has held unconstitutional on the grounds that:

1. The allocation procedure followed by the Screening Committee was arbitrary.
2. No objective criterion was used to determine the selection of companies.

The Supreme Court found:

The approach had been ad-hoc and casual. There was no fair and transparent procedure,...

The Supreme Court judgment gives much insight into the governance problems of RBI, and the concepts that drive the draft Indian Financial Code. It should be noted that the much criticised procedure of the coal ministry was ahead of RBI on good governance procedures in that they released the minutes of the meetings of the Screening Committee while RBI did not.

So far, we have discussed how RBI has run the licensing process. Now we turn to the RBI concept of a scarce number of licenses. Entry problems in public policy fall into two kinds:

1. Allocation problems: Where the resource is a limited resource and the number of claimants are more. The system preferred by governments in these types of situations is usually auctions with allocation to the highest bidder. This is used for situations like coal mines or spectrum allocation.
2. Licensing problems: These are activities where the entity carrying out the activity needs to have certain minimum qualifications. If such qualifications are absent then there is a risk that the activity may generate substantial negative externalities. As an example, an untrained driver may kill innocent bystanders.

Some problems are combination of the two, like the UPSC exams where you need skilled civil servants and the number of positions are limited. In those circumstances, scoring plus comparative ranking is the standard method.

The payments bank licenses are not an allocation problem. There would be no problem if there were 30 payment banks instead of 11. It would have promoted more competition and reduced costs to consumers. The arbitrary setting of numbers of payment banks harks back to the days of the Aluminium Control Order or Steel Control Order or the other Control Orders and Industrial Licenses which plagued the Indian economy before 1992. These laws set arbitrary production quotas and licenses for manufacturing without any rationale. A crucial part of our reforms of 1992 was removing the quotas and industrial licenses. Anyone can start a steel mill now or produce aluminium. there is no limit to the number of steel mills the government will allow.

Even if you were to consider the entry of payments banks as an allocation problem, there is an important governance failure right at the outset. Nowhere in the guidelines for payments banks (or any document that we could find in the public domain) is there any mention of the magic number 11. This is akin to the UPSC deciding on the number of students to take in after the exam. The arbitrariness of the decision is inconsistent with the rule of law.

### A climate of fear

One may wonder why no one stands up against these violations of the rule of law. The problem is, without a statutory or regulatory right to appeal, it is dangerous to go against a regulator in India. Even if you win in court, you would then be under the regulation of the same entity you vanquished in the High Court. In a non-rule-of-law situation, where the financial agency has wide and arbitrary powers, that is not a condition you would enjoy. Unlike other regulators (SEBI, IRDA, PFRDA) there are no statutory appeals under the RBI Act against the decisions of the RBI. Regulated persons feel safe dragging SEBI to SAT as SEBI operates in a rule of law environment and is not able to exact retribution through devious methods.

On this subject, see the recent article The rule of law in the regulatory state by John Cochrane. For an example of better governance found within India, consider SEBI. When a person applies to become a stock broker SEBI grants the broker a hearing before rejecting the application and even allows the broker to apply for a reconsideration (See Regulation 8 here). After that, the broker can appeal to the Securities Appellate Tribunal.

### Subjects versus citizens

RBI gives hope to the industry through the following paragraph in the notification on approvals:

Going forward, the Reserve Bank intends to use the learning from this licensing round to appropriately revise the Guidelines and move to giving licences more regularly, that is, virtually "on tap". The Reserve Bank believes that some of the entities who did not qualify in this round, could well be successful in future rounds.

This has two shortcomings:

1. The minimum standard of objectivity required under Article 14 of the Constitution is not dependent on the convenience of RBI. A regulator has to meet a minimum level of objectivity under the Constitution. It may improve upon it, but that does not allow it to deny it to the applicants at the present.
2. The consolation to the "failed" participants is in itself an admission of the lack of faith of RBI in the process. A more inclusive and consultative mechanism may have developed a more objective criteria for selection. Even RBI itself seems not to be sure about the legality of the present process when it says that the failed entities may be successful in the future. It seems unsure about the reasons why these entities were rejected.

For an analogy, we would not accept a Coal Ministry which first gave out 11 coal mines through a clubby process, which violates the rule of law, and said that in the future it will adhere to the Constitution.

The promise of future "success" is clouded by dangers for the 30 unsuccessful aspirants. The first set of 11 may rapidly develop their business. In network industries, subsequent entrants face an uphill task of competing against the entrenched incumbents. In addition, RBI makes no promises about when the next round of licenses will be awarded. From 1949 till 2015, RBI was not able to learn how to do on-tap licenses for banks. Hence, there is no telling how many decades of a head start these 11 payments banks will get. The harm imposed upon the unsuccessful applicants could be arbitrarily large. Under the rule of law, State actors can only inflict harm after following certain sound processes.

### The problem of learning

As emphasised above, the Constitution of India does not give rope to any part of the State to violate the rule of law while it is learning.

It is hard to see what was difficult about developing a sound rule of law framework. The steps required are quite elementary:

1. RBI staff could have read the Indian Financial Code and used the procedure there.
2. SEBI has long had sound entry procedures. RBI staff could have just studied SEBI and emulated them.
3. These procedures could have been validated against ample international experience which is on the web. UK regulations are available here, in the U.S. it is usually governed under state laws and the New York regulation process is available here. Australia regulates some payment systems under deposit taking regulations which can be seen here.

### Saving capitalism from the capitalists

A nimble new set of technology companies want to challenge an inefficient incumbent: the existing sleepy world of payments as done by banks. Both industries have one regulator -- RBI -- which is captured by the interests of incumbents. It has not permitted the new players to compete with the incumbents. In an age of great concern about crony capitalism, uncomfortable questions are raised when six out of 11 new entrants are big business houses, and the competitive energy of technology companies has been largely shut out. Many names in the list of 11 which appear new to this field have partnerships with incumbent banks -- e.g. Reliance with SBI, Fino with ICICI Bank, Airtel with Kotak Bank, etc.

How do we save capitalism from the capitalists? The rule of law is a necessary, though not sufficient, condition for a competitive economy where rich families and dominant firms face vigorous competitive pressure. The Constitution of India is a great tool for addressing our problems of cronyism.

At the same time, mere emphasis on rule of law procedures is not a sufficient condition for doing sound economic policy. A great deal of thought and care needs to go into the work, so as to ensure that good outcomes are obtained. E.g. it's easy to do a formal rule of law process which insists that any applicant for a payments bank must be a software company with over Rs.1 trillion a year of revenue. All the formal processes can be easily run with this criterion. But it will give us the wrong outcome. The rule of law is a necessary, but not sufficient, condition to get sound outcomes.

While the RBI Act, 1934, predates free India, in 1949, the Constitution of India established the rule of law in India. SEBI has long had a more evolved system on the entry of mutual funds. In 2013, the draft IFC was released, and offered a substantial step forward where the rule of law was integrally brought into all aspects of financial economic policy. Sound entry procedures are in Chapter 9 of the FSLRC Handbook. In 2014, this critique of RBI permitting two persons to start a bank came out. In this backdrop, the process failures with payments banks constitute an avoidable failure.

### Conclusion

It is better to have 11 new payments banks than to have none. However, there are important intellectual failures in what was done, on questions of economics and on questions of law.

### References

Supreme Court of India, Manohar Lal Sharma vs The Principle Secretary & Others, judgment dated 25th August, 2014

Ministry of Finance, Revised Draft Indian Financial Code for Public Comments

John Cochrane, The Rule of Law in the Regulatory State, The Grumpy Economist, 1 August 2015.

Ajay Shah, Rule of law and competition in banking, Economic Times, 9 January 2014.

1. This is quite a post. Comparing bank licencing to coal allocation process is like really hitting below the belt. It takes the entire so called oomph and hype from the so called reform to revolutionise banking in India. We are just so biased about whatever RBI is doing these days. And as you scratch the hype, you realise it is the same story just packaged differently.

Frankly, one is tired of hearing the word reform and revolution in banking. It is all so historical. General Bank of Bengal and Bahar, the first attempt to build a limited central bank in India by Warren Hastings was also proposed to look at the payments problem in India. It was problematic to accept and make payments in India barring for those part of the limited indigenous banking system. Sounds familiar?

Your posts on SEBI are quite an eye-opener as well. There is hardly any discussion on SEBI and its superb yet silent role in shaping Indian capital markets. Even the FSLRC and IFC debates are just around whether it will reduce RBI's powers and not a word about what happens to SEBI. It is amazing to note how easy it is to open and shut brokerages in capital markets but so difficult to do the same for banks. Infact, we had a much freer banking system till RBI came in the game. Worldwide, central banks have limited competition in the banking sector. How is it that whereas yesterday's large industrial players have either died or changed their businesses but large banks have remained for a very long time? That too despite multiple crises which have impacted world economy periodically

And how interesting to bring the subtitle - Saving capitalism from the capitalists. It should be titled instead saving capitalism from central bankers! Historically, central banking has always been a clubby affair dominated by people having certain degrees and credentials. How do you expect them to allow others to join in this ultimate power game of money management and control?

2. Hi,

Did you see this news bit: http://www.business-standard.com/article/finance/pre-payment-issuers-petition-rbi-on-payments-bank-licence-issue-115090300734_1.html

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.