## Friday, December 19, 2014

### Policy puzzles of the digital nirvana

by Arjun Rajagopal, Renuka Sane, Somasekhar Sundaresan.

It has been a bad week for Uber. The effect of its automated surge-pricing during the hostage crisis in Sydney and the reaction to this, have exacerbated the publicity surrounding the alleged rape of an Uber customer in Delhi by a driver who was listed on the company's app, and the litigation it faces in San Francisco over not conducting effective background checks on criminal records of drivers. In India, the outrage has been accompanied by bans in New Delhi and calls to ban or suspend Uber in other states.

This is not the first time, or the first geography in which Uber has run into trouble. It has been criticised over its sexism, ethics and bro culture and over its record on passenger safety. The app has been recently banned in Spain and Thailand and has run into trouble with authorities in several other countries, mostly for being at odds with the traditional, regulated taxi companies.

Uber does not have a taxi license in Delhi. It is not a radio taxi service. It does not own the cars or employ the drivers. Fine print on its website suggests that it disclaims the suitability, safety or ability of third-party providers. Uber claims to not be a transportation provider, and only connect riders to drivers through its app. Yet, it claims that its service is a safe and secure one, adopting standards that go way beyond local standards that regulatory authorities may prescribe.

These claims allow one to litigate against Uber with allegations of misleading customers with particular standards of safety when in fact there were none. This is the stuff class action suits are made of. Indian customers could potentially be creative and sue Uber in the US, where control over its operations is headquartered.

However, the furore also raises larger questions on what the appropriate legal or regulatory response should be to such aggregators, and where the liability lies when things go wrong.

### Is more required? Liabilities for information aggregators

A defining feature of the early 21st century is businesses becoming powered by software and synonymous with services delivered online. Technology platforms, or aggregators, purely facilitate the sale between buyers and sellers. In this sense, they are neither originators of the product, nor distributors linked to specific manufacturers. Often touted as disruptive innovators, such businesses tread thin on requirements under the licensing and regulatory systems even while competing with similar services provided by the traditional licensed players. This regulatory arbitrage raises important questions on the obligations of such aggregators.

The fundamental question that governments need to ask themselves is what, if any, obligations should be placed on businesses such as Uber. Should a market aggregator be responsible for the quality, or safety of a product that is sold on its technology platform? There are well reasoned views on both sides.

### The analogy with a financial exchange

We might make an analogy between Uber and an electronic stock exchange, in which case the exchange aggregates information and enables transactions. The quality of the product being sold (i.e. whether the shares represent a "good" investment) is not guaranteed by the exchange. The transaction is guaranteed. Under this analogy, Uber's job is simply to ensure that double booking of cabs never occurs, and that payment transfers occur reliably and seamlessly.

### The analogy with a retailer

We might also make an analogy between Uber and the owner of a neighborhood mall. While a case may be made against regulatory intervention to make the mall owner responsible for quality of the goods and services, there would also be a case for the mall owner being obliged to ensure safety and security in the mall premises, and being liable for a customer injuring themselves on a broken step. Besides, if the mall owner knows that sale of some goods needs a license (for example, alcohol), and turns a blind eye, it would beg the question if the owner can effectively defend against a charge of aiding and abetting a violation of the licensing of sales.

### The puzzle

Where does an aggregator fall, on the spectrum? This will determine what aspect of the customer interface the aggregator is held responsible for. For example, companies may make business decisions on where on the safety spectrum they lie, and charge a premium for it. Over time there may emerge expensive aggregator companies and cheap aggregator companies and customers take a call on how much they are willing to pay for what quality of service. Regulators could step in and prescribe minimum standards - in much the same way, product warranties across jurisdictions are spelt out. Rules about information disclosure could help consumers make better decisions.

At the point of entry into the cab, the consumer is held hostage to that cab, and must trust that certain minimum safety and service standards are met. In fact, such requirements were the reasons for entry regulations in the traditional cab industry. The onus for these then lie in the State capacity that gives commercial licenses, and does police verification checks. This requires the State machinery to work, and is independently an issue apart from that of requiring anything of a technology platform that merely brings together buyers and sellers.

### Conclusion: Another kind of activism

The problem then, is two-fold. First, there is the tragedy of poor State capacity in preventing and prosecuting crime, which incentivises simply banning certain activities. This is a generic problem that bedevils the working of the Indian economy in numerous contexts. The solution to this issue can only be a long and hard one: of improving the functioning of our public institutions.

The other problem to address is the poor methods for shaping, regulating or nudging commercial conduct. While India is a common law country and action for tort indeed can shape commercial conduct, the problems in delivering timely justice under common law has led to regulators occupying the commercial space in many sectors. However, regulatory interventions with clarity of thought on what can and cannot be done by commercial parties, leaves much to be desired even in sectors that have been regulated for decades. When Uber began its coverage of India, no transport regulator raised as much of a whisper about the status of its regulatory compliance and whether at all intervention was warranted. Now, at the first sight of a crime, the same regulators are quick to consider imposing a complete ban.

Common law remedies coupled with penalties and damages offer recourse to the affected parties, and hit offenders where it hurts: on their balance sheets. Predictable regulations would not only make service providers answerable to society but also would make regulators answerable to the service providers. Is it a substitute for criminal prosecution of heinous crimes? Certainly not. But it is a useful complement. Regulatory activism is not just about ensuring effective prosecution, it also means bringing companies to book if they are lying about what they are providing.

There is an expectation that software will eat the world', that lightweight businesses like Uber will come to dominate the whole world. Perhaps conditions of State capacity in India will prove to be a bottleneck, both in respect of problems such as obtaining law and order, and in terms of navigating the subtle questions of public economics and coming out with the right answers.

## Sunday, December 07, 2014

### Have Indian banks been underpricing corporate loans?

In the 1990s, we had large scale defaults on corporate loans, which felled IDBI and IFCI. For some time, we thought that we had learned our lessons, that micro-prudential regulation had improved, so that such failures would not recur. Has this happened? There is cause for concern.

1. After these bailouts of slightly over a decade ago, we haven't had a big banking crisis with a collapse of banks such as IDBI or IFCI. But there is a worrisome scale of regular injections of capital into public sector banks by the Ministry of Finance. If government puts Rs.100,000 crore into an episode of bailing out banks, we say there is a banking crisis. This is not too different from putting Rs.10,000 crore every year for 10 years. Have we had a chronic sub-clinical banking crisis for a long time? Some will argue that all healthy banks raise equity capital as they grow. But there is an unmistakable element of bailout in the equity capital that has gone into PSU banks. On this subject, see Harsh Vardhan and me from October 2012, and this blog post from October 2011.
2. India did not have a big financial crisis in 2008. All we have had was a business cycle downturn that's come after a big credit boom. Yet, we've now got a serious mess in banking on our hands.

These two difficulties suggest that micro-prudential regulation has not improved adequately.

When we open the black box of Indian banking, two problems are visible. Most banks have little skill in credit risk assessment. What passes for risk management' in Indian banks, too often, is the mechanical adherence with RBI regulations -- regulations that micro-manage and are often wrong. There is low ability in the essence of credit analysis: to understand a firm, and make forward-looking forecasts about default. In addition, all banks suffer from high levels of loss given default owing to the lack of a bankruptcy code.

The objective of micro-prudential regulation of banks is to put down requirements through which the failure probability of banks stays at acceptable and low levels (though not zero). The heart of this is  ensuring that the accounting value of each loan is aligned with market value. This was not done. Banks have systematically failed to recognise and provide for bad loans. I feel a distressing deja vu when I hear stories today about what has gone wrong in Indian banking; we heard those exact same stories in 1999. We did not learn; the problems weren't fixed.

Given these weaknesses of micro-prudential regulation, banks have had a merry time, showing accounting profits while giving out loans at artificially low prices, and building up an ever larger inventory of loans where the book value is in excess of the market value. Better micro-prudential regulation would have created a set of rules through which bad loans were valued at market price. Better micro-prudential regulation would have hindered, and not helped, banks in covering up bad news.

Better micro-prudential regulation would have created incentives for banks to charge higher prices for corporate lending, with interest rates that better reflected their own weaknesses in corporate credit risk assessment, and the high values of loss given default. Mistakes in micro-prudential regulation gave us a systematic under-pricing of risk.

The great credit boom of 2004-2007 has traditionally been interpreted as mistakes of macro policy: This came from the pegged exchange rate. RBI bought dollars in order to prevent INR appreciation, with incomplete sterilisation, which gave low interest rates at a time of a boom in business cycle conditions. This gave us the biggest ever credit boom in India's history. I would add one more ingredient in our understanding of that credit boom: Mistakes in micro-prudential regulation of banks. Mistakes of macro and micro came together to give that party.

Looking forward, improving the thinking in micro-prudential regulation of banks is an important priority. It will take years for India to reverse public sector domination of banking, and the consequential weaknesses of credit risk evaluation. It will take years for India to reduce the loss given default, by enacting an Indian Bankruptcy Code. In the short term, these problems must be treated as given. For the coming two years, the agenda must be to break away from the failures of the last 20 years in banking regulation, while treating the presence of PSU banks and the lack of a bankruptcy code as a given. At present, the landscape of banking regulation is riddled with mistakes  [example, example]. The fair price of a bank loan to a corporation is probably much higher than what we've thought it should be.

## Friday, December 05, 2014

### Justice Srikrishna's speech at the ICSI/BSE show on FSLRC

Mr. Arun Jaitley, Honourable Union Minister of Finance, members of the Institute of Company Secretaries, ladies and gentlemen.

My thanks are due to the organisers of this seminar for giving me an opportunity to address this gathering of professionals and practitioners, who have come here today to meet and share knowledge regarding the Indian Financial Code recommended by the Financial Sector Legislative Reforms Commission (FSLRC) in its report submitted last March [report, law]. I am happy to be here today to speak to you about an important component of economic development that has engaged even the attention of the President of India, as reported on the front page of the Indian Express yesterday.

As all of you are aware by now that there are nine key components of the legal framework recommended by the Commission. These are:

1. Consumer protection and competition,
2. Micro Prudential Regulation,
3. Resolution,
4. Systemic Risk,
5. Capital Controls,
6. Development,
7. Monetary Policy,
8. Public Debt Management,
9. Foundations of Contracts and Property.
I will not dwell on the nitty gritty details of the report of the Commission, or the draft Indian Financial Code recommended by it, since that will be sufficiently done in the technical sessions to follow. I shall try and put before you the flavour of the recommendations and what they seek to address.

The principle of consumer protection is an important determinant of the way governments and regulators think about their job. It is part of the invisible legal architecture that frames our choices, and shapes our lives. And yet, very few regulators have a precise understanding of what constitutes consumer protection. Some believe consumer protection is best served by holding financial literacy camps. Others believe consumer protection is best served by an ombudsman or by setting up an investor protection fund. While one cannot gainsay that these are important activities, consumer protection is much larger and requires an attitudinal change towards regulation.

Since there are no precise definitions of consumer protection in existing law, regulators claim the flexibility to do many things, some necessary and many not, under the garb of consumer protection and claim that the regulated should have a `healthy respect' towards them. Also, because consumer protection is not precisely defined, many regulations framed by regulators do not do enough to protect consumers. Moreover, in a developing country like ours, regulators often feel the need to develop the market, and in their enthusiasm inadvertently establish low thresholds of consumer protection. The need of the hour is for clearly identifying what constitutes consumer protection in the field of finance, and what tools regulators should develop to protect consumers.

As you know, I was fortunate enough to be associated with the Financial Sector Legislative Reforms Commission, as its Chairman, and its report is already in the public domain. The FSLRC and its members spent two years of intense research, consultations, deliberations and drafting, supported by a technical team of 30. In all, 146 people were involved in the Commission's work, including lawyers, economists and other domain experts from India and abroad, bringing with them decades of accumulated experience in finance, regulation and public administration. The aim of the Commission was to examine the foundations of the Indian financial system, raise fundamental questions unasked hitherto, attempt answers to them and recalibrate the financial system to align with the aspirations of a liberal democracy, namely that it should be open, transparent, and governed by the rule of law. The Commission's work involved asking difficult questions about the purpose of regulation, the role of the state, and the checks and balances required to make the system fair and accessible while being efficient and effective.

Though, the report of the Commission was released last year, along with the draft Indian Financial Code, the work started by it is far from over. Forums such as this are important opportunities to continue those debates in the public domain. Those are debates about democracy, fairness and justice just as much as they are about incentives, systems and signals. That is why it is so important for all of us to dispassionately participate in the debate so as to arrive at a consensus unhindered by the drag of status quo-ism.

There were a number of critical topics that were discussed during the writing of the FSLRC report, and consumer protection was an important one. All financial laws in the country are intended for the consumers, and regulators are the arms of the State to ensure that benefits reach the ultimate beneficiaries. The FSLRC report recommends a decisive move away from the old concept of 'caveat emptor,' or 'buyer beware,' towards a 'seller beware' regime. In such a regime, financial service providers must assume greater responsibility and care when recommending or selling products and services to consumers. This move is motivated by the realities of modern financial services, on the one hand, and the realities of an emerging market like India on the other. The deleterious consequences of the buyer beware approach is seen by us in real life today. When consumers are duped on a mass-scale by private entities within a buyer-beware attitude to regulation, there are few efficient mechanisms for recompense to the consumers. Consequently, there is an overreaction from government agencies who go into an overdrive to overkill by banning entire sectors of economic activity. We thus get caught in an adverse cycle of under-protection and over-reaction. We must surely ensure that the baby is not thrown away with the bath water!

As a general experience, financial service providers are in a position of massive bargaining power relative to all but their largest clients. Consumers are generally less informed about finance than financial firms, and typically not able to band together to defend their interests. India is an emerging economy: it cannot be taken for granted that all consumers are well-informed, or free to reject a contract they feel is unfair. The FLSRC therefore recommends a system-wide commitment to consumer protection, in which all regulators build relevant protections into their respective regulations. One critical priority is to mandate that financial service providers make clear, adequate and relevant disclosures to the consumer, and ensure that - based on the consumer's expressed needs - the product or service being offered is suitable for that consumer. Allow me to remind you all of the critical role you have to play, as you work to bolster development, in making that development more rule-bound and fair, and thus less fragile than if it were to come at the expense of the Indian consumer.

Let me turn for a moment to the broader debates I alluded to earlier about the fundamentals of a democratic system. Financial regulation, inasmuch as it is regulation - that is the application of the coercive power of the State to private parties - is not easy. As we pointed out in our report, and as any regulator would agree, regulation is technical work, requiring care, attention and expertise. It is precisely because of that complexity that we have created regulators in the first place: specialised bodies, who will accumulate knowledge and human capital, and will build institutional knowledge and physical infrastructure and information networks that allow them to rise to the challenges at hand. In order to be effective, regulators must have the legal mandate to formulate regulations, to allocate resources as they see fit, and to act decisively. All this contributes to an environment in which private parties can operate freely, and with clarity and confidence. I mentioned how we worked to strike a balance, and design a system that is fair and accessible while being efficient and effective. Striking this balance is complicated because many of the things we do to make regulators more effective, also diminish their accountability to Parliament, and to the people. After all, we cannot forget that regulators - like the State that they represent - are for the people and not vice versa.

The problem is that we have created regulators that function as 'mini-States.' In our larger Constitutional framework, we have separated the critical functions of the government - legislative, executive and judiciary - and placed them with separate bodies. This system of checks and balances is meant to prevent the government from acting outside the law. When we create a regulatory body, we often find it necessary to give that body the ability to write regulations, conduct investigations into the violation of those regulations, and adjudicate on the violation afterwards, which implies a lack of separation of powers. One partial solution to this is to at least separate out those functions within the regulator, and assign separate staff to carry out each function.

Let us take the case of adjudication: When the investigation staff of a regulator takes action against a regulated entity, they must provide it with an opportunity for a hearing before the regulator. When it appears before the regulator to represent its case, it should not before the investigation staff, but someone who will act as a neutral third party. This requires building a separate adjudicatory capability within the regulator, with dedicated staff - headed by an Administrative law member and staffed by lawyers. The person hearing the case should show no bias towards the investigation staff. He needs to be able to give a fair hearing, and, if necessary, decide against the investigatory staff. It is work to build this internal separation of powers, but it is absolutely critical to the rule of law. The game must not be rigged.

It has been argued that these internal safeguards are the only ones that are required to ensure that regulators remain within the bounds of the law, and that any interference beyond this will compromise their work. It is argued that the work of regulators is nuanced, esoteric and complicated, and that regulators need to preserve flexibility and freedom of action to respond decisively to events in the world. Under this view, there is a fear that judges will second-guess the actions of regulators, which is a short step from encroaching on the domain of policymaking. We disagree. We believe that in a democracy wedded to the rule of law, no one, not even a regulator, is above the law. Our response to the difficulty of building checks and balances in a specialised area of regulation, is to build a specialised mechanism - not to give up on checks and balances!

The proposed Financial Services Appellate Tribunal, would fulfill this role. It will be staffed by personnel with relevant domain knowledge, training and experience, and it is designed to adjudicate openly, efficiently and in an informed manner: the disputes between regulators and regulated entities. Yes, one can argue that a new tribunal will not have the institutional knowledge required to appreciate the subtlety of a regulator's action, and rule fairly. But the first thing to keep in mind is that the regulator must be able to explain its actions. If it can explain its actions to judges of the High Court or Supreme Court, neither of whom can claim expert knowledge, it can surely explain its actions to a Tribunal that can lay claims to such expertise. The second thing to keep in mind is that institutions evolve, and they evolve together through their interaction. Sometimes, that interaction may not be smooth, but may involve some tension. But that is the point of democracy, which is a system of checks and balances. A healthy tension between institutions is what makes a liberal democracy grow. In the years since the Securities Appellate Tribunal was set up, to hear appeals from SEBI orders, the quality of SEBI orders has increased. And so has the quality of SAT decisions. Regulators should take comfort in this fact.

Thankfully, the independent academic Dr. Rajan also disagrees with the RBI Governor Dr. Rajan on this aspect. In his 2009 report, A Hundred Small Steps, Dr. Rajan wrote, "Regulatory actions should be subject to appeal to the financial sector appellate tribunal." This is the thinking that guided the FSLRC's deliberations, and it is the position we still take today.

Allow me to share a few thoughts from the experience of the FSLRC: Those of you who have looked at the Commission's report will know that it has recommended re-organising India's financial regulatory architecture. Again, it is a topic that may easily get passed over by many people as esoteric, but all those who are connected with the financial field you would acknowledge that it will profoundly affect the Indian financial ecosystem. What is the logic behind this reconfiguration? The proposed regulatory architecture allows regulators to look beneath the surface of a product or service and unmask its essentials, be it termed a loan, an insurance scheme, or an investment, and regulate it according to the risk it poses to the system, and the probable risk that a firm may not fulfill its promise to a consumer. I fear that criticism of the proposed framework glosses over this critical argument, which is actually about more than just operational synergy. It is the result of soul-searching rethinking of the way we draw the lines around various kinds of financial activity and compartmentalise them, leading to avoidable turf wars. It is driven by a more agnostic view of risk management, both at the systemic level and at the level of individual firms. It is intended to underline the consumer as the focal point of the debate.

Finally, is this a leap into unknown territory? Hardly. This is the approach that many countries have implemented already. And yet, it is a bold move. But it is one that the Indian financial sector and the Indian consumers deserve. Our financial architecture has become unwieldy and unresponsive. Provisions that were intended be temporary (the RBI Act of 1934 is an example) have become permanent. Sometimes this is a result of inertia, rather than careful planning. Arrangements that were the best we could do at the time, are now considered to be the only way to do things. One fears that as a result, parts of our regulatory machinery have become brittle, fragile, or worse, completely irrelevant and utterly useless, if not downright harmful to our interests. It is this concern which motivated the proposed reforms. This is not change simply for the sake of change. And indeed what we proposed is far from cosmetic. This is fundamental re-engineering of a sector that desperately needs it. To push ahead with our current arrangements and simply hope for the best, would indeed be reckless, despite our known reliance on jugaad.. We must reform in anticipation of future crises, not only in response to past ones. Our preparedness for the future requires that our institutions and our systems be world class, when we expect our GDP to be in the region of about 20/30 trillion USD in about 15/20 years, and we entertain the ambition to be one of the leading leading players in the world's financial arena. Only legislation can drive institutional change to yield the high performance institutions that we need. Change is necessarily disruptive; hence people abhor it. Reform unsettles the powerful; hence it ruffles feathers. But it is necessary and it must be done now.

Thank you.