Supervising what you cannot inspect

by Maninder Singh Juneja and Renuka Sane.

In traditional financial regulation, supervisors are able to inspect the thing being regulated. For example, a scorecard by a bank was usually a short list of factors that one could comprehend. These factors could be traced to key documents within the bank, banks had a stated rationale for why they were being used, and there was some common sense in making weighted averages. The scorecard itself was static - the same weighted factors were applied over a period of time. Inspecting this process of constructing the scorecard and using it for management decisions was how trust was built. The AI world is different. Models are rented rather than built, change continuously, and behave probabilistically. They cannot be fully inspected even by the institution deploying them let alone by the regulator.

How should we then think of regulation?

One approach is to intensify the traditional approach where regulators demand more explainability, more documentation, and more validation. This will drive up the costs of compliance. But more importantly, this is ill suited to the new world where the technology changes rapidly, where the bank does not control the AI it uses, where there is no clear artefact that the bank can give the supervisor such as a model or a document (Board of Governors et al 2026). When regulators push traditional approaches, banks will respond by choosing AI models which are easy to document rather than the ones best for them, or defer AI deployments altogether. All these are unhappy consequences. What we need are policy makers who understand the live systems of the new world.

In this article, we analyse these emerging problems from first principles. We start from scratch, understand the landscape of market failure in the world of AI in banking, and think about how regulators can grapple with this world.

Our key idea is that AI systems resist replicability. Regulatory strategies that demand replicability will flounder or choke technology deployment. We suggest the regulatory standard applied should be AI deployments that are "supervisable" - the outcomes should be observable, they should be attributable to causes, and reversible by humans.

How AI is actually deployed

Before we get to the puzzles faced by regulators, we need to describe what AI in banking is. This categorisation is not unique to banking. Banking is simply where their consequences are regulated. Five modes span the range.

AI as Tool: AI augments a step, a search, a calculation, a first draft that the human controls. The human is cognitively engaged.

AI as Collaborator: The human and AI co-produce iteratively, and the human participates at every stage.

AI as Recommender: AI generates scores or options, and the human makes the decision. This is the classic human-in-the-loop, but credible only if the human can meaningfully interrogate the recommendation and not degenerate into cognitive surrender.

AI as Preparer: AI does the work and the human signs off. The approval here is closer to a check by a supervisor rather than a re-derivation.

AI as Autonomous executor (agentic AI): AI executes autonomously inside guardrails and the human monitors on exception. Emerging forms of agentic AI for banking run from single-task agents (a payment released, a service query resolved) to multi-step workflows and customer-facing agents that transact.

A central issue here is the true (de facto) role of the human. A reviewer who approves a thousand recommendations a day is not overseeing a model, the model is overseeing her. A feature of any deployment is the measured divergence between the model recommendation (and estimated uncertainty) vs. the human decision. Managers of banks will need to worry about relapses of human behaviour inside the organisation, a bit like how hospital managers worry about bad behaviour by doctors within their organisation.

What AI does to market failure in banking

We now shift gears to look at the standard knowledge on market failure in banking. Regulation may be justified when (and only when) there exist market failures which cannot self-correct fast enough, and there is adequate state capability in banking regulation to be able to correctly identify them and intervene. AI's distinctive feature is that it can cure several classic failures. Better default prediction reduces credit rationing, better fraud detection cuts deadweight loss, richer risk assessment lets banks serve customers they previously could not price, AI advisors help customers avoid some malpractices by the bank. But there are also some new problems that are anticipated.

Information asymmetry: This happens in banking at two levels: borrower-to-lender (adverse selection, hence credit rationing) and firm-to-consumer (product complexity, hence mis-selling). AI narrows the first through alternative data and may widen the second. On one hand, the customer armed with AI can see through many things proposed by the bank which are not in her best interest. But the consumer cannot observe why they were shown a product, offered a price, or steered toward a specific insurance plan. The sales process itself becomes more opaque. Personalised pricing approaches first-degree price discrimination, extracting consumer surplus. A single flawed model can mis-sell to millions simultaneously, converting isolated conduct failures of the pre-AI world into a big correlated event. And redress weakens when a denial comes from a model the firm itself cannot explain.

Systemic externalities from shared infrastructure: Each bank chooses its models, data sources and vendors to optimise its own performance. When multiple entities choose the same ones, the sector's exposures become correlated, which is a cost no individual bank prices in. We list the vulnerabilities below:

Correlated model risk: banks on similar models and the same foundation providers respond identically to an event. The regulator, at the system level, has to manage what happens when institutions move together, because no single entity has experience of such behaviour or of the impact synchronisation adds.

Third party concentration: one vendor's failure propagates everywhere at once. India has already run this experiment, when a ransomware attack on one shared technology provider knocked roughly three hundred cooperative and regional rural banks off the payments network.

Correlated cyber breach: shared stacks mean one exploited vulnerability is every institution's vulnerability. AI lowers the attacker's costs (automated vulnerability discovery, deepfake social engineering) and adds new attack surfaces (data poisoning, model inversion, prompt injection against agents that can move money).

Runs on banks at level 3. The bank runs of old were a queue on the pavement. Then we got to Silicon Valley Bank where over a weekend, customers took away money from the bank. Now we can be at level 3: autonomous agents managing customer cash can turn a shared signal into a self-reinforcing run at machine speed.

The various market failures listed above behave differently across the five modes of AI use. For example, when a human constructs the offer, opaque pricing can get contained. However, when an agent personalises autonomously at scale, this may become severe. Systemic correlation is moderate when AI advises and severe when fleets of similar agents act simultaneously. Any regulation that grades by model type alone, or by use case alone, misses half the object. The next question is the mode of regulation itself.

Regulatory strategy

Regulation can work in two ways. Process-based regulation is ex ante: it prescribes how the firm must operate, defines required controls, mandates oversight, validation standards, limits on autonomy. Outcome-based regulation is ex post: it prescribes ends, fair treatment, solvency, and judges results, leaving the choice of methods to the firm.

Outcome-based regulation is the efficient default. It is technology-neutral, so it does not ossify as methods change; it lets firms find the least cost route to compliance; and it does not require the regulator to understand the firm's production function better than the firm does. But it has important preconditions: the outcome must be observable and measurable; it must be attributable to the firm and, ideally, to the cause; and the harm must be reversible or compensable. Process regulation is the right departure from the default when those preconditions fail and where outcomes are unobservable, harm is catastrophic or irreversible, or damage manifests only systemically or with a lag.

Traditional doctrine treats these preconditions as given: examine the activity, choose the mode. This does not work for AI. Whether an AI deployment's outcomes are observable, attributable, and reversible is an engineering choice, which needs to be settled at design time. We suggest that regulation should mandate observability. This makes it possible to have an "outcome-based supervision" model. There are three ways to ensure observability.

1. Telemetry implies that the institution keeps a track of every decision such that the system records which version of itself it was using, what information it was given, and any time a person stepped in to overrule it. This makes outcomes attributable. That way if a certain group starts getting more (or less) approvals than before, the organisation can evaluate what caused the shift - was it the model, or the group itself. Such a capability is being mandated elsewhere in the world for similar use-cases (European Parliament and Council of the European Union, 2024). The regulator should also consider if it wants to set a minimum common telemetry standard.

2. Boundaries and rollback include putting caps on what the system is allowed to do, rolling out new updates to just a small number of cases first (instead of everyone at once), and having a tested plan for switching back to the older version if something goes wrong. This ensures that if a bad update slips through, it only affects a small slice of decisions.

3. Probes make bias observable. One way is "paired testing": one submits two applications that are exactly the same except for details that hint at things like someone's race or gender, and see if they get treated differently. One can also compare approval rates against the company's own normal levels. Together, these checks can catch an unfair credit model in just a few weeks, instead of waiting years to see who actually pays back their loans. One can also check rejected applicants against credit-bureau data to see which ones got approved by someone else, and how they fared.

These are similar to the idea of decision receipts that record which rules were applied to which facts and in what sequence for every decision made by a government or public system (Srivastava, 2026).

Over time, the supervisor should also build a repertoire of its own test cases drawn from incidents, complaints and examinations across the system and run it against every material AI deployment, much as stress-test scenarios are run against every balance sheet today. This will ensure that what surfaces in one institution becomes a probe for all others. It thus allows the regulator to set its own observability layer.

Process regulation is then reserved for the harms that are systemic, correlated, or irreversible at machine speed. For such events there need to be protections such as circuit breakers that halt things before they cause damage, limits on how much the systems can do on their own, model diversity so they don't all fail the same way, and rehearsed back-up plans.

What follows for the supervisor, the board, and the customer

For the supervisor: The unit of examination shifts from the model to the deployment, and the examiner's question shifts from "show me the validation report" to "show me the behaviour": what boundaries were set, what exceptions were thrown, what overrides were exercised, how far the system drifted from its baseline. Supervisors will also need to find the intellectual clarity to avoid a wide variety of extraneous claims about regulation of AI, e.g. the push for economic nationalism which has nothing to do with market failure.

For the board: A board cannot certify systems it cannot inspect or understand. Its role is to govern the framework which includes the limits on what the system is allowed to do, and making the rules for when a decision must be escalated to a human. The board then needs to continuously check the exceptions and overall performance.

Exception-handling should also be written into policy. If a problem stays unresolved beyond a defined size or time limit, it automatically gets escalated to the board. Internal auditors should double-check that the numbers are real. The board sticks to this audited framework and does not inspect the system directly.

If a board is asked to approve something they can't understand, they will default to saying no. But if you let them govern the limits and the exceptions instead, they can say yes.

The customer: The customer becomes part of the supervisory architecture. The widening firm-to-consumer asymmetry has a structural corrective the pre-AI world lacked: the customer now has AI too. Mainstream assistants abroad have begun connecting directly to users' accounts. If product terms (rates, fees, eligibility criteria) are mandated to be structured and machine-readable, the customer's own AI does the comparing, the explaining, and the policing of mis-selling, continuously and at zero supervisory cost. The redress channel weakened by opacity is restored the same way: an adverse decision should carry its reason to the customer, and what would have had to be different for the answer to change. This restores the ability to contest. The same asymmetry that AI widened, AI-equipped customers can close, but only if regulation hands them the data.

An example

Consider a debt-collection example. When borrowers fall behind on payments, the bank has to decide how to chase each overdue account. Contact methods differ in cost. Automated SMS and IVR (the automated phone system - "press 1 to pay") are cheap; having an employee actually phone the customer is expensive. So the bank builds a model that allocates accounts: cheap automated nudges for most, and the costly human call reserved for the accounts where the model predicts that talking to a person will actually "cure" the account (get it back to paying). Human calling works better but it costs more. The model is rationing an expensive resource to where it thinks it'll pay off.

With AI, the collection systems would retrain continuously or run reinforcement-style optimisation against a live reward (cure rate per rupee spent). The model would be moving on its own, faster than the review cycle, toward a target that is a proxy for what the bank may actually want. Under the conventional "inspect-the-artifact" approach, the model is checked before deployment, signed off, and reviewed on a schedule (say quarterly or annually). However, if AI is an active optimiser, a quarterly review discovers deterioration only after it has touched thousands of accounts. It may drift toward a mis-specified target in a way no one can read off the model itself. Further, if the collections model is a shared vendor product, or several banks fine-tune the same foundation model on similar data, they all go bad the same way at the same time when borrower behaviour shifts. In contrast, in a "supervisability-built in by design" approach, a small slice of accounts is deliberately kept on the previous allocation method, running live alongside the new model. This allows comparison between the old and the new in real time.

Every decision should keep a record of a few things: which version of the AI was used, what action it chose, which group of customers it was dealing with, and how things turned out in the end. This is how it helps: Say the AI's overall success rate is going up, but one particular group is quietly getting fewer phone calls from real people and doing way worse because of it. With all that recorded, that gap shows up right as it's happening. You can compare it against what's going well elsewhere, trace it back to a specific version of the AI, and undo it that same day, because the older setup is still up and running as a backup.

The company should never have to dig into the AI's inner "thought process" to realize its behaviour has taken a bad turn. None of this replaces the normal testing that the firm would do anyway. The difference is that here, the ability to observe what's happening, pin down what caused it, and shut it down are all built right into the system from the start, instead of being things you have to go do to the AI afterward.

Conclusion

AI deployments sit uneasily in conventional mores of banking regulation. The intelligence is rented, the behaviour is probabilistic, and harms can move at machine speed. We suggest that the response should not be more inspection of what cannot be inspected, nor blind faith in outcomes that arrive too late. Efficient regulation will emerge from observability, attributability, and reversibility. Regulators should require banks to build these.

References

Board of Governors of the Federal Reserve System, Federal Deposit Insurance Corporation, and Office of the Comptroller of the Currency. "Supervisory Guidance on Model Risk Management." SR Letter 26-2. April 17, 2026. https://www.federalreserve.gov/supervisionreg/srletters/SR2602.pdf.

European Parliament and Council of the European Union. Regulation (EU) 2024/1689 (Artificial Intelligence Act), arts. 12, 26(6), and Annex III(5)(b). https://ai-act-service-desk.ec.europa.eu/en/ai-act/article-12.

Srivastava, Manish. "Digital Governance Needs Decision Receipts." Episode 73 of Big Ideas. XKDR Forum, June 1, 2026. Podcast, video, 13:06. https://youtu.be/WFX4ITb9yok

Trust in the Era of the AI-Informed Customer

by Maninder Singh Juneja.

A patient grappling with a longstanding eye problem was diagnosed with MGD (Meibomian Gland Dysfunction) by an eye specialist. To make sense of the medical jargon, she photographed her meibography report, the gland scan, and uploaded it to an AI, which confirmed the diagnosis and the line of treatment. A few weeks later, wanting to avoid buying shades of cosmetics she already owned, she uploaded a photograph of her stock and of the items she was about to buy to the same AI. Instead of commenting on the shades, the AI told her the waterproof eyeliner she had used for years was blocking the pores of the gland, like wax in a drain, and making her dry eyes worse. She stopped, and this one change brought immediate relief. The AI connected across domains, unasked. Neither the physician nor the cosmetics counter had connected the two. The ophthalmologist saw the eye. The counter saw the product. AI saw the person. Neither was wrong. Each was trained to look at one domain. Every institution serves its own; AI serves the customer. This is not a failure of individuals but a structural shift with strategic implications. In this article we examine these implications.

Verification in markets

For as long as markets have existed, buyers have had to trust sellers, because there was no way to verify their word, or the effort and cost were too high. The economist George Akerlof spent many days puzzling over the nature of the vegetable retail market in Delhi. In 1970, he offered new insights into the market failure that arises when the buyer cannot verify what the seller knows. He later won the Nobel prize in economics for this work which helps us understand asymmetric information.

The market economy combats asymmetric information using substitutes, brands, professional licensing, statutory audits, regulators, and guarantees. For example, the customer trusts the bank's brand name instead of reading the fine print, or the doctor's medical degree instead of evaluating the diagnosis. Half a century of institutional architecture in financial services is built on this logic.

The use of AI has brought about a foundational shift in trust. AI has made verification in many situations quite feasible. Customers who once accepted substitutes for trust in institutions can now check a claim, compare alternatives, or challenge a recommendation on an inkling of doubt. The barriers to specialist access, cost, time, language, jargon, have all collapsed at once. The result is not an end of trust, but the pillars on which trust rests have changed.

How it impacts the brand

The brand is an informational shortcut, one of Akerlof's substitutes for verification. It stood in for the customer's inability to verify. The bank's or insurer's reputation stood in place of the customer's understanding of provider soundness, and the terms and conditions of the contract. AI changes that, but the picture is complicated, because the brand connotes two things at once: A promise of what we will do for you, and an aspiration of who you become or which tribe you join by choosing us.

The promise of what the brand will do is now more verifiable. The customer's AI checks every promise before purchase, searching the wider internet, cross-referencing user reviews and triangulating from multiple public sources. The bank that claims to be customer-first has its complaint-resolution record extracted from annual reports or instantly summarised from X. The insurer that promises easy claims has its claim-rejection ratio surfaced against peers. Promises that survive verification strengthen the brand. Promises that do not are revealed in seconds.

The aspiration is not in the product, it lives in the customer. People buy Apple products partly because Apple-ness signals something about themselves. Customers bank with a private bank not for any major service benefit but for who else banks there. People ride Royal Enfield partly because owning one says something no specification sheet captures. AI audits the promise. It cannot interpret the tribe.

The split deepens in the AI-to-AI world, where the customer's AI transacts with the institution's AI. The customer only experiences the outcomes, settlement speed, dispute rate, complaint-resolution time and median application-to-approval. It does not watch advertisements. In this perspective, brands need to invest in advertising that bolsters tribal loyalty, but advertising that is supposed to bolster the promise is now less important.

How it impacts labour

Like the brand, the professional is going to be hit by the AI wave. The professional of the old world was, by training, organised around the domain. The cardiologist gave the right answer to the heart in front of her, not to the medicine cabinet or the financial situation at home. The mutual fund agent recommended within his manufacturer's product set, not against the seven existing funds in the customer's portfolio. The AI can see the landscape comprehensively and the professional has to now compete with it.

In most domains the gap between the bottom and the top decile of professionals has been wide. AI compresses the gap from below. The advisor who tests his recommendation with AI before delivering it catches the portability clause he had not considered. The relationship manager who has the model argue the customer's case against his own pitch will close more often. Every word the professional says can now be cross verified; the smartest professionals will go up against 'unsophisticated consumers' with more respect.

In recent research, Brynjolfsson, Li and Raymond (2023) studied 5,179 customer-support agents at a Fortune 500 software firm and found that average productivity rose 14% with access to a generative AI assistant. Within this overall average, novice and low-skilled workers improved by 34%, while experienced and highly skilled workers showed minimal gain. By this reasoning, AI-powered unskilled labour will be tough competition against skilled practitioners.

The pattern is visible in India also. At one NBFC, the productivity of fresh-college LAP underwriters rose 40% with AI assistance, while experienced underwriters showed minimal gain. The bottom of the labour quality distribution comes closer to the top.

As the floor rises, the implications are twofold. First, the customer's worst experience disappears, and with it the customer's reason to switch providers. Second, the professional's competence becomes the table stakes. They have to now provide what AI cannot supply: Trust earned over time, judgement under ambiguity, the willingness to take a customer-friendly call when things are not going well with the business. These are the qualities that live on the aspiration side of the brand, and they are the real moats of professional competence.

When the asymmetry reverses

So far the shift has run one way, AI in the customer's hand against the institution. The same architecture runs the other way too. Institutions have always known what customers did. But AI interactions reveal something deeper, what the customer considered doing. The questions asked, the scenarios tested, the decisions abandoned. These are cognitive traces, and they sit closer to intent than anything an institution has had access to before. If applied to underwriting, pricing or customer acquisition, they create a new informational advantage that did not exist a year ago.

The trust consequence runs deeper than the privacy one. Behaviour is what the customer did once it was done. Intent is what she rehearsed before she was ready to be seen. A customer can absorb the knowledge that her behaviour was logged. When she learns the institution priced her, or declined her, on the strength of her question, the breach is of a different order. The first asymmetry was about information she did not have. This one is about information she did not know she was giving. This may lead the customer to stop being candid with the one tool that was working for her, because she now suspects it is also working for the firms. The independence this piece began with is the first thing she loses when the architecture turns around. We will have to face a new world of consumer protection complexity, going beyond the simpler questions of data privacy.

What then survives in trust? Not the part that rested on the customer's inability to check. What survives is what AI cannot manufacture. Judgement under ambiguity, the call no model will take responsibility for. A relationship proven over time, the banker who backed the customer through a bad cycle and was proved right. The human presence in a hard moment, the advisor who delivers difficult news with care. None of this can be read off a document, so none of it can be verified, and so none of it can be commoditised. The trust that survives is the trust that was never about information in the first place.

The institutions that Akerlof described were built on substitutes for verification. Those substitutes served a purpose. They filled a gap the customer could not fill herself. That gap is closing. What remains when the substitute is no longer needed is the thing the substitute was always standing in for. Genuine expertise, honestly applied, in the customer's interest. The institutions that had that all along have nothing to fear from the informed customer. The institutions that were selling the substitute will find, quietly and permanently, that the customer has stopped calling back.

The scarce asset is the question

If one risk is that the institution reads the customer's question, the other is that she asks a wrong question. While AI can reduce the information asymmetry, friction has not entirely disappeared. It rests on arriving at the right question to ask. Speed and convenience applied to the wrong question produce a confident wrong answer faster. The customer who compares home loans on interest rate alone misses the prepayment clause. The customer who has already decided to switch insurers asks questions that confirm the decision. The informed customer is powerful. The misinformed customer with AI is powerfully wrong.

The question itself is the unclaimed opportunity. No bank will build a question set that surfaces its own prepayment clause weaknesses. No insurer will build a question set that exposes its own claim-rejection ratio. This needs independent actors: non-profits, researchers, consumer bodies, who can make a GitHub for the questions consumers should be asking financial institutions.

What the Boardroom should debate

Every earlier wave of technology was institution first. Computing, internet, mobile was adopted and absorbed by the organisation and then passed onto the customer on their terms. AI exploded in consumers' hand, 100 million users in two months. The Board is now governing businesses where customers will know as much if not more than the organisation.

The first institutional response to AI has been operational, adding chatbots, analytical tools, dashboards, voice bots, service automation, that are good cost-saving initiatives. The strategic question is the viability of the business model itself.

What happens when the customer arrives informed? Which elements of the value proposition survive verification? Which revenue streams depend on customer ignorance or high search costs? Which promises would survive an AI audit? Which parts of the sales process assume an information asymmetry that no longer exists? These are not technology questions for the CTO. They are business-model questions for the Board.

The institutions that emerge stronger will not be those that adopted AI fastest. They will be those whose value remains after verification becomes cheap.

References

Akerlof, George A. (1970). "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism". Quarterly Journal of Economics, 84(3), 488-500. https://www.jstor.org/stable/1879431

Brynjolfsson, Erik, Danielle Li and Lindsey R. Raymond (2023). "Generative AI at Work". NBER Working Paper 31161. https://www.nber.org/papers/w31161

Reuters / Similarweb (2023). ChatGPT user-growth figures. https://www.reuters.com/technology/chatgpt-sets-record-fastest-growing-user-base-analyst-note-2023-02-01/

Reserve Bank of India. Annual Report of the Ombudsman Scheme. https://www.rbi.org.in/Scripts/AnnualReportPublications.aspx

Maninder Singh Juneja is a partner at True North. He serves on the boards of Pine Labs, Nivara Home Finance and Integrace, and has previously served on the boards of Niva Bupa Health Insurance, Federal Bank Financial Services and HomeFirst Finance. The author thanks participants at a talk at XKDR Forum for myriad good ideas, and Ajay Shah, Renuka Sane and Aditi Mascarenhas for comments on earlier drafts.

Examining grievances and redress for pension products

by Vimal Balasubramaniam, Aishwarya Gawali, Nancy Gupta, Renuka Sane and Srishti Sharma.

In a previous article, Examining grievances and redress for banking products, we studied the nature and extent of grievances for banking and payment products in India. We also evaluated whether grievance redress mechanisms worked, and what impact grievances had on the usage of products. In this article we study similar questions for contributory pension products. The analysis is based on a survey of 21,355 respondents that we conducted in five states including Maharashtra, Bihar, Haryana, Madhya Pradesh and Andhra Pradesh.

Measuring grievances

In the survey, we first ask if the respondent is using or has ever used contributory pension products. These would typically include the General Provident Fund (GPF), the Public Provident Fund (PPF), Employee Provident Fund (EPF) and the New Pension System (NPS). The study explicitly excluded various defined benefit pension plans, such as old age pensions, widow pensions, and disability pensions. We then ask the following questions:

1. If the respondent had faced an issue with any of the contributory pension products in the last 12 months?
2. If yes, what was the latest/most recent issue with the pension product?
3. Did the respondent complain after encountering the grievance?
4. Was the issue resolved after the first complaint?
5. If no, was the complaint escalated further?
6. If the complaint was escalated, was the issue resolved upon escalation?
7. If the respondent did not complain, what was the reason for not complaining?
8. Finally, what was the impact of the grievance on their usage of pension products?

We only consider complaints registered with the financial service provider or pension regulators. We do not include complaints filed in the police station or consumer courts as this is not in the ambit of the regulatory grievance redress system. Our questions do not pertain to any specific pension product. The results, therefore, are a reflection of the overall system of grievance redress, and not of any particular scheme.

Before we describe our results, it is useful to present the existing grievance redress mechanisms in the pension ecosystem. It is also important to note that while certain pension regulatory bodies report the incidence and resolution of grievance, there is no official consolidated statistic on the number of grievances for contributory pension schemes in India. Table 1 provides a snapshot of the governing regulatory bodies and grievance redress mechanisms (GRMs) of some of the major pension products that are relevant to our study.

Table 1: Pension products, regulatory body and GRMs

Pension Product	Regulatory Body	Grievance Redress Mechanisms (GRMs)
General Provident Fund (GPF)	Department of Pension and Pensioner's Welfare under the Ministry of Personnel, Public Grievances and Pensions	Online Grievance Lodging and Monitoring System at the Office of the Comptroller and Auditor General of India
Public Provident Fund (PPF)	Department of Post of India	Centralised Public Grievance Redress and Monitoring System (CPGRAMS) along with a dedicated Grievance Handling Cell accessible via call and email
Employee Provident Fund (EPF)	Employees' Provident Fund Organisation (EPFO)	EPF i-Grievance Management System (EPFiGMS)
New Pension System (NPS) and annuity schemes	Pension Fund Regulatory and Development Authority (PFRDA)	A multi-leveled Grievance Redressal System

Our analysis thus pertains to the use of contributory plans, which include, but are not limited to the schemes mentioned above.

Results

Our sample comprises of 21,355 respondents. 622 (2.92%)individuals reported having used a pension product. This is not surprising given that coverage through mandatory occupational pensions is low. However, there appears to be growing demand for micropension among poor families. This is reflected in our sample as well. Respondents with annual family income of less than one lakh rupees formed the largest share of pension users. 50% (314 out of 622) of pension users had an annual family income of less than one lakh rupees. 30% (187 out of 622) of the pension users had an annual family income between 1 to 3 lakh rupees. The remainder 20% had an annual family income of more than 3 lakh rupees.

Extent and nature of grievances

Of the pension users, about 11.4% (71 out of 622) reported having faced grievances related to pensions in the last 12 months. 61% (43 out of 71) of the grievances pertained to irregular or delayed pension payments, while 34% (24) of individuals claimed having not received their monthly pension during the last 12 months. About 6% (4) respondents faced grievance related to paper work issues.

From grievance to complaining and resolution

Table 2 presents the life cycle of pension related grievances - this helps us understand the working of the redress mechanisms, both at the level of financial service providers (FSPs) as well as the regulators. As described earlier, 622 respondents owned a pension product, while 71 had a grievance. Out of the 71, 59 (83%) complained to the FSP. The FSP was able to resolve 33 (56%)complaints. This implies that 26 complaints were not resolved. Of these only 8 (31%) were escalated to a higher authority, leading to a resolution of 5 (63%). Overall, this suggests that 33 grievances (46%) were not resolved - either because the respondent didn't complain at all, or because the problem was not resolved either at the FSP or regulatory level.

Table 2: Grievances, complaints and resolution

Pension Product	N
Own the product	622
Had a grievance	71
Complained to FSP	59
Resolved by FSP	33
Escalated to higher authority	8
Resolved upon escalation	5

We also explore the reasons why people do not complain when faced with a grievance with pension products. We focus on those respondents who didn't complain to the financial service provider/regulator. This doesn't include those who did not escalate their complaint after it was not resolved by the FSP. As seen in Table 2, 12 out of the 71 respondents who had a grievance chose not to complain. Of these 12 respondents, four felt that their problem would not get resolved sometimes because they didn't know if their problem was valid in the first place, while four were reluctant to access the process - either because they didn't have enough knowledge of the same, or because they felt the process was too costly and complex. These are very small sample sizes, and hence the results may not be generalisable.

Impact of grievance on usage

In Table 3, we present the impact on usage for those who had faced any grievance while using the pension products.

Table 3: Impact on usage of pension schemes

Impact on usage of pensions	N	%
Changed the provider	36	51
Stopped using the product	11	15
No change	11	15
Reduced the use of product	5	7
Do not know/wish to answer	5	7
Increased the use	3	5

Regardless of the respondent's course of action, the experience of having faced a grievance is bound to have an impact on the usage. As a result of encountering grievance, a majority of the respondents chose to change their service provider. 51% of those who had a grievance (36 out of 71) changed their service provider. 7% (5 out of 71) users having faced grievance reduced the usage of the pension products, while 15% (11 out of 71)stopped using the product.

Conclusion

GRMs in the pensions sector seem to be performing better than the banking and payments sectors. The incidence of grievances is lower, and the complaint rate is higher. However, banking and payments have a substantially larger number of users, and the products also get used more frequently than a pensions product. So it is not surprising that frictions in the banking space are higher. While the incidence of complaints may be lower in pensions, the impact of poor service may be higher on users, especially as the nature of grievances suggests that these occur later in life, when people may have limited means to solve the problem. This makes the numbers reported in the survey large enough to matter.

Vimal Balasubramaniam is a researcher at Queen Mary University, London. Aishwarya Gawali and Nancy Gupta are researchers at NIPFP. Renuka Sane is a researcher at Trustbridge. Srishti Sharma is a PhD student at Texas A&M University.

Examining grievances and redress for banking products

by Vimal Balasubramaniam, Aishwarya Gawali, Renuka Sane and Srishti Sharma.

Banks are witnessing persistent consumer complaints. These range from high service charges, lack of transparency in pricing and mis-selling. The regulatory system is currently not designed to capture the total number as well as nature of grievances. Without this information, it becomes difficult to design a policy solution.

We examine the grievance and redress experience for a heterogeneous set of consumers, with a large scale survey in five major states of India. Through this survey, we map the journey of a consumer's experience-- from usage and grievance to resolution. We also study the impact that a grievance has on subsequent usage of the product. In this article, we present information on the consumer's experience for a comprehensive set of banking and payment products that fall under the regulatory ambit of the Reserve Bank of India. These include:

1. Banking deposits.
2. Bank credit.
3. ATM/Debit Cards.
4. Net banking/ Phone Banking(including NEFT/IMPS/RTGS).
5. NBFC.
6. UPI Wallets.
7. Microfinance institutions.
8. Co-operative credit societies.

The evidence from our study can form the basis of a more responsive system of grievance redress for retail consumers.

The survey design

Our survey was conducted in Maharashtra, Bihar, Haryana, Madhya Pradesh and Andhra Pradesh. We used a multi-stage stratified sampling method to draw the sample of households. The Primary Sampling Units (PSUs) were the villages for rural areas and census enumeration blocks (CEBs) for urban areas. The Ultimate Sampling Units (USUs) were the households from these PSUs.

The 2011 Census served as the sampling frame for the identification of the districts within each state. All the districts in a state were divided into terciles on the basis of distribution of households using banking deposits, curated from the RBI data across four quarters of 2020-21. To ensure proportionate distribution in each tercile, two districts were picked from each tercile using systematic random sampling. This exercise was repeated for each of the five states. In states such as Maharashtra, Bihar & Haryana where one district held a substantially high proportion of deposits, it was treated as two districts and over sampled to account for the large proportion of deposits. Accordingly, we got a sample of six districts each from Madhya Pradesh and Andhra Pradesh, and five districts each from Maharashtra, Bihar and Haryana which gave us a total of 27 districts.

Within the district, the allocation of the sample between villages and CEBs was proportional to the rural-urban distribution of the population. Villages in a district were stratified on the basis of distance from district headquarters and CEBs were stratified on the basis of share of CEB in the district's urban population. Three strata were created on the basis of the above mentioned criteria, for both villages and CEBs. The number of households from each strata was selected in proportion with the population share of each strata. So if one village strata has 40% of the rural population, then 40% of the rural sample of the district came from that strata.

We collected information on the demographics, physical and financial assets, and liabilities of the household. The core module of the questionnaire focused on experience of consumers with grievances & redress regarding financial products. The total sample size was 21,355 respondents.

Measuring grievances in the banking system

According to the current grievance redress system for banks, NBFCs and Prepaid Payment Instruments, consumers must first lodge a complaint with their service provider. If the service provider is unable to provide resolution in a satisfactory manner in 30 days, the consumer may escalate the complaint to the Ombudsman. The Ombudsmen for banks, NBFCs and digital payments have been harmonised under the Integrated Ombudsman Scheme in 2021. Consumers can lodge their complaints with the Ombudsman using the Complaint Management System (CMS) portal or by using a complaint form. We asked the following questions to understand the consumer's experience at each step of the grievance redress process:

• First we asked if the respondent is using/has ever used the mentioned product. This ensures that we also capture past users of a financial product.
• We asked if they have faced an issue with the mentioned financial product in the last 12 months. This helps us capture the grievances faced by consumers.
• We then asked the respondent what their latest/most recent issue was for the mentioned financial product.
• We also asked them their first course of action after encountering the grievance. Through this question, we understand how many people complain after facing a grievance. We consider that a respondent has complained if they went to the service provider or regulator with their issue. We do not include complaints to the police or consumer courts as this is not in the ambit of the regulatory grievance redress system.
• Additionally, we asked if their issue was resolved after their first complaint.
• For those who did not receive resolution at this stage, we asked if they escalated the complaint to a higher authority.
• For those who escalated their complaint, we asked if they finally received resolution.
• To all those who faced a grievance, we asked what the impact of the grievance was, on their usage of the product.

Overview of grievances in the banking system

In Table 1, we describe the extent of usage and grievance for the various banking and payment products. Columns (1) and (2) provide the number and percentage of the sample who have used a particular financial product. Columns (3) and (4) describe the number and proportion of users that reported that they faced an issue/grievance related to the financial product in the last 12 months.

Table 1: Usage and incidence of grievance

Product	Usage		Incidence of grievance
	(1)	(2)	(3)	(4)
	N	%	N	%
Banking deposits	17407	81.51	2112	12.13
ATM/Debit Card	8625	40.39	1279	14.83
Netbanking (IMPS/NEFT/RTGS)	3161	14.80	503	15.91
UPI Wallets	2825	13.23	531	18.80
Bank credit	1640	7.68	242	14.76
MFI	961	4.50	104	10.82
NBFC	448	2.10	82	18.30
Cooperative credit society	386	1.81	72	18.65

Banking deposits were the most used product, followed by ATM/Debit cards. Netbanking was the third most used product, followed closely by UPI wallets. All the others were used by less than 10% of the sample each. Co-operative credit societies had the smallest share of users.

The incidence of grievance ranged between 10% to 19%. Even though bank deposits were the most used product, they had the second lowest incidence of grievances. The highest grievance rate was for UPI wallets, at about 19%. Co-operative credit societies had the second highest incidence of grievances, even though the usage of these products was the lowest in our sample. The same is true for NBFCs as well - only 2% of the sample used NBFC products, but 18% of these had faced a grievance.

Nature of grievance

In Table 2, we describe the nature of grievances faced by consumers. We select the top 3 grievances for each product and present the number and proportion of consumers who faced the given issue.

For banking deposits, 28% of those with a grievance had an issue related to transaction failure. 26% of the issues were related to charges being deducted without information. More worryingly, 12% of the issues were related to difficulties with opening a bank account. This has implications for financial inclusion as such issues may dissuade people from participating in the formal financial system.

Table 2: Nature of grievances

Product	N	%
Banking deposits
Failures or delays in transactions	585	28
Deductions or charges without information	555	26
Difficulty in opening bank account	244	12
Bank credit
Loan/interest rates	130	54
Fraud, hidden charges etc	102	42
Other, specify	10	4
ATM/Debit card
Lack of cash in ATM	490	38
ATM closed/non functional	259	20
Server down	156	12
Netbanking
Server down	204	44
Delays in services	92	20
Money deducted but transaction failed	85	18
UPI Wallets
Server down	188	35
Delays/Failure of transactions	115	22
Transaction failed, but money debited	99	19
NBFC
Complex terms and conditions/no adequate notice	36	44
Non transparency in contract/ loan	23	28
No communication about loan sanctioned	7	9
MFI
Charged higher interest rate than informed	56	54
Threat to increase interest rates	27	26
Painful recovery process	12	12
Co-operative credit society
Charged higher interest rate than informed	42	58
Threat to increase interest rates	23	32
Painful recovery process	6	8

For bank credit, the dominant issues were interest rate related, however, 42% of consumers reported having faced fraud which is a far more serious nature of grievance.

For ATM cards, 38% of the issues are related to lack of cash in ATMs, 20% were related to dysfunctional ATMs and 12% of the issues came up because the server was down. 44% of netbanking issues were also because the server was down. While transaction, server and service infrastructure related issues dominate in case of payment products such as UPI and net banking, interest rate and loan contract related issues are significant for credit products.

From grievance to complaints and resolution

In Table 3, we report the number of people who complained to either the service provider or regulator when faced with the grievance. Column 2 presents the total number of grievances for the given product. In Column 3, we report the number of users who complained to either their service provider or regulator. In Column 4, we report the number of consumers whose complaint was resolved at the first stage itself. In column 5, we report how many of those who escalated their complaint reported that their issue was resolved.

Table 3: Complaining, resolution, escalation and final resolution

	(1)	(2)	(3)	(4)	(5)
Product	Had a grievance	Complained to FSP	Resolved by FSP	Escalated to higher authority	Resolved upon escalation
Banking deposits	2112	1064	661	88	47
Bank credit	242	201	104	61	6
ATM/Debit Card	1279	521	410	39	22
Netbanking(IMPS/ NEFT/RTGS)	503	252	180	24	13
NBFC	82	72	43	8	4
UPI Wallets	531	187	141	9	6
MFI	104	69	34	8	4
Cooperative credit society	72	64	37	7	5

Out of the 2112 consumers who faced a grievance regarding banking deposits, 1064 (50%) complained, which is how we define the complaint rate. Of these, 661 (62%) reported that their problem was resolved after their first complaint, which is how we define the resolution rate. This suggests that 403 users' complaints were not resolved in the first instance. Of these, only 88 (21%) escalated the complaint. Of these 88, 47 (53%) reported that their problem was resolved after escalation. The other half of the complaints that were escalated remain unresolved.

Banking deposits have the highest number of complaints. However, co-operative credit societies had the highest complaint rate at 88% followed by NBFCs at 87%. ATM card complaints have the highest rate of resolution at the first stage -- 410 out of 521 complaints (78%) got resolved at the first stage. This is followed by UPI wallets at 75% and netbanking at 71%. High resolution rates suggest that grievance redress at the first point of contact, which is usually the Financial Service Providers (FSPs) is performing efficiently. MFIs have the lowest resolution rate (49%) at the first stage, and points to the deficiencies in the redress system.

Reasons for not complaining

In the previous section we examined what happens to the complaints that enter the official grievance redress system. However, it is evident that not all grievances turn into complaints. What about the users who do not lodge a complaint?Table 4 shows the reasons why people do not complain when faced with a grievance. The rows show the number of users who did not complain for the reason given in the column.

For banking deposits, the main reason for not complaining was that users did not know the process of grievance redress. 38% of those who did not complain, did so because they did not know the process. For bank credit, the costly and complex nature of the process was the main reason for not complaining with 34% users not complaining due to this reason. 36% ATM/debit card users who did not complain did so due to the complicated and expensive nature of the grievance redress process. 26% did not complain because they did not know the redress process. Another 15% did so because they were not sure about whether their problems would be resolved. For netbanking and phone banking, 33% users did not complain because the redress process is too costly and complex, 15% did not complain because they were not sure about their issue being resolved and 17% because they didn't know the process.

Product	Did not complain to FSP	Costly and complex process	Did not know validity of complaint	Do not know/wish to answer	Fear of retribution	Resolution unlikely	Unknown process	Was advised not to by friends family
Banking deposits	1048	177	88	96	38	188	408	19
Bank credit	41	14	6	-	2	6	5	2
ATM/Debit card	758	278	71	31	17	120	198	17
Netbanking	251	85	25	18	7	43	39	12
UPI wallets	344	114	41	9	5	60	94	15
NBFC	10	1	-	-	2	2	2	-
MFI	35	-	1	3	2	8	19	-
Co-operative credit society	8	-	-	1	1		2	-

In the case of UPI wallets, the expensive nature of the process, lack of information about redress procedures and the prospect of resolution being unlikely, were the main reasons for not complaining. For NBFCs, 20% of users who did not complain, did so because of the fear of retribution, another 20% did not complain because they were not sure about whether their problems would be resolved. Finally, 20% did not complain because they did not know the process. For MFIs, 54% of users did not complain because they were unaware of the process and for co-operative credit societies, this number was 25%.

Impact of grievance on usage

The experience of having faced a grievance is bound to have some impact on the consumer's usage of the product. In Table 3, we present the number and proportion of people who either changed their provider, reduced usage of the product or stopped using the product after facing a grievance. These actions indicate that the grievance had an adverse impact on the user. Grievances related to deposit or payment products lead about 30% of consumers to take some action. The response by consumers is higher for credit related products.

Table 4: Changed provider/reduced/stopped usage

Product	Had a grievance	Took action
(1)	(2)	(3)	(4)
	N	N	%
Co-operative credit society	72	63	88
NBFC	82	65	80
MFI	104	65	62
Bank credit	242	129	52
Netbanking	503	166	34
UPI Wallets	531	173	31
Banking deposits	2112	666	31
ATM/Debit card	1279	388	29

In the case of co-operative societies, 88% of those who faced a grievance either changed their provider or reduced or stopped using the product as a result of it. This indicates the co- operative society members who faced a grievance may not have had a satisfactory experience with the grievance redress process. This number stands at 80% for NBFCs and at 62% for MFIs. The impact of grievances for non-bank lending institutions is far more adverse than for any of the other products.

Conclusion

Improving outcomes for consumers is one of the core goals of finance. It is important to understand how the system deals with grievances of consumers, and where there is scope for improvement. Our results present a heterogenous picture. We find that the usage of deposit and payment products is higher than credit products, while the grievances are higher for credit products. Deposits are the most used product but have an incidence of grievance of 12%. Co-operative societies are used by less than 2% of the sample, yet almost 19% of its users have faced a grievance. NBFCs are used by just about 2% of the sample, and have an incidence of grievance of 18%. Deposit and payment related services are able to resolve grievances faster - more than 70% of the complaints were resolved in the first instance. This is not true of credit related products - for example, only 49% of the complaints were resolved for MFIs. Further research could explore the possible reasons for this heterogeneity.

Renuka Sane and Aishwarya Gawali are researchers at NIPFP. Vimal Balasubramaniam is a researcher at Queen Mary University, London. Srishti Sharma is a PhD student at Texas A&M University.

Grievance Redress by Courts in Consumer Finance Disputes

by Karan Gulati and Renuka Sane.

India has made progress on financial inclusion through the use of digital payments and fintech. As more and more consumers interact with the consumer finance industry, there will invariably be greater frictions and an increasing number of grievances. In an environment with a good consumer complaints system, these should get resolved by the financial service provider (FSP), and if not the FSP, then the regulator. However, this is not so in India. Courts are often the preferred recourse for retail consumers. For example, in the ongoing dispute regarding Yes Bank's written off AT-1 bonds, consumer courts seem like the last remaining alternative for retail investors. Unless grievances are satisfactorily resolved, we may hurt the progress made on financial inclusion. While India needs to set up good regulator-based grievance redress mechanisms such as a Financial Redress Agency, it also needs to improve the functioning of courts to provide effective relief in consumer finance (and other)disputes. In a recent paper, Grievance Redress by Courts in Consumer Finance Disputes, we review 60 judgments on consumer finance to study the position that courts have taken on these disputes. We also describe the challenges in court functioning that have a bearing on the efficiency of courts in dealing with issues of grievance redress.

The structure of courts

In 2020, India enacted a new Consumer Protection Act (CPA). The Act aims to protect consumers' interests and provide timely and effective settlement of disputes. It entrusts courts to redress consumer grievances. A complainant can approach specialised courts i.e. consumer commissions established by the CPA. However, these are additional remedies. Cases may also be decided by the High Court of various States and the Supreme Court of India.

The powers to grant relief depend on which court the complainant approaches. Consumer commissions are bound by the CPA. They may order a party to: (i) remove defects, (ii) return the price of the goods or the charges for the services along with interest, (iii) pay compensation or punitive damages, and (iv) withdraw the goods or services from the market. High Courts are bound to decide cases either within the confines of a statute under which they are approached or the constitution. Going one step further, the Supreme Court has held itself not restricted in any way to grant adequate relief.

Banking and insurance disputes

Litigation is disproportionately costly and troublesome for small consumers. Very rarely can an ordinary consumer go through the prolonged ordeal of fighting with a bank. For this reason, courts have granted relief to individual consumers, given that they come with clean hands.

This has not been the case when interpreting insurance contracts. If consumers knew about the terms, courts have enforced the terms of the contract, regardless of whether the terms themselves were unfair, one-sided, or opaque. On the other hand, if the terms were kept hidden from the consumer, courts have granted relief to consumers. This is true both while entering the contract and settling claims.

Several consumers have been introduced to complex products and contracts, but these consumers have insufficient know-how. They are vulnerable to mis-selling. The strategy in Indian finance has historically focused on the caveat emptor doctrine -- let the buyer beware. Though the new CPA gives consumer commissions the power to declare certain unfair terms as void, it does not address the ability to understand the terms. Thus, consumers have been left to their own devices, and unaware consumers are unlikely to get their desired remedy if they approach a court.

Challenges to court functioning

We find the following challenges in court functioning as they deal with consumer finance disputes.

1. Low Compensation: Courts tend to award low compensation that does not adequately compensate the complainant. For example, in Dr Virendra Pal Kapoor v. Union of India and Ors, a senior citizen had invested INR 50,000 in a unit-linked product in 2007. Upon payout in 2012, he had lost the entire sum except INR 248 on account of hidden charges. Though the insurer was directed to repay the original Rs. 50,000, no interest was awarded. The reason for low compensation seems to be that there are no guidelines for courts to follow. There is no expert analysis of the loss. In the absence of financially prudent legislation, courts often tend to award compensation that only makes sense when the legislation is enacted.

2. Delay: Low compensation becomes more severe when it takes too long to settle disputes. The CPA provides that cases should be decided in no more than five months. However, as per the case management system of the National Commission, it takes 1.99 and 2.38 years to settle banking and insurance disputes, respectively, i.e. more than five times the statutory guideline. In fact, in February 2020, the National Commission adjourned a matter till January 2021 - almost a year after the hearing.

3. No Class Action: If consumers cannot understand complex financial agreements, they may benefit from pooling their knowledge and approaching courts as a class. Plaintiffs can share evidence, expert witnesses, and litigation costs. However, unlike other countries, such suits are few and far between in India. This may be because of unclear substantive law and strict rules on financing litigation. This makes it difficult for class members to come together. Courts have left it to their discretion to evaluate whether the class is adequately represented and whether financing agreements are fair. Moreover, the legislature had prohibited contingency fees. This creates a system that either prohibits or disincentives class actions.

4. Specialisation: Consumer courts in India resolve all consumer disputes. Though the members are highly qualified individuals, they lack specialization in finance. This is unlike other common law countries where sectoral experts adjudicate finance disputes. They have adopted extensive adjudicatory legislation regarding financial products and services. On the other hand, laws in India regarding finance have been restricted, leaving courts to start from a clean slate. If timeliness and predictability can make India's finance regime more appealing, specialization by adjudicators could prove valuable.

Way forward

One obvious way to improve the system is by general improvements in the judiciary's capacity and knowledge on matters related to finance. This will, however, take a long time. Policymakers should also consider adopting certain targeted interventions.

There are two types of interventions that are required. The first is on the legislative front. Like the targeted legislation in other countries, the legislature could enact separate rules for financial transactions mandating clear and understandable disclosures. Policymakers may also consider prescribing adequacy requirements in class action suits and transitioning towards contingency fees for lawyers and third-party investors. Any such changes in legislation would also benefit from an advisory council on consumer finance. The council may be responsible for making representations about policies; reviewing, monitoring, and reporting their effectiveness; and highlighting its views on new rules and regulations.

The second is on the judicial front. One problem we identify is low compensation. This may be addressed by updating and consolidating the rules governing compensation considering modern market understanding. Other jurisdictions often order disgorgement (surrender of profits earned through illegal means) or grant a remedy of restitution. This seeks to measure actual damages. On the question of delays, courts may also separate their judicial and administrative functions. This will likely reduce the time it takes to conclude hearings since members of the commission would have more time to focus on their judicial tasks. The National Commission can also exercise its power to call for statistics from State Commissions and conduct systematic reviews.

These solutions can have significant consequences, especially in India, where financial literacy is low and regulatory enforcement appears weak. Though they were developed after studying consumer finance disputes, they may have consequences outside this domain and yield better functioning courts. Market-oriented compensation, without delay, when parties can come together as a class would be beneficial in any dispute. In a growing financial landscape such as India, redress bodies such as the judiciary become increasingly important. A specialized consumer protection law is a step in the right direction, but it can benefit from targeted interventions.

References

Department of Economic Affairs, Report of the Financial Sector Legislative Reforms Commission: Volume 1, March 2013.

Dhirendra Swarup, Establishing the Financial Redress Agency, January 27 2017, The Leap Blog.

Dr Virendra Pal Kapoor v. Union of India and Ors, May 29 2014, Allahabad High Court.

Karan Gulati and Renuka Sane, Why do we not see class-action suits in India? The case of consumer finance, May 03 2020, The Leap Blog.

Karan Gulati and Shubho Roy, India's low interest rate regime in litigation, March 11 2020, The Leap Blog.

Murali Krishnan, Supreme Court urges consumer forum to look into grievance of year-long adjournments, August 16 2020, Hindustan Times.

National Informatics Centre, Computerization and Computer Networking of Consumer Forum in the Country.

Neil Borate, Those mis-sold Yes Bank AT1 bonds face long haul, May 11 2020, LiveMint.

Pratik Datta, Mehtab Hans, Mayank Mishra, and others, How to Modernise the Working of Courts and Tribunals in India, March 25 2019, NIPFP Working Paper No 258.

Reserve Bank of India, National Strategy for Financial Inclusion, January 10 2020.

Supreme Court Bar Association v. Union of India, April 17 1998, Supreme Court of India.

Tinesh Bhasin, RBI sees 387% rise in complaints against NBFCs, 58% rise against banks, February 08 2021, LiveMint.

The authors are researchers at NIPFP.

Why do we not see class-action suits in India? The case of consumer finance

by Karan Gulati and Renuka Sane

Mis-selling of financial products is pervasive in India and across the world. Sound grievance redress systems are one path to ensuring a degree of consumer protection. For example, complaints by customers to the Financial Ombudsman Service in the UK on Payment Protection Insurance paved the way for redress.

Class-action suits are another important means of seeking redress. For example, Bank of America was accused of charging excessive overdraft fees. Consumers of the bank got paid USD 410 million in 2011 as a result of the class-action suit on this issue. J P Morgan also had to settle a case on similar allegations for USD 110 million. Citizen Bank agreed to pay USD 137.5 million.

In India, too, we have seen several instances of mis-selling. The sale of Yes Bank's risky AT1 bonds as guaranteed return bonds is a recent example. To the best of our knowledge, consumers have not initiated a class-action suit for any of the mis-selling episodes in India.

At best, courts have taken it upon themselves to grant a class-wide remedy. For example, in Dr Virendra Pal Kapoor v. Union of India and Ors, a senior citizen had invested INR 50,000 in a unit-linked product in 2007. Upon payout in 2012, he had lost the entire sum except INR 248 on account of hidden charges. He had been mis-sold the policy without any caution. The court declared the policy to be void. It also directed the regulator, the IRDAI, to re-examine all policies issued by the specific insurance provider. If it detected regulatory breaches, it was to wind up the business of the firm. The apex forum, however, dismissed the class remedy without offering a reason.

In this article, we examine the reason behind the lack of class action suits in India. We argue that this is because of two issues. First, the substantive law is not clear. This makes it difficult for class members to come together. Second, procedural issues limit the financing of such cases. The issues we raise are pertinent to all aspects of consumer protection: from health to the environment. In this article, we combine the general treatment of class action with features specific to financial consumer protection.

Why is class action important?

Civil litigation is important for two reasons. First, the threat of litigation serves as a deterrence from injuring others. Second, it provides insurance to the injured when deterrence has failed.

When claims are small, plaintiffs may not be able to undertake individual litigation. In such a case, the plaintiffs do not get a chance to seek a remedy. This collective action problem is solved using class action litigation. As Fitzpatrick, 2010 describes, class-action allows claims to get aggregated. This is especially important when parties do not enjoy an equal bargaining power, as is the case in consumer finance. Plaintiffs can share resources such as evidence, expert witnesses, and the costs of litigation. To the extent that class actions permit disputes to go forward that might not have done so individually, they provide the possibility of insurance to the plaintiffs. Class-action suits also help ultimate recoveries to be close to the cost of injuries. This is because plaintiffs can keep more of their awards for themselves.

The ability to go to courts for private resolution between different parties reduces the need for the administrative state. This is because if people can solve disputes in courts, the rationale for concentrating power in the hands of a regulator, and the subsequent creation of mini-states does not remain (Kelkar and Shah, 2019). Class action suits, thus, serve an important function over and beyond the relief that is made available through the suit.

Institutional framework required for class action suits

For class action to work, the institutional design has four pre-requisites.

• Identifying members: The first is the possibility to identify members of the class. The burden of identification is usually placed on the plaintiffs, which courts later certify. Identification is non-trivial and varies from case to case. Members connected through a transactional relationship are easier to identify. Fitzpatrick, 2010 showed that more than three-fourths of all class actions were based on cases where it was possible to identify the class by back tracing the contract. Identifying members aggrieved by mis-selling is thus easier than identifying those who have suffered health issues in an environmental dispute. Courts may have a concern about how a class has been identified. In this case, courts could allow the plaintiffs to draft a workable definition of members of the class. It need not be important to identify every single member at the time of certification. This determination can be made when new members join the suit.
  
  
• Aggregation of claims: The aggregated claims should represent a substantial portion of the full class (also termed as the adequacy of the class). This is because a class action by its very nature is "representative". The question of whether a suit represents a substantial portion does not have one easy answer. As a practical matter, courts should rarely need to worry about it. Few lawyers would want to waste their time pursuing class certification (with its hurdles) for a small number of claimants. Hence, the instances in which adequacy is a valid reason to reject the claim should be rare. If courts are unhappy about the adequacy of the class, they should allow plaintiffs to make a representation in this regard.
  
  
• Incentive alignment: In individual cases, clients approach the lawyer. In a class-action, it is more likely that lawyers solicit work from a class. Victims of a class seldom have much in common besides the injury. As a result, an informed referral process may not develop. The principals (the members of the class) may not be able to act as good monitors of the agent (the lawyer). The lawyer may have an incentive to engage in self-dealing (Lahav, 2003). Contingency fees solve the incentive problem by linking the lawyer's fees to the amount of benefit she provides to the class. This is especially important in consumer class actions where client cohesion is unusual.
  
  
• Meeting expenses: Lawsuits can be both expensive and risky. A class-action does not guarantee that members will be able to bear all expenses. Legal requirements may mean that members have to provide specific evidence individually. Litigation may also carry on for a long time leading to an increase in expenses. And it is always possible that members lose the suit. The legal system should allow expenses to be borne through "third-party funding". Contingency fees, discussed above is one element of it. A second element is raising finances from companies (such as Vanin Capital, IMF Bentham) specializing in investing in class-action litigation. If members win, they share their proceeds with the firm in return for financing the suit. The companies are in a better position to manage the risk of loss of the suit than class members.
  
  

The law in India

The Code of Civil Procedure, 1908 provides for representative suits where one or more persons can sue on behalf of all those who have a common interest or grievance. Such suits are also provided for under several other laws with varying scope. Shareholders and depositors may file a case for oppression and mismanagement under the Companies Act of 2013. Under the Consumer Protection Act, 1986, a consumer can file an action on behalf of all other interested consumers before a consumer court. A suit may also be filed under the Competition Act, 2002 to challenge anti-competitive agreements and market positions. The scheme of class actions suits may hence be summarized as follows:

Table 1: Scope of Laws governing Class Actions

Law	Subject Matter	Class	Example
Code of Civil Procedure	There are no limits on the subject matter except for actions that cannot be filed in the civil courts at all, such as mismanagement suits.	Persons having the 'same interest' in the suit	Excess demand by housing board
Companies Act	A suit can only be brought for oppression and mismanagement of the company but does not include a banking company.	Shareholders and Depositors in the Company	Depriving shareholders of their right to dividends
Competition Act	A class may dispute an agreement which causes an appreciable adverse effect on competition within India or abuse of dominant position by an enterprise.	Any person, consumer, or their association	price-fixing, output limitation, market sharing, and bid-rigging
Consumer Protection Act	The suit is restricted to goods and services sold/provided or delivered or agreed to be sold/provided or delivered.	Consumers of the goods or services	Mis-selling of products by a banking or insurance company

As the table shows, the subject matter and class depend on the law under which the suit is sought to be filed. However, there are two problems with this system:

1. A Representative Class: Persons who approach the court in a class-action suit need to represent an adequate portion of the class. The National Consumer Dispute Redressal Commission (NCDRC) has said that it would not permit a case if only 10 persons out of a class of 100 wish to litigate. They argue that if they accept the case, the other 90 would have to either file individual complaints or file on behalf of another class (Ambrish Kumar Shukla v. Ferrous Infrastructure). One could, however, argue that the other 90 could always opt-in to the action already initiated, or the court could club matters if two class-actions are initiated. This standard is also difficult where the class is likely to be millions of customers. For example, consider a dispute between a bank and its million customers over fees charged by a bank. While 10 out of 100 injured parties may seem inadequate, it is hard to argue the same if 100,000 customers out of a million formed a class. This issue is not unique to consumer disputes. The Companies Act prescribes a high adequacy standard if shareholders want to initiate class actions for oppression or mismanagement. The class needs to include at least 5% or 100 shareholders of the company. This may be difficult to meet since such cases are usually filed by minority shareholders.
   
   
2. The new Consumer Protection Law, 2019: India enacted a new consumer protection law in 2019. Unlike the erstwhile law which permitted a class to initiate a case before a consumer commission in cases of mis-selling, the 2019 law establishes a new regulator in the regime of consumer protection i.e. the Central Consumer Protection Authority (CCPA). The CCPA is tasked with protecting and enforcing the rights of consumers as a class. As per section 17 of the new Act, a complaint relating to violations of consumer rights prejudicial to the interests of consumers as a class is to be forwarded to the CCPA. It would then conduct a preliminary inquiry as to whether there exists a prima facie case of violation of consumer rights and instruct for an investigation to be conducted. This has taken away the power to initiate class actions from individuals and vested them into the hands of the regulator. Unlike earlier, where a class of consumers could approach consumer commissions with their common grievance, they are now required to meet the subjective satisfaction of the CCPA. This is then meant to result in an investigation, and consequent orders, if any. The difficulties of public management now impact the enforcement process in consumer grievances. Persons who have suffered harm are now supplicants before the regulator, requesting it to enforce consumer law. Several steps have been added in the process, which could lead to a lesser filing of class action suits.

Banking companies have been given additional protection against class actions. Though consumers of such companies can initiate class actions in cases of mis-selling subject to the above challenges, shareholders have been restricted from bringing any class actions. The Companies Act introduced in 2013 provides for class action suits by shareholders for oppression and mismanagement of a company. However, the Act explicitly bars any class action against a banking company in such cases. Interestingly, this is the case even when there is no bar on an individual shareholder of a banking company from bringing a claim of oppression and mismanagement. Hence, shareholders have to bring multiple cases such as "A v. Banking Co", "B v. Banking Co", so on and so forth. They cannot file a case as a class such as "Shareholders of Banking Co v. Banking Co". Thus, all that the law has done is to make sure that shareholders of banking services are unable to pool their resources.

Procedural and financial hindrances

Solving the substantive issues listed above will not lead to class-action suits. This is because of the incapacity of people to finance such disputes and regulations on how to do so.

1. Stamp Duties: Litigation is expensive. One reason for this is the stamp duty payable for the same. Stamp duty is a tax on the value of instruments used in various business transactions. There are two kinds of stamp duties: (i) judicial stamp duties, and (ii) non-judicial stamp duties. Judicial stamp duties are fees collected from litigants in courts. These are best viewed as court fees and act as the cost of bringing an action. They may be prohibitive. For example, the fees payable in Delhi for a plaint (the first document submitted in court for a case) has been set at 4% of the value claimed. Fees are also to be paid in cases of review or appeals. There may be charges for obtaining copies, translations, additional applications, etc. The law of evidence requires the payment of non-judicial stamp duty for all documents submitted in court. These costs add up and would become prohibitive for a million customers. In a case like that of the Bank of America mentioned above, a claim for USD 410 million would need a fee of at least USD 16.4 million.
   
   
2. Third-Party Funding: Third-party funding ("TPF") is the act of a party outside the litigation paying for its cost. If the litigation is successful, the party gets a share in the award. This becomes important on account of the increased costs of litigation. When parties are not able to afford the dispute themselves, they should be able to turn to third-parties for funding. In 2018, the Supreme Court in Bar Council of India v. AK Balaji noted that there was no limitation on third-party funding. The Code of Civil Procedure, as amended by some Indian states including Gujarat, Karnataka, Madhya Pradesh, and Maharashtra, explicitly recognizes the role of a financier of litigation costs of a plaintiff. It also sets out the circumstances when such a financier may be made a party to the dispute. However, there is no central law on TPF in India. As a result, there is considerable uncertainty on whether the courts will hold the TPF agreement as "just". As early as 1876, the court held in Ram Coomar Coondoo v. Chunder Canto Mookerjee that:
   
   
   

   "agreements of this kind ought to be carefully watched, and when found to be extortionate and unconscionable, so as to be inequitable against the party; or to be made, not with the bona fide object of assisting a claim believed to be just, and of obtaining a reasonable recompense therefore, but for improper objects, as for the purpose of gambling in litigation, or of injuring or oppressing others by abetting and encouraging unrighteous suits, so as to be contrary to public policy, [the] effect ought not to be given to them."

   
   Courts have left it to their own discretion to examine whether the financing agreement is just and fair. In the absence of statutory requirements, courts usually lay down legal tests to determine a question of law. This allows parties to predict the behavior of the courts and make appropriate arrangements. There are no tests to determine the appropriateness of TPF agreements. Their validity is entirely up to a judge's concept of just-ness, leading to the TPF market not evolving.
   
   
3. Contingency Fees: Contingency fees is the fees of the legal counsel as a stake in the outcome. This is prohibited in India. This is problematic as lawyers do not have an incentive to argue unless their fees are paid. This means that fees must be paid upfront. A class action with a high claim is likely to be argued by a senior member of the bar. Appearance costs may be to the tune of INR 1.2 million. With an average of one hearing every two months, this would be INR 64.8 million (USD 850,000) for nine years (the average time of a civil case in India). Further, TPF funders usually seek contingency fees of legal counsel as this ensures alignment of interests. The lack of contingency fees also has an adverse effect on TPF.
   

Besides these reasons, India follows the loser-pays rule in litigation (Law Commission of India, 2012). The unsuccessful party is ordered to pay the costs to the successful party. There is some merit in this as it restricts vexatious litigation. But for class-actions, the class has to worry about paying the defendant's attorney's fees and adjoining costs if it loses the case, even though Indian courts award low costs (Law Commission of India, 2012).

Conclusion

The laws in India create a system which either prohibits or disincentives class actions. This article is not a definitive finding on how to cure such a situation; however, our analysis shows that the two reasons for the absence of class action in India require independent solutions.

To achieve a sound law on class action, two changes have to be brought to Indian legislation. Laws that allow for such suits may provide for what constitutes an adequate portion of the class to approach a court. Further, the new consumer protection law could give more clarity on what constitutes a prima facie case of violation of consumer rights and the elements of the investigation thereon. We need to explore the possibility of transitioning away from the loser-pays principle in class actions and toward contingency fees for lawyers and third-party investors.

These reforms have the potential to pave the way for class action suits in a wide range of areas. They are also an extremely important pillar in the system of grievance redress in consumer financial markets to protect millions of customers against egregious behavior by financial firms.

References

Ambrish Kumar Shukla & 21 Ors v. Ferrous Infrastructure Pvt Ltd, January 19 2016, NCDRC.

Bar Council of India v. AK Balaji, March 13 2018, Supreme Court of India.

Coral Gables, $137.5 Million Settlement Announced In Citizens Bank Overdraft Fee Class Action, Lexis Nexis.

Dr Virendra Pal Kapoor v. Union of India and Ors, May 29 2014, Allahabad High Court.

Fitzpatrick, 2010, An empirical study of class action settlements and their fee awards. Journal of Empirical Legal Studies, 7 (4), pp 811-846.

Fitzpatrick, 2010, Do Class Action Lawyers Make Too Little?, University of Pennsylvania Law Review, 158 (7), pp 2043-2083.

Jonathan Stempel, BofA $410 million overdraft settlement wins court OK, May 24 2011, Reuters.

Jonathan Stempel, JPMorgan settles overdraft fee case for $110 million, February 07 2012, Reuters.

Kelkar and Shah, 2019, In Service Of The Republic: The Art And Science of Economic Policy, Penguin Random House India Private Limited.

Lahav, 2003, Fundamental Principles for Class Action Governance, Ind. L. Rev., 37, p 65.

Costs in Civil Litigation - Report No 240, May 2012, Law Commission of India.

Ram Coomar Coondoo v. Chunder Canto Mookerjee, June 30 1876, Privy Council.

Shreeja Sen and Deepti Bhaskaran, SC stays Allahabad HC order on scrutinizing SBI Life policies, July 15 2014, LiveMint.



Karan Gulati is a consultant at NIPFP and Renuka Sane is researcher at NIPFP. We thank Sudipto Banerjee, Aditi Dimri, Pratik Dutta and Ajay Shah for useful comments.