A week ago, Governor Raghuram Rajan delivered a speech in Mumbai, the theme of which was the Financial Sector Legislation Reforms Committee’s (FSLRC) report. Some sections of the media that reported the event seized upon the word “schizophrenic” that Rajan had used to describe some of its key recommendations.

The RBI governor had found some of the suggestions useful. But his criticism of two key FSLRC suggestions is noteworthy.

Rajan locates two fundamental areas of tension. One, the FSLRC wants to introduce judicial supervision as a tool to check and balance the activities of regulators, over and above those exercised by the constitutional courts such as the high court.

At the same time, the committee wishes to rewrite laws so that they do not micromanage regulators! The governor, perhaps wryly, calls for “a proper balance” between the two, a balance that would vary with the level of development.

The second area of tension is the “appropriate size and scope of regulators”. Or what he refers to as the regulatory architecture.

Why checks are needed

The need for regulation clearly flows from market failure or the “bad behaviour” of the entities to be regulated.

The FSLRC claims that such behaviour arises from poor incentives or incomplete information. But the RBI governor directs our attention to the most important reason for bad behaviour: “incomplete contracts”. The behaviour of a firm or a bank towards its customers or the public or the market cannot be completely specified in a formal and observed or verified contract by the regulator in real time. So, can a regulator detect bad behaviour?

Only by evaluating behaviour across contracts. Thus, if a bank’s customers complain about its credit card service, the regulator can look across similar complaints (if any) and determine something is wrong and banks’ “need to shape up”.

An investment product becomes bad (toxic) and requires regulation after the regulator has gauged its behaviour across entities offering that service.

Alternatively, a regulator can decide (on the basis of past experience or in anticipation of a systemic risk) that certain products or contracts — such as the collaterised debt obligations that lay at the centre of the financial crisis of 2008 — are bad and need to be banned even if they are contracted by consenting adults.

Rajan reiterates that, basically, regulatory action stems from the regulator’s exercise of judgment on products or services based on experience. In fact, the heart of regulatory exercise lies here; in its capacity to fill in “the gaps in laws, contracts and even regulations.”

In this context, the FSLRC’s suggestion for a Financial Sector Appellate Tribunal to exercise checks and balances over the regulator is an excessive and ill-timed legal oversight that could and, most likely, would cut the ground from under the regulators’ feet.

As it is, regulatory decisions can be challenged by writ petitions in the constitutional courts.

Rajan contests the capabilities of an appellate tribunal to evaluate decisions of a regulator that may not necessarily be based on precise evidence but on the agency’s judgment of “bad” behaviour.

The tribunal may or may not uphold the decision. But to say that tribunals would be fair and uphold sound judgments is really to set double standards.


“We trust the tribunal’s judgment but not that of the regulator.” Rajan sees more generalised dangers for policy formulation itself.

In a developing country like India with new regulations and legislations yet to be framed and new institutions yet to mature on the basis of experience, extending the right to appeal to more than administrative decisions (say about the size of the fines imposed) could “paralyse the system and create distortions, as needed regulations are held up and participants exploit loopholes”.

And from this follows the attendant danger: the lack of respect for the regulator “that serves to keep participants on the straight and narrow”.

Many or single regulator?

The second major area of tension the governor of the RBI sees is in the architecture of regulation. The FSLRC recommends dissolving various regulators into one unified financial agency (UFA).

The assumption is that the central synergy binding all under one roof is the fact that it will deal with products that are traded — commodities, bonds and the like.

But Rajan points to the existence of other synergies that are as important if not more. For instance, commodity trading relies on real commodities and warehouses where they are delivered. So would not the Forward Markets Commission, for instance, be better off linked to the ministries “overseeing the real commodities”?

Then again, while recommending a unification of regulatory activities under one roof, the committee also engages in what Rajan calls “balkanisation”.

Trading comes under one regulatory roof; consumer protection under another. But the regulation of credit is split with banks staying under the RBI and non-banking financial companies moving into the UFA’s stable.

For the Central bank such an arrangement would hamper the smooth operation of regulatory authority and more generally monetary policy.

And who can predict with certainty that “silos” would not grow within the huge bureaucratic behemoth of the UFA?

At another place, points out Rajan, the FSLRC suggests a strengthened Financial Sector Development Council as a venue for inter-regulatory functioning. Would not this serve to conserve and strengthen the synergies that the FSLRC at other places thinks a spanking new behemoth, the UFA, would?

The arguments that Rajan puts forth raises another crucial question: Unless an existing system is shown to have broken or turned dysfunctional is it necessary to replace it?

The FSLRC does not sound convincing in its implicit view, that extant institutions are inefficient and need to be re-organised to prevent systemic risks and market failures.

What matters is not whether a country has a single or multiple regulators. Both structural arrangements failed to prevent the financial crisis of 2008, the single one in the UK and the multiple one in the US.

What counts is the quality and integrity of regulation itself.