Big Bank Theory: Must tighten norms that delay NPA, fraud disclosure

Informist, Tuesday, Apr 6, 2021


With technology and database enabling better visibility on loans across the system, RBI should further tighten the norms on disclosure of stressed assets to end delayed reporting of frauds and divergence between actual and reported stressed loans. 


By T. Bijoy Idicheriah


In Hanoi, in what is Vietnam today, there was a massive surge in the population of rats in 1902, which was linked to a rise in plague in the city. To incentivise the population to join efforts to tackle this, the state offered a reward for anyone who produced the tail of a rat. 


The bounty programme was eventually scrapped, as this led to a surge in tail-less rats. While most people cut off only the tails, others started importing rats, and the more enterprising ones began breeding rats to collect the reward.

A similar predicament faced the British in India when they wanted to bring down the population of cobras in Delhi. On realising that cobras were being bred to collect the bounty, the reward programme was scrapped. With little incentive to hold on to the snakes any longer, they were released and this sparked an even steeper rise in the population of cobras.


This phenomenon of perverse incentives is popularly called the 'cobra effect'–a solution actually making the problem worse. 


Rarely did Indian banks declare the actual picture of their asset quality, that is till the Reserve Bank of India led industry-wide Asset Quality Review in 2014-15, which exposed the true picture and left banks providing for bad loans for years.


A large chunk of these loans were linked to the RBI window to restructure the debt of Indian companies post the global financial crisis. With the regulator offering space to provide less for such recast loans, banks aggressively restructured these loans even when the companies were in no shape to service them.


The result: For almost a decade, banks managed to avoid identifying such dud loans, thanks to misuse of forbearance.


It was a common practice that if a loan was unpaid on the 89th day since it was due, no bank would identify it or provide against it to guard against future losses. This is because RBI's norms asked banks to identify and provide only for a loan unpaid for 90 days.


The fact that RBI insisted on reporting one-day default data has tempered this practice to an extent.


For years, the discussion was about 'below the line' or 'actual' asset quality of Indian banks, especially state-owned ones, compared to the numbers declared publicly. This was also reflected in the weaker valuations and stock prices of these banks compared to their private peers. 


Although such practices were more common at public sector banks and old generation private banks, larger private banks were not averse to such practices. This has only been underlined by recent developments involving some top officials at ICICI Bank, YES Bank and even Lakshmi Vilas Bank.


Anybody tracking the banking sector closely would have noticed a spate of frauds being reported by banks since early December, linked to stressed assets that have been identified years ago. 


As reported by Informist on Dec 16, this has come on the back of a diktat by the Reserve Bank of India, asking banks to come clean on their disclosure on frauds linked to bank loans. 


Why do banks prefer to delay recognition of frauds? 


A fraudulent account has to be provided for over four quarters, while a non-performing account gets provided for over the course of almost two years. So, a bank can initially provide less and then, once fully provided for over two years, finally bite the bullet and declare it a fraud. 


In most cases declared as frauds over the last few months, banks often say they are fully provisioned against the exposure. 


The well-intended RBI regulation that looks to separate an asset stressed by genuine business failure from loans that turn bad due to frauds by company managements, has been subverted to protect bottomlines. 


Therefore, a bank aware of fraud in a large account will delay declaring it so that provision costs are spread out. By doing so, it can declare inflated profits based on muted bad loan numbers and, thus, artificially drive up stock prices.


The irony is that early detection and action against a fraudulent loan may actually lead to faster and higher recovery for a bank, but that means booking the pain upfront and not let the tail wag. It may also prevent other banks from taking on more exposure.


When RBI's annual inspection of bank books started showing a wide divergence in the bad loans and provisions reported by the lender, the central bank in April 2017 asked banks to declare when the divergence exceeds 15% of the reported number. Two years later, it further tightened the norms to factor in under-reporting of provisions on profits. 


Divergence in a large loan account could have been acceptable earlier, when data was not centrally available. For the last few years, the RBI has had a Central Repository on Information of Large Credits. Despite this, the incidence of divergences exceeding 15% continues at some banks.  


Why does this happen?


In a way, banks are interpreting RBI's leeway of up to 15% divergence for disclosures as tacit approval. In a perverse twist, banks are trying to use the regulatory space provided to not declare the actual bad loan data, but that data which will ensure it does not have to make a disclosure.


At press meetings following the announcement of earnings, chief executives of banks go out of their way to point out that divergences were below the 15% mark for making a disclosure; they don't share whether their divergences were 1% or 14.9%, as that isn't mandatory. 


The RBI, in its wisdom, wanted to give banks some breathing space to move to a more transparent reporting process on the divergences detected in supervisory audits. The intent, as articulated by the RBI management, was to stop such divergences in reporting by banks. 


The 15% divergence norm was never meant to be an end in itself, which it seems to have become for banks. It may be time to ask banks to declare complete data on divergence or shift to such disclosure in a phased manner, by initially bringing down the threshold to 10% and later, to 5%. 


RBI must impose a commensurate penalty or mandate bullet provisioning in cases where it detects delayed reporting of loan frauds, and ask banks to come clean on divergences.


It is time to do away with well-intentioned supervisory efforts that have been used perversely to delay full disclosure on asset quality or banks.  End

(When he's not breaking a leg on stage, he's breaking news and views on banks and regulatory issues. In Big Bank Theory, Assistant Editor (Money) T. Bijoy Idicheriah talks shop on matters that matter to the sector.)


Cogencis news is now Informist. This follows the acquisition of Cogencis Information Services Ltd by NSE Data & Analytics Ltd, a 100% subsidiary of the National Stock Exchange of India Ltd. As a part of the transaction, the news department of Cogencis has been sold to Informist Media Pvt Ltd.


Informist Media Tel +91 (11) 4220-1000

Send comments to


© Informist Media Pvt. Ltd. 2021. All rights reserved.

Other News

IPO Alert: Shriram Properties files draft prospectus with SEBI

Informist, Saturday, Apr 10, 2021 NEW DELHI – Shriram Properties Ltd filed its draft red herring prospectus with the Securities and Exchange Board of India on Friday for its initial public offering, the company today said in a press release. The issue comprises shares with a face value of 10 rupees each aggregating up to 8 bln rupees. The fresh […]

EDITORIAL: RBI’s gilt purchase plan quantitative easing by ‘jugaad’

Informist, Friday, Apr 9, 2021 Indians are experts at working with the resources at their disposal and coming up with the sort of life hacks which has given form to the word 'jugaad'. The Reserve Bank of India seems to be no exception, going by its new gilt purchase programme. The first version of the gilt acquisition programme, […]

Govt source says FY21 fisc gap seen lower than revised aim of 9.5%

Informist, Friday, Apr 9, 2021 –Govt source: See higher than projected revenues, spending in FY21 By Adrija Chatterjee and Shubham Batra NEW DELHI – The fiscal deficit for 2020-21 (Apr-Mar) is likely to be lower than the revised projection of 9.5% of GDP, a top finance ministry official told Informist. As per provisional estimates with the ministry, both the revenues […]