Banking sources say RBI red-flags loan accounts for fraud probe

Banking sources say RBI red-flags loan accounts for fraud probe

Wednesday, Dec 16, 2020

 

--Banking sources: RBI red-flags set of large loan accounts

--RBI asks banks to investigate red-flagged accts for fraud

--RBI seeks red-flagged accounts' status from banks by Dec 31

--RBI not happy with early warning systems of some banks

 

By T. Bijoy Idicheriah and Alekh Archana

 

MUMBAI – The Reserve Bank of India has identified a set of large red-flagged loan accounts and set a hard deadline of Dec 31 for banks to report if such accounts classify as "fraud", according to two people aware of the development.

 

"The deadline for banks is to do a proper classification and report the status of these red-flagged accounts. One of the key reasons for the RBI directive is the fact that the early warning system placed by some banks is not as robust as expected," a banking industry source told Cogencis.

 

The source added the regulator has also pulled up banks for not moving quickly enough on declaring such accounts as fraud, and failing to tag them as upgraded in case the loans have been regularised by the borrower.

 

Banks have to make 100% provision in four quarters for accounts tagged in the fraud category. In case of non-performing assets without delayed recovery, 100% provisioning effectively happens over eight quarters. 

 

Many banks have already tagged and declared some accounts as "fraud", but others are yet to act in the matter--which the RBI wants to put an end to, the second source told Cogencis.

 

Lax internal control mechanisms and delayed reporting of frauds by brand or mid-level staff to the head office and from the bank to the credit bureaus and large credit registry are key problems in identifying and classifying such loan accounts, the source said.

 

Noticing delay in reporting of fraud by banks, the RBI had, in 2015, introduced the concept of "red flagged accounts" wherein early warnings such as default on the payment of dues or resignation of key management personnel, indicating a potential fraud, are noticed. 

 

These signs are aimed at alerting the bank, which in turn is expected to launch a detailed investigation. All red-flagged or fraud accounts in which the banking sector’s exposure is over 500 mln rupees must be reported to the central bank’s database of large loans, known as Central Repository of Information on Large Credits.

 

Delays in identifying such frauds, though, persists.

 

Take the case of Bhushan Power and Steel Ltd, among the 12 companies for which the RBI, in June 2017, had directed banks to file insolvency proceedings.

 

Even as the case was under insolvency proceedings, in July 2019, the lead lender Punjab National Bank detected an over 38-bln-rupee fraud and reported it to the RBI. The bank had said that the company has misappropriated bank funds, and manipulated its books of accounts to raise funds from a consortium of lenders.

 

Similar cases of fraud were detected in loan accounts that had been resolved under the Insolvency and Bankruptcy Code.

 

The increasing number of such cases prompted the government and the RBI to nudge banks to act swiftly without delays. The central bank has also asked the lenders to declare such frauds in public, as they believe this is price-sensitive information for both the bank and borrowers. The practice also effectively ensures that other lenders to the same borrower revisit their exposures and classification of such loans.

 

In October 2019, bankers to the beleaguered Dewan Housing Finance initiated a forensic audit. Subsequently, an audit by KMPG revealed the diversion of large funds to third-party private entities linked to promoters.

 

The housing financier is facing proceedings under the National Company Law Tribunal. On Monday, the administrator in charge of the lender said Grant Thornton India LLP, which is auditing the books of Dewan Housing Finance Corp, has found fresh fraudulent transactions worth 10.6 bln rupees.

 

The RBI's annual report for 2019-20 (Jul-Jun), had said the average lag between the date of occurrence of frauds and their detection by banks and financial lenders was 24 months during the previous financial year. In case of large frauds of 1 bln rupees and above, the average lag was as much as 63 months.

 

“The sanction of the credit facility in many of these accounts was much older. Weak implementation of early warning signals by banks, non-detection of early warning signals during internal audits, non-cooperation of borrowers during forensic audits, inconclusive audit reports and lack of decision-making in Joint Lenders' meetings account for delay in detection of frauds,” the central bank had said in a report.

 

It had also said the mechanism for early warning signals was being revamped along with strengthening of the concurrent audit function, with timely and conclusive forensic audits of borrower accounts under scrutiny.

 

The industry source said that the RBI has also conveyed its displeasure over the quality of the early warning signals used by banks to identify such stressed loans and frauds.

 

"There is a wide divergence between the systems used by some banks, which are very sophisticated and customised, and those deployed by a few lenders which act more as post facto indicator checks rather than lead indicators. This means some banks are faster at detecting frauds, while others are followers," the second source said.

 

The source said that, for instance, when a company is going through stress and lays off staff or does not pay them on time, their posts on social media platforms such as Twitter or LinkedIn are captured by good systems, while stock systems only track official published data and hence work with a lag.

 

Another example is that a good early warning system may pick up problems at even a group-level entity, while a basic system will not be able to catch such signals.  End 

 

Edited by Charumathi Sankaran

 

Cogencis Tel +91 (22) 6619-0000

Send comments to feedback@cogencis.com

This copy was first published on the Cogencis WorkStation

© Cogencis Information Services Ltd. 2020. All rights reserved.