New bad loan provision rules are good for banks
4 min read . Updated: 27 Feb 2023, 01:20 AM IST
PremiumThe ‘expected credit loss’ method would go by internal anticipation of asset quality and allow accuracy in making provisions
The Reserve Bank of India (RBI) recently proposed that we adopt the Expected Credit Loss (ECL) approach under the International Financial Reporting Standard (IFRS-9). The central bank has released a discussion paper on the ECL model for loan loss provisioning by commercial banks in India. To classify an asset as “impaired" or “non-performing", our bank regulations consider non-payment 90-plus days past the due date as a cut-off. Banks are currently making provisions after assets are identified as impaired or non-performing as per this regulatory definition. Additionally, for provisioning, Indian banks are subjected to a gradual age-wise provision rule for sub-standard assets, starting from 15% in the first year to 100% in the fourth year, irrespective of whether collateral is available or not. In the US, provisioning norms are purely discretion-based and provided for by banks as per estimated credit losses associated with their loan portfolios. In case of a commercial loan, the fair value of collateral is taken into consideration to account for such provisioning, if any.
Against this background, RBI’s proposed approach for introducing ECL-based bank provisioning would be to formulate guidelines supplemented by regulatory backstops wherever felt necessary. Expected loss is applicable both for assets that are currently impaired (but exact loss amount is uncertain) and those which are performing but may face future impairment. The provision amount would be based on the current internal estimate of a potential future credit loss.
An Expected Credit Loss is defined as a loss anticipated on a credit exposure or credit portfolio due to defaults expected to occur during the normal course of business. The major inputs of ECL are: a) Probability of Default (PD); b) Exposure at Default (EAD); and c) Loss Given Default (LGD). The PD is an estimate of the likelihood of default over a given time horizon, and ECL is the product of PD, EAD and LGD estimates.
EAD provides an estimate of the exposure at a future default date, taking into account expected changes in the exposure after the reporting date, including repayments of principal and interest and expected drawdowns on committed facilities. The third driver, LGD is an estimate of the percentage loss arising from default. It is based on the difference between the contractual cash flows due and those that the lender would expect to receive, including from any collateral. ECL-based provisioning norms under IFRS-9 require institutions to use point-in-time projections (PIT) of PDs, LGDs and EADs. The new financial accounting system requires banks and other financial institutions to internally model the key elements of their credit risk loss, stay forward-looking and thereby derive more risk-sensitive measures for loan-loss provisions.
Conceptually, ECL is the bank’s internal estimate of what a loan going bad means financially. It is an integral component of credit risk estimation. IFRS-9 or Ind-AS-109 accounting standards explicitly require provisions and loss allowances to be made as per ECL data. RBI’s prescribed forward looking expected credit loss principal is in line with the IFRS-9 standard. ECL-based provisions are to be applied at origination and for all subsequent reporting periods of loan assets till their de-recognition. Three stages have been specified under the new accounting standard to determine the amount of impairment to be recognized as ECL at each reporting date.
A 12-month ECL allowance is applicable for Stage 1 (performing) assets that at initial recognition show low credit risk on the reporting date. Banks need to assess at each reporting date whether the credit risk on a corporate loan facility has increased significantly since initial recognition, and thus the asset reaches Stage 2. At this stage, allowances are to be made based on lifetime analysis of any expected loss. If the loan is credit impaired, it will be put under Stage 3, and the standard requires that provisions be based on lifetime expected losses with the probability of default taken as 100%.
The estimation of borrower-level default probabilities is the highlight of the ECL-based provisioning method for Stage 1 and Stage 2 accounts. For this, lenders need to derive forward-looking and more recent PD estimates. Table 1 provides a summary of industry-wise default probabilities. Point-in-time PDs are based on current conditions, adjusted for forward-looking scenarios. These PDs are to be used for ECL estimation purposes.
The new accounting standards aim to simplify and strengthen risk measurement and the reporting of financial instruments in an efficient and forward-looking manner. The ECL based provision measure will enable banks to more proactively identify credit impairment and make necessary loss provisions. Early detection of a significant increase in credit risk may incentivize banks to go in for better credit portfolio planning and lower their prospective non-performing asset (NPA) burdens.
The ECL methodology takes into account historical PD trends as well as current and future economic scenarios and predictions. Thus, it significantly changes the incentives of banks by inclining them to manage and dispose of bad loans much more actively. Banks with robust data management systems, capturing collateral and cash-flow data, for example, can estimate ECL easily.
Arindam Bandyopadhyay & Soumya Kanti Ghosh are, respectively, professor, National Institute of Bank Management, and group chief economic advisor, State Bank of India. These are the authors’ personal views.
