The shift to the ECL framework is expected to increase provisioning requirements across the sector, thereby exerting pressure on profitability.AI Quick ReadShares of PSU banks declined on Tuesday after the Reserve Bank of India (RBI) reiterated its stance on transitioning to the expected credit loss (ECL)-based provisioning framework, rejecting industry requests for an extension.
The central bank on Monday made it clear that the new norms will come into effect from April 1 next year, offering no additional transition time to lenders.
The Nifty PSU Bank index fell nearly 2%, with all its constituents trading in the red. Bank of India, Union Bank of India, Punjab & Sind Bank and Canara Bank were the top index losers, falling more than 2% each.
State Bank of India (SBI), Bank of Baroda, Punjab National Bank (PNB), Bank of Maharashtra, Indian Bank and UCO Bank also suffered losses.
The RBI has introduced a revised framework for asset classification, provisioning, and income recognition, anchored in a forward-looking Expected Credit Loss (ECL) model.
These tighter norms will require banks to set aside higher provisions for potential losses across their loan portfolios, aligning domestic regulatory standards with global practices.
Under the new framework, loans will be categorised into three stages based on evolving credit risk, necessitating earlier recognition of stress. The shift to the ECL framework — first proposed by the RBI in early 2023 — is expected to increase provisioning requirements across the sector, thereby exerting pressure on profitability.
The key shift lies in the transition from an “incurred loss” model to a forward-looking Expected Credit Loss (ECL) framework. Under this approach, banks will be required to recognise and provide for potential credit losses in advance, rather than waiting for a loan to become non-performing.
The framework introduces a three-stage asset classification system based on the extent of deterioration in credit risk:
Stage 1: Standard assets with no significant increase in credit risk; provisioning based on 12-month expected credit losses
Stage 2: Assets that have witnessed a significant increase in credit risk; provisioning based on lifetime expected credit losses
Stage 3: Credit-impaired assets; provisioning based on lifetime expected losses, with more stringent treatment
The RBI clarified that existing non-performing asset (NPA) classification norms will continue to apply, with loans being tagged as non-performing if repayments remain overdue for more than 90 days.
Read all Stock Market news here
Ankit Gohel is the Deputy Chief Content Producer at Livemint, specialising in financial markets, macroeconomics, and regulatory developments. With a strong focus on equity markets, primary issuances, and policy-driven market movements, he brings clarity to complex financial developments for investors and market participants.
With nine years of experience in business and financial journalism, Ankit’s approach is rooted in the belief that market reporting should go beyond headlines — connecting data, policy, and ground realities to deliver actionable insights. His work consistently bridges the gap between institutional analysis and investor understanding.
Ankit has spent three years at Livemint, where he currently helps drive market coverage, editorial strategy, and high-impact financial stories. Prior to this, he worked with leading business news networks such as CNBC-TV18, ET Now, TickerPlant News Service where he built deep expertise in stock market analysis, macroeconomic trends, primary markets, and coverage of key regulators including the RBI and SEBI.
Over the years, he has covered market cycles across bull and bear phases, IPO booms, liquidity shocks, and major policy shifts that reshaped investor sentiment. He has interviewed fund managers, corporate leaders, and policymakers, translating their perspectives into sharp, data-backed narratives. Ankit combines speed with accuracy — ensuring timely, credible, and insight-driven financial journalism that empowers both retail and institutional audiences.