ECL Will Require Auditors to Understand the Models Behind the Numbers

The implementation of expected credit loss, or ECL, is changing not only how banks and non-banking financial companies calculate credit impairment, but also what external auditors must know, examine, and challenge. The traditional audit of a provision is giving way to an audit of a number generated by credit-risk models, data, economic forecasts, and management judgement.

This is perhaps one of the least discussed, yet potentially most consequential, aspects of India’s transition to ECL-based impairment.

ECL is fundamentally different from the relatively straightforward provisioning norms that Indian banking has followed for decades. Under the earlier framework, provisions were largely driven by prescribed regulatory categories and relatively simple rules. Under ECL, the impairment estimate must reflect losses that can reasonably be anticipated from a financial asset, to reflect risks including that of the future economic conditions.

That is considerably harder.

A realistic estimate of expected losses requires an institution to draw upon many of the advances in credit risk assessment and measurement that have evolved over the past 15 years. Probability of default, loss given default, exposure at default, credit migration, forward-looking economic information, scenario analysis, and model-based estimation have traditionally belonged more to the domain of credit risk and risk management than to financial accounting. ECL has brought these disciplines directly into the accounting process.

In that sense, ECL has brought accounting and risk management closer than they have ever been. The framework is now more closely connected to the Basel risk architecture and to the quantitative techniques used in modern credit risk management.

This also changes the nature of the challenge for external auditors.

Beyond Accounting
The complexity of the process, therefore, is not simply a matter of mathematics. It is a matter of governance and auditability.

Where the framework permits judgement, that judgement cannot be allowed to become arbitrary. It must be supported by evidence, governed by defined processes, and subjected to systematic challenge. Models must be appropriately designed and validated. Data must be demonstrated as reliable. Assumptions must be reasonable and consistently applied. Changes in methodology must be explainable. Management overlays, where used, must have a defensible basis.

This is where the role of the external auditor becomes particularly important and considerably more demanding.

Internationally, the Basel framework has recognised that the scrutiny of ECL cannot necessarily be confined to traditional financial statement audit procedures. The assessment of accounting treatment and disclosures remains within the conventional audit domain, but the quantitative and risk dimensions of ECL may require specialist expertise. Basel guidance therefore envisages auditors drawing on appropriate risk, credit, and quantitative specialists where necessary.

The Indian regulatory environment is moving in the same direction. The Reserve Bank of India has emphasised that responsibilities for managing and overseeing the ECL framework should be divided between the finance and risk functions. The National Financial Reporting Authority (NFRA), too, has recognised the importance of the Basel approach to auditors assessing ECL.

The timing makes this particularly relevant for the Indian banking sector. ECL has already been applicable to NBFCs for more than six years. Banks, however, are scheduled to come under the ECL framework from April 2027. The implementation in banks is likely to attract considerably more structured scrutiny, given the scale, complexity, and systemic importance of the banking sector.

NBFCs should not assume, however, that the less structured scrutiny they have experienced to date will continue indefinitely. The Reserve Bank’s scale-based regulatory and supervisory framework is itself moving towards differentiated but increasingly risk-sensitive supervision. As ECL matures, NBFCs should expect the scrutiny of their impairment estimates to become progressively more sophisticated and, in many respects, more bank-like.

Capability Gap
This has an important implication for the external audit community.

There is little time for auditors to build the necessary capabilities before ECL becomes a mainstream banking audit issue. The concern is not merely that an audit may be challenged by management or an audit committee. More seriously, deficiencies in the auditor’s assessment of ECL may later be identified during regulatory or supervisory scrutiny, raising questions about the quality of the audit itself.

The stakes, therefore, are high.

International audit ecosystems have already travelled far down this road. The implementation of ECL under IFRS 9 and comparable frameworks required them to invest substantially in credit-risk specialists, quantitative analysts, model experts, and other technical capabilities. The Indian audit profession will need to make similar investments.

ECL is, after all, not simply a new provisioning formula. It represents a change in the architecture of impairment measurement. When the accounting number increasingly depends on models, data, credit-risk judgements, and economic forecasts, the audit of that number must evolve accordingly.

For external auditors, the message is clear: the era when ECL could be audited primarily as an accounting number is coming to an end. Auditors increasingly need to understand the risk models that generate the number.