For more updates follow Grant Thornton Bharat on WhatsApp
Predicting credit risk has always been a complex puzzle for financial institutions around the world. The global financial crisis made this painfully clear when even the most sophisticated banks miscalculated the risk of mortgage-backed securities, leading to a collapse in financial markets. India, too, has had its missteps. The 2018 IL&FS debacle highlighted just how difficult it can be to spot credit risk, even in large, established firms. It’s clear that credit risk estimation is complicated, overconfidence in growth cycles is hard to resist, and the consequences can be severe.
But is India’s regulatory framework robust enough for banks to accurately model credit risk, and for outsiders to analyse it effectively?
The honest answer, as of today, is no. India significantly trails behind international best practices when it comes to credit risk modelling. For a long time, the Reserve Bank of India (RBI) has prescribed how the industry conducts its credit risk assessments, with assets categorised as standard or non-performing (NPA) based on fixed rules. This system works to an extent, and the mix of conservatism and active supervision has helped maintain relative stability. However, this approach doesn't provide the dynamic, real-time risk assessment that modern markets demand. Most large economies have now moved on to having banks themselves conduct active forward-looking credit risk assessments of their portfolios, creating provisions and capital charges based on those.
Recognising this, the RBI is now fast-tracking the introduction of the ECL framework and new credit risk model management guidelines. Recent news suggests that it is expected to come in this year, with a 12-month implementation period.
At its core, the ECL model aims to bring more transparency by requiring banks to forecast credit losses over the life of a loan, factoring in expected macroeconomic conditions. Unlike the current backward-looking system, ECL is dynamic. It accounts for changes in economic environments, asset quality, and other evolving risks.
For comparison, the existing framework is like flying a fighter jet without advanced sensors. You’re relying only on your visual cues and past experiences, which might work for basic manoeuvres, but it leaves you vulnerable to unseen threats and rapid changes in the battle environment. In contrast, a forward-looking model, like the ECL framework, is akin to equipping your jet with state-of-the-art radar and targeting systems—allowing you to anticipate incoming risks, make faster decisions, and stay ahead of potential dangers.
Hence, the promise of ECL is real. Banks will be compelled to do a lot more analysis and modelling of their credit risk, to create provisions. Also very importantly, banks will need to disclose much more about their internal risk models, expected scenarios, and the credit risk in those situations. This could be transformative for investors, analysts, and depositors, who have often had to rely on vague executive statements or incomplete data. With the ECL framework, there will be more transparency and accountability in how banks manage their risks.
But how do we ensure that these benefits are realised? ECL models will not be without challenges. Credit risk modelling is inherently complex, and the quality of predictions is only as good as the data being used. Moreover, real-world events can often outpace even the best models. Take the current boom in unsecured personal loans and credit cards, for example. Encouraged by strong demand, many Indian banks have ramped up lending in this area. While this has been quite profitable in the short term, it also brings higher risks. Non-performing loans (NPLs) in this sector are now on the rise and suggest that some cracks are already appearing in lenders’ portfolios. The test of ECL modelling will be whether it helps banks recognise these risks at the right time and adjust their provisioning, capital management, pricing, and risk management strategies accordingly. Controlling biases and applying prudent judgement will be key to achieving this.
It won’t come cheap. Banks will need to invest significant resources into technology, data analytics, and expertise to properly implement the ECL framework. Larger banks may be able to manage this shift smoothly, but smaller institutions could struggle to keep up. If banks cut corners, they risk exacerbating the very problems that ECL is designed to solve—much like not everyone was able to afford an F-35 fighter jet.
Ultimately, while the ECL framework can close significant gaps in India’s credit risk landscape, it won’t eliminate the inherent complexity of predicting credit risk. What it will do is make the process more transparent, more accountable, and more informed by real-world scenarios. For India Inc., this could be a game-changer, fostering a more mature and resilient financial system. So, the road ahead will be challenging, requiring not just regulatory reform, but also a cultural shift in how credit risk is managed.
The journey is promising—but far from easy.
This article first appeared in Money control on 12 December 2024.