Developed an Expected Credit Loss (ECL) forecasting model for a leading Indian bank for their entire portfolio
Overview
Our client is one of India’s prominent private sector banks with a growing national footprint, offering specialized services across five key business sectors: Corporate Banking, Commercial Banking, Branch & Business Banking, Retail Assets, and Treasury & Financial Markets Operations. In line with Indian Accounting Standards (Ind AS), the client sought to establish a methodology for assessing expected credit losses. This method would involve calculating the probability-weighted estimate of credit losses throughout the expected lifespan of financial instruments. The primary goal is to quantify expected loss, distinguishing it from incurred losses.
Solution
- The three primary components i.e. Probability of default (PD), Loss given default (LGD) and Exposure at default (EAD) are derived based on empirical evidence, macro-economic factors and management judgement
- The ECL is calculated separately for all the portfolio segments as ECL= PD ∗LGD ∗EAD
- For calculating PD, flow rate approach is utilised fitting Weibull distribution to derive forecasted PDs
- LGD% are taken as discussed with subject matter expertise within the bank
- Approach for EAD reflects expected changes in the balance outstanding over the lifetime of the loan exposure
Impacts
The bank was able to set aside an expense as provisions to meet the potential future loss due to credit risk
Prompt loss recognition model, measured over a variety of data, helped the bank to reduce the procyclicality of bank lending
