Stage Migration Dynamics Under IFRS 9: Examining the Determinants of Significant Increase in Credit Risk and Its Impact on Expected Credit Loss Calculations in Retail Banking Portfolios

الملخص

The forward-looking approach to the measurement of credit risk is included in IFRS 9. What SICR, not to mention dynamic migration mechanisms, maybe, needs more detailed scrutiny. It is one of the most radical changes to accounting policing in the present days. Key issues include how credit losses should be accounted for, measured and disclosed by banks as well as the impact on regulatory capital requirements and risk-management practices. Now, as it is in the future, for banks and their overseers and any other believer that accurate financial reporting standards and sound risk management capabilities are a Good Thing to be done under whatever new framework we devise – it is of crucial importance that they understand what causes migration across stages. This initial up of several implementation outcome studies focuses on the key drivers for movement in accounts under IFRS 9, and what impact material increases in credit risk have upon expected loss calculations at retail banks and thus addresses a lack of available empirical data in the practical application results to date for IFRS 9. Drawing on a unique dataset of 2.4 million retail banking accounts from five large European banks covering the period between 2018 to 2023, this article utilizes logistic regressions models, survival analysis and machine learning methods to identify the key drivers that lead to migration into diverse phases of the customer relationship lifecycle. It uses both numerical and non-numerical characteristics of credit deterioration. The findings indicate that payment delinquency aggravated is the most important factor, with an odds ratio of 4.23 (p < p < 0.001). Following these other changes in debt-income ratio and their increase on Rem stage (odds ratio of 2.17, p8000). Last but not least come the macroeconomic stress indicators, and since we already know that they are used by every bank, even outside IFRS its odds ratios for taking such decisions exceeded 1.89 (p-value<0.001). Stage 2 classifications produce expected loss provisions 3.2 times the size of those from Stage 1 provisions with an accurate rate of 89% in predicting future defaults. But: it varies widely in how much it is used by product type and location. It should be noted that this research is focused on the European retail lending markets, and other kinds of lenders in Europe and in other areas of the banking industry might not be generalized to. However, these results contain useful recommendations for banks who may need to re-calibrate their thresholds of a high increase in credit risk or enhance the accuracy of expected loss by including into their macroeconomic scenarios indicators related to client behavior that are forward looking. This study is among the first extensive analyses on migration behavior across various European banking enterprises and provides some interesting practice that helps understand practical implications of IFRS 9 implementation as well as regulatory compliance. These findings will inform future policy work at the European Capital Adequacy system or national regulators and also general banking industry best-practice.