Abstract
For many years, the banking sector has been using International Accounting Standard (IAS) 39 to recognize credit losses. However, IAS 39 was heavily criticized for causing procyclicality due to recognizing credit losses after the impairment had already occurred, contributing to the global financial crisis (GFC). Procyclicality occurs when ECL and other macroeconomic variables fluctuate rapidly throughout the business cycle. These challenges led to the introduction of IFRS 9, which recognizes credit losses using the expected credit loss (ECL) model based on forward-looking information, as recommended by the Financial Advisory Group.
This study aimed to examine the procyclical effect of IFRS 9 ECL on banks listed on the Johannesburg Stock Exchange (JSE). The dependent variable in this study was represented by ECL, while the independent variables were proxied by the gross domestic product (GDP) and common equity tier 1 (CET1). Hence, the objectives of the study were twofold: 1. to examine the effect of GDP on ECL, and 2. to assess the effect the CET1 on ECL.
A quantitative research method was employed because the data collected was recorded in numeric format. The study adopted a casual-comparative research design was adopted. The population consisted of the eight banks listed on the JSE, with a sample of six banks selected based on purposive sampling. Secondary data was collected from annual financial statements and integrated reports of the selected banks covering a period of five-year period (2018–2022). Multiple linear regression analysis to determine the significant influence between the variables.
The findings revealed a positive and statistically significant influence of the CET 1 ratio on ECL. These results indicate that higher levels of CET1 are associated with the increased ECL, suggesting that IFRS 9 ECL may contribute to procyclicality in banks. This procyclicality could result in a reduced credit supply when needed, potentially contributing to the GFC. In contrast, the influence of GDP on ECL was found to be statistically insignificant, implying that there is insufficient evidence signalling a cyclical relationship between GDP and ECL under IFRS 9. This study contributes to the literature that assesses the cyclical effect of the new credit losses model under IFRS 9 post the implementation date. The finding could have implications for financial regulators and policymakers in emerging economies like South Africa as well as
international accounting standard setters, such as the International Accounting Standard Board and Financial Accounting Standards Board.