The Expected Credit Loss Model's Impact on the Cyclicality of Credit Supply: A Study of the Implementation of IFRS 9

Abstract: The accounting standard for recognizing loan loss provisions changed in 2018 from IAS 39 to IFRS 9. IFRS 9 introduced the expected credit loss model (ECL), intended to be an improved alternative to its predecessor, the incurred credit loss model (ICL), which was criticized for the "too little, too late" provisioning during the 2008 financial crisis. The ECL model was expected to mitigate procyclicality through a timelier recognition of loan losses. However, concerns have been raised that the ECL model will have the opposite effect. Therefore, this paper aims to provide empirical evidence on the implications of the ECL model on credit supply procyclicality by researching the question: Does the switch to the ECL model affect banks' lending sensitivity to changes in regulatory capital ratios when credit risk increases? Using a sample of listed banks in the EEA, we conduct regression analysis and find that when credit risk increases, the lending sensitivity to changes in capital ratios is lower for banks with assets greater than EUR 1,000 million after the implementation of IFRS 9. However, for banks with total assets between EUR 300 million and 1,000 million, we see an increased lending sensitivity to changes in capital ratios when credit risk increases. Several potential explanations are proposed, among which the greater resources available to larger banks for an accurate implementation of the ECL model, the higher regulatory and market scrutiny that these banks receive, as well as their superior risk-taking discipline. Our study contributes to the discussion regarding potential negative effects of the ECL model, which is of particular importance in the current conditions of economic uncertainty.