To analyze the impact of accounting standards and bank characteristics on provisioning behavior, the paper employs three types of regression tests:
Determinants of Provisioning Ratio: This analysis focuses on understanding the general drivers of provisioning behavior. It assesses changes in the provisioning ratio at the bank-firm level, with variables like accounting standard (IFRS9 vs. nGAAP), bank capitalization etc.
Provisioning Dynamics around Idiosyncratic Credit Events: By examining provisioning dynamics around individual loan-level events, it aims to identify potential cyclical implications of different accounting standards and bank practices. In other words, whether provision dynamics are generally different between loans using different accounting standards or loans issued by different banks which derive insight on cyclical implications.
Provisioning Dynamics around Macro-Economic Shocks: This analysis investigates provisioning dynamics during macro-economic shocks, such as those triggered by the COVID-19 pandemic and the outbreak of war in Ukraine. These events have broader systemic implications, and understanding their impact on provisioning is crucial.
The results of the analysis indicate that, as intended, provisions under IFRS9 are higher than those under nGAAP. During the COVID-19 pandemic, non-performing loans improved, but there was a noticeable increase in Stage 2 provisions. Banks with more capital headroom tend to make higher provisions, consistent with capital management motives. The analysis also highlights the influence of discretion on provisioning behavior.