Assessing the impact of IFRS 9’s Expected Credit Loss model on capital allocation in Jordanian banks

  • 13 Views
  • 2 Downloads

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

This study investigates the empirical effects of implementing the Expected Credit Loss (ECL) model under IFRS 9 on capital budgeting decisions within the Jordanian banking sector. The analysis is based on a full population of all 13 Jordanian commercial banks listed on the Amman Stock Exchange from 2013 to 2023. Using panel data regression models, the study evaluates changes in three key financial ratios: Capital to Assets (CA), Equity to Assets (EA), and Loans to Assets (LA).
The findings reveal that adopting the ECL model led to a statistically significant increase in CA by 0.3% (p = 0.04), suggesting that banks have strengthened capital buffers in anticipation of higher provisioning requirements. Conversely, the EA ratio declined sharply by 1.1% (p < 0.01), indicating equity reallocation to absorb credit risks. Most notably, the LA ratio fell by 3% (p = 0.006), highlighting a more conservative lending approach post-ECL implementation. Each model exhibited strong explanatory power (R² values between 0.79 and 0.87), supporting the robustness of the results.
These outcomes confirm that IFRS 9 has triggered a structural shift in how Jordanian banks manage capital and credit risk. The study underscores the critical need for adaptable capital strategies in emerging markets, where regulatory changes like IFRS 9 can significantly reshape financial behavior and resource allocation.

view full abstract hide full abstract
    • Table 1. Descriptive analysis
    • Table 2. Multicollinearity results
    • Table 3. Augmented Dickey-Fuller test results
    • Table 4. First regression model results
    • Table 5. Autocorrelation tests/Model 1
    • Table 6. Second regression model results
    • Table 7. Autocorrelation tests/Model 2
    • Table 8. Third regression model results
    • Table 9. Autocorrelation tests/Model 3
    • Table A1. Study sample
    • Conceptualization
      Mohammad Fawzi Shubita, Faez Hlail Srayyih, Sinan Abdullah Harjan, Dua’a Shubita, Majd Munir Iskandrani
    • Data curation
      Mohammad Fawzi Shubita
    • Formal Analysis
      Mohammad Fawzi Shubita
    • Funding acquisition
      Mohammad Fawzi Shubita, Faez Hlail Srayyih
    • Investigation
      Mohammad Fawzi Shubita, Majd Munir Iskandrani
    • Methodology
      Mohammad Fawzi Shubita
    • Resources
      Mohammad Fawzi Shubita, Faez Hlail Srayyih, Sinan Abdullah Harjan, Dua’a Shubita
    • Writing – original draft
      Mohammad Fawzi Shubita
    • Writing – review & editing
      Mohammad Fawzi Shubita, Faez Hlail Srayyih, Sinan Abdullah Harjan, Dua’a Shubita, Majd Munir Iskandrani
    • Supervision
      Faez Hlail Srayyih, Sinan Abdullah Harjan, Dua’a Shubita, Majd Munir Iskandrani
    • Validation
      Dua’a Shubita, Majd Munir Iskandrani