In the banking sector, problems in repaying customers’ credits can increase credit risk and fragility. Therefore, it is of great importance for banks to monitor the status of non-performing loans (NPLs) closely. This study analyzes the macroeconomic factors affecting NLPs in the Turkish banking sector. It used ARDL and QARDL approaches and data for 2011M5-2024M9 in the study. According to the long-run estimation results of the ARDL model, inflation and industrial production affect the NLPs in the opposite direction. In contrast, unemployment, the exchange rate, and interest rates affect it in the same direction. The estimation results are consistent with economic theory and the literature. The QARDL estimation results show that lnCPI (τ=0.2 to τ=0.8) has negative and significant coefficients in most quantiles (τ). The coefficients for lnPMI are generally negative and statistically insignificant. The lnUNE variable has positive and significant coefficients at most levels τ (τ=0.1 to τ=0.8). The estimation results for lnEXC show that the overall effect of the variable on NPL is positive and significant. The coefficients of interest rates are generally positive and significant. For the increase in the NLPs to remain at an acceptable threshold level for the banking sector and the Turkish economy, it is critical that the credit risk assessment system at the banking level works effectively and efficiently on the one hand and that macroeconomic indicators in the Turkish economy are supportive of the credit repayment conditions of economic agents on the other.
Primary Language | English |
---|---|
Subjects | Time-Series Analysis |
Journal Section | RESEARCH ARTICLE |
Authors | |
Publication Date | June 25, 2025 |
Submission Date | March 16, 2025 |
Acceptance Date | March 27, 2025 |
Published in Issue | Year 2025 |