An Analysis of Non-Performing Loan Dynamics in the Turkish Banking Sector Using the ARDL Method: A Comparative Assessment of Public, Private, and Foreign Capital Banks

This study examines the credit risk dynamics of public, private, and foreign-owned banks operating in Turkey within a time series analysis framework. The analysis employs monthly data covering the period from December 2004 to March 2025 and constructs separate time series for each bank ownership group in order to assess the determinants of risk. The Autoregressive Distributed Lag (ARDL) approach is adopted, allowing the joint analysis of short- and long-run relationships among the variables. The findings indicate that risk behavior differs across bank ownership types. In the short run, capital structure and market concentration play a stabilizing role in risk dynamics, while indicators related to loan portfolio quality move in line with risk as an expected balance-sheet effect. In the long run, leading indicators reflecting the course of economic activity are found to have a more pronounced impact on risk dynamics, particularly for foreign-owned banks. Overall, the results suggest that risk management in the banking sector should account for ownership-specific characteristics and dynamics.