Efficiency analysis of the banking sector: Application in Turkey with DEA and TOBIT methods

In our study, the concepts of performance, financial performance, efficiency and effectiveness are explained theoretically and the differences between them are revealed. Parametric methods (Regression Analysis, Stochastic Frontier Analysis, Thick Frontier Approach, Free Distribution Approach) and non-parametric methods used in efficiency measurement, the features of these methods, their advantages and disadvantages and their suitability for the banking sector are examined in detail. The history, definition, and basic models (CCR, BCC, etc.) of Data Envelopment Analysis (DEA) have been explained, emphasizing its advantages such as managing multiple inputs and outputs, not requiring a priori functional form assumptions, and determining relative efficiency. Using data from 22 deposit banks operating in the Turkish banking sector between 2009 and 2023, technical efficiency, pure technical efficiency, and scale efficiency values were calculated with DEA. According to DEA results, banks in the Turkish banking sector generally have high efficiency scores. On a bank group basis, public owned banks achieved, on average, the highest efficiency scores and demonstrated performance closer to full efficiency, especially during crisis periods. Privately-owned banks also exhibited high efficiency but generally lagged public capital banks. Foreign owned banks, while showing good performance in some periods, displayed more volatile efficiency performance during cyclical shocks. ANOVA and Tukey's HSD tests revealed a statistically significant difference in average efficiency scores between public owned banks and privately-owned banks and foreign owned banks, while no significant difference was found between privately-owned banks and foreign owned banks. The results of the TOBIT analysis indicate that the equity/assets ratio has a strong, positive, and statistically significant effect on banks' technical and pure technical efficiency, while the loans/assets ratio has a positive and significant effect on technical and scale efficiency. Market share was found to have a positive effect on technical and scale efficiency but no effect on pure technical efficiency. The capital adequacy ratio (CAR) was found to have no significant effect on technical efficiency but a negative effect on pure technical efficiency. The GDP growth rate, in general, did not have a significant effect on bank efficiency values.