Application Of The Two Stage Network Data Envelopment Analysis Model Using The Bootstrap Simulation Model: Case Of Turkish Banks

Authors : Rama Kayali, İhsan Alp

Pages : 14-32

Doi:10.55026/jobaf.1560724

View : 77 | Download : 63

Publication Date : 2025-01-13

Article Type : Research Paper

Abstract :This study aims to assess the efficiency of the banking sector in Turkey. To achieve this objective, a two-stage Network Data Envelopment Analysis (NDEA) model was applied. To enhance the accuracy of the results and calculate the bias in each stage, the bootstrap method was used in the NDEA. Specifically, an input-oriented under the constant return to scale (CRS) model was employed to evaluate efficiency. The study included data from the 2022 fiscal year for 13 commercial banks operating in Turkey. In the first stage, three input variables were considered: total assets, number of employees, and number of branches. The output for this stage was the general collected resources, which simultaneously served as the input for the second stage. In the second stage, two output variables were used: Net interest profit and other operating incomes. According to the results, the average efficiency of the banking sector in Turkey was 88.9% in the production stage and 80.7% in the intermediation stage, while the overall average efficiency was also 70.9%, as determined by the two-stage NDEA model. When applying the bootstrap method for the NDEA analysis, the average efficiencies for the three stages were 78.5%, 73.5%, and 56.4%, respectively. This indicates weak performance in the overall efficiency of the banking sector.
Keywords : Network Data Envelopment, Turkish Banks, Bootstrap, Efficiency.