Determinants of Bank Efficiency in Turkey: A Two Stage Data Envelopment Analysis

Financial industry plays an important role in the economy and banks are indispensable players in the financial industry. Therefore, the evaluation of banks’ performance became a popular research topic in all over the world, and also in Turkey. There are different techniques to determine the banks’ performance. Among those techniques, Data Envelopment Analysis (DEA), which is a non-parametric technique, has been widely used in the banking sector. In this research, we analyzed the efficiency of Turkish Banking Industry with Data Envelopment Analysis (DEA) methodology between 2002 and 2007. All the banks that constantly operated in the years between 2002 and 2007, excluding investment and development, participation banks, get into the analysis. So there are four groups of banks in the research, those are state-owned deposit banks, privately-owned deposit banks, foreign banks founded in Turkey, and foreign banks having branches in Turkey. In the research model, number of employees, interest expenses, non-interest expenses and total deposit are determined as input, total credits, interest revenue and non-interest revenue are determined as output. This analysis aims to explain the variation in efficiency scores with a set of explanatory variables, such as size, ownership type, nationality, being publicly held. According to results, the efficiency levels do not change very much between 2002 and 2007. The efficiency scores reached top level in 2005 and 2006. The results of regression application denote that all of the explanatory variables have a significant effect on banks’ efficiency levels. According to regression analysis results, size negatively affects the efficiency levels of banks. Publicly listed banks operate more efficient than not publicly listed banks. Foreign owned banks operate more efficient than their domestic peers. Furthermore, state owned banks are less efficient than non-state banks