عوامل تعیین کننده عملکرد بخش بانکداری در بازارهای مالی جهانی سازی شده؛ مورد کشور ترکیه

This study attempts to give an insight into the trend in the performance of the Turkish banking sector by conducting a panel data fixed effects regression analysis. The results reveal that efficiency change is negatively related to the number of branches. We find a positive relationship between the loan ratio and the performance indices efficiency and efficiency change. Furthermore, bank capitalization is positively related to efficiency change. Interestingly however, return on equity is not statistically significant in explaining any of the efficiency measures. There is also no robust relationship between foreign ownership and efficiency. Finally, restructuring attempts in post-crises epoch robustly account for the improvement in efficiency scores in recent years.

The 1990s, characterized by unstable macroeconomic performance, were the lost decade for Turkey. The financial sector and specifically the banking industry, which makes up around three-fourths of the financial system, experienced a period of high and volatile inflation and interest rates. Political pressures were felt considerably in the banking sector throughout the 1990s. The main motivation behind the banking sector activities and opening up of new banks was to satisfy the financial needs of group companies in a crowding-out environment in which funds were mainly channeled to the public sector. Consequently, around one-fourth of the banks in the sector were taken under the control of the Saving Deposit Insurance Fund. The group lending and the “connected lending” were two important reasons leading to the banking crisis in Turkey.1 Following the crises, the May 2001 Rehabilitation Program was launched by the Turkish Banking Regulation and Supervisory Agency (BRSA) [2]. With the help of this program, state and private banks were restructured. Moreover, the profitability and stability of the Turkish banking system increased [17] and [3] Although the sample period in this study covers the period 1990–2006, we are mainly concerned with determinants of the bank performance during the post-crisis era. Grigorian and Manole [10] conducted one of the studies that estimated the efficiency of the banking sector in transition countries. Utilizing the Data Envelopment Analysis (DEA) methodology,2 they ran the regression of the efficiency scores on variables related to macroeconomic environment, regulatory environment and bank-specific variables. Aysan and Ceyhan [5], Isık and Hassan [11], Isık and Hassan [12], Ozkan-Gunay and Tektas [16], among others, examined the performance of the Turkish banking sector. These studies focused on how the efficiency and productivity of the Turkish banking sector evolved over time, but not on the underlying reasons. Isık and Hassan [11] found the correlation of the efficiency values with such indicators of financial performance as “total cost/total assets, total assets/number of employees, net income/total assets and net income/total equity”. However, the study covered the period between 1988 and 1996. Yıldırım [21] investigated the relationship between efficiency and variables such as asset quality, profitability and bank size during the period 1988–1999. Hence, there exists no study covering the most recent period, and this study aims to fill this gap in the literature by identifying the determinants of the performance of the Turkish banking sector between 1990 and 2006 also comparing the performance levels of pre- and post-crises periods.3 In this study, we regress some performance indices (technical input efficiency, Malmquist Total Factor Productivity Change (TFPC) Index and its mutually exclusive and exhaustive components of efficiency change and technological change) on the foreign–domestic dummy, number of branches, bank capitalization, loan ratio, return on equity (ROE), dummies for the 1994 and 2001 crises and dummy for the reform period. We include all the banks in the Turkish banking industry except the state banks, development banks, investment banks, and the banks with insufficient report of data. This study suggests that the number of branches is negatively related to efficiency change. Moreover, bank capitalization is positively related to efficiency change. Furthermore, loan ratio is positively related to efficiency and efficiency change.4 Interestingly however, return on equity is not statistically significant in explaining any of the efficiency measures. There is also no robust relationship between foreign ownership and efficiency. Finally, restructuring attempts in post-crises epoch robustly account for the improvement in efficiency scores in recent years. This paper is organized as follows. The next section explains the performance indices used in this study and describes the dependent and independent variables as well as the data used. The third section describes the model and provides the intuition about the regression results. The last section concludes the paper.

The Turkish banking sector experienced a performance improvement after the restructuring process following the 2001 crisis. Many banks that were operating inefficiently closed down or merged with stronger banks. As a result, average performance indices of the sector increased. This study attempts to find out how different performance indicators are affected by bank-specific characteristics with the help of fixed effects panel data regression analysis. The dependent variables are technical input efficiency, Malmquist Total Factor Productivity Change (TFPC) Index, efficiency change and technological change. The independent variables are number of branches, bank capitalization, loan ratio, return on equity, foreign–domestic dummy, dummies for the 1994 and 2001 crises and dummy for the reform period. The sample period is 1990–2006 while special emphasis is given to the period after 2000. The sample consists of all banks in the Turkish banking industry except the state, development and investment banks. The regression results reveal that the number of branches is negatively related to efficiency. We explain this with the fact that opening up new branches increases costs and results in lower efficiency levels. One other result from the regression analysis is that there is a positive relationship between bank capitalization and efficiency. The justification for this result is that bank capital is like a deposit insurance that increases the amount of deposits at a bank. Loan ratio is positively related to efficiency and efficiency change. This finding explains the fact that a bank which gives higher percentage of its assets as loans is more likely to have a higher return on assets than other banks. Hence, these banks also have higher performance indices. Interestingly however, return on equity is not statistically significant in explaining any of the efficiency measures. There is also no robust relationship between foreign ownership and efficiency. Finally, restructuring attempts in post-crises epoch robustly account for the improvement in efficiency scores in recent years.