Banks' loan quality, loan growth and social capital: The case of Turkey banking sector

The purpose of this study is to reveal the effect of social capital on the credit expansion strategy and credit risk of banks in Turkey, a developing country. In the study, during the 12-year period covering the period of 2007-2018, the analysis was carried out by taking the credit data of all public, private and foreign capital deposit banks in Turkey on a provincial basis. In the study, the change in the total loan growth on a provincial basis, the change in the housing loan growth, the change in the growth in the commercial and industrial loans, the change in the growth in the construction loan, the change in the non-performing loans, the ratio of loans to total loans and finally the ratio of non-performing loans to total assets were used as dependent variables. The social capital variable was used as the independent variable. In addition, province-based organ donation rate was used as an alternative social capital variable. While creating the social capital index, the social capital value obtained from the 2 networks and 2 norm indicators used in previous studies was used. It is important to use an appropriate estimator according to whether there is autocorrelation, varying variance and correlation between units in fixed and random effects models examined in panel data applications. The occurrence of one, more or all of these conditions may result in biased results regarding the models to be predicted. For this reason, after the necessary tests were carried out, it was determined that there was a problem of varying variance, autocorrelation and correlation between units in the models, so it became necessary to estimate the coefficient with a resistive estimator. Under these conditions, the Beck-Katz estimator was used to obtain deviation-resistant standard errors from all three assumptions. When N>T, the Beck-Katz estimator can be used. In this study, the results were reported according to the Beck-Katz estimator, which gives more accurate results when T/N<1. According to the results of the analysis, there is a negative and negative correlation between the growth rate of social capital and total loans, the growth rate of housing loans, the growth rate of commercial and industrial loans, and the growth rate of construction loans. It was found that there is a significant relationship. It was also revealed that there is a negative and significant relationship between the social capital variable and the rate of change in non-performing loans, which is the credit risk indicator of banks, the ratio of non-performing loans to total loans and the ratio of non-performing loans to total assets. out. As the level of social capital rises in the provinces in Turkey, the opportunistic behavior of bank employees and managers, as well as their customers, decreases. For this reason, banks' credit growth is slower and credit risk is at a lower level in provinces with high social capital levels. These results support the Principal-Agent Theory and Morald Hazard Theory, which constitute the theoretical background of the study.