Assessment the Impact of Lending Relationships on Lending Transaction Costs: An Artificial Neural Network Model

According to the economics of transaction cost, trades in markets include side costs and philosophy of forming entities is minimizing these costs. Many studies have been conducted in developing countries suggesting that high transaction costs in credit market and this reduces the efficiency of credit institutions in financial markets. This study in first stage, describes the transaction costs imposed on credit institutions­ and the factors affecting them from the traditional model of transaction cost economics and also introduced the "relationship lending with the customers" as a variable that points to reduce the information asymmetry between the borrower and the credit institute. The impact of these variables on the Williamson model will be examined. This study methodology is based on discussion from the experts, questionnaire and neural network model for data collection and analysis. Credit institutes under this study have been collected as a random sample from banks in Tehran .These banks had been authorized by central bank of Iran till March 2012. The results shows that with the introduction of the independent variable "lending relationships" the effect of the independent variables "investment in specific assets", "special collaterals", "degree of uncertainty for the credit institute" and "difficulty in measuring employee performance" on variable of transaction costs of lending will be reduced.