Faculty of Economic Sciences and Management of Sfax, Tunisia (email@example.com)
Faculty of Economic Sciences and Management of Sfax, Tunisia (firstname.lastname@example.org)
The peculiarity in lending methodology in Microfinance institution characterized by unchallenged dominance of the loan officers in the decision-making prompted us to investigate the predictive accuracy of their subjective judgement. In addition, we investigate if the accuracy of this information depends on the strength of lenders-borrowers relationship. The objective of this paper is to understand the loan officer behaviour in default prediction task. Using credit file data from Tunisian Microfinance bank, we have found evidence that subjective judgment of loan officers has statistically significant predictive capability of default risk. In addition, our results suggest that novice loan officers have incorrect prediction of future default events because they are prone to different behavioral biases. Novice loan officers may overweight any information that is consistent with their existing beliefs and overestimate the precision of their own private information. However, through learning experienced loan officers have less behavior bias and more accurate feeling allowing them to make an accurate default prediction. To make correct decision in relationship lending approach, credit committee can rely on subjective judgment of experienced loan officers mainly when the latter judged the creditworthiness of frequent borrowers. As soft information is not easily and accurately transferable, Microfinance institutions have more interest to retain their experienced loan officers than their borrowers.
Keywords: subjective judgment; loan officer; predictive accuracy; expertise; microfinance institution