Abstract
We investigate differences in the use of credit guarantees among banks by using a bank-firm matched dataset from Japan. By following Khwaja and Mian (2008) to control for borrower characteristics and the supply factors of guarantees using firm fixed effects, we extract the part of the ratio of guaranteed loans to total loans that depends solely on banks’ factors for the demand for guarantees. We find significant differences in this after-control ratio for some banks, and the distribution of the ratio is significantly different from that of an uncontrolled ratio based on publicly available data. We further find that the controlled ratio does not depend on the financial conditions of the respective banks such as the capital ratio of the banks, which indicates that other observable and/or unobservable bank characteristics determine these differences in the use of credit guarantees across banks.