YEH-NING CHENHuang, Rachel J.Rachel J.HuangTsai, JohnJohnTsaiTzeng L.Y.2019-07-242019-07-242013-01-0109208550https://scholars.lib.ntu.edu.tw/handle/123456789/414933© 2013, Springer Science+Business Media New York. Using data from a Taiwanese finance company, this paper empirically investigates the value of soft information, information that requires the subjective interpretation by the loan officers who collect it and cannot be credibly transmitted to others, for making small business loans. It finds that the use of soft information significantly improves the power of default prediction models. It also identifies the types of soft information that are helpful for predicting loan defaults. In addition, it shows that borrowers with more favorable soft information enjoy lower interest rates. These results imply that soft information is important for small business lending.Credit scoring | Default prediction | Small business lending | Soft information[SDGs]SDG8[SDGs]SDG9journal article10.1007/s10693-013-0187-x2-s2.0-84887881788WOS:000348977400005https://api.elsevier.com/content/abstract/scopus_id/84887881788