In this paper, we investigate how the use of internal credit rating systems impacted the lending activity in Italy, during the 2010-2014 downturn. We argue that Irb models, both validated by supervisory authorities or adopted only for management purpose, contributed to reduce the offering of credit, given their rigidity that do not consider – or very little – soft information. Further, we explore the relation between Irb systems and not-performing loans, to test the claimed superiority of Irb methods in measuring credit risk. We find a negative correlation between Irb and credit reduction, whereas a positive correlation with Npls.
