The effect of 2020 lockdown on bank internal ratings

Abstract We study how bank internal ratings were affected by the first lockdown shock in Italy (March-June 2020). Using monthly proprietary data from one regional bank between October 2019 and January 2022, and adding credit registry data and balance sheet data for the client firms, we estimate the lockdown effect on the change in six different ratings that the bank assigns to client firms at June 2020 and December 2020. We obtain three main findings. First, controlling for both supply and demand factors, the lockdown is associated to a decrease in four out of six ratings. Second, the lockdown effect depends also on specific firm characteristics, that is the lockdown effect appears to be mitigated in the presence of high sales or for firms that belong to essential sectors. However, these interaction effects do not hold for all ratings suggesting that each type of rating is influenced by different firm characteristics, so that they complement each other. Third, once we account for a full-interaction model, the lockdown effect disappears in the medium run.