This study aims to provide a fresh look at banks as lenders in and extending past the COVID-19 crisis, with a particular focus on examining the results of recent work by Lei et al. (2020).
The authors’ replication, as well as the original paper, uses a fixed-effects model on panel data. The authors discuss issues regarding data sources as well as use an array of panel data robustness checks to help ascertain an appropriate empirical specification for continued research of this type.
The authors show that the results of Lei et al. (2020) are sensitive to the data source, as well as the construction of the standard errors in their regression framework, with an appropriate specification uncovered through panel data statistical tests. The authors also provide some extensions to the original work by including interacted fixed-effects models and extending the sample period from 2020Q1 to 2021Q1, noting some changes in results.
The authors provide novel results on banks’ lending constraints both at the onset of the COVID-19 pandemic and shortly thereafter. The study also provides an empirical framework for future studies conducted on similar panel data sets.
