My work examines the lending behavior of U.S. banks to SMEs in the aftermath of natural disasters. The study is based on 2016-2019 panel data at bank-county-neighborhood-year level. To identify a causal effect of natural disasters on bank lending, I use an ordinary least squares regression model with group fixed effects. My results demonstrate that there are differences in post shock lending behavior of affected local versus non-local banks. In normal times a strong positive relationship can be observed between SME lending and local bank presence, my results show robust evidence for the existence of a negative disaster lending effect. After a natural disaster, local banks lend significantly less to firms in affected areas tha...