An empirical study of depositor sensitivity to bank risk

Using a cross-sectional data set on U.S. commercial banks, the argument that during the mid 1980s to early 1990s depositors were insensitive to bank risk is empirically examined. Bank risk is measured by the predicted probability of bank failure as a function of contributing factors. The natural log of bank deposits is then regressed on bank risk and other control variables. The coefficient on bank risk in this equation measures the sensitivity of deposits to bank risk. A close examination of the estimated coefficient on bank risk offers little support for the argument that depositors were insensitive to bank risk.