Contagious Bank Runs: Evidence from the 1929-1933 Period

Abstract

This paper analyzes the behavior of deposit flows in failed banks and (a control) sample of non-failed banks over the 1929-1933 period. Evidence of significant contagion effects were found for the 1930-1932 period. No apparent contagion effects were found for the 1930-1932 period. No apparent contagion effects existed in the 1929 or 1933. It was also found that the pace of contagion accelerated between 1929 and 1932, indicative of a learning effect among depositors. Interestingly, even in the period of contagion there were a significant number of informed depositors who could distinguish among good and bad banks. This is despite the fact that this period preceded the establishment of the FDIC and the SEC and their associated information production requirement for banks. Finally, our results suggest that in the pre-1933 period there was a significant amount of depositor discipline on bad banks.