Additional evidence on the information-based contagion effects of bank failures

Recent theoretical models addressed the question of the nature of bank runs and what triggers them. Two competing hypotheses emerged: pure panic and information-based contagion. This study provides additional evidence consistent with the latter hypothesis. Three observable bank characteristics are examined as proxy measures for the interim private information used by rational depositors in assessing the riskiness of a bank's long-lived assets that may trigger bank runs. The three factors are (1) the distance of the solvent banks' headquarters from the headquarters of each failed bank; (2) the size of the solvent banks; and (3) the capital ratio as a proxy for their solvency. The analysis is conducted in the context of the five large bank failures that occurred in the Southwest region of the US during the mid-1980s. Weekly abnormal returns of 33 Southwestern BHCs, in ten critical failure-related event dates are regressed on the three observable bank characteristics. Our findings suggest that distance and capital adequacy are negatively related to the magnitude of the contagion effect, whereas size is positively related.