More on the Link between Deposit Insurance and Depositor Behavior: Experimental Results

Very Preliminary; Please do not quote 2 Abstract: More on the Link between Deposit Insurance and Depositor Behavior: Experimental Results This paper employs an experimental model to provide additional insight into the effect explicit deposit insurance has on depositor behavior in the time of a bank crisis. The strength of the conjoint approach used in this study is that it combines the control of a laboratory experiment with the external validity of a field-based survey. Respondent samples are selected from banking markets with and without an established system of deposit insurance. Given an extensive amount of literature in this field, our core hypothesis is that the respondents from the latter market will exhibit more significant monitoring behavior even after allowing for insurance features. We use an empirical model with both demanded interest cost and percentage withdrawal rate as correlated dependent variables. The results first indicate that depositors in this study did not place importance on the varying deposit insurance components in their response to the banking crisis. The primary influence on their behavior is the magnitude of deposits at risk as defined by both the level of the insured limit and the size of the deposit relative to that limit. While the role of bank fundamentals, as measured by the level of buffer capital, was generally significant, there was no significant difference among the samples. These results raise questions about how explicit deposit insurance does affect the monitoring intensity of depositors given a credible implicit insurance system. Important differences in the subsamples also provide added insight into the motivations associated with variations in depositor monitoring.