What's Different About Banks? Depositor Discipline and Active Monitoring

Abstract

In this paper we provide an empirical link from Fama (1985) and James' (1987) observation that banks are special to Diamond and Rajan's (2001) model of the banking firm as an active monitor financed with demandable debt. Drawing on a sample of 423 bank loan announcements made during 1988-1996, we document that borrower abnormal announcement returns are positively related to proxies for the bank's ability (or incentives) to monitor. Using a sample of 4,534 bank loans originated during 1987-2003, we find that an index of superior monitoring ability based on these factors is positively related to the loan yield spreads paid by bank borrowers. These two findings are evidence that 1) bank borrowers pay a premium for bank monitoring services; and 2) the excess returns associated with bank loan announcements are in fact a reflection of the monitoring benefits generated by the bank. The proxies that are related to both abnormal announcement returns and borrower spreads are 1) the ratio of the bank's uninsured deposits to total loans, 2) a risk-adjusted measure of recovered charge offs, and 3) the bank's relative capital ratio. Thus, our results reveal a heretofore empirically undocumented benefit of uninsured demand deposits: they appear to improve the bank's incentives to monitor borrowers. The first two proxies are strikingly consistent with the theory of banking laid out in Diamond and Rajan (2001). As predicted by Diamond and Rajan (2001), our empirical results indicate that, for the best bank monitors, monitoring skill and fragile (uninsured) demand deposits co-exist.