We examine the monitoring and ex-post influence of depositors on risk-taking of U.S. bank holding companies (BHCs) from September 1986 to December 2013. As the basis for our empirical analysis, we develop a theoretical model which shows that under risky lending and deposit insurance, a bank’s liability and asset choices are interrelated through its probability of insolvency. Our empirical results are as follows. First, for the sub-sample of the ten largest (Top10) BHCs, deposit risk pricing only exists over some sub-periods prior to the 2007 financial crisis. However, interest rates on insured deposits and uninsured deposits for the Non-Top10 BHCs increase with bank risk over the whole sample period. Moreover, the growth rates of insured and uninsured deposits tend to decrease as bank risk increases for Non-Top10 BHCs over the entire sample period, but only in some sub-periods for the Top10 institutions. Second, although Top10 BHCs do not increase the insured deposits-to-liabilities ratio to weaken market discipline over the entire sample period, all other institutions engage in such regulatory arbitrage in some sub-periods. Third, higher risk premium embedded in current deposit interest rates is more likely to reduce future insolvency risk of troubled BHCs. This suggests that depositors monitor the riskiness of BHCs while also exerting strong ex-post influence on risk-taking of problem institutions. Fourth, in the post-Dodd-Frank Act/Basel III period, the interest rates, the shares, and the growth rate of insured deposits for the Top10 BHCs are significantly negatively related to bank insolvency risk. This could be due to strengthened regulatory oversight on the largest high-risk institutions and is consistent with a substitution relationship between depositor discipline and regulatory oversight.

We examine a sample of 670 firms that announce asset purchases. We hypothesize that buyer announcement returns should be higher in the presence of better monitoring and better governance. Consistent with the monitoring hypothesis, we find that buyers with higher private debt make purchase decisions that increase shareholder value. Consistent with the governance hypothesis, we find that returns are higher for buyers that have lower antitakeover provisions in place. Consistent with the managerial discretion hypothesis, buyer announcement-period returns increase with buyer leverage. Consistent with the liquidity hypothesis, we find that announcement-period returns decrease with the seller's Z-score, suggesting that buyers benefit from the lower liquidity of assets sold by sellers with lower debt capacity and higher financial distress. We also find that buyer announcement-period returns are directly related to their operating performance in the post-purchase year.