This paper exploits the 1990 change in capital adequacy regulations to construct more powerful tests of capital and earnings management effects on bank loan loss provisions. We find strong support for the hypothesis that loan loss provisions are used for capital management. We do not find evidence of earnings management via loan loss provisions. We also document the reasons for the conflicting results on these effects observed in prior studies. Additionally, we find that loan loss provisions are negatively related to both future earnings changes and contemporaneous stock returns contrary to the signaling results documented in prior work.

Overconfident managers overestimate future returns from their firms’ investments. Thus, we predict that overconfident managers will tend to delay loss recognition and generally use less conservative accounting. Furthermore, we test whether external monitoring helps to mitigate this effect. Using measures of both conditional and unconditional conservatism respectively, we find robust evidence of a negative relation between CEO overconfidence and accounting conservatism. We further find that external monitoring does not appear to mitigate this effect. Our findings add to the growing literature on overconfidence and complement the findings by Schrand and Zechman [2011] that overconfidence affects financial reporting behavior.