臺大管理論叢 NTU Management Review VOL.29 NO.1

CEO Incentives and Bank Liquidity Management 306 S&P 1,500 banks Independent Variable 1992-2012 1992-2006 2007-2012 1992-2012 1992-2006 2007-2012 (7) (8) (9) (10) (11) (12) Constant -0.004 0.008 -0.067 -0.031 -0.023 -0.046 (-0.12) (0.17) (-0.92) (-0.82) (-0.50) (-0.64) CashBonus i,t-1 -0.011 -0.018 -0.002 -0.007 -0.017 0.009 (-0.98) (-1.28) (-0.12) (-0.63) (-1.26) (0.47) EquityIncentive ($) i,t-1 0.012 0.007 0.041 (1.26) (0.61) (2.04)** EquityRiskExposure ($) i,t-1 0.031 0.027 0.044 (3.45)*** (2.59)*** (2.48)** EquityIncentive (%) i,t-1 0.061 0.064 0.041 (6.16)*** (5.43)*** (1.95)* EquityRiskExposure (%) i,t-1 0.022 0.022 0.025 (2.53)** (2.08)** (1.43) LoanCommitment i,t-1 0.592 0.671 0.356 0.633 0.698 0.440 (12.07)*** (11.66)*** (3.78)*** (13.57)*** (12.76)*** (4.85)*** DepositBase i,t-1 1.860 2.025 1.236 1.675 1.823 1.166 (14.64)*** (13.94)*** (4.75)*** (13.22)*** (12.51)*** (4.45)*** LoanCommitment i,t-1 * DepositBase i,t-1 -5.983 -6.891 -3.316 -5.343 -6.107 -3.073 (-12.23)*** (-11.91)*** (-3.84)*** (-11.13)*** (-10.65)*** (-3.56)*** BankCapital i,t-1 -0.185 -0.196 -0.139 -0.275 -0.282 -0.178 (-1.78)* (-1.71)* (-0.47) (-2.73)*** (-2.56)** (-0.57) Share of Deposit Financing i,t-1 0.030 0.012 0.116 0.050 0.040 0.090 (0.69) (0.24) (1.48) (1.17) (0.79) (1.17) BusinessCycle t-1 -0.331 -0.174 -0.855 -0.456 -0.460 -0.706 (-1.39) (-0.36) (-2.07)** (-1.99)** (-0.97) (-1.73)* Fixed effect Yes Yes Yes Yes Yes Yes N 299 227 72 288 217 71 R-square 0.555 0.589 0.507 0.589 0.622 0.512 F test of excluded instruments (40.02)*** (34.49)*** (7.09)*** (44.23)*** (37.79)*** (7.10)*** 4.3.2 CEO Incentives, Bank Liquidity Holding, and Bank Performance Finally we conducted 2-stage least square regressions to explore the relation between CEO incentives, bank performance, and liquidity holding policies. We used the predicted value of liquidity holding from regression estimation (1) to run the regression of bank performance. Tables 11 and 12 show the results for bank performance measured as ROA and ROE, respectively. Similar to the previous results, we reported the results for the whole sample, S&P 1,500 banks, and non-S&P 1,500 banks. In addition, the empirical relation was