臺大管理論叢 NTU Management Review VOL.27 NO.2S

276 /342

美國產險業 CEO 更迭與再保險需求
276
Table 9 Regressions of Reinsurance Demand on CEO Turnover with SOX Act
Dependent variables ∆Reins ratio ∆Reins_aff_ratio ∆Reins_nonaff_ratio
Independent Variables Estimate P value Estimate P value Estimate P value
SOX -0.007 0.738 -0.018 0.408 0.011 0.162
Turnover (t-1) 0.027 0.285 0.012 0.646 0.015* 0.095
SOX×Turnover (t-1) 0.019 0.523 0.025 0.403 -0.006 0.562
Mutual (t-1) -0.001 0.984 0.017 0.801 -0.018 0.445
SOX×Mutual (t-1) 0.000 0.995 0.000 0.991 0.000 0.989
Duality (t-1) -0.035* 0.090 -0.022 0.303 -0.013* 0.073
SOX×Duality (t-1) -0.041* 0.054 -0.050** 0.023 0.008 0.287
∆ Board size 0.002 0.570 0.002 0.650 0.000 0.765
SOX× ∆ Board size -0.003 0.285 -0.004 0.259 0.000 0.834
∆ Independent director 0.072 0.147 0.016 0.752 0.056*** 0.002
SOX× ∆ Independent director -0.071 0.202 -0.019 0.731 -0.051*** 0.010
Big 4 auditor (t-1) 0.011 0.643 0.003 0.883 0.007 0.385
SOX×Big 4 auditor (t-1) -0.017 0.545 -0.018 0.518 0.001 0.886
∆ Ln(na) -0.278*** 0.000 -0.288*** 0.000 0.009 0.521
∆ Herfindahl 0.019 0.826 0.067 0.448 -0.048 0.127
∆ Geoherfindahl -0.473*** 0.000 -0.483*** 0.000 0.011 0.775
∆ Leverage 0.033 0.780 -0.168 0.155 0.201*** 0.000
∆ Underwriting risk -0.306*** 0.000 -0.225*** 0.005 -0.081*** 0.005
∆ 2yearLossDevelopment 0.090** 0.012 0.136*** 0.000 -0.046*** 0.000
∆ Coastal prem -0.355** 0.017 -0.304** 0.043 -0.051 0.339
∆ Long-tail 0.249*** 0.007 0.235** 0.012 0.014 0.673
∆ Tax_ex -0.021 0.405 -0.024 0.350 0.003 0.748
∆ ROA 0.048 0.747 -0.043 0.778 0.092* 0.092
Group (t-1) 0.010 0.769 0.024 0.472 -0.014 0.227
Intercept 0.027 0.678 -0.018 0.408 0.011 0.162
Hausman Test 13.07 12.08 36.79
R-Square 0.090 0.088 0.128
N 2,772 2,772 2,772
Note: The table shows the regression results of reinsurance demand on CEO turnover, organizational
structure and corporate governance variables with SOX Act. “∆x” means change in x. Specifically
it sugggets that ∆xi,t means x i,t minus xi,t-1 . Please see definition of all variables in Appendix A.
***significant at 1%, ** significant at 5%, * significant at 10%.