Corporate spinoffs and executive compensation

Frontiers of Business Research in China

Table 6 Change in governance & post spinoff pay-performance elasticity regressions

	Dependent Variable: Log Difference in Total Compensation
	Component 1 Board Structure	Component 2 Committee Independence	Component 3 Board and Committee Meetings	Component 4 Institutional Ownership	Component Total
Intercept (b1)	0.866***	1.080***	0.454**	0.662***	1.046***
Intercept (b1)	(3.45)	(6.45)	(2.28)	(5.65)	(3.23)
Rtn _t (b2)	0.227	−0.349	− 0.587	0.122	− 0.608
Rtn _t (b2)	(0.41)	(−1.34)	(− 1.57)	(0.59)	(− 1.88)
Rtn _t-1 (b3)	0.318	−0.075	0.398	0.305**	−0.086
Rtn _t-1 (b3)	(0.57)	(−0.32)	(1.24)	(2.01)	(−0.22)
∆RCG (b4)	−0.002	− 0.006***	0.003	−0.002	− 0.001
∆RCG (b4)	(−1.08)	(−2.57)	(1.35)	(−0.55)	(−1.12)
Rtn _t × ∆RCG (b5)	0.000	0.007*	0.008*	0.002	0.002**
Rtn _t × ∆RCG (b5)	(−0.09)	(1.88)	(1.85)	(0.46)	(2.34)
Rtn _t-1 × ∆RCG (b6)	0.002	0.006	−0.002	0.001	0.001
Rtn _t-1 × ∆RCG (b6)	(0.31)	(1.63)	(−0.44)	(0.25)	(0.93)
N	178	169	162	196	151
R ²	0.102	0.089	0.089	0.041	0.052

This table reports the coefficients on the post-spinoff pay-performance elasticity regression interacted with the change in composite corporate governance score, ∆RCG, for the post-spinoff parent firms from Year + 1 to Year + 3. The dependent variable for all regressions is the first difference in the log of total CEO compensation. Shareholder wealth is measured by stock return. ∆RCG is a score based on the rank of change in the average corporate governance variables of parent firms for two years before spinoff and three years after spinoff in the 4 corporate governance components as previously defined in Table 5. Component Total is a composite score which denotes the sum of all ∆RCG scores for the 4 corporate governance components. A higher value ∆RCG denotes stronger governance improvement. Other variables are previously defined. The t-statistics, which are based on White standard errors robust to within firms’ cluster correlation, are reported in parenthesis. ***, ** and * denote significance at the 1%, 5%, and 10% level respectively. Variation in sample size is due to data availability