381 / 414
臺大管理論叢
第
27
卷第
1
期
381
Variable
Mean Std. Dev.
25
Percentile
Median
75
Percentile
Financial-Control Variables
Sales (in millions)
7286.68 25574.64 608.17 1596.14 4803
Interest expense to sales
0.015
0.022
0.002
0.008
0.020
Stock return
0.010
0.036
0.008
0.011
0.029
R&D expense to sales
0.000
0.002
0
0
0
Table 4 shows the complete correlation matrix between the variables in the regression.
There is a significant positive correlation between the incidence of equity-based
compensation and board size, which suggests that if there are more directors on the board,
firms tend to compensate audit committee members with stock and options. The correlation
between the occurrence of equity-based compensation and the proportion of cash fees to total
compensation is negative, suggesting that if a company gives directors more cash, given a
fixed total compensation, the proportion of compensation in stock and options is less.
Internal corporate governance is significantly negatively correlated with equity-based
compensation, while external governance is significantly positively correlated with equity-
based compensation. The correlation between equity-based compensation (
EQUCOMP
) and
the proportion of audit committee members receiving compensation from other firms
(
OTHER_EXE
) is significantly positive, which is different from our expectation. The
insignificance of the correlation between
EQUCOMP
and
FREE_TOBIN
and
OVERLAP_
AUCOMP
and the contradictory results on the correlation between
OTHER_EXE
and
EQUCOMP
may result from the influence of omitted factors on the univariate correlation
test. In section 5.2, we conduct a multivariate regression analysis to control for other factors
that might affect audit committee members’ equity-based compensation.
We also compare characteristics of firms that compensate their audit committee
members with stocks and options to those of firms that do not. Table 5 presents the results of
this univariate analysis. The first and the second column present mean estimates of board,
governance, and firm characteristics for the two subsamples respectively. The third column
presents the absolute differences between the two subsamples and their significance.
The result shows that firms without equity pay tend to compensate their audit
committee members with cash. Also, CEOs in firms without equity-based compensation tend
to own more shares and have higher probability to be the chairman of the board than those in
firms with equity-based compensation. Stock return seems to be higher in firms that do not
pay equity-based compensation to audit committee members.