30 /274
公司盈餘平穩化行為與盈餘資訊性之關係－合格境外機構投資者角色之檢測
30
5.2 Alternative Influential Criteria of QFIIs’ Shareholdings Examinations
This study uses the continuous variable to measure the influential power of QFIIs on a
firm’s earnings reporting in the IS_QFII model. A potential concern is whether the QFIIs
with small ownership can play a role with their influential power on firms’ income
smoothing strategies. Therefore, this study uses a dummy variable for the QFIIs with high
ownership to replace the initial continuous variable for QFIIs’ownership in Equation (2) and
reruns the regressions. To gain confirmatory support for the initial results, this study uses
both 5% and industry-year median cutoff criteria to measure QFIIs’ influential power.
Namely, the dummy variable of QFIIs is denoted as 1 for firms with QFIIs’ ownership of
more than (or equal to) 5% (or the industry-year median), otherwise 0. The extracted results
are reported in Panel B of Table 7 (denoted as IS_QFII model (5%) and IS_QFII model
(median)). From Table 7, the coefficients of QFII
t
*IS
t
*X
t3
in the 5% and industry-year
median cutoff rate models are -0.074 (
t
= -1.72) and -0.109 (
t
= -3.07) in the post-
deregulation subperiod, and they are negative and statistically significant at the 10% and 1%
levels. The coefficients of QFII
t
*IS
t
*X
t3
in the pre-deregulation subperiod are again
statistically insignificant. Thus, the findings are unlikely to be impacted by the influential
power measurement of QFIIs’ ownership.
We also use the industry-year median as alternative cutoff criteria to calculate the
dummy variable of HH and HL. The QFII_HH (QFII_HL) variable is denoted as firms with
high QFIIs’ ownership which are more than the industry-year median with high (low)
shareholdings volatility. The extracted results are denoted as the IS_HH_HL model (median)
and reported in Table 7. It is found that the coefficients of QFII_HH
t
*IS
t
*X
t3
and QFII_
HL
t
*IS
t
*X
t3
are -0.142 (
t
= -6.31) and -0.106 (
t
= -2.06) in the post-deregulation subperiod,
respectively, and are negative and statistically significant at the 1% and 5% levels.
Hypothesis 2a (2b) is again supported (not supported) in the post-deregulation subsample
testing. The result in the low shareholdings volatility case to the extent can be attributed to
the inherent characteristic (i.e., focusing on information gathering and trading for private
gain) after the deregulation. The coefficients of QFII_HH
t
*IS
t
*X
t3
and QFII_HL
t
*IS
t
*X
t3
are
-0.007 (
t
= -0.26) and -0.007 (
t
= -0.13) in the pre-deregulation subperiod, both statistically
insignificant. Thus, the deregulation of QFIIs’ ownership restrictions does indeed enhance
the opportunistic role of QFIIs and reduces the informative component of earnings for firms
with income smoothing.