

9
臺大管理論叢
第
28
卷第
1
期
protection or higher levels of institutional holding. Hanlon and Slemrod (2009) find a
negative market reaction to news about a firm’s involvement in tax shelters, which is
consistent with the notion that investors are concerned about managers using tax shelters
for diversion and earnings manipulation, even though reducing tax payment can increase
shareholders’ benefits. The negative reaction is more pronounced in firms with weak
corporate governance.
Finally, two studies have recently examined the relation between tax avoidance and
agency costs between controlling shareholders and non-controlling shareholders. Chen,
Chen, Cheng, and Shevlin (2010) argue that as family owners have bigger agency
conflicts between dominant and small shareholders, family owners' greater influence
provide them more opportunities to seek rents, through transactions such as tax avoidance
activities and related-party transactions. Thus, the potential private benefits from rent
extraction can be bigger for family owners than for managers in non-family firms.
However, in the U.S., the potential penalty imposed by the IRS and the litigation costs are
likely more substantial to family owners than to CEOs. Thus, Chen et al. (2010) find that
family owners exhibit lower tax aggressiveness because they are concerned with potential
damage on family reputation and the potential penalty imposed by IRS. Our results differ
from Chen et al. (2010) in one main aspect. Chen et al. (2010) examines the agency
conflicts between controlling and non-controlling shareholder using family firms in the
U.S., where the agency conflicts are not so pronounced compared to those in East Asia.
One main reason is that family owner's greater reputation and litigation concerns in the
U.S. may mitigate the conflicts. For example, Chen, Chen, and Cheng (2008) find that
family firms provide less voluntary disclosure than nonfamily firms, but they are more
likely to give earnings warnings to preempt the negative publicity that can result from not
issuing warnings. Thus, we would like to re-examine whether a high level of tax
aggressiveness is associated with higher agency conflicts between controlling shareholder
and non-controlling shareholders.
The second paper that link tax avoidance with the agency costs between controlling
shareholder and non-controlling shareholder is McGuire, Wang, and Wilson (2014), who
investigate the effect of dual class ownership structure on the level of firms’ tax
avoidance. The dual class ownership structure entails agency problems between
controlling shareholder and non-controlling shareholder, because the superior class of
stock allows insiders to control a majority of the votes but have claims to a minority of the
firms' cash flows (Gompers, Ishii, and Metrick, 2010). Using the quiet life theory