4 /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
4
markedly between firms due to unobservable firm-specific traits (Henderson and Kaplan,
2000). With unbalanced-panel data analysis, especially as the estimation focuses on within-
firm variations, omitted variables bias and survivorship bias are avoided, allowing this study
to gain confirmatory empirical results to support our conjectures.
This paper is organized as follows: Section 2 describes the institutional characteristics,
related studies, and develops the empirical hypotheses; Section 3 presents the empirical
design; Section 4 presents and discusses the empirical results; Section 5 presents the
robustness test; and Section 6 concludes the study.
2. Institutional Characteristics, Related Research and Hypotheses
2.1 Institutional Characteristics
Like many emerging Asian capital markets, Taiwan has experienced rapid economic
development in past decades. According to the published data provided by the Taiwan Stock
Exchange Corporation, the percentage of trading dollar amounts by QFIIs increased from
5.9% in 2001 to an impressive 18.4% in 2010. Consequently, the role of QFIIs in Taiwan’s
capital market has garnered a great deal of attention. Taiwan has weak corporate governance,
inadequate shareholder protection, and heightened stock market volatility (Bekaert and
Harvey, 2003). This institutional characteristic provides an adaptive circumstance to breed
managerial strategic earnings reporting, such as income smoothing. It is worth a try to
understand the role of QFIIs on a firm’s income smoothing decision in Taiwan.
Starting in 1991, QFIIs were allowed to participate in Taiwan’s stock market and since
then, QFIIs have been denoted as qualified foreign institutional investors. QFIIs were
restricted to owning 5% of shares for each stock at the beginning of the deregulation. The
fraction of ownership by each QFII for each stock was increased to 50% in 1999. There was
further deregulation in 2003. Following the gradual deregulation of QFIIs’ ownership
restrictions, trading volume by these investors has increased. The annual firm level month-
average ownership ratios by the QFIIs have gradually been increasing between the years
2000~2007, yet the QFIIs only account for a 7.027% of firm-level ownership on average.
This relatively low ownership to some extent imposes restriction on the monitoring role
played by the QFIIs. Moreover, some degrees of herding among QFIIs were also expected,
(Lien, Tseng, and Wu, 2013) for they are unfamiliar with the Taiwan capital market. For
example, Chen, Wang, and Lin (2008) found that QFIIs follow and/or herd each other into
and out of the same securities in Taiwan’s stock market. They also found that characteristic
herding and investigative herding explain QFIIs’ trading behavior in Taiwan. Recently, Ting