Description
Information disclosure is a necessary activity in corporate governance; information transparency plays a
unique role in corporate governance in the era of knowledge-based economy. Lack of transparency can
lead to confusion, misinformation, and distrust. With this in mind, we examine the factors that influence
corporate information transparency in terms of two dimensions: technology intensity and institutional
ownership. Drawing on data from a 2005e2012 cross-section sample of 1391 public firms evaluated by
the official ‘information disclosure and transparency ranking system’ (IDTRS), we find that increases in
domestic institutional ownership for firms in high-tech industries, relative to foreign institutional
ownership, lead to a current-year upgrade in information transparency, but not for firms in other industries.
We also find that firms with increased foreign institutional ownership and high-tech firms with
both increased governmental institutional or corporate ownership and high R&D intensity can sustain a
longer-run upgrade in corporate transparency. Pushing further we also investigate whether corporate
transparency in high-tech industries is negatively affected if governmental institutional or corporate
shareholders are involved in corporate governance, but cannot find strong evidence for such a tendency.
Our results suggest that institutional shareholders promote good corporate governance practices which
gradually improve at the pace of high technology development.
Institutional ownership and information transparency: Role of technology
intensities and industries
Chung-Hao Hsu
a, b
, Syou-Ching Lai
a, *
, Hung-Chih Li
b
a
Department of Accounting and Information System, Chang Jung Christian University, Taiwan, ROC
b
Department of Accountancy, National Cheng Kung University, Taiwan, ROC
a r t i c l e i n f o
Article history:
Received 16 March 2015
Accepted 8 June 2015
Available online xxx
Keywords:
Information transparency
Institutional ownership
Institutional monitoring
R&D intensity
a b s t r a c t
Information disclosure is a necessary activity in corporate governance; information transparency plays a
unique role in corporate governance in the era of knowledge-based economy. Lack of transparency can
lead to confusion, misinformation, and distrust. With this in mind, we examine the factors that in?uence
corporate information transparency in terms of two dimensions: technology intensity and institutional
ownership. Drawing on data from a 2005e2012 cross-section sample of 1391 public ?rms evaluated by
the of?cial ‘information disclosure and transparency ranking system’ (IDTRS), we ?nd that increases in
domestic institutional ownership for ?rms in high-tech industries, relative to foreign institutional
ownership, lead to a current-year upgrade in information transparency, but not for ?rms in other in-
dustries. We also ?nd that ?rms with increased foreign institutional ownership and high-tech ?rms with
both increased governmental institutional or corporate ownership and high R&D intensity can sustain a
longer-run upgrade in corporate transparency. Pushing further we also investigate whether corporate
transparency in high-tech industries is negatively affected if governmental institutional or corporate
shareholders are involved in corporate governance, but cannot ?nd strong evidence for such a tendency.
Our results suggest that institutional shareholders promote good corporate governance practices which
gradually improve at the pace of high technology development.
© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier
Taiwan LLC. All rights reserved.
1. Introduction
A good corporate governance framework should ensure that
timely and accurate disclosure is made on all material matters
regarding the corporation, including the ?nancial situation, per-
formance, ownership, and governance of the company (OECD,
2004
1
). With an eye towards corporate governance framework
and technological knowledge characteristics we investigate the
joint effects of research and development (R&D) intensity and the
four categories of institutional ownership on corporate information
transparency for the three tiers of industries (i.e., high-tech, me-
dium-high-tech and traditional) and the three categories of ?rms
(i.e., government-involved, group-af?liated and manager-
governed) over a sample period from 2005 to 2012.
Proponents of ‘Asian value’ argue that companies led by
governmental institutions have better performance in terms of
in?uencing the course of economic development and income dis-
tribution, especially in knowledge-based innovative services and
industries (Lee, 2003). Asian countries have a long tradition of
governments exerting a lot of authority in various matters. Do
governmental institutions intervening in business affairs of high-
tech ?rms in?uence corporate transparency? By contrast,
following the democratization of Taiwan and Korea, universalism
claims that the Asian experience is not exceptional at all. Stock
markets in Taiwan are ‘plate-form’ markets that are signi?cantly
in?uenced by the international contagion effect. The stock price
?uctuations are seen to exhibit signs of increasing foreign
* Corresponding author.
E-mail address: [email protected] (S.-C. Lai).
Peer review under responsibility of College of Management, National Cheng
Kung University.
1
In conjunction with national governments, other relevant international orga-
nizations and the private sector, the Organization for Economic Cooperation and
Development (OECD) has developed a set of corporate governance standards and
guidelines, ‘The OECD Principles of Corporate Governance’. We quote the foreword
of its principle V: ‘Disclosure and Transparency’, one of the six main principles, as
our writing motivation.
HOSTED BY
Contents lists available at ScienceDirect
Asia Paci?c Management Review
j ournal homepage: www. el sevi er. com/ l ocat e/ apmrvhttp://dx.doi.org/10.1016/j.apmrv.2015.06.001
1029-3132/© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
Asia Paci?c Management Review xxx (2015) 1e12
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
investments acting as a critical factor in swaying capital markets.
Does foreign institutional ownership have signi?cant in?uence on
disclosure quality of ?rms in the plate-form stock markets? Prior
studies have provided inconsistent explanations for the mixed ev-
idence on the in?uence of institutional ownership. Since every
category of institutional ownership has its different competing ef-
fects on information asymmetry, it is unclear whether it is practi-
cable to generalize the results of prior studies to other institutional
shareholders.
Using the IDTRS evaluation from 2005 to 2012, we investigate
the direct causal relations between institutional ownership and
corporate transparency in Taiwan, where national policies are seen
as serving as a hub for the technology industries. Based on the
different characteristics in diversi?ed industries, we correct stan-
dard errors for industry-level clustering because information
transparency is likely to be correlated within an industry since
some industry-speci?c attributes are mandated by accounting and
disclosure rules (De Franco, Kothari, & Verdi, 2011). For instance,
the accounting standards regarding R&D expenditures, intangible
assets, and impairment of assets are highly correlated with
technology-oriented industries. We preliminarily ?nd that ?rms
and industries operating at a higher technology level tend to have
higher R&D intensity and information transparency. We also ?nd
that domestic institutional ownership is positively correlated with
corporate transparency. However, in pursuit of their maximum
interest, the dual roles of stakeholder and shareholder may even-
tually con?ict by controlling managers' decisions that are in con-
?icts with shareholders' interests including transparent
information. We ?nd the correlation coef?cient of institutional
ultimate controlling ownership is signi?cantly negative, suggesting
that a possible role duality con?ict may affect the own-
ershipetransparency relationship.
R&D intensity is more characteristic of industrial technology
?rms than other indicators (Hsu, Lai, & Li, 2015). We use both
R&D-based industry classi?cation and a ?rm's R&D intensity as
the technology factors to moderate the linkages between institu-
tional ownership and corporate transparency. We further imple-
ment multivariate changes regressions to examine the effect of
ownership on transparency at the level of both industry and ?rm.
Our regression analysis of the above relationship leads to four
?ndings. First, we ?nd that technology industry classi?cation is, in
general, more indicative than technology intensity magnitude.
Second, increases in domestic institutional ownership help high-
tech ?rms increase their current-year information transparency,
while the positive effects of increases in foreign institutional
ownership on corporate transparency are deferred to the
following year. Third, ?rms with a higher R&D intensity which are
experiencing increased governmental institutional or corporate
ownership tend to continuously improve their information
transparency in the following year. Finally, by comparing changes
in the controlling ownership of government-involved ?rms and
group-af?liated ?rms with manager-governed ?rms, we cannot
?nd a negative effect of increased controlling institutional
ownership on corporate transparency.
The contributions which this research makes to the ?eld are
summarized in the following. First, we integrate the IDTRS evalu-
ation and the OECD industry classi?cation (OECD, 2011) into our
research; this includes fairly newdata in the literature on corporate
governance. Second, this is the ?rst paper that uses industry-level
and ?rm-level technology intensity together to determine the
moderating effect of various institutional ownership on informa-
tion transparency. Third, though related research is abundant, in
this paper we perform an analysis through a multivariate changes
regression model to provide a deeper insight into the causal rela-
tion between institutional ownership and corporate transparency.
We believe that our focus on of?cial data and standards, direct
causality relationship, and counterevidence can provide more
convincing results than previous research has managed to achieve.
Research background and hypotheses development are explained
in the next section; data and methodology in the third section;
empirical research ?ndings in the fourth section; and the conclu-
sion in the ?nal section.
2. Background and hypotheses
2.1. Background description
In this paper, institutional shareholders are operationally
de?ned as the shareholders of domestic institutions which are
governed by the Taiwan Securities Exchange Act,
2
and foreign in-
stitutions and funds.
3
Depending upon their backgrounds, trading
tendencies and regulatory restrictions, various economic conse-
quences can be caused by institutional ownership (Choi, Lam, Sami,
& Zhou, 2013). Shareholders who focus on a ?rm's short-term
performance tend to be concerned with market-price volatility
but not the ?rm's prospects and management, and therefore, no
positive impact of their shareholdings on corporate governance is
expected. Inversely, expectations regarding a ?rm's future devel-
opment, long-term institutional shareholders will have a greater
in?uence on ?rms. They use different means to monitor the ?rm's
decision-making. Transparent accounting information facilitates
institutional shareholder monitoring and the effective exercise of
shareholder rights under existing securities laws (Bushman &
Smith, 2003).
According to Black (1991), institutional shareholders can be
categorized into ‘Institutional voice’ and ‘Institutional control’. The
former does not participate in management but can pressure and
challenge ?rm managers. By contrast, the latter participates in the
?rm's business by holding key positions, such as directors or
managers. Many prior studies (i.e., Chen, Harford, & Li, 2007;
Ferreira & Matos, 2008; Aggarwal, Erel, Ferreira, & Matos, 2011)
classify institutional investors according to the potential for busi-
ness ties to a ?rm, referring to those with strong ties as “grey
institutions”
4
and those with weak ties as “independent
institutions”.
Corporate shareholders have a strong underlying purpose for
building business ties to a corporation. Their tight business and
monitoring capability can result in ef?cient monitoring or strategic
alignment (Pound, 1988). The shareholdings of governmental in-
stitutions usually can represent governmental policy objectives and
implementation. In the same vein, ?nancial institutional share-
holders may play a role like governmental institutions, or may be
able to make direct fund transfers to ?rms just as corporate
shareholders and grey institutions are able to.
2
In this paper, our domestic institutions comprise all institutions other than
foreign ones. The composition of institutional shareholders is different according to
the country's related law. According to the trading proportion statistics of share-
holder categories of the Taiwan ‘Securities Exchange Act’, the institutional share-
holders in a broad sense are the category relative to human shareholders. However,
in a narrow sense, the Act also categorizes ?nancial institutions, investment and
trust companies and some speci?c funds as the ‘professional institutional
shareholders’.
3
The quali?ed foreign institutional shareholders (QFII) in this paper include all
institutional shareholders who are not domiciled in Taiwan, including foreign in-
stitutions, foreign mutual funds, and foreign corporate shareholders and other
institutions.
4
Grey institutions refer to those investors who have, whether actual or only
potential, business relationships with ?rms in which they invest. The ownership is
mainly the percentage of shares held by bank trust departments, insurance com-
panies, and other institutions (e.g., pension funds, endowments).
C.-H. Hsu et al. / Asia Paci?c Management Review xxx (2015) 1e12 2
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
Domestic mutual fund and foreign institutional shareholders
have independent institutional ownership, because they are more
likely to collect corporate information and are subject to fewer
regulatory restrictions. They have the highest shareholder sophis-
tication but fewer potential business relationships with the cor-
porations. Based on their greater independence, these two groups
are more involved in monitoring corporate management. However,
foreign institutional shareholders seldom participate in local ?rm's
operations because of greater restrictions, fewer information
channels and unfavorable environmental factors.
5
They know that
excessive intervention in invested ?rms can negatively in?uence
?rm's operation and stock price, and may cause a loss for inde-
pendent shareholders (Palmiter, 2002).
Isolating two distinct factors, Bushman, Piotroski, and Smith
(2004) ?nd that the governance transparency factors are primar-
ily related to a country's legal/judicial regime, whereas the ?nancial
transparency factors are primarily related to political economy.
Hence, we argue that the of?cial evaluation is a direct measure of
information transparency since the systematic review is based on
de?nite evaluation metrics at the class and method level.
6
It pro-
vides another type of comparable data for transparency, relative to
the measures based on market reactions (e.g., Choi &Seo, 2008; Liu,
Hsu, & Li, 2014).
In 2002, the Taiwan Securities & Futures Institute (SFI) estab-
lished an independent council for professional evaluation and
launched the IDTRS, through which a two-stage screening process
is conducted. First, all information provided is given a preliminary
screening by the SFI's in-house ranking team based on each
disclosure item in an evaluation year. Second, all companies are
entitled to review the preliminary results via the internet and
respond directly to SFI regarding ambiguous issues in the following
year. Upon receiving the different opinions expressed by com-
panies, the ranking committee assesses the correspondences,
makes adjustments if needed; and the consulting committee de-
cides the ?nal ranking results. About half of the indicators are
concerned with ?nancial statement information, and about forty
percent are designed for voluntary disclosure (including R&D-
related information). The latter causes the major differences in
ratings. The IDTRS thus help all stakeholders monitor corporate
governance.
In Fig. 1 we can see that two subsequent events are scheduled
for the following year. The ?rst is the procedure of review and re-
evaluation. It is possible for evaluated ?rms to change the ?rst-
stage results by improving their website disclosure and ?nancial
forecasts. The second is the year t annual and the year t þ1 quar-
terly reports issuance. However, we ?nd that both the two ?nancial
reports are evaluated through the year t þ1 IDTRS (their issuance
year) and only impact the year t þ1 evaluation result.
2.2. Hypothesis development
Relative to retail shareholders, large external shareholders,
block owners, and institutional shareholders have strong in-
centives to monitor domestic public ?rms (Allen, Bernardo, &
Welch, 2000; El-Gazzar, 1998; Hotchkiss & Strickland, 2003;
Shleifer & Vishny, 1986). Prior studies have found that monitoring
by large external shareholders reduces the agency costs of equities
by effectively controlling managers' decisions when they are in
con?icts with shareholders' interests. Thus, ?rms with long-term
supervision have better prospects for long-term earnings growth
and increases in value. However, some studies (e.g., Porter, 1992;
Bushee, 2001) claim that institutional ownership with short in-
vestment horizons will maximize short-term results at the
expense of long-term equity value or ?rm value in expected near
term earnings.
In the present work we analyze the in?uence of various external
shareholders on corporate transparency. Governmental institutions
are concerned with governmental institutional service and policy
implementation. In Taiwan, Chen, Chen, and Chen (2005) indicate
that ?rms with a higher percentage of ownership by the govern-
ment, ?nancial institutions, and foreign institutions have superior
long-term supervision and better disclosure. Ku (2013) asserts that
a larger board and higher proportion of institutional ownership will
strengthen the information transfer effect. The so-called ‘three
legal entities’
7
in Taiwan's stock markets have signi?cant in?uence
on ?rm's market price, which is why both ?rm managers and se-
curities analysts keep a close eye on their trading tendency. Based
on the in?uence of domestic mutual funds on the trading in sec-
ondary markets, ?rms concerned about the effect of the three legal
entities on market prices may manage to promote their informa-
tion transparency.
Based on the higher economic value of high technology, the
bene?ts accruing from the ?rm's increased value serve as an
important driving force for institutional shareholders to enhance
their in?uence for self-interest. Bearing greater cost and risk in
their investments, institutional shareholders will select ?rms with
high information transparency and require the ?rms to disclose
more information relevant to ?nancial analysis, risk evaluation and
operational plans. Hence, we postulate that institutional share-
holders also have a higher investment preference for information
about ?rm's technology level, especially in high-tech industries.
Financial accounting year t
year t annual report
& year t+1 quarterly report
issuance in year t+1
Month: 1~12 1-2 3-4 5 6-7 8-12
IDTRS evaluation year t
(Stage I)
Review & Re-evaluation
(Stage II)
year t evaluation report
issuance in year t+1
Fig. 1. Timeline of annual reports and IDTRS implementation in Taiwan.
5
In the US, institutional control shareholders of investment ?rms and funds are
also rare, due to the Investment Company Act of 1940 No.17. In Taiwan, few internal
positions are held by representatives of foreign institutions due to compliance with
laws enacted to protect national defense, national communication security, local
industries, local culture, public interests, market price, etc.
6
The IDTRS broadly identi?es 109 disclosure items regarding regulation
compliance, disclosure timeliness, ?nancial forecast publication, and annual reports
as evaluation criteria and gauges the level of corporate transparency by searching
public information disclosed in various ways. For full ranking results, please visit
website:http://www.s?.org.tw.
7
They are the professional institutional shareholders including securities in-
vestment companies (dealers), domestic investment trust funds, and foreign in-
stitutions and funds.
C.-H. Hsu et al. / Asia Paci?c Management Review xxx (2015) 1e12 3
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
Consistent with the monitoring argument (Shleifer & Vishny, 1986,
1997), we establish our hypothesis as follows:
Hypothesis 1. High-tech ?rms experiencing increases in institu-
tional shareholdings have a higher possibility of increasing their in-
formation transparency.
By splitting the sample into high-tech and low-tech com-
panies, Czarnitzki and Thorwarth (2012) ?nd a large premium of
R&D for ?rms in high-tech industries, but no premium in low-
tech sectors. This suggests that rational institutional share-
holders want more relevant information disclosed, especially
about high-tech ?rm's R&D activities and capital-intensive pro-
jects, which concern virtually all of the ?rm's operational results
(Bushee, 1998). On the contrary, too much technology-originated
information disclosed will impair the vested interests of insiders.
A ?rm's insiders will conservatively disclose information to
decrease information transparency. Owing to the ?rm-speci?c
risks of high-tech industries, we expect that corporate trans-
parency will be affected by the percentage ownership of external
institutional shareholders.
Pindado, Queiroz, and Torre (2015) ?nd that effective corporate
governance allows the market to better assess a ?rm's R&D in-
vestments because effective control mechanisms reinforce the
positive effect of R&D on a ?rm's market value. Using order backlog
as a predictor, Jiambalvo, Rajgopal, and Venkatachalam (2002)
assert that institutional investors may be attracted to ?rms in
richer information environments where stock prices tend to lead
earnings. Using R&D expenses as a proxy for forward-looking in-
formation, Chen and Wang (2006) ?nd that institutional investors
can incorporate R&D information that re?ects future earnings,
especially for group-af?liated ?rms. Hence, we know that larger
institutional shareholders are better able to use enhanced corpo-
rate transparency to acquire relevant information. This leads to the
following hypothesis:
Hypothesis 2. As their R&D intensity increases, high-tech ?rms
experiencing increases in institutional shareholdings have a higher
possibility of increasing their information transparency.
However, the associations between corporate characteristics
and disclosures level are moderated by differences in disclosure
index construction, differences in de?nition of the explanatory
variables, and differences in research settings (Ahmed & Courtis,
1999). There are alternative views which are diametrically
opposed to our above two hypotheses. First, Fan and Wong (2002)
argue that, in a region of high-developed knowledge economics,
the holding percentage and voting right enforcement of govern-
mental institutions indicate strategic economic purposes. Morck,
Shleifer, and Vishny (1988) claim that institutional shareholders
place too much emphasis on ?rm's technological capacity and
leadership will increasingly push ?rm managers to seek value-
maximization over all other goals. As a result, in compliance
with institutional shareholder's expectations, ?rms tend to over-
look the importance of disclosure quality and information
transparency.
Second, others argue that the frequent trading and short-term
focus of institutional shareholders encourages managers to
engage in myopic investment behavior. Firm managers may take
advantage of transient institutional shareholders such that they
make decisions (including accounting/reporting decisions) so as to
maximize their own self-interest. As the level of institutional
ownership increases, ?rm managers' opportunistic behavior is
likely to increase even as their ?rm's disclosure level decreases
(Laverty, 1996).
Third, Lambert (1986) indicates that the con?ict of interest
between the executive and the principal may lead to different
selections in risky projects. Forker (1992) shows a model of
optimal disclosure decision which is presented in terms of
managerial incentives and the impact of corporate governance
structures. He argues that a threat to monitoring quality exists
where the roles of chief executive and chairman are combined.
Choi and Seo (2008) indicate that risk-averse managers may have
incentives to reduce the ?rm-speci?c risks of ?rms' activities,
including reduction of information transparency, and such in-
centives of managers are in con?icts with the incentives of
external shareholders owning controlling ownership or having
intimate business relationship.
In addition, Graves (1988) argues for the detrimental effect of
institutional stock ownership on corporate R&D spending in the
computer industry, whereas Baysinger, Kosnik, and Turk (1992)
argue the exact opposite, claiming to ?nd that high insider rep-
resentation on a board and concentration of equity among insti-
tutional shareholders positively affects corporate R&D spending.
Thus the institutional shareholders can in?uence or even deter-
mine the R&D spending when they have signi?cant in?uence on
?rms.
Taken together, an increase in the ?rms' institutional share-
holdings has a detrimental effect on corporate transparency
because of con?icts of interest, especially for high-tech ?rms and
?rms with high R&D intensity. Accordingly, Hypothesis 3 is pro-
posed as follows:
Hypothesis 3. As their R&D intensity increases, high-tech ?rms
whose institutional shareholdings are becoming larger will have
greater role duality con?icts and therefore decrease their information
transparency.
3. Data and methodology
3.1. Sample collection
We obtain our information transparency data from the IDTRS
annual evaluation reports; our institutional percentage owner-
ship and ?nancial information data come from the Taiwan Eco-
nomic Journal database (TEJ). We use the variable term,
Institutions, to represent four individual institutional ownership
and total institutional shareholdings, generally based upon data
we gather from of?cial information websites. These shareholding
combinations include: (1) Govt: denotes the shareholdings of
governmental and ?nancial institutions in Taiwan; (2) Corp: de-
notes the shareholdings of domestic business corporations; (3)
Fund: denotes the shareholdings of domestic mutual funds; (4)
Foreign: denotes the shareholdings of foreign institutions and
funds; and (5) Inst: denotes shareholdings of all above institu-
tional shareholders.
We collect a sample of 9200 ?rm-year evaluated observations.
After 219 ?nancial institutions (due to odd variables or a lack of
R&D data) and 144 inappropriate observations (those with un-
usual and/or irregular conditions) are deleted, our sample is
reduced to 8837 ?rm-year observations (including 1391 evaluated
?rms) from 2005 (the year scaling to rank was ?rst used) to 2012
(the year before the International Financial Reporting Standards
(IFRS) was ?rst adopted in Taiwan). As shown in Table 1 about half
of the evaluated ?rm-years are classi?ed as high-tech ?rms,
whose R&D intensity is higher, relatively, than those in the other
two industries. The mean and correlation coef?cient of R&D in-
tensity with high-tech industries tend to be higher, suggesting
that R&D intensity can characterize the industry-speci?c tech-
nology level.
C.-H. Hsu et al. / Asia Paci?c Management Review xxx (2015) 1e12 4
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
In order to mitigate the in?uence of outliers, we winsorize all
of the scaled ?nancial independent variables (e.g., R&D, Turnover,
RET, RETVar, Leverage, Pro?t) at the top and bottom .1% of their
respective distributions. After deleting the inadequate observa-
tions (i.e., missing-variables, without prior-year data) and those
with no changes in transparency grade, we are left with 2154
?rm-years in our test sample. Matching the TEJ with the OECD
industry classi?cation codes,
8
we set three industry dummy
variables, Hightech, Medium and Traditional, for high-tech, me-
dium-high-tech and traditional industries, respectively, in our
regressions.
Because the TEJ database categorizes four different types of ul-
timate ownership, we regroup the evaluated ?rms into three cat-
egories: (1) Government-involved (labeled as GA ?rms
9
): 1197
?rm-years directly or indirectly controlled by central, local au-
thorities, their peripheral organizations, or cooperative gover-
nance; (2) Group-af?liated (also called ‘family business’; labeled as
FA ?rms); 5523 ?rm-years held by a single group consisting of an
individual and his or her family members; (3) Manager-governed
(labeled as MG ?rms), 2117 ?rm-years governed by professional
executives or managers.
3.2. The measure of variables
In compliance with the measures used in prior literature (e.g.,
Agrawal & Mandelker, 1990), the ownership of a substantial stake
in a ?rm by an institution is expressed as a percentage of the ?rm's
year-end outstanding shares. We test the relation between changes
in institutional shareholdings, DInstitutions, and changes in trans-
parency grade, Upgrade
10
, in an attempt to examine the relationship
between shareholders' intention and corporate transparency. The
dependent variable, Upgrade, is equal to one if the yearly change in
Grade indicates an improvement in information transparency, and
zero otherwise. The other test variable, R&D, is formed by creating a
ratio of annual R&D expense to annual reported net operating
income.
11
We then estimate equation (1) for individual industries and
whole industries. We directly compare the coef?cients on DIn-
stitutions across industries and institutional ownership. We express
the following control variables in terms of changes, including: (1)
MVE: equal to the natural logarithmof market value of equity at the
beginning of the sample year; (2) Nanalyst: equal to the number of
analysts following the ?rm at year-end. (3) Turnover: equal to the
annual share of trading volume divided by shares outstanding; (4)
RET: equal to stock returns over the ?scal year; (5) RETVar: equal to
volatility of stock return equal to annual stock returns rate variance;
(6) Leverage: equal to ratio of total liabilities to total assets; (7)
Pro?t: equal to ratio of prior-three-year average net pro?t margin
fromsales; (8) SGrowth: equal to ratio of incremental sales to prior-
year sales; (9) Grade: equal to the lagged level of transparency and
used as a regressor to account for situations in which changes are
limited, and to capture any changes in response to existing levels;
and (10) DYear: a dummy variable capturing year ?xed-effect. To
solve the omitted-variables problem, we also include ?rm ?xed-
effects with standard errors adjusted for ?rm-level clustering in
our regressions to control for unobserved sources of ?rm
heterogeneity.
We also use DInstitutions to interact with R&D and regress Up-
grade on the interaction term, R&D Â DInstitutions, to observe
whether the level of R&D intensity strengthens the effect of
Table 1
Sample distributions by industry, transparency grade, and institutional ownership.
Grade C
À
C B A A
þ
Best voluntary disclosure
Score range
Information disclosure is a necessary activity in corporate governance; information transparency plays a
unique role in corporate governance in the era of knowledge-based economy. Lack of transparency can
lead to confusion, misinformation, and distrust. With this in mind, we examine the factors that influence
corporate information transparency in terms of two dimensions: technology intensity and institutional
ownership. Drawing on data from a 2005e2012 cross-section sample of 1391 public firms evaluated by
the official ‘information disclosure and transparency ranking system’ (IDTRS), we find that increases in
domestic institutional ownership for firms in high-tech industries, relative to foreign institutional
ownership, lead to a current-year upgrade in information transparency, but not for firms in other industries.
We also find that firms with increased foreign institutional ownership and high-tech firms with
both increased governmental institutional or corporate ownership and high R&D intensity can sustain a
longer-run upgrade in corporate transparency. Pushing further we also investigate whether corporate
transparency in high-tech industries is negatively affected if governmental institutional or corporate
shareholders are involved in corporate governance, but cannot find strong evidence for such a tendency.
Our results suggest that institutional shareholders promote good corporate governance practices which
gradually improve at the pace of high technology development.
Institutional ownership and information transparency: Role of technology
intensities and industries
Chung-Hao Hsu
a, b
, Syou-Ching Lai
a, *
, Hung-Chih Li
b
a
Department of Accounting and Information System, Chang Jung Christian University, Taiwan, ROC
b
Department of Accountancy, National Cheng Kung University, Taiwan, ROC
a r t i c l e i n f o
Article history:
Received 16 March 2015
Accepted 8 June 2015
Available online xxx
Keywords:
Information transparency
Institutional ownership
Institutional monitoring
R&D intensity
a b s t r a c t
Information disclosure is a necessary activity in corporate governance; information transparency plays a
unique role in corporate governance in the era of knowledge-based economy. Lack of transparency can
lead to confusion, misinformation, and distrust. With this in mind, we examine the factors that in?uence
corporate information transparency in terms of two dimensions: technology intensity and institutional
ownership. Drawing on data from a 2005e2012 cross-section sample of 1391 public ?rms evaluated by
the of?cial ‘information disclosure and transparency ranking system’ (IDTRS), we ?nd that increases in
domestic institutional ownership for ?rms in high-tech industries, relative to foreign institutional
ownership, lead to a current-year upgrade in information transparency, but not for ?rms in other in-
dustries. We also ?nd that ?rms with increased foreign institutional ownership and high-tech ?rms with
both increased governmental institutional or corporate ownership and high R&D intensity can sustain a
longer-run upgrade in corporate transparency. Pushing further we also investigate whether corporate
transparency in high-tech industries is negatively affected if governmental institutional or corporate
shareholders are involved in corporate governance, but cannot ?nd strong evidence for such a tendency.
Our results suggest that institutional shareholders promote good corporate governance practices which
gradually improve at the pace of high technology development.
© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier
Taiwan LLC. All rights reserved.
1. Introduction
A good corporate governance framework should ensure that
timely and accurate disclosure is made on all material matters
regarding the corporation, including the ?nancial situation, per-
formance, ownership, and governance of the company (OECD,
2004
1
). With an eye towards corporate governance framework
and technological knowledge characteristics we investigate the
joint effects of research and development (R&D) intensity and the
four categories of institutional ownership on corporate information
transparency for the three tiers of industries (i.e., high-tech, me-
dium-high-tech and traditional) and the three categories of ?rms
(i.e., government-involved, group-af?liated and manager-
governed) over a sample period from 2005 to 2012.
Proponents of ‘Asian value’ argue that companies led by
governmental institutions have better performance in terms of
in?uencing the course of economic development and income dis-
tribution, especially in knowledge-based innovative services and
industries (Lee, 2003). Asian countries have a long tradition of
governments exerting a lot of authority in various matters. Do
governmental institutions intervening in business affairs of high-
tech ?rms in?uence corporate transparency? By contrast,
following the democratization of Taiwan and Korea, universalism
claims that the Asian experience is not exceptional at all. Stock
markets in Taiwan are ‘plate-form’ markets that are signi?cantly
in?uenced by the international contagion effect. The stock price
?uctuations are seen to exhibit signs of increasing foreign
* Corresponding author.
E-mail address: [email protected] (S.-C. Lai).
Peer review under responsibility of College of Management, National Cheng
Kung University.
1
In conjunction with national governments, other relevant international orga-
nizations and the private sector, the Organization for Economic Cooperation and
Development (OECD) has developed a set of corporate governance standards and
guidelines, ‘The OECD Principles of Corporate Governance’. We quote the foreword
of its principle V: ‘Disclosure and Transparency’, one of the six main principles, as
our writing motivation.
HOSTED BY
Contents lists available at ScienceDirect
Asia Paci?c Management Review
j ournal homepage: www. el sevi er. com/ l ocat e/ apmrvhttp://dx.doi.org/10.1016/j.apmrv.2015.06.001
1029-3132/© 2015 College of Management, National Cheng Kung University. Production and hosting by Elsevier Taiwan LLC. All rights reserved.
Asia Paci?c Management Review xxx (2015) 1e12
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
investments acting as a critical factor in swaying capital markets.
Does foreign institutional ownership have signi?cant in?uence on
disclosure quality of ?rms in the plate-form stock markets? Prior
studies have provided inconsistent explanations for the mixed ev-
idence on the in?uence of institutional ownership. Since every
category of institutional ownership has its different competing ef-
fects on information asymmetry, it is unclear whether it is practi-
cable to generalize the results of prior studies to other institutional
shareholders.
Using the IDTRS evaluation from 2005 to 2012, we investigate
the direct causal relations between institutional ownership and
corporate transparency in Taiwan, where national policies are seen
as serving as a hub for the technology industries. Based on the
different characteristics in diversi?ed industries, we correct stan-
dard errors for industry-level clustering because information
transparency is likely to be correlated within an industry since
some industry-speci?c attributes are mandated by accounting and
disclosure rules (De Franco, Kothari, & Verdi, 2011). For instance,
the accounting standards regarding R&D expenditures, intangible
assets, and impairment of assets are highly correlated with
technology-oriented industries. We preliminarily ?nd that ?rms
and industries operating at a higher technology level tend to have
higher R&D intensity and information transparency. We also ?nd
that domestic institutional ownership is positively correlated with
corporate transparency. However, in pursuit of their maximum
interest, the dual roles of stakeholder and shareholder may even-
tually con?ict by controlling managers' decisions that are in con-
?icts with shareholders' interests including transparent
information. We ?nd the correlation coef?cient of institutional
ultimate controlling ownership is signi?cantly negative, suggesting
that a possible role duality con?ict may affect the own-
ershipetransparency relationship.
R&D intensity is more characteristic of industrial technology
?rms than other indicators (Hsu, Lai, & Li, 2015). We use both
R&D-based industry classi?cation and a ?rm's R&D intensity as
the technology factors to moderate the linkages between institu-
tional ownership and corporate transparency. We further imple-
ment multivariate changes regressions to examine the effect of
ownership on transparency at the level of both industry and ?rm.
Our regression analysis of the above relationship leads to four
?ndings. First, we ?nd that technology industry classi?cation is, in
general, more indicative than technology intensity magnitude.
Second, increases in domestic institutional ownership help high-
tech ?rms increase their current-year information transparency,
while the positive effects of increases in foreign institutional
ownership on corporate transparency are deferred to the
following year. Third, ?rms with a higher R&D intensity which are
experiencing increased governmental institutional or corporate
ownership tend to continuously improve their information
transparency in the following year. Finally, by comparing changes
in the controlling ownership of government-involved ?rms and
group-af?liated ?rms with manager-governed ?rms, we cannot
?nd a negative effect of increased controlling institutional
ownership on corporate transparency.
The contributions which this research makes to the ?eld are
summarized in the following. First, we integrate the IDTRS evalu-
ation and the OECD industry classi?cation (OECD, 2011) into our
research; this includes fairly newdata in the literature on corporate
governance. Second, this is the ?rst paper that uses industry-level
and ?rm-level technology intensity together to determine the
moderating effect of various institutional ownership on informa-
tion transparency. Third, though related research is abundant, in
this paper we perform an analysis through a multivariate changes
regression model to provide a deeper insight into the causal rela-
tion between institutional ownership and corporate transparency.
We believe that our focus on of?cial data and standards, direct
causality relationship, and counterevidence can provide more
convincing results than previous research has managed to achieve.
Research background and hypotheses development are explained
in the next section; data and methodology in the third section;
empirical research ?ndings in the fourth section; and the conclu-
sion in the ?nal section.
2. Background and hypotheses
2.1. Background description
In this paper, institutional shareholders are operationally
de?ned as the shareholders of domestic institutions which are
governed by the Taiwan Securities Exchange Act,
2
and foreign in-
stitutions and funds.
3
Depending upon their backgrounds, trading
tendencies and regulatory restrictions, various economic conse-
quences can be caused by institutional ownership (Choi, Lam, Sami,
& Zhou, 2013). Shareholders who focus on a ?rm's short-term
performance tend to be concerned with market-price volatility
but not the ?rm's prospects and management, and therefore, no
positive impact of their shareholdings on corporate governance is
expected. Inversely, expectations regarding a ?rm's future devel-
opment, long-term institutional shareholders will have a greater
in?uence on ?rms. They use different means to monitor the ?rm's
decision-making. Transparent accounting information facilitates
institutional shareholder monitoring and the effective exercise of
shareholder rights under existing securities laws (Bushman &
Smith, 2003).
According to Black (1991), institutional shareholders can be
categorized into ‘Institutional voice’ and ‘Institutional control’. The
former does not participate in management but can pressure and
challenge ?rm managers. By contrast, the latter participates in the
?rm's business by holding key positions, such as directors or
managers. Many prior studies (i.e., Chen, Harford, & Li, 2007;
Ferreira & Matos, 2008; Aggarwal, Erel, Ferreira, & Matos, 2011)
classify institutional investors according to the potential for busi-
ness ties to a ?rm, referring to those with strong ties as “grey
institutions”
4
and those with weak ties as “independent
institutions”.
Corporate shareholders have a strong underlying purpose for
building business ties to a corporation. Their tight business and
monitoring capability can result in ef?cient monitoring or strategic
alignment (Pound, 1988). The shareholdings of governmental in-
stitutions usually can represent governmental policy objectives and
implementation. In the same vein, ?nancial institutional share-
holders may play a role like governmental institutions, or may be
able to make direct fund transfers to ?rms just as corporate
shareholders and grey institutions are able to.
2
In this paper, our domestic institutions comprise all institutions other than
foreign ones. The composition of institutional shareholders is different according to
the country's related law. According to the trading proportion statistics of share-
holder categories of the Taiwan ‘Securities Exchange Act’, the institutional share-
holders in a broad sense are the category relative to human shareholders. However,
in a narrow sense, the Act also categorizes ?nancial institutions, investment and
trust companies and some speci?c funds as the ‘professional institutional
shareholders’.
3
The quali?ed foreign institutional shareholders (QFII) in this paper include all
institutional shareholders who are not domiciled in Taiwan, including foreign in-
stitutions, foreign mutual funds, and foreign corporate shareholders and other
institutions.
4
Grey institutions refer to those investors who have, whether actual or only
potential, business relationships with ?rms in which they invest. The ownership is
mainly the percentage of shares held by bank trust departments, insurance com-
panies, and other institutions (e.g., pension funds, endowments).
C.-H. Hsu et al. / Asia Paci?c Management Review xxx (2015) 1e12 2
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
Domestic mutual fund and foreign institutional shareholders
have independent institutional ownership, because they are more
likely to collect corporate information and are subject to fewer
regulatory restrictions. They have the highest shareholder sophis-
tication but fewer potential business relationships with the cor-
porations. Based on their greater independence, these two groups
are more involved in monitoring corporate management. However,
foreign institutional shareholders seldom participate in local ?rm's
operations because of greater restrictions, fewer information
channels and unfavorable environmental factors.
5
They know that
excessive intervention in invested ?rms can negatively in?uence
?rm's operation and stock price, and may cause a loss for inde-
pendent shareholders (Palmiter, 2002).
Isolating two distinct factors, Bushman, Piotroski, and Smith
(2004) ?nd that the governance transparency factors are primar-
ily related to a country's legal/judicial regime, whereas the ?nancial
transparency factors are primarily related to political economy.
Hence, we argue that the of?cial evaluation is a direct measure of
information transparency since the systematic review is based on
de?nite evaluation metrics at the class and method level.
6
It pro-
vides another type of comparable data for transparency, relative to
the measures based on market reactions (e.g., Choi &Seo, 2008; Liu,
Hsu, & Li, 2014).
In 2002, the Taiwan Securities & Futures Institute (SFI) estab-
lished an independent council for professional evaluation and
launched the IDTRS, through which a two-stage screening process
is conducted. First, all information provided is given a preliminary
screening by the SFI's in-house ranking team based on each
disclosure item in an evaluation year. Second, all companies are
entitled to review the preliminary results via the internet and
respond directly to SFI regarding ambiguous issues in the following
year. Upon receiving the different opinions expressed by com-
panies, the ranking committee assesses the correspondences,
makes adjustments if needed; and the consulting committee de-
cides the ?nal ranking results. About half of the indicators are
concerned with ?nancial statement information, and about forty
percent are designed for voluntary disclosure (including R&D-
related information). The latter causes the major differences in
ratings. The IDTRS thus help all stakeholders monitor corporate
governance.
In Fig. 1 we can see that two subsequent events are scheduled
for the following year. The ?rst is the procedure of review and re-
evaluation. It is possible for evaluated ?rms to change the ?rst-
stage results by improving their website disclosure and ?nancial
forecasts. The second is the year t annual and the year t þ1 quar-
terly reports issuance. However, we ?nd that both the two ?nancial
reports are evaluated through the year t þ1 IDTRS (their issuance
year) and only impact the year t þ1 evaluation result.
2.2. Hypothesis development
Relative to retail shareholders, large external shareholders,
block owners, and institutional shareholders have strong in-
centives to monitor domestic public ?rms (Allen, Bernardo, &
Welch, 2000; El-Gazzar, 1998; Hotchkiss & Strickland, 2003;
Shleifer & Vishny, 1986). Prior studies have found that monitoring
by large external shareholders reduces the agency costs of equities
by effectively controlling managers' decisions when they are in
con?icts with shareholders' interests. Thus, ?rms with long-term
supervision have better prospects for long-term earnings growth
and increases in value. However, some studies (e.g., Porter, 1992;
Bushee, 2001) claim that institutional ownership with short in-
vestment horizons will maximize short-term results at the
expense of long-term equity value or ?rm value in expected near
term earnings.
In the present work we analyze the in?uence of various external
shareholders on corporate transparency. Governmental institutions
are concerned with governmental institutional service and policy
implementation. In Taiwan, Chen, Chen, and Chen (2005) indicate
that ?rms with a higher percentage of ownership by the govern-
ment, ?nancial institutions, and foreign institutions have superior
long-term supervision and better disclosure. Ku (2013) asserts that
a larger board and higher proportion of institutional ownership will
strengthen the information transfer effect. The so-called ‘three
legal entities’
7
in Taiwan's stock markets have signi?cant in?uence
on ?rm's market price, which is why both ?rm managers and se-
curities analysts keep a close eye on their trading tendency. Based
on the in?uence of domestic mutual funds on the trading in sec-
ondary markets, ?rms concerned about the effect of the three legal
entities on market prices may manage to promote their informa-
tion transparency.
Based on the higher economic value of high technology, the
bene?ts accruing from the ?rm's increased value serve as an
important driving force for institutional shareholders to enhance
their in?uence for self-interest. Bearing greater cost and risk in
their investments, institutional shareholders will select ?rms with
high information transparency and require the ?rms to disclose
more information relevant to ?nancial analysis, risk evaluation and
operational plans. Hence, we postulate that institutional share-
holders also have a higher investment preference for information
about ?rm's technology level, especially in high-tech industries.
Financial accounting year t
year t annual report
& year t+1 quarterly report
issuance in year t+1
Month: 1~12 1-2 3-4 5 6-7 8-12
IDTRS evaluation year t
(Stage I)
Review & Re-evaluation
(Stage II)
year t evaluation report
issuance in year t+1
Fig. 1. Timeline of annual reports and IDTRS implementation in Taiwan.
5
In the US, institutional control shareholders of investment ?rms and funds are
also rare, due to the Investment Company Act of 1940 No.17. In Taiwan, few internal
positions are held by representatives of foreign institutions due to compliance with
laws enacted to protect national defense, national communication security, local
industries, local culture, public interests, market price, etc.
6
The IDTRS broadly identi?es 109 disclosure items regarding regulation
compliance, disclosure timeliness, ?nancial forecast publication, and annual reports
as evaluation criteria and gauges the level of corporate transparency by searching
public information disclosed in various ways. For full ranking results, please visit
website:http://www.s?.org.tw.
7
They are the professional institutional shareholders including securities in-
vestment companies (dealers), domestic investment trust funds, and foreign in-
stitutions and funds.
C.-H. Hsu et al. / Asia Paci?c Management Review xxx (2015) 1e12 3
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
Consistent with the monitoring argument (Shleifer & Vishny, 1986,
1997), we establish our hypothesis as follows:
Hypothesis 1. High-tech ?rms experiencing increases in institu-
tional shareholdings have a higher possibility of increasing their in-
formation transparency.
By splitting the sample into high-tech and low-tech com-
panies, Czarnitzki and Thorwarth (2012) ?nd a large premium of
R&D for ?rms in high-tech industries, but no premium in low-
tech sectors. This suggests that rational institutional share-
holders want more relevant information disclosed, especially
about high-tech ?rm's R&D activities and capital-intensive pro-
jects, which concern virtually all of the ?rm's operational results
(Bushee, 1998). On the contrary, too much technology-originated
information disclosed will impair the vested interests of insiders.
A ?rm's insiders will conservatively disclose information to
decrease information transparency. Owing to the ?rm-speci?c
risks of high-tech industries, we expect that corporate trans-
parency will be affected by the percentage ownership of external
institutional shareholders.
Pindado, Queiroz, and Torre (2015) ?nd that effective corporate
governance allows the market to better assess a ?rm's R&D in-
vestments because effective control mechanisms reinforce the
positive effect of R&D on a ?rm's market value. Using order backlog
as a predictor, Jiambalvo, Rajgopal, and Venkatachalam (2002)
assert that institutional investors may be attracted to ?rms in
richer information environments where stock prices tend to lead
earnings. Using R&D expenses as a proxy for forward-looking in-
formation, Chen and Wang (2006) ?nd that institutional investors
can incorporate R&D information that re?ects future earnings,
especially for group-af?liated ?rms. Hence, we know that larger
institutional shareholders are better able to use enhanced corpo-
rate transparency to acquire relevant information. This leads to the
following hypothesis:
Hypothesis 2. As their R&D intensity increases, high-tech ?rms
experiencing increases in institutional shareholdings have a higher
possibility of increasing their information transparency.
However, the associations between corporate characteristics
and disclosures level are moderated by differences in disclosure
index construction, differences in de?nition of the explanatory
variables, and differences in research settings (Ahmed & Courtis,
1999). There are alternative views which are diametrically
opposed to our above two hypotheses. First, Fan and Wong (2002)
argue that, in a region of high-developed knowledge economics,
the holding percentage and voting right enforcement of govern-
mental institutions indicate strategic economic purposes. Morck,
Shleifer, and Vishny (1988) claim that institutional shareholders
place too much emphasis on ?rm's technological capacity and
leadership will increasingly push ?rm managers to seek value-
maximization over all other goals. As a result, in compliance
with institutional shareholder's expectations, ?rms tend to over-
look the importance of disclosure quality and information
transparency.
Second, others argue that the frequent trading and short-term
focus of institutional shareholders encourages managers to
engage in myopic investment behavior. Firm managers may take
advantage of transient institutional shareholders such that they
make decisions (including accounting/reporting decisions) so as to
maximize their own self-interest. As the level of institutional
ownership increases, ?rm managers' opportunistic behavior is
likely to increase even as their ?rm's disclosure level decreases
(Laverty, 1996).
Third, Lambert (1986) indicates that the con?ict of interest
between the executive and the principal may lead to different
selections in risky projects. Forker (1992) shows a model of
optimal disclosure decision which is presented in terms of
managerial incentives and the impact of corporate governance
structures. He argues that a threat to monitoring quality exists
where the roles of chief executive and chairman are combined.
Choi and Seo (2008) indicate that risk-averse managers may have
incentives to reduce the ?rm-speci?c risks of ?rms' activities,
including reduction of information transparency, and such in-
centives of managers are in con?icts with the incentives of
external shareholders owning controlling ownership or having
intimate business relationship.
In addition, Graves (1988) argues for the detrimental effect of
institutional stock ownership on corporate R&D spending in the
computer industry, whereas Baysinger, Kosnik, and Turk (1992)
argue the exact opposite, claiming to ?nd that high insider rep-
resentation on a board and concentration of equity among insti-
tutional shareholders positively affects corporate R&D spending.
Thus the institutional shareholders can in?uence or even deter-
mine the R&D spending when they have signi?cant in?uence on
?rms.
Taken together, an increase in the ?rms' institutional share-
holdings has a detrimental effect on corporate transparency
because of con?icts of interest, especially for high-tech ?rms and
?rms with high R&D intensity. Accordingly, Hypothesis 3 is pro-
posed as follows:
Hypothesis 3. As their R&D intensity increases, high-tech ?rms
whose institutional shareholdings are becoming larger will have
greater role duality con?icts and therefore decrease their information
transparency.
3. Data and methodology
3.1. Sample collection
We obtain our information transparency data from the IDTRS
annual evaluation reports; our institutional percentage owner-
ship and ?nancial information data come from the Taiwan Eco-
nomic Journal database (TEJ). We use the variable term,
Institutions, to represent four individual institutional ownership
and total institutional shareholdings, generally based upon data
we gather from of?cial information websites. These shareholding
combinations include: (1) Govt: denotes the shareholdings of
governmental and ?nancial institutions in Taiwan; (2) Corp: de-
notes the shareholdings of domestic business corporations; (3)
Fund: denotes the shareholdings of domestic mutual funds; (4)
Foreign: denotes the shareholdings of foreign institutions and
funds; and (5) Inst: denotes shareholdings of all above institu-
tional shareholders.
We collect a sample of 9200 ?rm-year evaluated observations.
After 219 ?nancial institutions (due to odd variables or a lack of
R&D data) and 144 inappropriate observations (those with un-
usual and/or irregular conditions) are deleted, our sample is
reduced to 8837 ?rm-year observations (including 1391 evaluated
?rms) from 2005 (the year scaling to rank was ?rst used) to 2012
(the year before the International Financial Reporting Standards
(IFRS) was ?rst adopted in Taiwan). As shown in Table 1 about half
of the evaluated ?rm-years are classi?ed as high-tech ?rms,
whose R&D intensity is higher, relatively, than those in the other
two industries. The mean and correlation coef?cient of R&D in-
tensity with high-tech industries tend to be higher, suggesting
that R&D intensity can characterize the industry-speci?c tech-
nology level.
C.-H. Hsu et al. / Asia Paci?c Management Review xxx (2015) 1e12 4
Please cite this article in press as: Hsu, C.-H., et al., Institutional ownership and information transparency: Role of technology intensities and
industries, Asia Paci?c Management Review (2015),http://dx.doi.org/10.1016/j.apmrv.2015.06.001
In order to mitigate the in?uence of outliers, we winsorize all
of the scaled ?nancial independent variables (e.g., R&D, Turnover,
RET, RETVar, Leverage, Pro?t) at the top and bottom .1% of their
respective distributions. After deleting the inadequate observa-
tions (i.e., missing-variables, without prior-year data) and those
with no changes in transparency grade, we are left with 2154
?rm-years in our test sample. Matching the TEJ with the OECD
industry classi?cation codes,
8
we set three industry dummy
variables, Hightech, Medium and Traditional, for high-tech, me-
dium-high-tech and traditional industries, respectively, in our
regressions.
Because the TEJ database categorizes four different types of ul-
timate ownership, we regroup the evaluated ?rms into three cat-
egories: (1) Government-involved (labeled as GA ?rms
9
): 1197
?rm-years directly or indirectly controlled by central, local au-
thorities, their peripheral organizations, or cooperative gover-
nance; (2) Group-af?liated (also called ‘family business’; labeled as
FA ?rms); 5523 ?rm-years held by a single group consisting of an
individual and his or her family members; (3) Manager-governed
(labeled as MG ?rms), 2117 ?rm-years governed by professional
executives or managers.
3.2. The measure of variables
In compliance with the measures used in prior literature (e.g.,
Agrawal & Mandelker, 1990), the ownership of a substantial stake
in a ?rm by an institution is expressed as a percentage of the ?rm's
year-end outstanding shares. We test the relation between changes
in institutional shareholdings, DInstitutions, and changes in trans-
parency grade, Upgrade
10
, in an attempt to examine the relationship
between shareholders' intention and corporate transparency. The
dependent variable, Upgrade, is equal to one if the yearly change in
Grade indicates an improvement in information transparency, and
zero otherwise. The other test variable, R&D, is formed by creating a
ratio of annual R&D expense to annual reported net operating
income.
11
We then estimate equation (1) for individual industries and
whole industries. We directly compare the coef?cients on DIn-
stitutions across industries and institutional ownership. We express
the following control variables in terms of changes, including: (1)
MVE: equal to the natural logarithmof market value of equity at the
beginning of the sample year; (2) Nanalyst: equal to the number of
analysts following the ?rm at year-end. (3) Turnover: equal to the
annual share of trading volume divided by shares outstanding; (4)
RET: equal to stock returns over the ?scal year; (5) RETVar: equal to
volatility of stock return equal to annual stock returns rate variance;
(6) Leverage: equal to ratio of total liabilities to total assets; (7)
Pro?t: equal to ratio of prior-three-year average net pro?t margin
fromsales; (8) SGrowth: equal to ratio of incremental sales to prior-
year sales; (9) Grade: equal to the lagged level of transparency and
used as a regressor to account for situations in which changes are
limited, and to capture any changes in response to existing levels;
and (10) DYear: a dummy variable capturing year ?xed-effect. To
solve the omitted-variables problem, we also include ?rm ?xed-
effects with standard errors adjusted for ?rm-level clustering in
our regressions to control for unobserved sources of ?rm
heterogeneity.
We also use DInstitutions to interact with R&D and regress Up-
grade on the interaction term, R&D Â DInstitutions, to observe
whether the level of R&D intensity strengthens the effect of
Table 1
Sample distributions by industry, transparency grade, and institutional ownership.
Grade C
À
C B A A
þ
Best voluntary disclosure
Score range