Description
The purpose of this paper is to report the impact of the Chinese capital market split equity
(SE) reform in 2005 on the corporate financial transparency of Chinese listed companies
Accounting Research Journal
The split equity reform and corporate financial transparency in China
Wendy Green Richard D. Morris Haiping Tang
Article information:
To cite this document:
Wendy Green Richard D. Morris Haiping Tang, (2010),"The split equity reform and corporate financial
transparency in China", Accounting Research J ournal, Vol. 23 Iss 1 pp. 20 - 48
Permanent link to this document:
http://dx.doi.org/10.1108/10309611011060515
Downloaded on: 24 January 2016, At: 21:10 (PT)
References: this document contains references to 42 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 930 times since 2010*
Users who downloaded this article also downloaded:
Raghavan J . Iyengar, J udy Land, Ernest M. Zampelli, (2010),"Does board governance improve
the quality of accounting earnings?", Accounting Research J ournal, Vol. 23 Iss 1 pp. 49-68 http://
dx.doi.org/10.1108/10309611011060524
Katherine Uylangco, Steve Easton, Robert Faff, (2010),"The equity and efficiency of the Australian share
market with respect to director trading", Accounting Research J ournal, Vol. 23 Iss 1 pp. 5-19 http://
dx.doi.org/10.1108/10309611011060506
Minna Yu, (2010),"Analyst following and corporate governance: emerging-market evidence", Accounting
Research J ournal, Vol. 23 Iss 1 pp. 69-93 http://dx.doi.org/10.1108/10309611011060533
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The split equity reform and
corporate ?nancial transparency
in China
Wendy Green and Richard D. Morris
School of Accounting, University of New South Wales, Sydney, Australia, and
Haiping Tang
Pricewaterhousecoopers, Sydney, Australia
Abstract
Purpose – The purpose of this paper is to report the impact of the Chinese capital market split equity
(SE) reform in 2005 on the corporate ?nancial transparency of Chinese listed companies.
Design/methodology/approach – Using an International Financial Reporting Standards-based
checklist, the paper investigates whether the post-reform 2005 annual reports of reformed companies
improved transparency compared to pre-reform 2004 reports. The transparency of the reformed
companies was also compared to a control group of companies unreformed on December 31, 2005.
Findings – Results indicate that the SE reform increased corporate disclosures. Reformed companies
had higher mandatory and voluntary disclosures in their post-reform 2005 annual reports compared to
their pre-reform 2004 annual reports. In addition, the improvement in mandatory and voluntary
disclosures for reformed companies is greater than that of the unreformed control group.
Research limitations/implications – The SE reform provides a unique natural experimental
setting in which to examine the impact of the SE reform, with its associated change in ownership
structure and corporate governance, on corporate disclosure.
Practical implications – The results of this paper suggest that the SE reform has had a positive
effect on corporate ?nancial transparency in China, thereby indicating the positive response to
regulation in this emerging market. Further, the results suggest that as the proportion of government
ownership falls, management has increased incentive to voluntarily supply additional information to
the market.
Originality/value – The SE reform is unique to China and this paper is the ?rst to report on
?nancial reporting disclosure implications of this reform.
Keywords Equity capital, Disclosure, China
Paper type Research paper
1. Introduction
In 1978, the Chinese Government initiated their “economic reform and opening up”
policy aimed at attracting foreign investments and encouraging experiments in
enterprise autonomy. Since then, the Chinese economy has experienced strong growth
with gross domestic product averaging 10 percent per annum. In contrast, the growth
of the domestic share market has been much more volatile. For example, between
2001 and 2005, the benchmark Shanghai Composite Index fell more than 50 percent
from its peak even though the domestic economy was strong. The existence of the split
equity (SE) structure in the Chinese share market is said to have been a major hurdle to
share market development (Ang and Ma, 1999; Chen et al., 2001; Feng and Xu, 2007).
The SE structure, established in the early 1990s upon the formation of the Shanghai
and Shenzhen Stock Exchanges, was intended to help state-owned enterprises (SOEs)
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
ARJ
23,1
20
Accounting Research Journal
Vol. 23 No. 1, 2010
pp. 20-48
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611011060515
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
raise ?nance, other than by bank borrowing (Walter and Howie, 2006), without the
state losing control over key assets. Under this structure, only one-third of the shares of
listed companies were tradable on public markets, with the remaining two-thirds being
non-tradable. In 2005, non-tradable shares accounted for more than 60 percent of all
issued shares, while tradable shares accounted for less than 40 percent of total shares
(Feng and Xu, 2007).
On April 29, 2005, the China Securities Regulatory Commission (CSRC), the national
capital market regulator, launched the SE reform. Its aim was to ?oat previously
non-tradable shares into the market, allowing a shift in the balance of share ownership
to public ownership by institutional investors and existing tradable shareholders,
while in some cases still retaining state control. The importance of this reform can be
seen in the comments of Shan Fuling, CSRC Chairman, who stated in May 2005 that
“The Split Equity reform marks the most fundamental reform ever undertaken in the
Chinese capital market and it has huge implications for the market in the long run.”
He further noted that “One of the aims of the Split Equity reform is to improve
corporate governance in China and to enhance the transparency of listed companies.”
The SE structure was a major corporate governance device unique to China.
Whether its abolition had the intended effect of increasing transparency of listed
companies is therefore of interest to both regulators and the academic community.
The sheer volume of previously non-tradable shares becoming tradable after the
reform, together with the implications for capital markets and demand for information,
warrants an investigation of that change. There is unlikely to be any other setting
where there has been such a potentially large transfer of company shares from public
to private ownership in such a short period of time. The Chinese Government’s decision
to conduct the reform in batches of companies, with the ?rst stage completed by
December 31, 2005, and the fact that by 2007 around 95 percent of the listed companies
on the Shanghai and Shenzhen stock exchanges had completed the SE reform, provides
a natural experimental setting of short duration to examine the impact of the reform on
corporate ?nancial transparency. Speci?cally, our study investigates whether the
reform, and the shift in ownership structure it entails, results in an improvement in
transparency for the ?rst stage reformed companies in their post-reform 2005 annual
reports when compared to their pre-reform 2004 reports, and also when compared to a
matched control group of then (in 2005) unreformed companies.
Transparency is measured by adapting to the Chinese environment the
International Financial Reporting Standards (IFRS) disclosure checklist used by
Morris and Gray (2007). Since Chinese generally accepted accounting principles
(GAAP) provided the mandated reporting requirements during the period covered by
this study, and since these requirements were not identical to IFRS, three transparency
scores are calculated:
(1) a mandatory transparency score, based on the disclosure of items required by
both IFRS and Chinese GAAP in 2004 and 2005[1];
(2) a voluntary transparency score, based on the level of disclosure of items
covered only by IFRS in 2004 and 2005; and
(3) an overall score for mandatory and voluntary transparency in 2004 and 2005.
Our results indicate that the SE reform had a positive impact on the transparency of
the reformed companies. In the post-reform 2005 annual reports, both mandatory and
Split equity
reform
21
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
voluntary disclosures of reformed companies improved compared to their pre-reform
2004 reports. The improvement in the reformed company disclosures was also greater
than that of the matched control group of companies that had not commenced the
reform by December 31, 2005.
The study makes several contributions. First, it adds to the literature by linking the
impact of the SE reform with corporate disclosure in China, thereby showcasing
the unique institutional setting found in China. Second, given the changes to corporate
ownership following the reform, the study also provides insights into the effect of
ownership structure changes, particularly decreased government ownership, on
corporate disclosure in a natural experimental setting not otherwise available. Third,
the study provides evidence that the high levels of mandatory disclosure observed in
China, despite the presence of factors such as high government ownership that are
normally indicative of low disclosure, can be explained by the presence of numerous
templates contained in Chinese GAAP. Finally, the ?nding of a positive impact of the
SE reform on corporate ?nancial transparency in China has practical implications for
Chinese policymakers. The regulators claimed that one of the aims of the SE reform
was to improve corporate governance in China, and ultimately to enhance the
transparency of listed companies. Our study suggests that the reform has indeed
achieved this aim through improved transparency in the reformed companies.
2. The Chinese institutional environment
China provides a unique institutional setting in which to examine the effects of corporate
governance reforms. Prior research has noted that an understanding of the speci?c
institutional setting is vital in improving corporate governance and corporate disclosure
(Enriques and Volpin, 2007). China has traditionally been characterized by strong
government ownership, along with limited trading of certain share classes because of
the SE system. Reform of the SE system provides an opportunity to study the impact of
the move fromprimarily government ownership to more diverse ownership on one of the
basic components of corporate governance: corporate ?nancial transparency.
2.1 The SE structure in China
The SE structure divided the market into two segments: tradable and non-tradable
shares. Tradable shares held by non-state investors made up one-third of the shares of
listed companies while the remaining two thirds of the listed shares were non-tradable
shares typically held by the state[2]. Tradable shares comprise A-, B-, and H-shares.
A-shares are traded in Chinese currency (RMB) and were initially only open to domestic
investors in the Shanghai and Shenzhen Stock Exchanges. From 2003, quali?ed foreign
institutional investors (QFII) could also invest in A-shares. B-shares are US dollar
denominated and were originally intended only for foreign investors, but from 2001,
domestic investors could also invest in B-shares using US dollars. H-shares are shares
issued by Chinese companies that are dual listed on the Hong Kong Stock Exchange,
traded in Hong Kong dollars; they are also traded in Chinese RMB currency on the
Shanghai Stock Exchange.
Although the SE structure resolves the state’s concerns about loss of control,
problems arose as public markets grew. In particular, it was dif?cult for holders of
non-tradable shares to dispose of their shares. Non-tradable shares could only be
transferred between non-tradable shareholders, but any transfer of state-owned shares
ARJ
23,1
22
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
required government approval and the transfer price could not be below net book value
per share, to prevent the “loss” of state assets. Such treatment could bear little
relationship to the value the share market might place on a company (Walter and Howie,
2006) and thereby preventing holders of non-tradable shares from directly bene?tting
from improvements in corporate performance. In short, the market mechanism was
distorted.
Also, the SE structure created a divergence of interests between non-tradable and
tradable shareholders. While tradable shareholders’ interests were aligned with the
company’s share price, the latter remained of little interest to non-tradable shareholders.
Hence, the SE structure created incentives for controlling shareholders to seek private
gains through transactions that were often at the cost of tradable shareholders ( Jian and
Wong, 2003; Liu and Lu, 2003). Further, with a huge number of non-tradable shares
controlled by the state or controlling shareholders, the tradable shareholders were
minority shareholders. Thus, market disciplining mechanisms, including the threat of
takeover by another company, did not fully operate in the SE structure.
In order to address these issues, the CSRC launched the SE reform on April 29, 2005.
The aim of this reform was to ?oat the previously non-tradable shares into the market,
allowing a shift in the balance of share ownership from the government to public
ownership by institutional investors and existing tradable shareholders. The reform
was conducted in batches of companies, with the ?rst batch completed by December
31, 2005. Once companies completed the reform, they were referred to as reformed
G-share companies to distinguish them from unreformed companies (G stands for
reformed in Chinese Pingying).
The 2005 SE reform contained various features to ensure its success. First, in order to
protect the rights of existing tradable shareholders, the CSRC required non-tradable
shareholders to pay compensationto existingtradable shareholders to offset anypotential
loss that the latter would incur due to the ?oating of a potentially large number of
non-tradable shares. Second, the CSRCimposeda temporary lock-up, or sellingrestriction,
of up to three years on non-tradable shareholders to prevent the large increase in the
number of non-tradable shares causing a sharp decline in share price (Inoue, 2005).
Non-tradable shareholders could not sell more than 5 percent of the companies’ shares in
the ?rst six months after the reform; no more than 10 percent in the following 12 months,
with the restriction lifted from the third year onwards. Third, a ?exible approach was
adopted with regard to the reform proposal (Inoue, 2005). For example, instead of a
“one-size-?ts-all” approach imposed by the government, companies were able to design
their own compensation proposals (such as additional shares, cash or European warrants)
through negotiation with existing tradable shareholders; these were only approved once
there was agreement bymore than two-thirds of the existing tradable shareholders (Inoue,
2005). Fourth, after the initial pilot companies selected by the CSRC had completed the
reformprocess, the reformcontinued inbatches, with companies self-selecting themselves
into the reformprocess. Finally, while unsuccessful SE reformattempts in 1999 and 2001
took place in bull markets[3], the SE reform in 2005 was introduced when the market had
already hit a low point with the Shanghai Benchmark Index falling below 1,000 points
fromits peakof 2,000 points in2001. Giventhat the market was alreadythought to be at its
lowest, it was suggested that the reformwas less likely to lead to a further decline in share
prices due to this perceived “?oor effect.”
Split equity
reform
23
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
A number of other regulatory changes took place either before or immediately after
the SE reform in 2005 that also have the potential to improve ?nancial reporting
transparency in China. These changes involved improved corporate governance
requirements and the opening up of domestic A-shares to QFII, domestic institutional
investors and foreign strategic investors. These changes are outlined in Table I and
their potential effects are addressed in our model examining transparency.
3. The concept of transparency
Since the 1997 Asian ?nancial crisis, improving transparency to enhance investors’
con?dence has been one of the priority tasks for policy makers in the Asian region. More
recent corporate collapses in the USA and Australia have also emphasized to stakeholders
the importance of transparency. However, despite its wide usage, there is no consensus on
the de?nition of transparency. Transparency can exist at two levels: the corporate and the
country levels. Bushman et al. (2004) conceptualise transparency at the country level as the
joint output of a multi-faceted system whose components collectively produce, gather,
validate, and disseminate information to market participants. Bushman and Smith (2003)
de?ne transparency at the company level as the widespread availability of
company-speci?c information concerning publicly listed companies to those outside the
companies. Our focus is on corporate ?nancial transparency, that is, transparency at the
corporate level.
Compliance with accounting standards is commonly used to measure ?nancial
transparency at the corporate level, based on the premise that such compliance is
a signi?cant element in ?nancial reporting quality (Susilowati, 2008). Examples
Regulatory change Description
Corporate governance regulation The CSRC required all listed companies to comply with a code of
corporate governance by the end of 2003 which required: at least
two independent directors; an audit committee with at least
one-third independent directors; and separation of roles of CEO
and Chairman
a
. Remuneration and nomination committees were
also recommended
In December 2006, the CSRC further required independent
directors to have relevant ?nancial literacy and possess
necessary quali?cations
QFII During 2002, foreigners were able to invest for the ?rst time in
the domestic A-shares market directly rather than through the
B-share market
Domestic institutional investor
growth
The domestic institutional investor base was expanded in
November 2002 to allow domestic insurance companies to invest
in domestic A-shares without restriction; with social securities
and pension funds were approved in 2004
Foreign strategic investors In January 2006, the CSRC allowed selected foreign strategic
investors to acquire strategic stakes (a minimum of 10 percent of
outstanding shares) in reformed G share companies for at least
three years
Note:
a
These requirements were enhanced on December 8, 2006, with the CSRC requiring that the
independent directors appointed must also have demonstrated relevant ?nancial literacy and possess
the necessary quali?cations
Table I.
Regulatory changes in
China potentially
improving corporate
disclosure
ARJ
23,1
24
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
include: compliance with domestic accounting standards (Morris et al., 2004, 2009);
voluntary compliance with IFRS (Ashbaugh and Pincus, 2001; Ho and Wong,
2001; Morris et al., 2004, 2009); and voluntary compliance with US GAAP (Morris et al.,
2004, 2009).
Our study uses IFRS as the benchmark accounting standards rather than US GAAP
because IFRS have been used as the basis for developing Chinese GAAP since the ?rst
Chinese standards were issued in 1993; and on November 25, 2006, the Chinese
regulators announced that China would fully comply with IFRS from January 1, 2007.
Also, many countries in the Asian region including Singapore, Indonesia, Malaysia,
and Australia use IFRS as the basis of their domestic accounting standards.
3.1 Associations between transparency and corporate governance
The association between corporate ?nancial transparency and corporate governance is
premised on the idea that good corporate governance can promote corporate ?nancial
transparency through the effectiveness of corporate governance mechanisms[4],
and that transparent ?nancial reports can reduce the information asymmetry and
the con?ict of interest between companies’ insiders and outsiders, thus assisting the
effectiveness of corporate governance mechanisms. Some of the most common
corporate governance mechanisms include the separation of the roles of CEO and
board chairman, the presence of an audit committee, and independent non-executive
directors on the board.
While prior literature reports a positive association between the corporate
governance mechanisms and corporate ?nancial transparency in developed countries,
the results in emerging countries are mixed. For example, for Indonesia, Susilowati
(2008) reports a positive association between percentage of independent directors and
the presence of an audit committee and corporate ?nancial transparency. In the Chinese
context, Tower and Yan (2006) have also found dual leadership (where CEO and Chair
are separate people) and independent directors were negatively associated with
transparency. It is possible that these ?ndings result fromthe fact that audit committees
are comprised of independent directors who are alleged to be reluctant to stand up
against boards’ decisions because they are often appointed by controlling shareholders
(Lin, 2001). Further, we note that dual leadership may be more of a perception than a
reality in China because, although CEOand chair are separated at the company level, the
CEO or chair is often also the chair or CEO of the group to which the company belongs.
In contrast, Tower and Yan (2006) suggest that the reason for the negative relation
between independent directors and transparency could be a lack of “quali?ed”
independent directors with strong ?nancial and business experience in China. Further
they suggest that in the case of dual leadership, there is a lack of incentives to ensure
dual leadership leads to improved corporate operating decisions since the majority of
listed companies in China are SOEs and managerial jobs are not openly contestable,
but rather, are appointed by the government. In addition, Tower and Yan (2006) note
that the remuneration system is ?xed and often standardized so that remuneration is
only weakly linked to the performance of SOEs. On the other hand, where the CEO and
Chairman is the same person there may be more incentive to improve corporate
disclosure since the person is often appointed by the government and is subject to the
direct scrutiny of the CSRC (Tower and Yan, 2006).
Split equity
reform
25
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.2 Mandatory disclosure
Mandatory disclosure occurs where companies disclose information in accordance with
legislative requirements, including domestic accounting standards and any other
regulations. When examining mandatory disclosure, the level of enforcement is
important in determining compliance with accounting standards. While the level of
enforcement can be rigorous in developed countries such as the USAand the UK, it is not
always so in emerging countries.
In the Chinese context, the CSRC is in charge of the enforcement of corporate
disclosure. But it has been suggested that the CSRC lacks the independence and
necessary power in enforcing disclosures in China (Walter and Howie, 2006). In addition,
previous studies suggest that the investor protection and legal system are weak in
China, and that there is little litigation risk or risk of corporate takeover in China (Haw
et al., 2001; Xiao, 1999).
In combination, these factors suggest that the expected level of compliance with
mandatory disclosure would be low. However, previous studies examining disclosure
levels for large Chinese companies have found that mandatory disclosure compliance in
China is, in fact, high (Xiao, 1999). Such high-disclosure levels are positively associated
with company characteristics such as size, overseas listing, and the employment of Big N
auditors, andcorporate ?nancial transparencyinChina (measuredbyinternet disclosures)
is negatively associated with government (state) ownership (Xiao et al., 2004a, b)[5].
An additional factor leading to high levels of mandatory disclosures in China is that
Chinese GAAP includes numerous “templates” or “formats” embedded in the
Accounting System for Business Enterprises (ASBE), accounting standards and other
regulatory requirements issued by various parties[6]. To comply with Chinese GAAP,
listed companies are automatically forced to comply with the “templates.” The
implication of such templates is that in situations where the company does not have
certain items (e.g. intangible assets), the items are still disclosed in annual reports, but
with amounts shown as zeros, because of the required template details. Since reporting
an item(albeit zero) is coded as a disclosure for this study, this means that the templates
tend to result in high mandatory disclosure scores (see Xiao et al., 2004a, b for a
discussion of the development of accounting regulation in China).
3.3 Voluntary disclosure
Voluntary disclosure occurs where companies disclose information beyond mandatory
disclosure requirements. In the Chinese context, previous studies suggest that Chinese
companies provide little voluntary disclosure (DeFond et al., 2000; Tower and Yan, 2006;
Wang et al., 2007). This ?nding is in contrast to the positive relationship between
government ownership and voluntary disclosures noted by Eng and Mak (2003) in
Singapore which also has high-government levels of corporate ownership. The SE
structure in China, by limiting outside tradable shareholders, may limit demand for
voluntary disclosures. It is also possible that while the “templates” contained in Chinese
GAAP ensure high mandatory disclosure, they may work against voluntary disclosure
in that the templates can restrict the way in which companies disclose information.
4. Hypothesis development
Positive accounting theory (Watts and Zimmerman, 1986) deals with the con?icts of
interest that arise between management and the (outside) providers of equity and
ARJ
23,1
26
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
debt ?nance. To reduce these con?icts of interest, ?rms have incentives to disclose
information to outsiders and to have these disclosures audited. Accounting standards
(and GAAP generally) assist these monitoring activities. Similarly, companies that
make voluntary disclosures can reduce information asymmetry in a contractual setting
and further facilitate communication between the companies’ insiders and outsiders. As
managers’ share ownership falls, their incentives to increase disclosures to reduce
agency costs will increase. Evidence fromprior research shows this positive relationship
between ownership structure and disclosure (Eng and Mak, 2003; Ruland et al., 1990).
Similarly, signaling theory (Spence, 1973) focuses on how companies release
value-revealing information to signal to the market. Those companies that make credible
voluntary disclosures canenable investors to evaluate the company’s riskandits potential
future returns more accurately, which could lead to a more appropriate level of cost of
equity capital (Botosan, 1997; Botosan and Harris, 2000; Botosan and Plumlee, 2002).
In the aftermath of the SE reform, the proportion of outside equity in Chinese ?rms
will increase. Therefore, reformed companies are expected to have more incentive to
enhance their transparency via greater disclosure. In particular, as the interests of
controlling shareholders are now more aligned with the market performance of the
company, increased disclosure will be made to maintain the company’s share price.
The reform also creates additional incentives for corporate managers in China. With
the SE structure removed, companies are allowed to implement stock option schemes to
further align the interests of managers with outside shareholders. Consistent with
positive accounting theory (Watts and Zimmerman, 1986), corporate managers in those
reformed companies are expected to push for greater disclosure to enhance transparency
as the market value of the entity will be used as a basis for assessing their operational
performance. Further, since the reform will reduce the state’s administrative
involvement in Chinese listed ?rms, managerial career concerns will be switched
more from government/political concerns (with close government scrutiny) to be
somewhat closer to managerial career concerns in a market system, where managers are
concerned about their reputation and human capital in the managerial labor market. It is
expected that these implicit incentives will motivate managers to be more responsive to
the market and thereby provide more disclosures.
Furthermore, as more non-tradable shares become tradable, the in?uence of holders
of tradable shares, especially institutional investors (both domestic and foreign), should
increase and the in?uence of the controlling shareholder should decrease. These outside
shareholders are likely to demand the disclosure of more information from companies.
Therefore, in the aftermath of the SE reform, the reformed companies will feel increased
pressure to be more transparent than before the reform. However, because of the
moderating effect on the mandatory disclosure of the reporting templates included in
the Chinese GAAP, we expect the increase in mandatory disclosure to be smaller than
the increase in voluntary disclosure of the post-SE reform. These expectations are
addressed through the following hypotheses:
H1a. There will be an increase in mandatory disclosure between the pre-reform
2004 annual reports and the post-reform 2005 annual reports for companies
completing the SE reform.
H1b. The increase in mandatory disclosure between the pre-reform 2004 annual
reports and the post-reform 2005 annual reports will be greater for
Split equity
reform
27
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
companies completing the SE reform compared to companies not completing
the reform.
H2a. There will be an increase in voluntary disclosure between the pre-reform
2004 annual reports and the post-reform 2005 annual reports for companies
completing the SE reform.
H2b. The increase in voluntary disclosure between the pre-reform 2004 annual
reports and the post-reform 2005 annual will be greater for companies
completing the SE reform compared to companies not completing the
reform.
H3a. The post-reformincrease inmandatory disclosure will be smaller thanthe post-
reform increase in voluntary disclosure for companies completing the
SE reform.
H3b. The positive difference between the post-reform voluntary disclosure and the
post-reformmandatory disclosure will be greater for companies completingthe
SE reform compared to companies not completing the reform.
5. Research method
5.1 Sample selection
Our sample comprises all companies listed on the China Securities Index (CSI)
300 index in 2005 that completed the SE reform in 2005 (reformed companies, n ¼ 58),
and a control group of companies, matched by size and industry, that had not
commenced the reform in 2005 (unreformed companies, n ¼ 58). The resulting sample
size of 116 (58 £ 2) is more than suf?cient to detect large effect sizes in regressions with
up to 20 explanatory variables (Field, 2009, pp. 56-7 and 222-3)[7].
The matched pairs design addresses concerns that company size or industry are
associated with companies selected for reform by the government, or with companies
that self-selected for early reform. The CSI 300 Index covers more than 60 percent of
the market value of shares listed on the Shanghai and Shenzhen Stock Exchanges.
Consistent with prior research, only companies listed on the Shanghai Stock Exchange
are selected, since this exchange is the major stock exchange in China and has been the
benchmark used in prior research (Feng and Xu, 2007).
The years 2004 and 2005 are our focus because, for those years, the SEreformcreates
a natural experimental setting. The year 2004 is the year immediately before the SE
reform took place (pre-reform year), while 2005 represents the post-reform period as the
reform commenced in April 2005. Thus, by examining transparency scores, pre- and
post-reform for the reformed companies and comparing those to a (then) unreformed
control group, we expect to capture the impact of the reform on reformed companies.
Of the 58 companies which announced their participation in the reform before
December 31, 2005, 37 companies completed the reform before December 31, 2005[8]
and 21 companies completed the reform by February 2006 (prior to release of their 2005
annual reports at the end of March 2006)[9]. Matching of each reformed company with
an unreformed control company was based on their pre-reform size and industry (as on
December 31, 2004). Each reformed and matched unreformed company had to be in the
same industry, based on the Global Industry Classi?cation System developed by
Morgan Stanley and Standard & Poors. Within that same industry classi?cation,
ARJ
23,1
28
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the company closest in size (by natural logarithm of total assets) to the reformed
company was selected as the matched unreformed company. As reported in Table VIII,
the matching process resulted in no signi?cant difference in the average size of the
reformed and unreformed groups of companies. Other differences between the two
samples are addressed in the “Results” section.
Companies’ 2004 and 2005 annual reports (in Chinese language) were obtained from
the Shanghai Stock Exchange web site. Other ?nancial data for the sample for 2004
and 2005 were obtained from the China Center for Economic Research (CCER)
database. CCER is a major database provider in China and has been used in previous
studies (Feng and Xu, 2007).
5.2 Corporate ?nancial transparency measurement
The mandatory disclosure requirements for listed companies (Chinese GAAP) are to be
found in accounting standards, the requirements of the ASBE[10], and additional
disclosure requirements issued by the Ministry of Finance, the CSRC and the Shanghai
Stock Exchange. To measure corporate ?nancial transparency, we developed an
extensive checklist of 270 items from IFRS that could have been adopted by Chinese
companies. To develop the measure, we started with the disclosure checklist of 441
IFRS items in Morris and Gray (2007), itself based on a much larger list of items
prepared by Deloitte Touche Tohmatsu (2001). Morris and Gray removed items which
could not reasonably be observed by an annual report reader. We modi?ed their
checklist to make it fully applicable to Chinese companies, by removing items which
are not relevant in China. For example, the IFRS checklist requires disclosure of
employee entitlements in relation to de?ned bene?ts; however, de?ned bene?t pension
schemes are not applicable or used by Chinese companies. Further, the IFRS
requirements for ?nancial instruments in relation to accounting for hedges are not
applicable since the fair value and revaluation model are not allowed in Chinese GAAP,
only the cost model is allowed. As a result, it would be unrealistic to expect that any
company would voluntarily disclose any ?nancial instrument in relation to hedged
items where it uses fair value as the basis. Our resulting checklist contains 270 items.
Of these items, 178 were required by Chinese GAAP on December 31, 2004 and also on
December 31, 2005 – we treat them as mandatory disclosures. The remaining 92 items
were not required by Chinese GAAP in 2004 or 2005, but are included in the IFRS
regime in 2004 and 2005 – they are treated as voluntary disclosures. Therefore,
voluntary disclosures are IFRS-based checklist items not required by Chinese GAAP
but which could have been disclosed voluntarily by Chinese companies[11].
Each itemin the checklist is given equal weight based on the assumption that all items
of disclosure are equally important. This method eliminates any bias inherent in a
weighting system (Chau and Gray, 2002; Chow and Wong-Boren, 1987). Each checklist
itemwas coded 1 if disclosed, 0 if not disclosed and NAif not applicable to that company.
Three transparency scores were calculated: mandatory, voluntary, and overall.
The calculations are illustrated in the following example. As the mandatory
disclosure score has 178 items in total, assume company ABC has 142 items disclosed,
six items not disclosed, and 30 items not-applicable to the company. Thus, the
mandatory score is calculated as 142/(142 þ 6) ¼ 0.96. Similarly, for the 92 voluntary
disclosure items, assume company ABC has 40 items disclosed, 28 items not voluntarily
disclosed, and 24 items not applicable to the company. Hence, the voluntary score is
Split equity
reform
29
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
calculated as 40/(40 þ 28) ¼ 0.59. Finally, the overall score is ð142 þ 40Þ=
ð142 þ 6 þ 40 þ 28Þ ¼ 0:84.
5.3 Research design
The measure used in this study to determine the impact of the SE reform on corporate
?nancial transparency in China is the DiD estimator (Wooldridge, 2006). The DiD
estimator measures the difference in average outcome in a treatment group before and
after treatment minus the difference in average outcome in the control group before and
after treatment (Wooldridge, 2006). Essentially, the DiD estimator captures changes in
transparency of the 58 reformed companies and compares them with changes in
transparency of the 58 unreformed control companies. The DiD estimator can capture
the reform’s impact on transparency through the following basic model:
DTRANSP
i
¼ d
0
þb
1
DReformed
i
þDu
i
The DiD estimator[12]:
b
1
¼ DTRANSPi
treatment
2DTRANSPi
control
ði ¼ year 2004 or 2005Þ
.
Dependent variable. DTRANSP
i
is the change in transparency score between
2004 and 2005.
.
Variable of interest. DReformed
i
is a variable coded 1 for the reformed group as
they change from 0 in 2004 (pre-reform) to 1 in 2005 (post-reform) while the
change in score for the unreformed group is 0 to 0 from 2004-2005.
An important assumption underlying the DiD estimator is the parallel trend
assumption, namely that the treatment group (reformed companies) and the control
group (unreformed companies) should be reasonably similar before the “treatment”
event happens[13]. In other words, if the parallel trend assumption holds, the “starting
points” (pre-reform 2004) transparency scores for the two groups before the 2005 SE
reform should be reasonably similar. The parallel trend assumption also assumes that
the impact on the transparency of the two groups in 2004 and between 2004 and 2005
of any other factors associated with transparency should be reasonably similar. These
other factors (suggested by studies such as Tower and Yan, 2006) are:
Control variables
(1) Financial variables:
.
DUltimate control
i
measures the change in the ultimate controller of the
company between 2004 and 2005. It is assigned 1 if the ultimate controller
changes from privately controlled to government (state) controlled, and 0
otherwise.
.
DSize
i
measures the change in company size between 2004 and 2005. Size is
measured by the natural logarithm of total assets of the company.
.
DLeverage
i
measures the change in capital structure between 2004 and 2005.
This is proxied by total liabilities divided by total assets (TL/TA).
.
DNPAT
i
measures the change in pro?tability de?ated by the amount of total
assets between 2004 and 2005.
ARJ
23,1
30
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
.
DNon-tradable
i
measures the change in the percentage of non-tradable
shares between 2004 and 2005[14].
.
DForeign
i
measures the change in percentage foreign ownership between
2004 and 2005. Foreign investors include QFII and Hong Kong investors as
reported in annual reports.
(2) Corporate governance variables:
.
DAudit committee
i
equals 1 if the company set up an audit committee
between 2004 and 2005, and 0 otherwise.
.
DRemuneration committee
i
equals 1 if the company set up a remuneration
committee between 2004 and 2005, and 0 otherwise.
.
DNomination committee
i
equals 1 if the company set up a nomination
committee between 2004 and 2005, and 0 otherwise.
.
DDual
i
equals 1 if the company separated the role of CEO and chairman
between 2004 and 2005, and 0 if no change, and 21 if separation in 2004 was
removed in 2005.
(3) Additional control variables used in prior studies:
.
DInstitutional
i
measures the change in the proportion of ownership held by
institutional investors within the top 20 shareholders as disclosed in the
annual reports between 2004 and 2005. It is the sum of the domestic
institutional investors and the foreign institutional investors such as QFII.
.
DAudit ?rm
i
equals 1 if the company switched from non-Big Four to Big
Four auditor between 2004 and 2005, and 0 otherwise.
.
DIndependent directors
i
measures the change in the percentage of
independent directors on the board between 2004 and 2005. The percentage
of independent directors is the number of independent directors divided by
the total board size.
However, the DiD approach only requires these control variables if the parallel trend
assumption does not hold between the reformed and unreformed groups. Paired
samples t-tests reported in the results section indicate that size, leverage, NPAT,
non-tradable, foreign, audit committee, remuneration committee, and nomination
committee differed signi?cantly between 2004 and 2005 and therefore require control
variables in our model. In addition, independent t-tests reported in the results section
indicate that ultimate control, non-tradable and transparency scores differ signi?cantly
between the reformed and unreformed companies in the pre-reform (2004) period and
therefore also require control variables in our model. The resulting model tested is
shown as model (1)[15]:
DTRANSP
i
¼ d
0
þb
1
DReformed
i
þb
2
DUltimate control
i
þb
3
DSize
i
þb
4
DLeverage
i
þb
5
DNPAT
i
þb
6
DNon–tradable
i
þb
7
DForeign
i
þb
8
DAudit committee
i
þb
9
DRemuneration committee
i
þb
10
DNomination committee
i
þb
11
DDual
i
þb
12
Disc04
þb
13
Non–tradable04 þb
14
Ultimate control04 þDu
i
Model ð1Þ
Split equity
reform
31
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
6. Results and discussion
Tables II-V provide descriptive statistics for all variables for the 58 reformed and 58
unreformed companies in the pre-reform 2004 period (Tables II and IV) and the
post-reform 2005 period (Tables III and V). Most continuous variables do not exhibit
any pronounced skewness. Some binary variables are skewed, however as discussed
later, they do not bias our central ?ndings. Consistent with Walter and Howie (2006),
there are few institutional investors to demand voluntary corporate disclosure in China
from either the reformed or unreformed companies in the pre-SE reform period. The
level of institutional shareholding is small compared to the percentage of tradable
shares available, and the percentage of companies listed in Hong Kong (and thus
issuing H-shares) is low.
Table VI shows that in 2004, the mandatory, voluntary, and overall average scores
for the reformed companies (0.964; 0.677; 0.912) are lower than for the unreformed
companies (0.973; 0.690; 0.921), the differences for mandatory and overall scores being
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory score 04 0.890 0.990 0.964 0.017 21.952
Voluntary score 04 0.460 0.850 0.677 0.092 20.116
Overall score 04 0.830 0.970 0.912 0.025 20.578
Independent variables
Ultimate control 0.000 1.000 0.120 0.329 22.391
B-share 0.000 1.000 0.070 0.256 3.493
H-share 0.000 1.000 0.030 0.184 5.239
Size 20.989 24.886 22.338 0.822 0.683
Leverage 0.161 0.762 0.446 0.157 0.212
NAPT 20.073 0.164 0.058 0.044 0.298
Non-tradable 0.243 0.846 0.627 0.120 20.924
Institutional 0.000 0.417 0.063 0.086 2.821
Foreign 0.000 0.414 0.020 0.078 4.519
Audit ?rm 0.000 1.000 0.170 0.381 1.781
Audit committee 0.000 1.000 0.530 0.503 20.142
Remuneration committee 0.000 1.000 0.450 0.502 0.214
Nomination committee 0.000 1.000 0.330 0.473 0.754
Dual 0.000 1.000 0.983 0.131 27.616
Independent directors 0.180 0.500 0.331 0.052 0.313
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table II.
Reformed companies
2004
ARJ
23,1
32
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
statisticallysigni?cant. However, in2005, the mandatory, voluntary and overall average
scores for the reformed companies (0.973; 0.694; 0.921) are higher than for the
unreformed companies (0.972; 0.666; 0.917), although the differences are not statistically
signi?cant. That is, between 2004 and 2005, the mandatory, voluntary and overall
average scores increased for the reformed, but fell slightly for the unreformed,
companies.
The results for 2004 (Table VI) show that mandatory and overall transparency
scores, but not the voluntary scores, are signi?cantly different between the reformed
and unreformed companies. This indicates that the parallel trend assumption in 2004
may not hold as the starting points for the mandatory and overall scores are different
for reformed and unreformed samples. In our multivariate tests, we address this
problem by including a control variable (Disc_04) for 2004 disclosure levels. There are
no signi?cant differences between reformed and unreformed companies’ disclosure
scores in 2005.
Table VI also shows that that reformed companies’ scores on mandatory and overall
disclosure signi?cantly increased from 2004 to 2005. While reformed companies’
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory Score 05 0.940 1.000 0.973 0.011 20.693
Voluntary Score 05 0.500 0.900 0.694 0.086 0.242
Overall Score 05 0.880 0.980 0.921 0.019 0.675
Independent variables
Ultimate control 0.000 1.000 0.900 0.307 22.674
Size 21.009 25.679 22.456 0.872 1.044
Leverage 0.136 0.834 0.475 0.160 0.034
NAPT 20.154 0.205 0.051 0.052 20.058
Non-tradable tradable 0.122 0.797 0.512 0.143 20.634
Institutional 0.000 0.405 0.070 0.078 2.487
Foreign 0.000 0.385 0.020 0.069 4.562
Audit ?rm 0.000 1.000 0.160 0.365 1.956
Audit committee 0.000 1.000 0.570 0.500 20.286
Remuneration committee 0.000 1.000 0.500 0.504 0.000
Nomination committee 0.000 1.000 0.330 0.473 0.754
DUAL 0.000 1.000 0.931 0.256 23.493
Independent directors 0.231 0.500 0.330 0.045 1.385
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table III.
Reformed companies
2005
Split equity
reform
33
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
voluntary disclosure scores increased from 2004 to 2005, the change is only signi?cant
at the 10 percent level (two-tailed). On the other hand, unreformed companies’
mandatory and overall disclosures do not signi?cantly vary between 2004 and 2005,
and their voluntary disclosure scores signi?cantly fall from 2004 to 2005. In sum, the
results in Table VI support H1a and H2a.
Table VII shows that the average changes in mandatory, voluntary and overall
scores between 2004 and 2005 are greater for the reformed companies than for the
unreformed, and all differences between reformed and unreformed companies are
statistically signi?cant. The results in Tables VI and VII are consistent with the SE
reform impacting as predicted (H1b and H2b) on the reformed companies’ transparency
scores. Tests reported in Tables VI and VII were repeated using non-parametric
Mann-Whitney U-tests and Wilcoxon signed ranked tests. Results were substantially
unchanged.
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory score 04 0.940 1.000 0.973 0.011 20.693
Voluntary score 04 0.520 0.880 0.690 0.084 20.081
Overall score 04 0.870 0.950 0.921 0.018 20.491
Independent variables
Ultimate control 0.000 1.000 0.280 0.451 21.030
B-share 0.000 1.000 0.070 0.256 3.493
H-share 0.000 1.000 0.070 0.256 3.493
Size 20.782 24.172 22.343 0.740 0.602
Leverage 0.033 0.916 0.454 0.182 20.270
NAPT 0.001 0.226 0.051 0.045 1.552
Non-tradable 0.000 0.850 0.546 0.187 21.203
Institutional 0.000 0.449 0.071 0.083 2.500
Foreign 0.000 0.431 0.030 0.079 3.746
Audit ?rm 0.000 1.000 0.220 0.421 1.358
Audit committee 0.000 1.000 0.550 0.502 20.214
Remuneration committee 0.000 1.000 0.570 0.500 20.286
Nomination committee 0.000 1.000 0.430 0.500 0.286
Dual 1.000 1.000 1.000 0.000 0.000
Independent directors 0.000 0.430 0.314 0.068 23.444
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table IV.
Unreformed companies
2004
ARJ
23,1
34
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
For the reformedcompanies, the increase inthe voluntary score (0.0178) is greater thanthe
increase in the mandatory score (0.0093), although the difference is not statistically
signi?cant ( p ¼ 0.395); thus, H3a is not supported. The unreformed group had decreases
in voluntary (20.0238) and mandatory disclosure (20.0040), with their voluntary
disclosure decrease being signi?cantly greater than their mandatory disclosure decrease.
Reformed companies had a greater increase, albeit not statistically signi?cant, in
voluntary disclosure than in mandatory disclosure, but they did not experience the same
signi?cant gap between the declines in mandatory and voluntary disclosures that the
unreformed group did; this result provides quali?ed support for H3b.
In 2004, the mandatory and overall scores for both the reformed companies and the
unreformed companies were quite high. This result is likely driven by the reporting
templates contained in the Chinese GAAP. Compliance with these templates can push
up mandatory disclosure, thereby creating a ceiling effect which allows little room for
improvement in the TRANSP score over time. Nevertheless, the reformed companies
had a signi?cant increase in mandatory disclosure from 2004 to 2005, the unreformed
companies did not.
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory Score 05 0.930 1.000 0.972 0.013 21.237
Voluntary Score 05 0.470 0.880 0.666 0.086 0.030
Overall Score 05 0.870 0.950 0.917 0.019 20.401
Independent variables
Ultimate control 0.000 1.000 0.240 0.432 1.241
Size 21.049 24.384 22.450 0.755 0.690
Leverage 0.048 0.887 0.482 0.176 20.475
NAPT 20.248 0.207 0.029 0.069 21.194
Non-tradable 0.000 0.850 0.542 0.186 21.182
Institutional 0.000 0.000 0.070 0.089 2.455
Foreign 0.000 0.484 0.030 0.083 3.836
Audit ?rm 0.000 1.000 0.240 0.432 1.241
Audit committee 0.000 1.000 0.590 0.497 20.359
Remuneration committee 0.000 1.000 0.660 0.479 20.670
Nomination committee 0.000 1.000 0.480 0.504 0.071
DUAL 0.000 1.000 0.897 0.307 22.674
Independent directors 0.000 0.533 0.317 0.077 21.710
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table V.
Unreformed companies
2005
Split equity
reform
35
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
As noted, the mandatory and overall disclosure scores of unreformed companies
exceeded those of the reformed companies before the reform, with the difference
reversing after the reform. This seemingly counterintuitive result can be explained as
follows. The CSRC selected the Pilot Reform companies after considering the likely
attitude of shareholders to the Pilot Reform; the CSRC selected those companies where
they believed the shareholders were likely come to an agreed position quite quickly.
Since reformed companies had a higher percentage of state ownership than unreformed
companies (88 percent reformed, 72 percent unreformed), the government may have
considered it could expedite the reform more smoothly in the companies chosen. Prior
research in China indicates that state ownership is negatively associated with
corporate ?nancial transparency (Morris et al., 2006). Therefore, the higher level of
government ultimate control over the reformed companies in 2004 may help explain
the lower opening disclosure level for these companies. We return to the difference in
control in the following discussion.
Reformed
(n ¼ 58)
Unreformed
(n ¼ 58)
Mean
difference:
reformed vs
unreformed t
Sig.
(two-tailed)
2004
Mandatory score 0.964
a
0.973
a
20.009 23.233 0.002
Voluntary score 0.677
b
0.690
b
20.013 20.810 0.420
Overall score 0.912
c
0.921
c
20.009 22.275 0.025
2005
Mandatory score 0.973
a
0.972
a
0.001 0.636 0.526
Voluntary score 0.694
b
0.666
b
0.028 1.774 0.079
Overall score 0.921
c
0.917
c
0.004 1.236 0.219
a
Sig. (two-tailed) mandatory 04 vs 05 0.000 0.650
b
Sig. (two-tailed) voluntary 04 vs 05 0.097 0.039
c
Sig. (two-tailed) overall 04 vs 05 0.003 0.133
Notes: Mandatory – checklist items required by Chinese GAAP; voluntary – checklist items not
required by Chinese GAAP; overall – combined mandatory and voluntary score
Table VI.
Mean differences between
reformed and unreformed
companies on disclosure
scores in 2004 and 2005
Mean change
04/05 reformed
(n ¼ 58)
Mean change
04/05
unreformed
(n ¼ 58) t
Sig.
(two-tailed)
Mandatory score 0.0093 20.0007 24.051 0.000
Voluntary score 0.0178 20.0238 22.693 0.008
Overall score 0.0095 20.0040 23.331 0.001
t-score mandatory change vs voluntary change 20.857 2.073
Sig. (two-tailed) 0.395 0.043
Notes: Mandatory – checklist items required by Chinese GAAP; voluntary – checklist items not
required by Chinese GAAP; overall – combined mandatory and voluntary score
Table VII.
Changes in disclosure
scores between 2004 and
2005 for reformed and
unreformed companies
ARJ
23,1
36
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table VIII indicates that, of the control variables, only ultimate control (which shows
the extent of state control) and non-tradable shares are signi?cantly different between
the reformed and unreformed companies in 2004. Therefore, as previously noted,
because of these two potential violations of the DiD equation parallel assumption, we
include control variables for the 2004 values of ultimate control and non-tradable shares
in our models.
In addition, paired sample t-tests were performed to see whether or not the control
variables used in previous studies changed signi?cantly between 2004 and 2005
(Table VIII). There were signi?cant differences between years for the following
variables: size, leverage, pro?t, ultimate control, non-tradable shares, foreign ownership,
and audit, remuneration and nomination committees. Since the changes in these
variables could affect the change in transparency they are included in our DiD model
(model (1)).
Correlation matrices (Tables IX-XI) indicate that while there are a number of
independent variables that are signi?cantly correlated in both 2004 and 2005, only two
correlation coef?cients are greater than 0.8: the correlation of institutional and foreign
(0.86 in 2004; 0.84 in 2005), and foreign and H-share (0.86 in 2004; 0.86 in 2005).
Reformed vs unreformed
in 2004
All ?rms: change
from 2004 to 2005
Mean
difference t
Sig.
(two-tailed)
Mean
change t
Sig.
(two-tailed)
Ultimate control 0.155 2.118 0.037 20.026 21.346 0.181
B-share 0.000 0.000 1.000 n/a
H-share 20.034 20.834 0.406 n/a
Size 20.005 20.033 0.974 20.112 27.091 0.000
Leverage 20.008 20.254 0.800 20.029 24.133 0.000
NPAT 0.006 0.774 0.440 0.014 3.325 0.001
Non-tradable 0.080 2.751 0.007 0.059 8.953 0.000
Institutional 20.008 20.497 0.620 20.004 21.450 0.150
Foreign 20.006 20.423 0.673 20.002 22.134 0.035
Audit ?rm 20.052 20.694 0.489 0.000 0.000 1.000
Audit committee 20.017 20.185 0.854 20.034 22.027 0.045
Remuneration committee 20.121 21.298 0.197 20.069 22.919 0.004
Nomination committee 20.103 21.145 0.255 20.026 21.747 0.083
Dual 0.017 1.000 0.322 0.078 3.110 0.002
Independent directors 0.017 1.531 0.129 20.001 20.237 0.813
Notes: Ultimate control – 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1
if the company has B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size –
natural logarithm of total assets; leverage – total liabilities divided by total assets;
NPAT – pro?tability de?ated by total assets; non-tradable – percentage of non-tradable shares
(as described in footnote [14] in 2005 this variable measures restricted tradable shares); institutional
– proportion of top 20 shareholder ownership held by institutional investors (includes domestic and
foreign institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and
Hong Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit
committee, 0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise;
nomination committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEOand chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table VIII.
Tests for signi?cant
differences among
potential control
variables
Split equity
reform
37
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
B
-
s
h
a
r
e
H
-
s
h
a
r
e
S
i
z
e
L
e
v
e
r
a
g
e
N
P
A
T
N
o
n
-
t
r
a
d
a
b
l
e
I
n
s
t
i
t
u
t
i
o
n
a
l
F
o
r
e
i
g
n
A
u
d
i
t
?
r
m
A
u
d
i
t
c
o
m
m
i
t
t
e
e
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
u
a
l
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
1
B
-
s
h
a
r
e
2
0
.
0
3
5
1
H
-
s
h
a
r
e
0
.
1
1
6
2
0
.
0
6
4
1
S
i
z
e
0
.
2
2
3
*
0
.
1
3
6
0
.
2
7
8
*
*
1
L
e
v
e
r
a
g
e
2
0
.
2
4
7
*
*
0
.
1
6
5
2
0
.
1
7
2
0
.
2
2
0
*
1
N
A
P
T
0
.
2
1
0
*
2
0
.
1
3
1
0
.
1
9
3
*
0
.
1
0
2
2
0
.
4
1
5
*
*
1
N
o
n
-
t
r
a
d
a
b
l
e
0
.
2
8
5
*
*
2
0
.
2
1
3
*
2
0
.
0
0
8
0
.
2
8
6
*
*
2
0
.
1
6
7
0
.
3
5
3
*
*
1
I
n
s
t
i
t
u
t
i
o
n
a
l
0
.
0
9
5
0
0
.
7
2
1
*
*
0
.
2
1
6
*
2
0
.
1
1
2
0
.
2
7
8
*
*
2
0
.
0
7
3
1
F
o
r
e
i
g
n
0
.
0
3
9
0
.
0
4
8
0
.
8
5
6
*
*
0
.
2
6
2
*
*
2
0
.
1
0
3
0
.
1
6
4
2
0
.
0
8
1
0
.
8
5
9
*
*
1
A
u
d
i
t
?
r
m
0
.
1
3
9
0
.
1
2
1
0
.
3
7
2
*
*
0
.
4
1
9
*
*
2
0
.
1
6
4
0
.
2
6
7
*
*
0
.
2
1
4
*
0
.
2
4
0
*
*
0
.
3
0
7
*
*
1
A
u
d
i
t
c
o
m
m
i
t
t
e
e
0
.
0
2
1
2
0
.
0
2
4
0
.
1
3
6
0
.
0
9
1
2
0
.
0
7
0
.
1
0
7
0
.
1
2
4
2
0
.
0
3
1
0
.
0
2
5
0
.
1
0
9
1
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
0
.
0
3
2
0
.
0
0
5
2
0
.
0
8
2
2
0
.
0
3
3
2
0
.
0
3
2
0
.
1
2
5
0
.
1
3
4
2
0
.
1
1
1
2
0
.
1
8
2
0
.
0
7
3
0
.
3
7
9
*
*
1
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
4
6
0
.
0
6
8
2
0
.
1
0
2
2
0
.
1
0
1
0
.
0
4
2
0
.
0
0
3
0
.
0
4
6
2
0
.
1
2
2
0
.
1
1
4
0
.
0
1
2
0
.
5
7
4
*
*
0
.
5
2
0
*
*
1
D
u
a
l
2
0
.
1
8
8
*
0
.
0
2
5
0
.
0
2
2
0
.
0
1
8
2
0
.
0
5
4
2
0
.
0
4
5
2
0
.
0
4
4
2
0
.
0
1
0
0
.
0
1
0
0
.
0
4
6
0
.
1
0
2
2
0
.
0
9
2
0
.
0
7
3
1
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
2
0
.
1
3
6
0
.
1
4
2
2
0
.
4
3
8
*
*
2
0
.
1
2
9
0
.
0
9
3
2
0
.
0
7
4
2
0
.
0
4
5
2
0
.
2
3
3
*
2
0
.
3
1
0
*
*
2
0
.
2
4
3
*
*
2
0
.
0
8
6
2
0
.
0
3
4
2
0
.
0
0
2
2
0
.
0
8
1
1
N
o
t
e
s
:
C
o
r
r
e
l
a
t
i
o
n
i
s
s
i
g
n
i
?
c
a
n
t
a
t
t
h
e
*
0
.
0
5
a
n
d
*
*
0
.
0
1
l
e
v
e
l
s
(
t
w
o
-
t
a
i
l
e
d
)
,
r
e
s
p
e
c
t
i
v
e
l
y
;
m
a
n
d
a
t
o
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
v
o
l
u
n
t
a
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
n
o
t
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
o
v
e
r
a
l
l
–
c
o
m
b
i
n
e
d
m
a
n
d
a
t
o
r
y
a
n
d
v
o
l
u
n
t
a
r
y
s
c
o
r
e
;
f
o
r
t
h
e
f
o
l
l
o
w
i
n
g
v
a
r
i
a
b
l
e
s
D
i
n
d
i
c
a
t
e
s
c
h
a
n
g
e
f
r
o
m
2
0
0
4
t
o
2
0
0
5
:
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
–
1
i
f
t
h
e
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
l
e
r
i
s
g
o
v
e
r
n
m
e
n
t
(
s
t
a
t
e
)
,
0
o
t
h
e
r
w
i
s
e
;
B
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
B
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
H
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
H
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
s
i
z
e
–
n
a
t
u
r
a
l
l
o
g
a
r
i
t
h
m
o
f
t
o
t
a
l
a
s
s
e
t
s
;
l
e
v
e
r
a
g
e
–
t
o
t
a
l
l
i
a
b
i
l
i
t
i
e
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
N
P
A
T
–
p
r
o
?
t
a
b
i
l
i
t
y
d
e
?
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
n
o
n
-
t
r
a
d
a
b
l
e
–
p
e
r
c
e
n
t
a
g
e
o
f
n
o
n
-
t
r
a
d
a
b
l
e
s
h
a
r
e
s
(
a
s
d
e
s
c
r
i
b
e
d
i
n
f
o
o
t
n
o
t
e
[
1
4
]
i
n
2
0
0
5
t
h
i
s
v
a
r
i
a
b
l
e
m
e
a
s
u
r
e
s
r
e
s
t
r
i
c
t
e
d
t
r
a
d
a
b
l
e
s
h
a
r
e
s
)
;
i
n
s
t
i
t
u
t
i
o
n
a
l
–
p
r
o
p
o
r
t
i
o
n
o
f
t
o
p
2
0
s
h
a
r
e
h
o
l
d
e
r
o
w
n
e
r
s
h
i
p
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
(
i
n
c
l
u
d
e
s
d
o
m
e
s
t
i
c
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
a
n
d
f
o
r
e
i
g
n
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
s
u
c
h
a
s
Q
F
I
I
)
;
f
o
r
e
i
g
n
–
p
e
r
c
e
n
t
a
g
e
f
o
r
e
i
g
n
o
w
n
e
r
s
h
i
p
(
i
n
c
l
u
d
e
s
Q
F
I
I
a
n
d
H
o
n
g
K
o
n
g
i
n
v
e
s
t
o
r
s
)
;
a
u
d
i
t
?
r
m
–
1
i
f
t
h
e
B
i
g
4
a
u
d
i
t
o
r
,
0
o
t
h
e
r
w
i
s
e
;
a
u
d
i
t
c
o
m
m
i
t
t
e
e
–
1
i
f
a
u
d
i
t
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
d
u
a
l
–
e
q
u
a
l
s
1
i
f
t
h
e
c
o
m
p
a
n
y
s
e
p
a
r
a
t
e
d
t
h
e
r
o
l
e
o
f
C
E
O
a
n
d
c
h
a
i
r
m
a
n
b
e
t
w
e
e
n
2
0
0
4
a
n
d
2
0
0
5
,
a
n
d
0
i
f
n
o
c
h
a
n
g
e
,
a
n
d
–
1
i
f
s
e
p
a
r
a
t
i
o
n
i
n
2
0
0
4
w
a
s
r
e
m
o
v
e
d
i
n
2
0
0
5
;
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
–
n
u
m
b
e
r
o
f
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
b
o
a
r
d
s
i
z
e
Table IX.
Pearson correlation
matrix between
independent variables
2004
ARJ
23,1
38
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
R
e
f
o
r
m
e
d
B
-
s
h
a
r
e
H
-
s
h
a
r
e
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
S
i
z
e
L
e
v
e
r
a
g
e
N
P
A
T
N
o
n
-
t
r
a
d
a
b
l
e
I
n
s
t
i
t
u
t
i
o
n
a
l
F
o
r
e
i
g
n
A
u
d
i
t
?
r
m
A
u
d
i
t
c
o
m
m
i
t
t
e
e
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
u
a
l
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
R
e
f
o
r
m
e
d
1
B
-
s
h
a
r
e
0
.
0
0
0
1
H
-
s
h
a
r
e
2
0
.
0
7
8
2
0
.
0
6
4
1
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
0
.
1
8
3
*
2
0
.
0
3
4
0
.
1
0
7
1
S
i
z
e
0
.
0
0
3
0
.
1
0
.
2
8
8
*
*
2
0
.
1
8
5
*
1
L
e
v
e
r
a
g
e
2
0
.
0
2
3
0
.
0
8
8
2
0
.
1
7
0
.
1
8
1
0
.
1
5
9
1
N
A
P
T
0
.
1
7
2
2
0
.
0
4
1
0
.
2
1
9
*
2
0
.
1
6
1
0
.
2
3
3
*
2
0
.
3
5
9
*
*
1
N
o
n
-
t
r
a
d
a
b
l
e
2
0
.
0
9
2
2
0
.
1
6
3
0
.
0
5
4
2
0
.
1
3
2
0
.
3
4
8
*
*
2
0
.
1
8
0
.
2
5
4
*
*
1
I
n
s
t
i
t
u
t
i
o
n
a
l
2
0
.
0
5
1
2
0
.
0
0
6
0
.
7
1
8
*
*
2
0
.
0
7
8
0
.
2
9
1
*
*
2
0
.
1
8
7
*
0
.
4
1
8
*
*
2
0
.
0
0
5
1
F
o
r
e
i
g
n
2
0
.
0
5
7
0
.
0
5
2
0
.
8
5
9
*
*
2
0
.
0
3
8
0
.
2
8
3
*
*
2
0
.
1
6
1
0
.
2
5
9
*
*
2
0
.
0
1
2
0
.
8
3
8
*
*
1
A
u
d
i
t
?
r
m
2
0
.
1
0
8
0
.
0
3
5
0
.
3
7
2
*
*
2
0
.
1
1
3
0
.
3
9
6
*
*
2
0
.
2
2
3
*
0
.
2
5
1
*
*
0
.
2
6
2
*
*
0
.
3
1
7
*
*
0
.
3
3
4
*
*
1
A
u
d
i
t
c
o
m
m
i
t
t
e
e
2
0
.
0
1
7
2
0
.
0
4
3
0
.
1
2
1
2
0
.
0
7
2
0
.
1
1
2
0
.
0
0
5
2
0
.
0
2
5
0
.
1
3
1
2
0
.
0
4
4
0
.
0
2
3
0
.
1
6
3
1
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
5
7
2
0
.
0
4
3
0
.
1
2
1
2
0
.
0
2
5
0
.
0
1
2
2
0
.
0
1
9
0
.
0
5
0
.
1
6
9
0
.
0
4
5
0
.
0
4
9
0
.
0
7
5
0
.
4
7
0
*
*
1
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
5
8
0
.
0
5
3
2
0
.
0
3
4
0
.
0
8
8
2
0
.
0
6
7
0
.
0
7
9
2
0
.
0
1
6
0
.
1
1
2
2
0
.
1
3
9
2
0
.
1
2
4
0
.
0
7
4
0
.
5
9
9
*
*
0
.
5
2
8
*
*
1
D
U
A
L
0
.
0
6
1
0
.
0
8
4
0
.
0
7
2
0
.
1
8
5
*
0
.
1
2
2
0
.
0
1
3
0
.
0
0
4
2
0
.
0
1
2
0
.
0
4
4
0
.
0
7
4
2
0
.
0
0
1
2
0
.
0
7
6
2
0
.
1
3
8
2
0
.
0
5
9
1
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
0
.
1
0
1
0
.
0
8
6
2
0
.
3
0
2
*
*
0
.
0
9
3
2
0
.
1
2
5
0
.
1
4
1
2
0
.
1
3
0
.
0
2
3
2
0
.
2
0
3
*
2
0
.
2
6
5
*
*
2
0
.
1
7
2
0
.
0
8
4
2
0
.
0
9
6
2
0
.
0
7
6
2
0
.
1
1
3
1
N
o
t
e
s
:
C
o
r
r
e
l
a
t
i
o
n
i
s
s
i
g
n
i
?
c
a
n
t
a
t
t
h
e
*
0
.
0
5
a
n
d
*
*
0
.
0
1
l
e
v
e
l
s
(
t
w
o
-
t
a
i
l
e
d
)
,
r
e
s
p
e
c
t
i
v
e
l
y
;
m
a
n
d
a
t
o
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
v
o
l
u
n
t
a
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
n
o
t
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
o
v
e
r
a
l
l
–
c
o
m
b
i
n
e
d
m
a
n
d
a
t
o
r
y
a
n
d
v
o
l
u
n
t
a
r
y
s
c
o
r
e
;
f
o
r
t
h
e
f
o
l
l
o
w
i
n
g
v
a
r
i
a
b
l
e
s
D
i
n
d
i
c
a
t
e
s
c
h
a
n
g
e
f
r
o
m
2
0
0
4
t
o
2
0
0
5
:
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
–
1
i
f
t
h
e
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
l
e
r
i
s
g
o
v
e
r
n
m
e
n
t
(
s
t
a
t
e
)
,
0
o
t
h
e
r
w
i
s
e
;
B
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
B
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
H
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
H
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
s
i
z
e
–
n
a
t
u
r
a
l
l
o
g
a
r
i
t
h
m
o
f
t
o
t
a
l
a
s
s
e
t
s
;
l
e
v
e
r
a
g
e
–
t
o
t
a
l
l
i
a
b
i
l
i
t
i
e
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
N
P
A
T
–
p
r
o
?
t
a
b
i
l
i
t
y
d
e
?
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
n
o
n
-
t
r
a
d
a
b
l
e
–
p
e
r
c
e
n
t
a
g
e
o
f
n
o
n
-
t
r
a
d
a
b
l
e
s
h
a
r
e
s
(
a
s
d
e
s
c
r
i
b
e
d
i
n
f
o
o
t
n
o
t
e
[
1
4
]
i
n
2
0
0
5
t
h
i
s
v
a
r
i
a
b
l
e
m
e
a
s
u
r
e
s
r
e
s
t
r
i
c
t
e
d
t
r
a
d
a
b
l
e
s
h
a
r
e
s
)
;
i
n
s
t
i
t
u
t
i
o
n
a
l
–
p
r
o
p
o
r
t
i
o
n
o
f
t
o
p
2
0
s
h
a
r
e
h
o
l
d
e
r
o
w
n
e
r
s
h
i
p
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
(
i
n
c
l
u
d
e
s
d
o
m
e
s
t
i
c
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
a
n
d
f
o
r
e
i
g
n
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
s
u
c
h
a
s
Q
F
I
I
)
;
f
o
r
e
i
g
n
–
p
e
r
c
e
n
t
a
g
e
f
o
r
e
i
g
n
o
w
n
e
r
s
h
i
p
(
i
n
c
l
u
d
e
s
Q
F
I
I
a
n
d
H
o
n
g
K
o
n
g
i
n
v
e
s
t
o
r
s
)
;
a
u
d
i
t
?
r
m
–
1
i
f
t
h
e
B
i
g
4
a
u
d
i
t
o
r
,
0
o
t
h
e
r
w
i
s
e
;
a
u
d
i
t
c
o
m
m
i
t
t
e
e
–
1
i
f
a
u
d
i
t
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
d
u
a
l
–
e
q
u
a
l
s
1
i
f
t
h
e
c
o
m
p
a
n
y
s
e
p
a
r
a
t
e
d
t
h
e
r
o
l
e
o
f
C
E
O
a
n
d
c
h
a
i
r
m
a
n
b
e
t
w
e
e
n
2
0
0
4
a
n
d
2
0
0
5
,
a
n
d
0
i
f
n
o
c
h
a
n
g
e
,
a
n
d
–
1
i
f
s
e
p
a
r
a
t
i
o
n
i
n
2
0
0
4
w
a
s
r
e
m
o
v
e
d
i
n
2
0
0
5
;
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
–
n
u
m
b
e
r
o
f
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
b
o
a
r
d
s
i
z
e
Table X.
Pearson correlation
matrix between
independent variables
2005
Split equity
reform
39
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
D
R
e
f
o
r
m
e
d
D
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
D
S
i
z
e
D
L
e
v
e
r
a
g
e
D
N
P
A
T
D
N
o
n
-
t
r
a
d
a
b
l
e
D
I
n
s
t
i
t
u
t
i
o
n
a
l
D
F
o
r
e
i
g
n
D
A
u
d
i
t
?
r
m
D
A
u
d
i
t
c
o
m
m
i
t
t
e
e
D
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
D
u
a
l
D
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
D
R
e
f
o
r
m
e
d
1
D
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
0
.
0
4
2
1
D
S
i
z
e
0
.
0
2
9
2
0
.
0
1
1
1
D
L
e
v
e
r
a
g
e
0
.
0
0
2
0
.
0
1
4
0
.
5
9
8
*
*
1
D
N
P
A
T
0
.
1
6
0
2
0
.
0
3
6
0
.
0
9
9
2
0
.
4
1
8
*
*
1
D
N
o
n
-
t
r
a
d
a
b
l
e
2
0
.
7
7
7
*
*
0
.
0
4
9
0
.
0
8
6
0
.
0
6
1
2
0
.
1
5
9
1
D
I
n
s
t
i
t
u
t
i
o
n
a
l
2
0
.
0
1
4
0
.
0
2
2
2
0
.
0
5
9
2
0
.
1
7
4
0
.
1
3
1
2
0
.
0
4
2
1
D
F
o
r
e
i
g
n
2
0
.
1
0
6
0
.
0
3
2
0
.
0
3
0
2
0
.
1
9
8
*
0
.
0
3
7
0
.
0
3
1
0
.
3
6
9
*
*
1
D
A
u
d
i
t
?
r
m
2
0
.
0
9
3
0
.
0
0
0
0
.
0
1
5
2
0
.
0
5
0
0
.
0
0
3
0
.
0
2
0
0
.
0
2
2
0
.
0
1
1
1
D
A
u
d
i
t
c
o
m
m
i
t
t
e
e
0
.
0
0
0
0
.
0
2
4
0
.
1
2
0
0
.
3
0
7
*
*
2
0
.
2
3
2
*
0
.
0
2
1
2
0
.
0
8
2
0
.
0
7
3
0
.
0
0
0
1
D
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
0
6
8
0
.
1
9
9
*
0
.
1
3
7
0
.
1
1
9
0
.
0
2
5
0
.
0
9
6
0
.
0
4
6
0
.
1
1
1
0
.
0
0
0
0
.
3
2
1
*
*
1
D
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
6
3
0
.
0
2
0
0
.
1
1
5
0
.
3
3
5
*
*
2
0
.
0
7
8
0
.
1
3
6
2
0
.
0
7
2
0
.
0
3
9
0
.
0
0
0
0
.
5
6
5
*
*
0
.
3
8
4
*
*
1
D
D
u
a
l
0
.
0
9
7
2
0
.
0
3
6
2
0
.
0
9
3
2
0
.
0
6
2
0
.
0
6
3
2
0
.
0
2
6
0
.
0
7
2
2
0
.
0
5
2
2
0
.
1
7
4
2
0
.
1
2
2
2
0
.
0
4
8
2
0
.
1
5
6
1
D
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
2
0
.
0
5
5
0
.
1
1
6
0
.
0
5
9
0
.
0
9
2
2
0
.
0
8
8
0
.
0
5
1
0
.
1
3
4
0
.
1
1
1
0
.
0
4
7
0
.
2
4
8
*
*
0
.
3
3
7
*
*
0
.
2
1
7
*
0
.
0
0
3
1
N
o
t
e
s
:
C
o
r
r
e
l
a
t
i
o
n
i
s
s
i
g
n
i
?
c
a
n
t
a
t
t
h
e
*
0
.
0
5
a
n
d
*
*
0
.
0
1
l
e
v
e
l
s
(
t
w
o
-
t
a
i
l
e
d
)
,
r
e
s
p
e
c
t
i
v
e
l
y
;
m
a
n
d
a
t
o
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
v
o
l
u
n
t
a
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
n
o
t
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
o
v
e
r
a
l
l
–
c
o
m
b
i
n
e
d
m
a
n
d
a
t
o
r
y
a
n
d
v
o
l
u
n
t
a
r
y
s
c
o
r
e
;
f
o
r
t
h
e
f
o
l
l
o
w
i
n
g
v
a
r
i
a
b
l
e
s
D
i
n
d
i
c
a
t
e
s
c
h
a
n
g
e
f
r
o
m
2
0
0
4
t
o
2
0
0
5
:
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
–
1
i
f
t
h
e
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
l
e
r
i
s
g
o
v
e
r
n
m
e
n
t
(
s
t
a
t
e
)
,
0
o
t
h
e
r
w
i
s
e
;
B
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
B
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
H
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
H
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
s
i
z
e
–
n
a
t
u
r
a
l
l
o
g
a
r
i
t
h
m
o
f
t
o
t
a
l
a
s
s
e
t
s
;
l
e
v
e
r
a
g
e
–
t
o
t
a
l
l
i
a
b
i
l
i
t
i
e
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
N
P
A
T
–
p
r
o
?
t
a
b
i
l
i
t
y
d
e
?
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
n
o
n
-
t
r
a
d
a
b
l
e
–
p
e
r
c
e
n
t
a
g
e
o
f
n
o
n
-
t
r
a
d
a
b
l
e
s
h
a
r
e
s
(
a
s
d
e
s
c
r
i
b
e
d
i
n
f
o
o
t
n
o
t
e
[
1
4
]
i
n
2
0
0
5
t
h
i
s
v
a
r
i
a
b
l
e
m
e
a
s
u
r
e
s
r
e
s
t
r
i
c
t
e
d
t
r
a
d
a
b
l
e
s
h
a
r
e
s
)
;
i
n
s
t
i
t
u
t
i
o
n
a
l
–
p
r
o
p
o
r
t
i
o
n
o
f
t
o
p
2
0
s
h
a
r
e
h
o
l
d
e
r
o
w
n
e
r
s
h
i
p
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
(
i
n
c
l
u
d
e
s
d
o
m
e
s
t
i
c
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
a
n
d
f
o
r
e
i
g
n
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
s
u
c
h
a
s
Q
F
I
I
)
;
f
o
r
e
i
g
n
–
p
e
r
c
e
n
t
a
g
e
f
o
r
e
i
g
n
o
w
n
e
r
s
h
i
p
(
i
n
c
l
u
d
e
s
Q
F
I
I
a
n
d
H
o
n
g
K
o
n
g
i
n
v
e
s
t
o
r
s
)
;
a
u
d
i
t
?
r
m
–
1
i
f
t
h
e
B
i
g
4
a
u
d
i
t
o
r
,
0
o
t
h
e
r
w
i
s
e
;
a
u
d
i
t
c
o
m
m
i
t
t
e
e
–
1
i
f
a
u
d
i
t
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
d
u
a
l
–
e
q
u
a
l
s
1
i
f
t
h
e
c
o
m
p
a
n
y
s
e
p
a
r
a
t
e
d
t
h
e
r
o
l
e
o
f
C
E
O
a
n
d
c
h
a
i
r
m
a
n
b
e
t
w
e
e
n
2
0
0
4
a
n
d
2
0
0
5
,
a
n
d
0
i
f
n
o
c
h
a
n
g
e
,
a
n
d
–
1
i
f
s
e
p
a
r
a
t
i
o
n
i
n
2
0
0
4
w
a
s
r
e
m
o
v
e
d
i
n
2
0
0
5
;
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
–
n
u
m
b
e
r
o
f
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
b
o
a
r
d
s
i
z
e
Table XI.
Pearson correlation
matrix between changes
in independent variables
2004-2005
ARJ
23,1
40
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
These correlations are not unexpected, since the institutional variable is the sum of
domestic and foreign investors and the H-share investors who are from Hong Kong are
treated as part of the foreign investors. No two of these highly correlated variables are
included in the same regression. Therefore, multicollinearity is not a concern in our
analysis.
6.1 DiD regression results
The DiD regressions reported in Table XII show that the reform variable is signi?cant
for mandatory, voluntary and overall disclosure scores between 2004 and 2005,
indicating support for hypotheses H1b and H2b. Change in disclosure is also
signi?cantly negatively associated with the 2004 disclosure score, meaning that
companies with lower 2004 disclosure scores tend to have signi?cantly higher 2004-2005
change scores. However, since none of the other control variables are consistently
signi?cant, the model was re-estimated as follows. The variables potentially violating
the equality of the opening position were retained (ultimate control; non-tradable; and
opening disclosure); while the control variables measuring change between 2004 and
2005, found insigni?cant in the model (1) estimation, were dropped. This model is
expressed as model (2):
DTRANSP
i
¼ d
0
þb
1
DReformed
i
þb
2
Ultimate control04
i
þb
3
Non–tradable04
i
þb
4
Disc04 þDu
i
Model ð2Þ
The model (2) DiD regression results are reported in Table XIII. The reform variable is
signi?cant in each regression, indicating that reformed companies had a greater change
in disclosure scores than unreformed companies did. These results support H1b and
H2b that, as a result of the SEreform, mandatory, voluntary and thus overall disclosures
would increase more for reformed companies than for unreformed companies. Given
that the support for H1b has occurred despite the ceiling effect on the mandatory score in
2004. These results suggest that the reform has had a strong effect, given that the full
effect of the reformis not likely to be fully captured in the ?rst year, as the reformperiod
could take up to three years, and given the SEreform’s share escrowconditions outlined
earlier. Since ultimate control and non-tradable for 2004 show some degree of skewness
(Tables II and IV) we reran model (2) without those variables. Our results are unchanged
from those reported in Table XIII.
The DiD reform coef?cient in model (2) shows that the reform improved the
mandatory disclosure score for reformed companies by 0.4 percent from the 2004 base
year. Hence, if reformed companies on average score 96.4 out of 100 for mandatory
disclosure score in 2004, the score has improved to 96.4 þ (96.4
*
0.004) ¼ 96.786 as a
result of the reform. While such an increase may not seem economically large, it is
constrained by the ceiling effect as the increase is based on the high mandatory
disclosure score of 96.4 in 2004. For voluntary disclosure, the reform coef?cient
indicates that voluntary disclosure improved by 3.5 percent from the 2004 base. In
other words, if the average voluntary disclosure for reformed companies is 67.7 out of
100 in 2004, the reform pushes up the voluntary disclosure in 2005 by 67.7
*
0.035 to
70.07 in 2005. While the increase may not seem economically large, the full effect of the
reform would not be expected to be captured until three years after the reform (in 2008).
These results provide support for H3a, that the increase in mandatory disclosures
would be smaller than the increase in voluntary disclosures for reformed companies.
Split equity
reform
41
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
6.2 Sensitivity analyses
A number of additional tests were performed as sensitivity analyses for potential
confounding factors. This section reports the results.
6.2.1 Early and late reformers. Recall that 37 companies completed the reform by
December 31, 2005, while 21 companies announced the reform before December 31,
2005 and completed it by February 2006. H1b and H2b were re-tested to see whether
Change in mandatory
disclosure
Change in voluntary
disclosure
Change in overall
disclosure
B t Sig. B t Sig. B t Sig.
(Constant) 0.588 8.938 0.000 0.303 4.721 0.000 0.592 8.424 0.000
DReformed 0.008 2.341 0.021 0.045 1.928 0.057 0.012 2.246 0.027
DUltimate control 0.001 0.109 0.913 0.032 0.857 0.394 0.006 0.634 0.528
DSize 20.010 21.062 0.291 0.123 1.895 0.061 0.021 1.400 0.156
DLeverage 0.024 1.001 0.319 20.228 21.331 0.186 20.033 20.829 0.409
DNPAT 0.041 1.473 0.144 20.059 20.288 0.774 0.025 0.534 0.595
DNon-tradable 0.033 1.478 0.142 0.084 0.522 0.603 0.043 1.154 0.251
DForeign 0.043 0.498 0.619 20.049 20.079 0.937 0.071 0.495 0.622
DAudit committee 0.002 0.269 0.788 0.007 0.138 0.890 0.003 0.282 0.779
DRemuneration committee 20.004 20.990 0.324 20.034 21.097 0.275 20.010 21.450 0.150
DNomination committee 0.004 0.484 0.629 0.027 0.446 0.657 0.009 0.679 0.499
DDual 0.009 2.386 0.019 0.014 0.508 0.613 0.011 1.759 0.082
Disc_04 20.602 28.953 0.000 20.478 25.984 0.000 20.646 28.566 0.000
Non-tradable 04 0.000 20.065 0.948 20.007 20.150 0.881 20.003 20.241 0.810
Ultimate control 04 0.000 0.025 0.980 0.012 0.598 0.551 0.004 0.945 0.347
F-statistic 8.650 3.703 7.226
Sig. of F 0.000 0.000 0.000
Adjusted R
2
0.482 0.248 0.431
Notes: Mandatory checklist items required by Chinese GAAP; voluntary checklist items not required
by Chinese GAAP; overall combined mandatory and voluntary score; for the following variables D
indicates change from 2004 to 2005; ultimate control 1 if the ultimate controller is government (state), 0
otherwise; DReformed
i
is a variable coded 1 for the reformed group as they change from 0 in 2004 (pre-
reform) to 1 in 2005 (post-reform) while the change in score for the unreformed group is 0 to 0 from
2004-2005; DUltimate control
i
measures the change in the ultimate controller of the company between
2004 and 2005; it is assigned 1 if the ultimate controller changes from privately controlled to
government (state) controlled, and 0 otherwise; DSize
i
measures the change in company size between
2004 and 2005; size is measured by the natural logarithm of total assets of the company; DLeverage
i
measures the change in capital structure between 2004 and 2005; this is proxied by total liabilities
divided by total assets (TL/TA); DNPAT
i
measures the change in pro?tability de?ated by the amount
of total assets between 2004 and 2005; DNon-tradable
i
measures the change in the percentage of non-
tradable shares between 2004 and 2005 (as described in note 15 in 2005 this variable measures
restricted tradable shares); DForeign
i
measures the change in percentage foreign ownership between
2004 and 2005; foreign investors include QFII and Hong Kong investors as reported in annual reports;
DAudit committee
i
equals 1 if the company set up an audit committee between 2004 and 2005, and 0
otherwise; DRemuneration committee
i
equals 1 if the company set up a remuneration committee
between 2004 and 2005, and 0 otherwise; DNomination committee
i
equals 1 if the company set up a
nomination committee between 2004 and 2005, and 0 otherwise; DDual
i
equals 1 if the company
separated the role of CEO and chairman between 2004 and 2005, and 0 if no change, and 21 if
separation in 2004 was removed in 2005; Disc_04 is the disclosure score for 2004; non-tradable 04 is the
percentage of tradable shares in 2004; ultimate control 04 is 1 if the ultimate controller in 2004 is
government (state), 0 otherwise
Table XII.
DiD regressions on
disclosure between 2004
and 2005 – model (1)
ARJ
23,1
42
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the impact of the reform on these two sub-samples is the same. Dividing the sample in
two considerably weakens the power of the tests and the results. For both sub-samples,
the models (1) and (2) regression results for the reform variable are consistent in
direction with the combined sample results for all disclosure measures. However, the
reform variable is only signi?cant in the voluntary disclosures regression for model (2)
of both sub-samples and weakly signi?cant.
6.2.2 Companies selected by the government. Recall that the reformwas completed in
batches, with 11 companies in the ?rst set of reformed companies selected by the
government to pilot the reform, and the remaining 47 companies self-selecting to reform.
A concern is that the ?rst 11 reformed companies might cover some of the “best” stocks
listed on the Shanghai Stock Exchange, and therefore the impact of the reform on these
11 companies could be greater than on the rest of the reformed companies. Thus, these 11
early reformed companies were dropped from the sample (together with their 11
matched controls) to see if the improved disclosure effect of the reform still holds for the
remaining 47 reformed companies. The models (1) and (2) results with the reduced
sample are very similar to those of the full sample.
7. Conclusions
The SE structure was seen as a major obstacle hindering the growth and stability of the
Chinese share market. On April 29, 2005, the Chinese Government launched the SE
reform to address this problem. One of the aims of this reform was to improve corporate
governance and thereby to improve corporate ?nancial transparency. This study
examines the impact of the SE reform on corporate ?nancial transparency in China,
investigating how the reform affected post reform transparency of companies that
commenced the reform (reformed G-share companies) before December 31, 2005.
Mandatory disclosure Voluntary disclosure Overall disclosure
B t Sig. B t Sig. B t Sig.
(Constant) 0.578 9.051 0.000 0.286 4.594 0.000 0.574 8.222 0.000
DReformed 0.004 2.155 0.033 0.035 2.479 0.015 0.008 2.320 0.022
DDual 0.009 2.615 0.010
Disc_04 20.593 29.064 0.000 20.470 26.023 0.000 20.631 28.399 0.000
Non-tradable 04 20.001 20.162 0.872 0.022 0.490 0.625 0.004 0.341 0.734
Ultimate control 04 20.001 20.374 0.709 0.009 0.506 0.614 0.003 0.640 0.524
F-statistic 23.937 11.726 22.411
Sig. of F 0.000 0.000 0.000
Adjusted R
2
0.499 0.272 0.427
Notes: Mandatory checklist items required by Chinese GAAP; voluntary checklist items not required
by Chinese GAAP; overall combined mandatory and voluntary score; for the following variables
D indicates change from 2004 to 2005; ultimate control 1 if the ultimate controller is government (state),
0 otherwise; DReformed
i
is a variable coded 1 for the reformed group as they change from 0 in 2004
(pre-reform) to 1 in 2005 (post-reform) while the change in score for the unreformed group is 0 to 0 from
2004-2005; DDual
i
equals 1 if the company separated the role of CEO and chairman between 2004 and
2005, and 0 if no change, and 21 if separation in 2004 was removed in 2005; Disc_04 is the disclosure
score for 2004; non-tradable 04 is the percentage of tradable shares in 2004; ultimate control 04 is 1 if
the ultimate controller in 2004 is government (state), 0 otherwise
Table XIII.
DiD regressions on
disclosure between 2004
and 2005 – model (2)
Split equity
reform
43
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
It was expected that the SE reform would improve the transparency of the reformed
companies after the reform(in 2005) compared to their pre-reformtransparency (in 2004)
and compared to a control group of unreformed companies in 2005. Transparency was
measured using a 270-item checklist based on IFRS but designed speci?cally for the
Chinese environment. Three transparency scores were derived from the checklist:
(1) a mandatory transparency score, based on the level of disclosure of items
required by Chinese GAAP in 2004 and 2005;
(2) a voluntary transparency score, based on the level of disclosure of items
required only by IFRS in 2004 and 2005; and
(3) an overall transparency score, based on the results of the mandatory and
voluntary transparency scores in 2004 and 2005.
The results of the studysupport the positive impact of the SEreformonthe transparency
of reformed companies. In the post-reform 2005 annual reports, mandatory voluntary
and overall disclosures of reformed companies improved compared to their pre-reform
levels and also compared to those of a matched group of unreformed companies.
Voluntary disclosures increased more than mandatory disclosures, but the mandatory
disclosures still increased signi?cantly despite the ceiling effect on them caused by the
disclosure templates in Chinese GAAP.
A limitation of our study is sample size as we focus on reformed companies listed on
the CSI 300 index. Although we include all such companies plus controls, the resulting
sample is relatively small (n ¼ 116). However, our primary focus is on the effect of the
SE reform, and since we ?nd the reform variable is signi?cant in all regression models,
our sample size is suf?ciently large to detect that effect.
Our study is also limited to companies that commenced the SE reform before
December 31, 2005 and examines the immediate impact of the reform on their corporate
?nancial transparency. Future research could also examine the reform’s long-term
effect on corporate ?nancial transparency in China. The impact of the reform is
expected to be greater over the three years after 2005 especially as the share escrow on
controlling shareholders was lifted by 2008. Therefore, it would be interesting to see if
the practical signi?cance of the reform becomes greater over those latter years on
reformed companies. Moreover, it would be interesting to examine whether the reform
has facilitated other important changes for Chinese companies in areas such as
improved corporate governance, and the attraction of more institutional investors,
especially QFII, to invest in the reformed companies. In addition, consistent with other
disclosure studies, our study only examines changes to the quantity of disclosures and
not the quality of these disclosures. Examining the quality of increased disclosures
would be a fruitful area for future research.
Notes
1. The report by Deloitte Touche Tohmatsu (2007) “A Comparison with current PRC GAAP and
IFRS”, compares the 2005 PRC GAAP, the (then) new2006 Accounting Standards to be applied
fromJanuary 1, 2007 and IFRS: (available at: www.iasplus.com/country/china.htm). There was
no change in the Chinese GAAP between December 31, 2004 and December 31, 2005.
2. The state-related shares also include legal-person shares that are owned by domestic
companies, but these domestic companies are typically other SOEs. Hence, the distinction
between state and legal shares is often just in name.
ARJ
23,1
44
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3. The SE reform attempts in 1999 and 2001 were quickly cancelled due to the negative market
reaction, caused possibly by the fear that a sudden ?oating of a huge number of non-tradable
shares would substantially affect the supplyanddemand balance, leadingto animmediate drop
in share price (Inoue, 2005). Both reform attempts were followed by bear market periods.
4. There is a debate in the corporate ?nancial transparency and corporate governance literature
whether the former is considered a subset of the latter or vice versa (Bai et al., 2004; Bushman
and Smith, 2001). This study treats corporate ?nancial transparency as part of corporate
governance, consistent with previous studies in this region (Kim, 2005; Susilowati, 2008).
5. Note that since these studies are restricted to large companies, the results may not be
representative of all Chinese companies. The current study addresses this through a more
representative sample of companies.
6. The presence of the templates is so “encompassing” their effects also appear in the corporate
governance section and directors’ report in annual reports where each company has the same
format in terms of disclosure and discussion on corporate governance mechanisms and
management discussions. The accounting model then adopted is called “fund accounting”
which is entirely rule-based meaning it has “an entire of code of extremely speci?c
accounting treatments developed to meet nearly every situation like the templates, thus, all
accountants need to do is just to ‘tick in the box’” (Walter and Howie, 2006).
7. Effect size refers to the impact that one variable has on another, and is measured using a
standardized variable such as the Pearson correlation coef?cient. Cohen (1988, 1992)
suggests that X has a large effect on Y if the Pearson correlation between them is at least 0.5.
8. Of the companies, 11 were chosen by the government as the ?rst two pilot reform batches,
the remaining 26 companies self-selected into the reform process. Sensitivity analyses
reported in the results section address potential differences between these two sub-samples.
9. The ?nancial year in China is from January 1 to December 31, with listed companies required
to submit their annual report between March and April in the following year. Therefore, the
second group of companies would still have incentives to improve disclosure in their 2005
annual report. Even so, it is possible that there may be a time lag effect for improved
disclosure after the reform, and if this were the case, the lag could be greater for the group of
companies completing the reform in February 2006 despite announcing the reform plan in
December 2005. Sensitivity analyses reported in the results section address this possibility.
10. These disclosure requirements are more comprehensive than accounting standards and are
applicable to all companies, not just listed companies.
11. Chinese companies issuing H-shares listed in Hong Kong are required to prepare ?nancial
statements using either IFRS or Hong Kong GAAP for their Hong Kong shareholders. Our
study focuses on the ?nancial statements (in Chinese) prepared for domestic Chinese
shareholders. For that purpose, checklist items not required by Chinese GAAP are
considered voluntary. Only six companies (two reformed, four unreformed) in the sample
have H-shares. Thus, the in?uence on our sample of the IFRS reporting requirement for
H-share issuers is minimal.
12. The equation is referred to as a ?rst-differenced equation. The strength of the equation is
that it overcomes “heterogeneity biases” (Wooldridge, 2006), because any omitted variable
problem or unobserved effect a
i
that can impact transparency scores remain constant or
?xed given the same companies are followed in two periods (2004 and 2005 in this study).
By taking the difference across two years, any omitted variable or unobserved effect is
cancelled out. It has been noted that the DiD methodology results in low R
2
(Wooldridge,
2006).
Split equity
reform
45
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
13. Small deviations from parallel assumptions are tolerated since resulting biases would be
small (Wooldridge, 2006).
14. After the 2005 reform, there is no distinction between non-tradable and tradable shares for
the reformed companies. However, since the previously non-tradable shareholders still face
temporary lockup on their shareholdings (share escrow), the non-tradable share category in
2005 is referred to as restricted tradable shares.
15. As discussedinthe results section, when this model was estimated, none of the control variables
measuring change between 2004 and 2005 were found to be signi?cant. Since these variables do
not account for the variation in disclosure between 2004 and 2005, they were dropped from the
model. The ?nal DiD model estimated (model (2)) is reported in the results section.
References
Ang, J.S. and Ma, Y.L. (1999), “Transparency in Chinese stocks: a study of earnings forecasts by
professional analysts”, Paci?c-Basin Finance Journal, Vol. 7 No. 2, pp. 129-55.
Ashbaugh, H. and Pincus, M. (2001), “Domestic accounting standards, international accounting
standards, and predictability of earnings”, Journal of Accounting Research, Vol. 39,
pp. 417-34.
Bai, C.E., Liu, Q., Lu, J., Song, F.M. and Zhang, J. (2004), “Corporate governance and market
valuation in China”, Journal of Comparative Economics, Vol. 32, pp. 599-616.
Botosan, C.A. (1997), “Disclosure level and the cost of equity capital”, The Accounting Review,
Vol. 72 No. 3, pp. 323-49.
Botosan, C.A. and Harris, M. (2000), “Motivations for a change in disclosure frequency and its
consequences: an examination of voluntary quarterly segment disclosures”, Journal of
Accounting Research, Vol. 38 No. 2, pp. 329-53.
Botosan, C.A. and Plumlee, M. (2002), “A re-examination of disclosure level and the expected cost
of equity capital”, Journal of Accounting Research, Vol. 40 No. 1, pp. 21-40.
Bushman, R.M. and Smith, A.J. (2001), “Financial accounting information and corporate
governance”, Journal of Accounting and Economics, Vol. 32, pp. 237-333.
Bushman, R.M. and Smith, A.J. (2003), “Transparency, ?nancial accounting information and
corporate governance”, FBRNY Economic Policy Review, April, pp. 65-87.
Bushman, R.M., Piotroski, J. and Smith, A.J. (2004), “What determines corporate transparency?”,
Journal of Accounting Research, Vol. 42 No. 2, pp. 207-52.
Chau, G.K. and Gray, S.J. (2002), “Ownership structure and corporate voluntary disclosure in
Hong Kong and Singapore”, The International Journal of Accounting, Vol. 37, pp. 247-65.
Chen, J.P., Chen, S.M. and Su, X.J. (2001), “Is accounting information value-relevant in the
emerging Chinese stock market?”, Journal of International Accounting, Auditing and
Taxation, Vol. 10 No. 1, pp. 1-22.
Chow, C.W. and Wong-Boren, A. (1987), “Voluntary ?nancial disclosure by Mexican
corporations”, The Accounting Review, Vol. 62 No. 3, pp. 533-41.
Cohen, J. (1988), Statistical Power Analysis for the Behavioural Sciences, 2nd ed., Academic Press,
New York, NY.
Cohen, J. (1992), “A power primer”, Psychological Bulletin, Vol. 112 No. 1, pp. 155-9.
DeFond, M.L., Wong, T.J. and Li, S. (2000), “The impact of improved auditor independence on
audit market concentration in China”, Journal of Accounting and Economics, Vol. 28,
pp. 269-305.
ARJ
23,1
46
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Deloitte Touche Tohmatsu (2001), “International accounting standards presentation and
disclosure checklist”, available at: www.iasplus.com/fs/iaschk01.pdf
Eng, L.L. and Mak, Y.T. (2003), “Corporate governance and voluntary disclosure”, Journal of
Accounting & Public Policy, Vol. 22 No. 4, pp. 325-45.
Enriques, L. and Volpin, P.F. (2007), “Corporate governance reforms in continental Europe”,
Journal of Economic Perspectives, Vol. 21 No. 1, pp. 117-40.
Feng, L.C. and Xu, W.H. (2007), “Has the reform of non-tradable shares raised prices?”, Emerging
Markets Finance and Trade, Vol. 43 No. 2, pp. 33-62.
Field, A. (2009), Discovering Statistics Using SPSS, 3rd ed., Sage, London.
Haw, I.M., Qi, D. and Wu, W. (2001), “The nature of information in accruals and cash ?ows in an
emerging capital market: the case of China”, The International Journal of Accounting,
Vol. 36, pp. 391-406.
Ho, S.S.M. and Wong, K.S. (2001), “A study of the relationship between corporate governance
structures and the extent of voluntary disclosure”, Journal of International Accounting,
Auditing and Taxation, Vol. 10, pp. 139-56.
Inoue, T. (2005), “Reform of China’s split-share structure takes shape”, Nomura Capital Market
Review, Vol. 8 No. 3, pp. 2-21.
Jian, M. and Wong, T.J. (2003), “Earnings management and tunneling through related party
transactions: evidence from Chinese corporate groups”, working paper, The Hong Kong
University of Science and Technology, Hong Kong.
Kim, T.E. (2005), “Determinants of corporate ?nancial reporting transparency: evidence from
South Korea”, thesis, University of New South Wales, Sydney.
Lin, T. (2001), “Corporatisation and corporate governance in China’s economic transition”,
Economics of Planning, Vol. 34, pp. 5-35.
Liu, Q. and Lu, Z.J. (2003), “Earnings management to tunnel: evidence from China’s listed
companies”, working paper, University of Hong Kong, Hong Kong.
Morris, R.D. and Gray, S.J. (2007), “Corporate transparency differences in the Asian region: the
feasibility of global standardization in practice?”, paper presented at the AFAANZ
Conference, Gold Coast, July.
Morris, R.D., Leung, Q. and Gray, S.J. (2006), “Corporate transparency in China: factors in?uencing
?nancial disclosure levels”, paper presented at the American Accounting Association
International Accounting Section Midyear Meeting, Los Angeles, CA, January 13.
Morris, R.D., Pham, T. and Gray, S.J. (2009), “Corporate governance and transparency in
Malaysia: before and after the Asian ?nancial crisis”, in Baxter, J. and Poullaos, C. (Eds),
Practices, Profession and Pedagogy in Accounting – Essays in Honour of Bill Birkett,
Chapter 5, Sydney University Press, Sydney.
Morris, R.D., Ho, B.U.S., Pham, T. and Gray, S.J. (2004), “Financial reporting practices of
Indonesian companies before and after the Asian ?nancial crisis”, Asia-Paci?c Journal of
Accounting and Economics, Vol. 11 No. 2, pp. 1-27.
Ruland, W., Tung, S. and George, N.E. (1990), “Factors associated with the disclosure of
managers’ forecasts”, The Accounting Review, Vol. 65 No. 3, pp. 710-21.
Spence, M. (1973), “Job marketing signaling: the informational structure of job markets and
related phenomena”, Quarterly Journal of Economics, Vol. 87, pp. 355-74.
Susilowati, I. (2008), International Differences in Corporate Governance and Financial
Transparency: A Comparative Study of In?uential Factors in Australia, India, Indonesia
and Japan, VDM, Saarbru¨cken.
Split equity
reform
47
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Tower, G. and Yan, Z.H. (2006), “Asset impairment disclosure in China”, working paper, Curtin
University of Technology, Perth.
Walter, C.E. and Howie, F.J.T. (2006), Privatizing China: The Stock Markets and their Role in
Corporate Reform, 2nd ed., Wiley, Singapore.
Wang, Q.T., Wong, J. and Xia, L.J. (2007), “State ownership, institutional environment and auditor
choice: evidence fromChina”, working paper, Chinese University of Hong Kong, Hong Kong.
Watts, R. and Zimmerman, J. (1986), Positive Accounting Theory, Prentice-Hall, Englewood Cliffs, NJ.
Wooldridge, J.M. (2006), Introductory Econometrics, 3rd ed., Thomson South Western, Mason, OH.
Xiao, J., Weetman, P. and Sun, M. (2004a), “Political in?uence and coexistence of a uniform
accounting system and accounting standards: recent developments in China”, Abacus,
Vol. 40 No. 2, pp. 193-218.
Xiao, J., Yang, H. and Chow, C. (2004b), “The determinants and characteristics of voluntary
internet-based disclosures by Chinese listed companies”, Journal of Accounting & Public
Policy, Vol. 23 No. 3, pp. 191-225.
Xiao, Z. (1999), “Corporate disclosure made by Chinese listed companies”, The International
Journal of Accounting, Vol. 34 No. 3, pp. 349-73.
About the authors
Wendy Green is a Senior Lecturer at the University of New South Wales who teaches
Auditing and Assurance Services at the undergraduate and postgraduate levels. Wendy Green
researches in the discipline of auditing. Wendy Green’s research to date has focused on factors
affecting audit quality and includes work on audit quali?cations in Australia and in emerging
markets, auditor switching, audit committees, auditor judgment in an analytical procedures
setting, and auditor specialisation. This work has appeared in both Australian and international
refereed journals and has been included in refereed conference proceedings both in Australia and
internationally. More recently, Wendy Green’s research has focused on informing the regulation
of auditing and assurance in the areas of partner rotation and the assurance of carbon emissions.
Wendy Green is the corresponding author and can be contacted at: [email protected]
Richard D. Morris is an Associate Professor at the University of New South Wales. His
research interests are in ?nancial accounting, international accounting, and accounting history.
They include the impact of IFRS adoption in Australia and the EU, international harmonisation
of ?nancial reporting, corporate transparency in the Asia Paci?c region, ?nancial reporting,
and the Asian ?nancial crisis of 1997-1998, the evolution of ?nancial reporting in Britain and
Australia during the nineteenth century, and the economic factors associated with accounting
choices made by ?rms. His work has appeared in international refereed journals, professional
journals, monographs, and chapters in books. Richard D. Morris’s teaching responsibilities are in
?nancial accounting at undergraduate and postgraduate levels.
Haiping Tang was a research student in the School of Accounting, University of New South
Wales, Australia.
ARJ
23,1
48
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Xiaoyan Chen, Xin Ling. 2015. Determinants of Chinese equity financing behaviours: traditional model
and the alternatives. Accounting & Finance n/a-n/a. [CrossRef]
2. Xiaoyan Chen. 2015. Improved corporate governance and Chinese seasoned equity offering announcement
effects. Accounting & Finance n/a-n/a. [CrossRef]
3. Haiyan Jiang, Ahsan Habib. 2012. Split-share reform and earnings management: Evidence from China.
Advances in Accounting 28, 120-127. [CrossRef]
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_916368570.pdf
The purpose of this paper is to report the impact of the Chinese capital market split equity
(SE) reform in 2005 on the corporate financial transparency of Chinese listed companies
Accounting Research Journal
The split equity reform and corporate financial transparency in China
Wendy Green Richard D. Morris Haiping Tang
Article information:
To cite this document:
Wendy Green Richard D. Morris Haiping Tang, (2010),"The split equity reform and corporate financial
transparency in China", Accounting Research J ournal, Vol. 23 Iss 1 pp. 20 - 48
Permanent link to this document:
http://dx.doi.org/10.1108/10309611011060515
Downloaded on: 24 January 2016, At: 21:10 (PT)
References: this document contains references to 42 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 930 times since 2010*
Users who downloaded this article also downloaded:
Raghavan J . Iyengar, J udy Land, Ernest M. Zampelli, (2010),"Does board governance improve
the quality of accounting earnings?", Accounting Research J ournal, Vol. 23 Iss 1 pp. 49-68 http://
dx.doi.org/10.1108/10309611011060524
Katherine Uylangco, Steve Easton, Robert Faff, (2010),"The equity and efficiency of the Australian share
market with respect to director trading", Accounting Research J ournal, Vol. 23 Iss 1 pp. 5-19 http://
dx.doi.org/10.1108/10309611011060506
Minna Yu, (2010),"Analyst following and corporate governance: emerging-market evidence", Accounting
Research J ournal, Vol. 23 Iss 1 pp. 69-93 http://dx.doi.org/10.1108/10309611011060533
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The split equity reform and
corporate ?nancial transparency
in China
Wendy Green and Richard D. Morris
School of Accounting, University of New South Wales, Sydney, Australia, and
Haiping Tang
Pricewaterhousecoopers, Sydney, Australia
Abstract
Purpose – The purpose of this paper is to report the impact of the Chinese capital market split equity
(SE) reform in 2005 on the corporate ?nancial transparency of Chinese listed companies.
Design/methodology/approach – Using an International Financial Reporting Standards-based
checklist, the paper investigates whether the post-reform 2005 annual reports of reformed companies
improved transparency compared to pre-reform 2004 reports. The transparency of the reformed
companies was also compared to a control group of companies unreformed on December 31, 2005.
Findings – Results indicate that the SE reform increased corporate disclosures. Reformed companies
had higher mandatory and voluntary disclosures in their post-reform 2005 annual reports compared to
their pre-reform 2004 annual reports. In addition, the improvement in mandatory and voluntary
disclosures for reformed companies is greater than that of the unreformed control group.
Research limitations/implications – The SE reform provides a unique natural experimental
setting in which to examine the impact of the SE reform, with its associated change in ownership
structure and corporate governance, on corporate disclosure.
Practical implications – The results of this paper suggest that the SE reform has had a positive
effect on corporate ?nancial transparency in China, thereby indicating the positive response to
regulation in this emerging market. Further, the results suggest that as the proportion of government
ownership falls, management has increased incentive to voluntarily supply additional information to
the market.
Originality/value – The SE reform is unique to China and this paper is the ?rst to report on
?nancial reporting disclosure implications of this reform.
Keywords Equity capital, Disclosure, China
Paper type Research paper
1. Introduction
In 1978, the Chinese Government initiated their “economic reform and opening up”
policy aimed at attracting foreign investments and encouraging experiments in
enterprise autonomy. Since then, the Chinese economy has experienced strong growth
with gross domestic product averaging 10 percent per annum. In contrast, the growth
of the domestic share market has been much more volatile. For example, between
2001 and 2005, the benchmark Shanghai Composite Index fell more than 50 percent
from its peak even though the domestic economy was strong. The existence of the split
equity (SE) structure in the Chinese share market is said to have been a major hurdle to
share market development (Ang and Ma, 1999; Chen et al., 2001; Feng and Xu, 2007).
The SE structure, established in the early 1990s upon the formation of the Shanghai
and Shenzhen Stock Exchanges, was intended to help state-owned enterprises (SOEs)
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
ARJ
23,1
20
Accounting Research Journal
Vol. 23 No. 1, 2010
pp. 20-48
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611011060515
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
raise ?nance, other than by bank borrowing (Walter and Howie, 2006), without the
state losing control over key assets. Under this structure, only one-third of the shares of
listed companies were tradable on public markets, with the remaining two-thirds being
non-tradable. In 2005, non-tradable shares accounted for more than 60 percent of all
issued shares, while tradable shares accounted for less than 40 percent of total shares
(Feng and Xu, 2007).
On April 29, 2005, the China Securities Regulatory Commission (CSRC), the national
capital market regulator, launched the SE reform. Its aim was to ?oat previously
non-tradable shares into the market, allowing a shift in the balance of share ownership
to public ownership by institutional investors and existing tradable shareholders,
while in some cases still retaining state control. The importance of this reform can be
seen in the comments of Shan Fuling, CSRC Chairman, who stated in May 2005 that
“The Split Equity reform marks the most fundamental reform ever undertaken in the
Chinese capital market and it has huge implications for the market in the long run.”
He further noted that “One of the aims of the Split Equity reform is to improve
corporate governance in China and to enhance the transparency of listed companies.”
The SE structure was a major corporate governance device unique to China.
Whether its abolition had the intended effect of increasing transparency of listed
companies is therefore of interest to both regulators and the academic community.
The sheer volume of previously non-tradable shares becoming tradable after the
reform, together with the implications for capital markets and demand for information,
warrants an investigation of that change. There is unlikely to be any other setting
where there has been such a potentially large transfer of company shares from public
to private ownership in such a short period of time. The Chinese Government’s decision
to conduct the reform in batches of companies, with the ?rst stage completed by
December 31, 2005, and the fact that by 2007 around 95 percent of the listed companies
on the Shanghai and Shenzhen stock exchanges had completed the SE reform, provides
a natural experimental setting of short duration to examine the impact of the reform on
corporate ?nancial transparency. Speci?cally, our study investigates whether the
reform, and the shift in ownership structure it entails, results in an improvement in
transparency for the ?rst stage reformed companies in their post-reform 2005 annual
reports when compared to their pre-reform 2004 reports, and also when compared to a
matched control group of then (in 2005) unreformed companies.
Transparency is measured by adapting to the Chinese environment the
International Financial Reporting Standards (IFRS) disclosure checklist used by
Morris and Gray (2007). Since Chinese generally accepted accounting principles
(GAAP) provided the mandated reporting requirements during the period covered by
this study, and since these requirements were not identical to IFRS, three transparency
scores are calculated:
(1) a mandatory transparency score, based on the disclosure of items required by
both IFRS and Chinese GAAP in 2004 and 2005[1];
(2) a voluntary transparency score, based on the level of disclosure of items
covered only by IFRS in 2004 and 2005; and
(3) an overall score for mandatory and voluntary transparency in 2004 and 2005.
Our results indicate that the SE reform had a positive impact on the transparency of
the reformed companies. In the post-reform 2005 annual reports, both mandatory and
Split equity
reform
21
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
voluntary disclosures of reformed companies improved compared to their pre-reform
2004 reports. The improvement in the reformed company disclosures was also greater
than that of the matched control group of companies that had not commenced the
reform by December 31, 2005.
The study makes several contributions. First, it adds to the literature by linking the
impact of the SE reform with corporate disclosure in China, thereby showcasing
the unique institutional setting found in China. Second, given the changes to corporate
ownership following the reform, the study also provides insights into the effect of
ownership structure changes, particularly decreased government ownership, on
corporate disclosure in a natural experimental setting not otherwise available. Third,
the study provides evidence that the high levels of mandatory disclosure observed in
China, despite the presence of factors such as high government ownership that are
normally indicative of low disclosure, can be explained by the presence of numerous
templates contained in Chinese GAAP. Finally, the ?nding of a positive impact of the
SE reform on corporate ?nancial transparency in China has practical implications for
Chinese policymakers. The regulators claimed that one of the aims of the SE reform
was to improve corporate governance in China, and ultimately to enhance the
transparency of listed companies. Our study suggests that the reform has indeed
achieved this aim through improved transparency in the reformed companies.
2. The Chinese institutional environment
China provides a unique institutional setting in which to examine the effects of corporate
governance reforms. Prior research has noted that an understanding of the speci?c
institutional setting is vital in improving corporate governance and corporate disclosure
(Enriques and Volpin, 2007). China has traditionally been characterized by strong
government ownership, along with limited trading of certain share classes because of
the SE system. Reform of the SE system provides an opportunity to study the impact of
the move fromprimarily government ownership to more diverse ownership on one of the
basic components of corporate governance: corporate ?nancial transparency.
2.1 The SE structure in China
The SE structure divided the market into two segments: tradable and non-tradable
shares. Tradable shares held by non-state investors made up one-third of the shares of
listed companies while the remaining two thirds of the listed shares were non-tradable
shares typically held by the state[2]. Tradable shares comprise A-, B-, and H-shares.
A-shares are traded in Chinese currency (RMB) and were initially only open to domestic
investors in the Shanghai and Shenzhen Stock Exchanges. From 2003, quali?ed foreign
institutional investors (QFII) could also invest in A-shares. B-shares are US dollar
denominated and were originally intended only for foreign investors, but from 2001,
domestic investors could also invest in B-shares using US dollars. H-shares are shares
issued by Chinese companies that are dual listed on the Hong Kong Stock Exchange,
traded in Hong Kong dollars; they are also traded in Chinese RMB currency on the
Shanghai Stock Exchange.
Although the SE structure resolves the state’s concerns about loss of control,
problems arose as public markets grew. In particular, it was dif?cult for holders of
non-tradable shares to dispose of their shares. Non-tradable shares could only be
transferred between non-tradable shareholders, but any transfer of state-owned shares
ARJ
23,1
22
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
required government approval and the transfer price could not be below net book value
per share, to prevent the “loss” of state assets. Such treatment could bear little
relationship to the value the share market might place on a company (Walter and Howie,
2006) and thereby preventing holders of non-tradable shares from directly bene?tting
from improvements in corporate performance. In short, the market mechanism was
distorted.
Also, the SE structure created a divergence of interests between non-tradable and
tradable shareholders. While tradable shareholders’ interests were aligned with the
company’s share price, the latter remained of little interest to non-tradable shareholders.
Hence, the SE structure created incentives for controlling shareholders to seek private
gains through transactions that were often at the cost of tradable shareholders ( Jian and
Wong, 2003; Liu and Lu, 2003). Further, with a huge number of non-tradable shares
controlled by the state or controlling shareholders, the tradable shareholders were
minority shareholders. Thus, market disciplining mechanisms, including the threat of
takeover by another company, did not fully operate in the SE structure.
In order to address these issues, the CSRC launched the SE reform on April 29, 2005.
The aim of this reform was to ?oat the previously non-tradable shares into the market,
allowing a shift in the balance of share ownership from the government to public
ownership by institutional investors and existing tradable shareholders. The reform
was conducted in batches of companies, with the ?rst batch completed by December
31, 2005. Once companies completed the reform, they were referred to as reformed
G-share companies to distinguish them from unreformed companies (G stands for
reformed in Chinese Pingying).
The 2005 SE reform contained various features to ensure its success. First, in order to
protect the rights of existing tradable shareholders, the CSRC required non-tradable
shareholders to pay compensationto existingtradable shareholders to offset anypotential
loss that the latter would incur due to the ?oating of a potentially large number of
non-tradable shares. Second, the CSRCimposeda temporary lock-up, or sellingrestriction,
of up to three years on non-tradable shareholders to prevent the large increase in the
number of non-tradable shares causing a sharp decline in share price (Inoue, 2005).
Non-tradable shareholders could not sell more than 5 percent of the companies’ shares in
the ?rst six months after the reform; no more than 10 percent in the following 12 months,
with the restriction lifted from the third year onwards. Third, a ?exible approach was
adopted with regard to the reform proposal (Inoue, 2005). For example, instead of a
“one-size-?ts-all” approach imposed by the government, companies were able to design
their own compensation proposals (such as additional shares, cash or European warrants)
through negotiation with existing tradable shareholders; these were only approved once
there was agreement bymore than two-thirds of the existing tradable shareholders (Inoue,
2005). Fourth, after the initial pilot companies selected by the CSRC had completed the
reformprocess, the reformcontinued inbatches, with companies self-selecting themselves
into the reformprocess. Finally, while unsuccessful SE reformattempts in 1999 and 2001
took place in bull markets[3], the SE reform in 2005 was introduced when the market had
already hit a low point with the Shanghai Benchmark Index falling below 1,000 points
fromits peakof 2,000 points in2001. Giventhat the market was alreadythought to be at its
lowest, it was suggested that the reformwas less likely to lead to a further decline in share
prices due to this perceived “?oor effect.”
Split equity
reform
23
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
A number of other regulatory changes took place either before or immediately after
the SE reform in 2005 that also have the potential to improve ?nancial reporting
transparency in China. These changes involved improved corporate governance
requirements and the opening up of domestic A-shares to QFII, domestic institutional
investors and foreign strategic investors. These changes are outlined in Table I and
their potential effects are addressed in our model examining transparency.
3. The concept of transparency
Since the 1997 Asian ?nancial crisis, improving transparency to enhance investors’
con?dence has been one of the priority tasks for policy makers in the Asian region. More
recent corporate collapses in the USA and Australia have also emphasized to stakeholders
the importance of transparency. However, despite its wide usage, there is no consensus on
the de?nition of transparency. Transparency can exist at two levels: the corporate and the
country levels. Bushman et al. (2004) conceptualise transparency at the country level as the
joint output of a multi-faceted system whose components collectively produce, gather,
validate, and disseminate information to market participants. Bushman and Smith (2003)
de?ne transparency at the company level as the widespread availability of
company-speci?c information concerning publicly listed companies to those outside the
companies. Our focus is on corporate ?nancial transparency, that is, transparency at the
corporate level.
Compliance with accounting standards is commonly used to measure ?nancial
transparency at the corporate level, based on the premise that such compliance is
a signi?cant element in ?nancial reporting quality (Susilowati, 2008). Examples
Regulatory change Description
Corporate governance regulation The CSRC required all listed companies to comply with a code of
corporate governance by the end of 2003 which required: at least
two independent directors; an audit committee with at least
one-third independent directors; and separation of roles of CEO
and Chairman
a
. Remuneration and nomination committees were
also recommended
In December 2006, the CSRC further required independent
directors to have relevant ?nancial literacy and possess
necessary quali?cations
QFII During 2002, foreigners were able to invest for the ?rst time in
the domestic A-shares market directly rather than through the
B-share market
Domestic institutional investor
growth
The domestic institutional investor base was expanded in
November 2002 to allow domestic insurance companies to invest
in domestic A-shares without restriction; with social securities
and pension funds were approved in 2004
Foreign strategic investors In January 2006, the CSRC allowed selected foreign strategic
investors to acquire strategic stakes (a minimum of 10 percent of
outstanding shares) in reformed G share companies for at least
three years
Note:
a
These requirements were enhanced on December 8, 2006, with the CSRC requiring that the
independent directors appointed must also have demonstrated relevant ?nancial literacy and possess
the necessary quali?cations
Table I.
Regulatory changes in
China potentially
improving corporate
disclosure
ARJ
23,1
24
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
include: compliance with domestic accounting standards (Morris et al., 2004, 2009);
voluntary compliance with IFRS (Ashbaugh and Pincus, 2001; Ho and Wong,
2001; Morris et al., 2004, 2009); and voluntary compliance with US GAAP (Morris et al.,
2004, 2009).
Our study uses IFRS as the benchmark accounting standards rather than US GAAP
because IFRS have been used as the basis for developing Chinese GAAP since the ?rst
Chinese standards were issued in 1993; and on November 25, 2006, the Chinese
regulators announced that China would fully comply with IFRS from January 1, 2007.
Also, many countries in the Asian region including Singapore, Indonesia, Malaysia,
and Australia use IFRS as the basis of their domestic accounting standards.
3.1 Associations between transparency and corporate governance
The association between corporate ?nancial transparency and corporate governance is
premised on the idea that good corporate governance can promote corporate ?nancial
transparency through the effectiveness of corporate governance mechanisms[4],
and that transparent ?nancial reports can reduce the information asymmetry and
the con?ict of interest between companies’ insiders and outsiders, thus assisting the
effectiveness of corporate governance mechanisms. Some of the most common
corporate governance mechanisms include the separation of the roles of CEO and
board chairman, the presence of an audit committee, and independent non-executive
directors on the board.
While prior literature reports a positive association between the corporate
governance mechanisms and corporate ?nancial transparency in developed countries,
the results in emerging countries are mixed. For example, for Indonesia, Susilowati
(2008) reports a positive association between percentage of independent directors and
the presence of an audit committee and corporate ?nancial transparency. In the Chinese
context, Tower and Yan (2006) have also found dual leadership (where CEO and Chair
are separate people) and independent directors were negatively associated with
transparency. It is possible that these ?ndings result fromthe fact that audit committees
are comprised of independent directors who are alleged to be reluctant to stand up
against boards’ decisions because they are often appointed by controlling shareholders
(Lin, 2001). Further, we note that dual leadership may be more of a perception than a
reality in China because, although CEOand chair are separated at the company level, the
CEO or chair is often also the chair or CEO of the group to which the company belongs.
In contrast, Tower and Yan (2006) suggest that the reason for the negative relation
between independent directors and transparency could be a lack of “quali?ed”
independent directors with strong ?nancial and business experience in China. Further
they suggest that in the case of dual leadership, there is a lack of incentives to ensure
dual leadership leads to improved corporate operating decisions since the majority of
listed companies in China are SOEs and managerial jobs are not openly contestable,
but rather, are appointed by the government. In addition, Tower and Yan (2006) note
that the remuneration system is ?xed and often standardized so that remuneration is
only weakly linked to the performance of SOEs. On the other hand, where the CEO and
Chairman is the same person there may be more incentive to improve corporate
disclosure since the person is often appointed by the government and is subject to the
direct scrutiny of the CSRC (Tower and Yan, 2006).
Split equity
reform
25
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.2 Mandatory disclosure
Mandatory disclosure occurs where companies disclose information in accordance with
legislative requirements, including domestic accounting standards and any other
regulations. When examining mandatory disclosure, the level of enforcement is
important in determining compliance with accounting standards. While the level of
enforcement can be rigorous in developed countries such as the USAand the UK, it is not
always so in emerging countries.
In the Chinese context, the CSRC is in charge of the enforcement of corporate
disclosure. But it has been suggested that the CSRC lacks the independence and
necessary power in enforcing disclosures in China (Walter and Howie, 2006). In addition,
previous studies suggest that the investor protection and legal system are weak in
China, and that there is little litigation risk or risk of corporate takeover in China (Haw
et al., 2001; Xiao, 1999).
In combination, these factors suggest that the expected level of compliance with
mandatory disclosure would be low. However, previous studies examining disclosure
levels for large Chinese companies have found that mandatory disclosure compliance in
China is, in fact, high (Xiao, 1999). Such high-disclosure levels are positively associated
with company characteristics such as size, overseas listing, and the employment of Big N
auditors, andcorporate ?nancial transparencyinChina (measuredbyinternet disclosures)
is negatively associated with government (state) ownership (Xiao et al., 2004a, b)[5].
An additional factor leading to high levels of mandatory disclosures in China is that
Chinese GAAP includes numerous “templates” or “formats” embedded in the
Accounting System for Business Enterprises (ASBE), accounting standards and other
regulatory requirements issued by various parties[6]. To comply with Chinese GAAP,
listed companies are automatically forced to comply with the “templates.” The
implication of such templates is that in situations where the company does not have
certain items (e.g. intangible assets), the items are still disclosed in annual reports, but
with amounts shown as zeros, because of the required template details. Since reporting
an item(albeit zero) is coded as a disclosure for this study, this means that the templates
tend to result in high mandatory disclosure scores (see Xiao et al., 2004a, b for a
discussion of the development of accounting regulation in China).
3.3 Voluntary disclosure
Voluntary disclosure occurs where companies disclose information beyond mandatory
disclosure requirements. In the Chinese context, previous studies suggest that Chinese
companies provide little voluntary disclosure (DeFond et al., 2000; Tower and Yan, 2006;
Wang et al., 2007). This ?nding is in contrast to the positive relationship between
government ownership and voluntary disclosures noted by Eng and Mak (2003) in
Singapore which also has high-government levels of corporate ownership. The SE
structure in China, by limiting outside tradable shareholders, may limit demand for
voluntary disclosures. It is also possible that while the “templates” contained in Chinese
GAAP ensure high mandatory disclosure, they may work against voluntary disclosure
in that the templates can restrict the way in which companies disclose information.
4. Hypothesis development
Positive accounting theory (Watts and Zimmerman, 1986) deals with the con?icts of
interest that arise between management and the (outside) providers of equity and
ARJ
23,1
26
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
debt ?nance. To reduce these con?icts of interest, ?rms have incentives to disclose
information to outsiders and to have these disclosures audited. Accounting standards
(and GAAP generally) assist these monitoring activities. Similarly, companies that
make voluntary disclosures can reduce information asymmetry in a contractual setting
and further facilitate communication between the companies’ insiders and outsiders. As
managers’ share ownership falls, their incentives to increase disclosures to reduce
agency costs will increase. Evidence fromprior research shows this positive relationship
between ownership structure and disclosure (Eng and Mak, 2003; Ruland et al., 1990).
Similarly, signaling theory (Spence, 1973) focuses on how companies release
value-revealing information to signal to the market. Those companies that make credible
voluntary disclosures canenable investors to evaluate the company’s riskandits potential
future returns more accurately, which could lead to a more appropriate level of cost of
equity capital (Botosan, 1997; Botosan and Harris, 2000; Botosan and Plumlee, 2002).
In the aftermath of the SE reform, the proportion of outside equity in Chinese ?rms
will increase. Therefore, reformed companies are expected to have more incentive to
enhance their transparency via greater disclosure. In particular, as the interests of
controlling shareholders are now more aligned with the market performance of the
company, increased disclosure will be made to maintain the company’s share price.
The reform also creates additional incentives for corporate managers in China. With
the SE structure removed, companies are allowed to implement stock option schemes to
further align the interests of managers with outside shareholders. Consistent with
positive accounting theory (Watts and Zimmerman, 1986), corporate managers in those
reformed companies are expected to push for greater disclosure to enhance transparency
as the market value of the entity will be used as a basis for assessing their operational
performance. Further, since the reform will reduce the state’s administrative
involvement in Chinese listed ?rms, managerial career concerns will be switched
more from government/political concerns (with close government scrutiny) to be
somewhat closer to managerial career concerns in a market system, where managers are
concerned about their reputation and human capital in the managerial labor market. It is
expected that these implicit incentives will motivate managers to be more responsive to
the market and thereby provide more disclosures.
Furthermore, as more non-tradable shares become tradable, the in?uence of holders
of tradable shares, especially institutional investors (both domestic and foreign), should
increase and the in?uence of the controlling shareholder should decrease. These outside
shareholders are likely to demand the disclosure of more information from companies.
Therefore, in the aftermath of the SE reform, the reformed companies will feel increased
pressure to be more transparent than before the reform. However, because of the
moderating effect on the mandatory disclosure of the reporting templates included in
the Chinese GAAP, we expect the increase in mandatory disclosure to be smaller than
the increase in voluntary disclosure of the post-SE reform. These expectations are
addressed through the following hypotheses:
H1a. There will be an increase in mandatory disclosure between the pre-reform
2004 annual reports and the post-reform 2005 annual reports for companies
completing the SE reform.
H1b. The increase in mandatory disclosure between the pre-reform 2004 annual
reports and the post-reform 2005 annual reports will be greater for
Split equity
reform
27
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
companies completing the SE reform compared to companies not completing
the reform.
H2a. There will be an increase in voluntary disclosure between the pre-reform
2004 annual reports and the post-reform 2005 annual reports for companies
completing the SE reform.
H2b. The increase in voluntary disclosure between the pre-reform 2004 annual
reports and the post-reform 2005 annual will be greater for companies
completing the SE reform compared to companies not completing the
reform.
H3a. The post-reformincrease inmandatory disclosure will be smaller thanthe post-
reform increase in voluntary disclosure for companies completing the
SE reform.
H3b. The positive difference between the post-reform voluntary disclosure and the
post-reformmandatory disclosure will be greater for companies completingthe
SE reform compared to companies not completing the reform.
5. Research method
5.1 Sample selection
Our sample comprises all companies listed on the China Securities Index (CSI)
300 index in 2005 that completed the SE reform in 2005 (reformed companies, n ¼ 58),
and a control group of companies, matched by size and industry, that had not
commenced the reform in 2005 (unreformed companies, n ¼ 58). The resulting sample
size of 116 (58 £ 2) is more than suf?cient to detect large effect sizes in regressions with
up to 20 explanatory variables (Field, 2009, pp. 56-7 and 222-3)[7].
The matched pairs design addresses concerns that company size or industry are
associated with companies selected for reform by the government, or with companies
that self-selected for early reform. The CSI 300 Index covers more than 60 percent of
the market value of shares listed on the Shanghai and Shenzhen Stock Exchanges.
Consistent with prior research, only companies listed on the Shanghai Stock Exchange
are selected, since this exchange is the major stock exchange in China and has been the
benchmark used in prior research (Feng and Xu, 2007).
The years 2004 and 2005 are our focus because, for those years, the SEreformcreates
a natural experimental setting. The year 2004 is the year immediately before the SE
reform took place (pre-reform year), while 2005 represents the post-reform period as the
reform commenced in April 2005. Thus, by examining transparency scores, pre- and
post-reform for the reformed companies and comparing those to a (then) unreformed
control group, we expect to capture the impact of the reform on reformed companies.
Of the 58 companies which announced their participation in the reform before
December 31, 2005, 37 companies completed the reform before December 31, 2005[8]
and 21 companies completed the reform by February 2006 (prior to release of their 2005
annual reports at the end of March 2006)[9]. Matching of each reformed company with
an unreformed control company was based on their pre-reform size and industry (as on
December 31, 2004). Each reformed and matched unreformed company had to be in the
same industry, based on the Global Industry Classi?cation System developed by
Morgan Stanley and Standard & Poors. Within that same industry classi?cation,
ARJ
23,1
28
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the company closest in size (by natural logarithm of total assets) to the reformed
company was selected as the matched unreformed company. As reported in Table VIII,
the matching process resulted in no signi?cant difference in the average size of the
reformed and unreformed groups of companies. Other differences between the two
samples are addressed in the “Results” section.
Companies’ 2004 and 2005 annual reports (in Chinese language) were obtained from
the Shanghai Stock Exchange web site. Other ?nancial data for the sample for 2004
and 2005 were obtained from the China Center for Economic Research (CCER)
database. CCER is a major database provider in China and has been used in previous
studies (Feng and Xu, 2007).
5.2 Corporate ?nancial transparency measurement
The mandatory disclosure requirements for listed companies (Chinese GAAP) are to be
found in accounting standards, the requirements of the ASBE[10], and additional
disclosure requirements issued by the Ministry of Finance, the CSRC and the Shanghai
Stock Exchange. To measure corporate ?nancial transparency, we developed an
extensive checklist of 270 items from IFRS that could have been adopted by Chinese
companies. To develop the measure, we started with the disclosure checklist of 441
IFRS items in Morris and Gray (2007), itself based on a much larger list of items
prepared by Deloitte Touche Tohmatsu (2001). Morris and Gray removed items which
could not reasonably be observed by an annual report reader. We modi?ed their
checklist to make it fully applicable to Chinese companies, by removing items which
are not relevant in China. For example, the IFRS checklist requires disclosure of
employee entitlements in relation to de?ned bene?ts; however, de?ned bene?t pension
schemes are not applicable or used by Chinese companies. Further, the IFRS
requirements for ?nancial instruments in relation to accounting for hedges are not
applicable since the fair value and revaluation model are not allowed in Chinese GAAP,
only the cost model is allowed. As a result, it would be unrealistic to expect that any
company would voluntarily disclose any ?nancial instrument in relation to hedged
items where it uses fair value as the basis. Our resulting checklist contains 270 items.
Of these items, 178 were required by Chinese GAAP on December 31, 2004 and also on
December 31, 2005 – we treat them as mandatory disclosures. The remaining 92 items
were not required by Chinese GAAP in 2004 or 2005, but are included in the IFRS
regime in 2004 and 2005 – they are treated as voluntary disclosures. Therefore,
voluntary disclosures are IFRS-based checklist items not required by Chinese GAAP
but which could have been disclosed voluntarily by Chinese companies[11].
Each itemin the checklist is given equal weight based on the assumption that all items
of disclosure are equally important. This method eliminates any bias inherent in a
weighting system (Chau and Gray, 2002; Chow and Wong-Boren, 1987). Each checklist
itemwas coded 1 if disclosed, 0 if not disclosed and NAif not applicable to that company.
Three transparency scores were calculated: mandatory, voluntary, and overall.
The calculations are illustrated in the following example. As the mandatory
disclosure score has 178 items in total, assume company ABC has 142 items disclosed,
six items not disclosed, and 30 items not-applicable to the company. Thus, the
mandatory score is calculated as 142/(142 þ 6) ¼ 0.96. Similarly, for the 92 voluntary
disclosure items, assume company ABC has 40 items disclosed, 28 items not voluntarily
disclosed, and 24 items not applicable to the company. Hence, the voluntary score is
Split equity
reform
29
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
calculated as 40/(40 þ 28) ¼ 0.59. Finally, the overall score is ð142 þ 40Þ=
ð142 þ 6 þ 40 þ 28Þ ¼ 0:84.
5.3 Research design
The measure used in this study to determine the impact of the SE reform on corporate
?nancial transparency in China is the DiD estimator (Wooldridge, 2006). The DiD
estimator measures the difference in average outcome in a treatment group before and
after treatment minus the difference in average outcome in the control group before and
after treatment (Wooldridge, 2006). Essentially, the DiD estimator captures changes in
transparency of the 58 reformed companies and compares them with changes in
transparency of the 58 unreformed control companies. The DiD estimator can capture
the reform’s impact on transparency through the following basic model:
DTRANSP
i
¼ d
0
þb
1
DReformed
i
þDu
i
The DiD estimator[12]:
b
1
¼ DTRANSPi
treatment
2DTRANSPi
control
ði ¼ year 2004 or 2005Þ
.
Dependent variable. DTRANSP
i
is the change in transparency score between
2004 and 2005.
.
Variable of interest. DReformed
i
is a variable coded 1 for the reformed group as
they change from 0 in 2004 (pre-reform) to 1 in 2005 (post-reform) while the
change in score for the unreformed group is 0 to 0 from 2004-2005.
An important assumption underlying the DiD estimator is the parallel trend
assumption, namely that the treatment group (reformed companies) and the control
group (unreformed companies) should be reasonably similar before the “treatment”
event happens[13]. In other words, if the parallel trend assumption holds, the “starting
points” (pre-reform 2004) transparency scores for the two groups before the 2005 SE
reform should be reasonably similar. The parallel trend assumption also assumes that
the impact on the transparency of the two groups in 2004 and between 2004 and 2005
of any other factors associated with transparency should be reasonably similar. These
other factors (suggested by studies such as Tower and Yan, 2006) are:
Control variables
(1) Financial variables:
.
DUltimate control
i
measures the change in the ultimate controller of the
company between 2004 and 2005. It is assigned 1 if the ultimate controller
changes from privately controlled to government (state) controlled, and 0
otherwise.
.
DSize
i
measures the change in company size between 2004 and 2005. Size is
measured by the natural logarithm of total assets of the company.
.
DLeverage
i
measures the change in capital structure between 2004 and 2005.
This is proxied by total liabilities divided by total assets (TL/TA).
.
DNPAT
i
measures the change in pro?tability de?ated by the amount of total
assets between 2004 and 2005.
ARJ
23,1
30
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
.
DNon-tradable
i
measures the change in the percentage of non-tradable
shares between 2004 and 2005[14].
.
DForeign
i
measures the change in percentage foreign ownership between
2004 and 2005. Foreign investors include QFII and Hong Kong investors as
reported in annual reports.
(2) Corporate governance variables:
.
DAudit committee
i
equals 1 if the company set up an audit committee
between 2004 and 2005, and 0 otherwise.
.
DRemuneration committee
i
equals 1 if the company set up a remuneration
committee between 2004 and 2005, and 0 otherwise.
.
DNomination committee
i
equals 1 if the company set up a nomination
committee between 2004 and 2005, and 0 otherwise.
.
DDual
i
equals 1 if the company separated the role of CEO and chairman
between 2004 and 2005, and 0 if no change, and 21 if separation in 2004 was
removed in 2005.
(3) Additional control variables used in prior studies:
.
DInstitutional
i
measures the change in the proportion of ownership held by
institutional investors within the top 20 shareholders as disclosed in the
annual reports between 2004 and 2005. It is the sum of the domestic
institutional investors and the foreign institutional investors such as QFII.
.
DAudit ?rm
i
equals 1 if the company switched from non-Big Four to Big
Four auditor between 2004 and 2005, and 0 otherwise.
.
DIndependent directors
i
measures the change in the percentage of
independent directors on the board between 2004 and 2005. The percentage
of independent directors is the number of independent directors divided by
the total board size.
However, the DiD approach only requires these control variables if the parallel trend
assumption does not hold between the reformed and unreformed groups. Paired
samples t-tests reported in the results section indicate that size, leverage, NPAT,
non-tradable, foreign, audit committee, remuneration committee, and nomination
committee differed signi?cantly between 2004 and 2005 and therefore require control
variables in our model. In addition, independent t-tests reported in the results section
indicate that ultimate control, non-tradable and transparency scores differ signi?cantly
between the reformed and unreformed companies in the pre-reform (2004) period and
therefore also require control variables in our model. The resulting model tested is
shown as model (1)[15]:
DTRANSP
i
¼ d
0
þb
1
DReformed
i
þb
2
DUltimate control
i
þb
3
DSize
i
þb
4
DLeverage
i
þb
5
DNPAT
i
þb
6
DNon–tradable
i
þb
7
DForeign
i
þb
8
DAudit committee
i
þb
9
DRemuneration committee
i
þb
10
DNomination committee
i
þb
11
DDual
i
þb
12
Disc04
þb
13
Non–tradable04 þb
14
Ultimate control04 þDu
i
Model ð1Þ
Split equity
reform
31
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
6. Results and discussion
Tables II-V provide descriptive statistics for all variables for the 58 reformed and 58
unreformed companies in the pre-reform 2004 period (Tables II and IV) and the
post-reform 2005 period (Tables III and V). Most continuous variables do not exhibit
any pronounced skewness. Some binary variables are skewed, however as discussed
later, they do not bias our central ?ndings. Consistent with Walter and Howie (2006),
there are few institutional investors to demand voluntary corporate disclosure in China
from either the reformed or unreformed companies in the pre-SE reform period. The
level of institutional shareholding is small compared to the percentage of tradable
shares available, and the percentage of companies listed in Hong Kong (and thus
issuing H-shares) is low.
Table VI shows that in 2004, the mandatory, voluntary, and overall average scores
for the reformed companies (0.964; 0.677; 0.912) are lower than for the unreformed
companies (0.973; 0.690; 0.921), the differences for mandatory and overall scores being
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory score 04 0.890 0.990 0.964 0.017 21.952
Voluntary score 04 0.460 0.850 0.677 0.092 20.116
Overall score 04 0.830 0.970 0.912 0.025 20.578
Independent variables
Ultimate control 0.000 1.000 0.120 0.329 22.391
B-share 0.000 1.000 0.070 0.256 3.493
H-share 0.000 1.000 0.030 0.184 5.239
Size 20.989 24.886 22.338 0.822 0.683
Leverage 0.161 0.762 0.446 0.157 0.212
NAPT 20.073 0.164 0.058 0.044 0.298
Non-tradable 0.243 0.846 0.627 0.120 20.924
Institutional 0.000 0.417 0.063 0.086 2.821
Foreign 0.000 0.414 0.020 0.078 4.519
Audit ?rm 0.000 1.000 0.170 0.381 1.781
Audit committee 0.000 1.000 0.530 0.503 20.142
Remuneration committee 0.000 1.000 0.450 0.502 0.214
Nomination committee 0.000 1.000 0.330 0.473 0.754
Dual 0.000 1.000 0.983 0.131 27.616
Independent directors 0.180 0.500 0.331 0.052 0.313
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table II.
Reformed companies
2004
ARJ
23,1
32
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
statisticallysigni?cant. However, in2005, the mandatory, voluntary and overall average
scores for the reformed companies (0.973; 0.694; 0.921) are higher than for the
unreformed companies (0.972; 0.666; 0.917), although the differences are not statistically
signi?cant. That is, between 2004 and 2005, the mandatory, voluntary and overall
average scores increased for the reformed, but fell slightly for the unreformed,
companies.
The results for 2004 (Table VI) show that mandatory and overall transparency
scores, but not the voluntary scores, are signi?cantly different between the reformed
and unreformed companies. This indicates that the parallel trend assumption in 2004
may not hold as the starting points for the mandatory and overall scores are different
for reformed and unreformed samples. In our multivariate tests, we address this
problem by including a control variable (Disc_04) for 2004 disclosure levels. There are
no signi?cant differences between reformed and unreformed companies’ disclosure
scores in 2005.
Table VI also shows that that reformed companies’ scores on mandatory and overall
disclosure signi?cantly increased from 2004 to 2005. While reformed companies’
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory Score 05 0.940 1.000 0.973 0.011 20.693
Voluntary Score 05 0.500 0.900 0.694 0.086 0.242
Overall Score 05 0.880 0.980 0.921 0.019 0.675
Independent variables
Ultimate control 0.000 1.000 0.900 0.307 22.674
Size 21.009 25.679 22.456 0.872 1.044
Leverage 0.136 0.834 0.475 0.160 0.034
NAPT 20.154 0.205 0.051 0.052 20.058
Non-tradable tradable 0.122 0.797 0.512 0.143 20.634
Institutional 0.000 0.405 0.070 0.078 2.487
Foreign 0.000 0.385 0.020 0.069 4.562
Audit ?rm 0.000 1.000 0.160 0.365 1.956
Audit committee 0.000 1.000 0.570 0.500 20.286
Remuneration committee 0.000 1.000 0.500 0.504 0.000
Nomination committee 0.000 1.000 0.330 0.473 0.754
DUAL 0.000 1.000 0.931 0.256 23.493
Independent directors 0.231 0.500 0.330 0.045 1.385
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table III.
Reformed companies
2005
Split equity
reform
33
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
voluntary disclosure scores increased from 2004 to 2005, the change is only signi?cant
at the 10 percent level (two-tailed). On the other hand, unreformed companies’
mandatory and overall disclosures do not signi?cantly vary between 2004 and 2005,
and their voluntary disclosure scores signi?cantly fall from 2004 to 2005. In sum, the
results in Table VI support H1a and H2a.
Table VII shows that the average changes in mandatory, voluntary and overall
scores between 2004 and 2005 are greater for the reformed companies than for the
unreformed, and all differences between reformed and unreformed companies are
statistically signi?cant. The results in Tables VI and VII are consistent with the SE
reform impacting as predicted (H1b and H2b) on the reformed companies’ transparency
scores. Tests reported in Tables VI and VII were repeated using non-parametric
Mann-Whitney U-tests and Wilcoxon signed ranked tests. Results were substantially
unchanged.
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory score 04 0.940 1.000 0.973 0.011 20.693
Voluntary score 04 0.520 0.880 0.690 0.084 20.081
Overall score 04 0.870 0.950 0.921 0.018 20.491
Independent variables
Ultimate control 0.000 1.000 0.280 0.451 21.030
B-share 0.000 1.000 0.070 0.256 3.493
H-share 0.000 1.000 0.070 0.256 3.493
Size 20.782 24.172 22.343 0.740 0.602
Leverage 0.033 0.916 0.454 0.182 20.270
NAPT 0.001 0.226 0.051 0.045 1.552
Non-tradable 0.000 0.850 0.546 0.187 21.203
Institutional 0.000 0.449 0.071 0.083 2.500
Foreign 0.000 0.431 0.030 0.079 3.746
Audit ?rm 0.000 1.000 0.220 0.421 1.358
Audit committee 0.000 1.000 0.550 0.502 20.214
Remuneration committee 0.000 1.000 0.570 0.500 20.286
Nomination committee 0.000 1.000 0.430 0.500 0.286
Dual 1.000 1.000 1.000 0.000 0.000
Independent directors 0.000 0.430 0.314 0.068 23.444
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table IV.
Unreformed companies
2004
ARJ
23,1
34
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
For the reformedcompanies, the increase inthe voluntary score (0.0178) is greater thanthe
increase in the mandatory score (0.0093), although the difference is not statistically
signi?cant ( p ¼ 0.395); thus, H3a is not supported. The unreformed group had decreases
in voluntary (20.0238) and mandatory disclosure (20.0040), with their voluntary
disclosure decrease being signi?cantly greater than their mandatory disclosure decrease.
Reformed companies had a greater increase, albeit not statistically signi?cant, in
voluntary disclosure than in mandatory disclosure, but they did not experience the same
signi?cant gap between the declines in mandatory and voluntary disclosures that the
unreformed group did; this result provides quali?ed support for H3b.
In 2004, the mandatory and overall scores for both the reformed companies and the
unreformed companies were quite high. This result is likely driven by the reporting
templates contained in the Chinese GAAP. Compliance with these templates can push
up mandatory disclosure, thereby creating a ceiling effect which allows little room for
improvement in the TRANSP score over time. Nevertheless, the reformed companies
had a signi?cant increase in mandatory disclosure from 2004 to 2005, the unreformed
companies did not.
Minimum Maximum Mean SD Skewness
Transparency index
Mandatory Score 05 0.930 1.000 0.972 0.013 21.237
Voluntary Score 05 0.470 0.880 0.666 0.086 0.030
Overall Score 05 0.870 0.950 0.917 0.019 20.401
Independent variables
Ultimate control 0.000 1.000 0.240 0.432 1.241
Size 21.049 24.384 22.450 0.755 0.690
Leverage 0.048 0.887 0.482 0.176 20.475
NAPT 20.248 0.207 0.029 0.069 21.194
Non-tradable 0.000 0.850 0.542 0.186 21.182
Institutional 0.000 0.000 0.070 0.089 2.455
Foreign 0.000 0.484 0.030 0.083 3.836
Audit ?rm 0.000 1.000 0.240 0.432 1.241
Audit committee 0.000 1.000 0.590 0.497 20.359
Remuneration committee 0.000 1.000 0.660 0.479 20.670
Nomination committee 0.000 1.000 0.480 0.504 0.071
DUAL 0.000 1.000 0.897 0.307 22.674
Independent directors 0.000 0.533 0.317 0.077 21.710
Notes: n ¼ 58; mandatory – checklist items required by Chinese GAAP; voluntary – checklist items
not required by Chinese GAAP; overall – combined mandatory and voluntary score; ultimate control
– 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1 if the company has
B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size – natural logarithm
of total assets; leverage – total liabilities divided by total assets; NPAT – pro?tability de?ated by
total assets; non-tradable – percentage of non-tradable shares (as described in footnote [14] in 2005
this variable measures restricted tradable shares); institutional – proportion of top 20 shareholder
ownership held by institutional investors (includes domestic institutional investors and foreign
institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and Hong
Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit committee,
0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise; nomination
committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEO and chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table V.
Unreformed companies
2005
Split equity
reform
35
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
As noted, the mandatory and overall disclosure scores of unreformed companies
exceeded those of the reformed companies before the reform, with the difference
reversing after the reform. This seemingly counterintuitive result can be explained as
follows. The CSRC selected the Pilot Reform companies after considering the likely
attitude of shareholders to the Pilot Reform; the CSRC selected those companies where
they believed the shareholders were likely come to an agreed position quite quickly.
Since reformed companies had a higher percentage of state ownership than unreformed
companies (88 percent reformed, 72 percent unreformed), the government may have
considered it could expedite the reform more smoothly in the companies chosen. Prior
research in China indicates that state ownership is negatively associated with
corporate ?nancial transparency (Morris et al., 2006). Therefore, the higher level of
government ultimate control over the reformed companies in 2004 may help explain
the lower opening disclosure level for these companies. We return to the difference in
control in the following discussion.
Reformed
(n ¼ 58)
Unreformed
(n ¼ 58)
Mean
difference:
reformed vs
unreformed t
Sig.
(two-tailed)
2004
Mandatory score 0.964
a
0.973
a
20.009 23.233 0.002
Voluntary score 0.677
b
0.690
b
20.013 20.810 0.420
Overall score 0.912
c
0.921
c
20.009 22.275 0.025
2005
Mandatory score 0.973
a
0.972
a
0.001 0.636 0.526
Voluntary score 0.694
b
0.666
b
0.028 1.774 0.079
Overall score 0.921
c
0.917
c
0.004 1.236 0.219
a
Sig. (two-tailed) mandatory 04 vs 05 0.000 0.650
b
Sig. (two-tailed) voluntary 04 vs 05 0.097 0.039
c
Sig. (two-tailed) overall 04 vs 05 0.003 0.133
Notes: Mandatory – checklist items required by Chinese GAAP; voluntary – checklist items not
required by Chinese GAAP; overall – combined mandatory and voluntary score
Table VI.
Mean differences between
reformed and unreformed
companies on disclosure
scores in 2004 and 2005
Mean change
04/05 reformed
(n ¼ 58)
Mean change
04/05
unreformed
(n ¼ 58) t
Sig.
(two-tailed)
Mandatory score 0.0093 20.0007 24.051 0.000
Voluntary score 0.0178 20.0238 22.693 0.008
Overall score 0.0095 20.0040 23.331 0.001
t-score mandatory change vs voluntary change 20.857 2.073
Sig. (two-tailed) 0.395 0.043
Notes: Mandatory – checklist items required by Chinese GAAP; voluntary – checklist items not
required by Chinese GAAP; overall – combined mandatory and voluntary score
Table VII.
Changes in disclosure
scores between 2004 and
2005 for reformed and
unreformed companies
ARJ
23,1
36
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table VIII indicates that, of the control variables, only ultimate control (which shows
the extent of state control) and non-tradable shares are signi?cantly different between
the reformed and unreformed companies in 2004. Therefore, as previously noted,
because of these two potential violations of the DiD equation parallel assumption, we
include control variables for the 2004 values of ultimate control and non-tradable shares
in our models.
In addition, paired sample t-tests were performed to see whether or not the control
variables used in previous studies changed signi?cantly between 2004 and 2005
(Table VIII). There were signi?cant differences between years for the following
variables: size, leverage, pro?t, ultimate control, non-tradable shares, foreign ownership,
and audit, remuneration and nomination committees. Since the changes in these
variables could affect the change in transparency they are included in our DiD model
(model (1)).
Correlation matrices (Tables IX-XI) indicate that while there are a number of
independent variables that are signi?cantly correlated in both 2004 and 2005, only two
correlation coef?cients are greater than 0.8: the correlation of institutional and foreign
(0.86 in 2004; 0.84 in 2005), and foreign and H-share (0.86 in 2004; 0.86 in 2005).
Reformed vs unreformed
in 2004
All ?rms: change
from 2004 to 2005
Mean
difference t
Sig.
(two-tailed)
Mean
change t
Sig.
(two-tailed)
Ultimate control 0.155 2.118 0.037 20.026 21.346 0.181
B-share 0.000 0.000 1.000 n/a
H-share 20.034 20.834 0.406 n/a
Size 20.005 20.033 0.974 20.112 27.091 0.000
Leverage 20.008 20.254 0.800 20.029 24.133 0.000
NPAT 0.006 0.774 0.440 0.014 3.325 0.001
Non-tradable 0.080 2.751 0.007 0.059 8.953 0.000
Institutional 20.008 20.497 0.620 20.004 21.450 0.150
Foreign 20.006 20.423 0.673 20.002 22.134 0.035
Audit ?rm 20.052 20.694 0.489 0.000 0.000 1.000
Audit committee 20.017 20.185 0.854 20.034 22.027 0.045
Remuneration committee 20.121 21.298 0.197 20.069 22.919 0.004
Nomination committee 20.103 21.145 0.255 20.026 21.747 0.083
Dual 0.017 1.000 0.322 0.078 3.110 0.002
Independent directors 0.017 1.531 0.129 20.001 20.237 0.813
Notes: Ultimate control – 1 if the ultimate controller is government (state), 0 otherwise; B-share – 1
if the company has B-shares, 0 otherwise; H-share – 1 if the company has H-shares, 0 otherwise; size –
natural logarithm of total assets; leverage – total liabilities divided by total assets;
NPAT – pro?tability de?ated by total assets; non-tradable – percentage of non-tradable shares
(as described in footnote [14] in 2005 this variable measures restricted tradable shares); institutional
– proportion of top 20 shareholder ownership held by institutional investors (includes domestic and
foreign institutional investors such as QFII); foreign – percentage foreign ownership (includes QFII and
Hong Kong investors); audit ?rm – 1 if the Big4 auditor, 0 otherwise; audit committee – 1 if audit
committee, 0 otherwise; remuneration committee – 1 if remuneration committee, 0 otherwise;
nomination committee – 1 if nomination committee, 0 otherwise; dual – 1 if CEOand chairman separate,
0 otherwise; independent directors – number of independent directors divided by total board size
Table VIII.
Tests for signi?cant
differences among
potential control
variables
Split equity
reform
37
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
B
-
s
h
a
r
e
H
-
s
h
a
r
e
S
i
z
e
L
e
v
e
r
a
g
e
N
P
A
T
N
o
n
-
t
r
a
d
a
b
l
e
I
n
s
t
i
t
u
t
i
o
n
a
l
F
o
r
e
i
g
n
A
u
d
i
t
?
r
m
A
u
d
i
t
c
o
m
m
i
t
t
e
e
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
u
a
l
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
1
B
-
s
h
a
r
e
2
0
.
0
3
5
1
H
-
s
h
a
r
e
0
.
1
1
6
2
0
.
0
6
4
1
S
i
z
e
0
.
2
2
3
*
0
.
1
3
6
0
.
2
7
8
*
*
1
L
e
v
e
r
a
g
e
2
0
.
2
4
7
*
*
0
.
1
6
5
2
0
.
1
7
2
0
.
2
2
0
*
1
N
A
P
T
0
.
2
1
0
*
2
0
.
1
3
1
0
.
1
9
3
*
0
.
1
0
2
2
0
.
4
1
5
*
*
1
N
o
n
-
t
r
a
d
a
b
l
e
0
.
2
8
5
*
*
2
0
.
2
1
3
*
2
0
.
0
0
8
0
.
2
8
6
*
*
2
0
.
1
6
7
0
.
3
5
3
*
*
1
I
n
s
t
i
t
u
t
i
o
n
a
l
0
.
0
9
5
0
0
.
7
2
1
*
*
0
.
2
1
6
*
2
0
.
1
1
2
0
.
2
7
8
*
*
2
0
.
0
7
3
1
F
o
r
e
i
g
n
0
.
0
3
9
0
.
0
4
8
0
.
8
5
6
*
*
0
.
2
6
2
*
*
2
0
.
1
0
3
0
.
1
6
4
2
0
.
0
8
1
0
.
8
5
9
*
*
1
A
u
d
i
t
?
r
m
0
.
1
3
9
0
.
1
2
1
0
.
3
7
2
*
*
0
.
4
1
9
*
*
2
0
.
1
6
4
0
.
2
6
7
*
*
0
.
2
1
4
*
0
.
2
4
0
*
*
0
.
3
0
7
*
*
1
A
u
d
i
t
c
o
m
m
i
t
t
e
e
0
.
0
2
1
2
0
.
0
2
4
0
.
1
3
6
0
.
0
9
1
2
0
.
0
7
0
.
1
0
7
0
.
1
2
4
2
0
.
0
3
1
0
.
0
2
5
0
.
1
0
9
1
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
0
.
0
3
2
0
.
0
0
5
2
0
.
0
8
2
2
0
.
0
3
3
2
0
.
0
3
2
0
.
1
2
5
0
.
1
3
4
2
0
.
1
1
1
2
0
.
1
8
2
0
.
0
7
3
0
.
3
7
9
*
*
1
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
4
6
0
.
0
6
8
2
0
.
1
0
2
2
0
.
1
0
1
0
.
0
4
2
0
.
0
0
3
0
.
0
4
6
2
0
.
1
2
2
0
.
1
1
4
0
.
0
1
2
0
.
5
7
4
*
*
0
.
5
2
0
*
*
1
D
u
a
l
2
0
.
1
8
8
*
0
.
0
2
5
0
.
0
2
2
0
.
0
1
8
2
0
.
0
5
4
2
0
.
0
4
5
2
0
.
0
4
4
2
0
.
0
1
0
0
.
0
1
0
0
.
0
4
6
0
.
1
0
2
2
0
.
0
9
2
0
.
0
7
3
1
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
2
0
.
1
3
6
0
.
1
4
2
2
0
.
4
3
8
*
*
2
0
.
1
2
9
0
.
0
9
3
2
0
.
0
7
4
2
0
.
0
4
5
2
0
.
2
3
3
*
2
0
.
3
1
0
*
*
2
0
.
2
4
3
*
*
2
0
.
0
8
6
2
0
.
0
3
4
2
0
.
0
0
2
2
0
.
0
8
1
1
N
o
t
e
s
:
C
o
r
r
e
l
a
t
i
o
n
i
s
s
i
g
n
i
?
c
a
n
t
a
t
t
h
e
*
0
.
0
5
a
n
d
*
*
0
.
0
1
l
e
v
e
l
s
(
t
w
o
-
t
a
i
l
e
d
)
,
r
e
s
p
e
c
t
i
v
e
l
y
;
m
a
n
d
a
t
o
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
v
o
l
u
n
t
a
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
n
o
t
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
o
v
e
r
a
l
l
–
c
o
m
b
i
n
e
d
m
a
n
d
a
t
o
r
y
a
n
d
v
o
l
u
n
t
a
r
y
s
c
o
r
e
;
f
o
r
t
h
e
f
o
l
l
o
w
i
n
g
v
a
r
i
a
b
l
e
s
D
i
n
d
i
c
a
t
e
s
c
h
a
n
g
e
f
r
o
m
2
0
0
4
t
o
2
0
0
5
:
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
–
1
i
f
t
h
e
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
l
e
r
i
s
g
o
v
e
r
n
m
e
n
t
(
s
t
a
t
e
)
,
0
o
t
h
e
r
w
i
s
e
;
B
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
B
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
H
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
H
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
s
i
z
e
–
n
a
t
u
r
a
l
l
o
g
a
r
i
t
h
m
o
f
t
o
t
a
l
a
s
s
e
t
s
;
l
e
v
e
r
a
g
e
–
t
o
t
a
l
l
i
a
b
i
l
i
t
i
e
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
N
P
A
T
–
p
r
o
?
t
a
b
i
l
i
t
y
d
e
?
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
n
o
n
-
t
r
a
d
a
b
l
e
–
p
e
r
c
e
n
t
a
g
e
o
f
n
o
n
-
t
r
a
d
a
b
l
e
s
h
a
r
e
s
(
a
s
d
e
s
c
r
i
b
e
d
i
n
f
o
o
t
n
o
t
e
[
1
4
]
i
n
2
0
0
5
t
h
i
s
v
a
r
i
a
b
l
e
m
e
a
s
u
r
e
s
r
e
s
t
r
i
c
t
e
d
t
r
a
d
a
b
l
e
s
h
a
r
e
s
)
;
i
n
s
t
i
t
u
t
i
o
n
a
l
–
p
r
o
p
o
r
t
i
o
n
o
f
t
o
p
2
0
s
h
a
r
e
h
o
l
d
e
r
o
w
n
e
r
s
h
i
p
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
(
i
n
c
l
u
d
e
s
d
o
m
e
s
t
i
c
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
a
n
d
f
o
r
e
i
g
n
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
s
u
c
h
a
s
Q
F
I
I
)
;
f
o
r
e
i
g
n
–
p
e
r
c
e
n
t
a
g
e
f
o
r
e
i
g
n
o
w
n
e
r
s
h
i
p
(
i
n
c
l
u
d
e
s
Q
F
I
I
a
n
d
H
o
n
g
K
o
n
g
i
n
v
e
s
t
o
r
s
)
;
a
u
d
i
t
?
r
m
–
1
i
f
t
h
e
B
i
g
4
a
u
d
i
t
o
r
,
0
o
t
h
e
r
w
i
s
e
;
a
u
d
i
t
c
o
m
m
i
t
t
e
e
–
1
i
f
a
u
d
i
t
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
d
u
a
l
–
e
q
u
a
l
s
1
i
f
t
h
e
c
o
m
p
a
n
y
s
e
p
a
r
a
t
e
d
t
h
e
r
o
l
e
o
f
C
E
O
a
n
d
c
h
a
i
r
m
a
n
b
e
t
w
e
e
n
2
0
0
4
a
n
d
2
0
0
5
,
a
n
d
0
i
f
n
o
c
h
a
n
g
e
,
a
n
d
–
1
i
f
s
e
p
a
r
a
t
i
o
n
i
n
2
0
0
4
w
a
s
r
e
m
o
v
e
d
i
n
2
0
0
5
;
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
–
n
u
m
b
e
r
o
f
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
b
o
a
r
d
s
i
z
e
Table IX.
Pearson correlation
matrix between
independent variables
2004
ARJ
23,1
38
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
R
e
f
o
r
m
e
d
B
-
s
h
a
r
e
H
-
s
h
a
r
e
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
S
i
z
e
L
e
v
e
r
a
g
e
N
P
A
T
N
o
n
-
t
r
a
d
a
b
l
e
I
n
s
t
i
t
u
t
i
o
n
a
l
F
o
r
e
i
g
n
A
u
d
i
t
?
r
m
A
u
d
i
t
c
o
m
m
i
t
t
e
e
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
u
a
l
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
R
e
f
o
r
m
e
d
1
B
-
s
h
a
r
e
0
.
0
0
0
1
H
-
s
h
a
r
e
2
0
.
0
7
8
2
0
.
0
6
4
1
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
0
.
1
8
3
*
2
0
.
0
3
4
0
.
1
0
7
1
S
i
z
e
0
.
0
0
3
0
.
1
0
.
2
8
8
*
*
2
0
.
1
8
5
*
1
L
e
v
e
r
a
g
e
2
0
.
0
2
3
0
.
0
8
8
2
0
.
1
7
0
.
1
8
1
0
.
1
5
9
1
N
A
P
T
0
.
1
7
2
2
0
.
0
4
1
0
.
2
1
9
*
2
0
.
1
6
1
0
.
2
3
3
*
2
0
.
3
5
9
*
*
1
N
o
n
-
t
r
a
d
a
b
l
e
2
0
.
0
9
2
2
0
.
1
6
3
0
.
0
5
4
2
0
.
1
3
2
0
.
3
4
8
*
*
2
0
.
1
8
0
.
2
5
4
*
*
1
I
n
s
t
i
t
u
t
i
o
n
a
l
2
0
.
0
5
1
2
0
.
0
0
6
0
.
7
1
8
*
*
2
0
.
0
7
8
0
.
2
9
1
*
*
2
0
.
1
8
7
*
0
.
4
1
8
*
*
2
0
.
0
0
5
1
F
o
r
e
i
g
n
2
0
.
0
5
7
0
.
0
5
2
0
.
8
5
9
*
*
2
0
.
0
3
8
0
.
2
8
3
*
*
2
0
.
1
6
1
0
.
2
5
9
*
*
2
0
.
0
1
2
0
.
8
3
8
*
*
1
A
u
d
i
t
?
r
m
2
0
.
1
0
8
0
.
0
3
5
0
.
3
7
2
*
*
2
0
.
1
1
3
0
.
3
9
6
*
*
2
0
.
2
2
3
*
0
.
2
5
1
*
*
0
.
2
6
2
*
*
0
.
3
1
7
*
*
0
.
3
3
4
*
*
1
A
u
d
i
t
c
o
m
m
i
t
t
e
e
2
0
.
0
1
7
2
0
.
0
4
3
0
.
1
2
1
2
0
.
0
7
2
0
.
1
1
2
0
.
0
0
5
2
0
.
0
2
5
0
.
1
3
1
2
0
.
0
4
4
0
.
0
2
3
0
.
1
6
3
1
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
5
7
2
0
.
0
4
3
0
.
1
2
1
2
0
.
0
2
5
0
.
0
1
2
2
0
.
0
1
9
0
.
0
5
0
.
1
6
9
0
.
0
4
5
0
.
0
4
9
0
.
0
7
5
0
.
4
7
0
*
*
1
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
5
8
0
.
0
5
3
2
0
.
0
3
4
0
.
0
8
8
2
0
.
0
6
7
0
.
0
7
9
2
0
.
0
1
6
0
.
1
1
2
2
0
.
1
3
9
2
0
.
1
2
4
0
.
0
7
4
0
.
5
9
9
*
*
0
.
5
2
8
*
*
1
D
U
A
L
0
.
0
6
1
0
.
0
8
4
0
.
0
7
2
0
.
1
8
5
*
0
.
1
2
2
0
.
0
1
3
0
.
0
0
4
2
0
.
0
1
2
0
.
0
4
4
0
.
0
7
4
2
0
.
0
0
1
2
0
.
0
7
6
2
0
.
1
3
8
2
0
.
0
5
9
1
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
0
.
1
0
1
0
.
0
8
6
2
0
.
3
0
2
*
*
0
.
0
9
3
2
0
.
1
2
5
0
.
1
4
1
2
0
.
1
3
0
.
0
2
3
2
0
.
2
0
3
*
2
0
.
2
6
5
*
*
2
0
.
1
7
2
0
.
0
8
4
2
0
.
0
9
6
2
0
.
0
7
6
2
0
.
1
1
3
1
N
o
t
e
s
:
C
o
r
r
e
l
a
t
i
o
n
i
s
s
i
g
n
i
?
c
a
n
t
a
t
t
h
e
*
0
.
0
5
a
n
d
*
*
0
.
0
1
l
e
v
e
l
s
(
t
w
o
-
t
a
i
l
e
d
)
,
r
e
s
p
e
c
t
i
v
e
l
y
;
m
a
n
d
a
t
o
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
v
o
l
u
n
t
a
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
n
o
t
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
o
v
e
r
a
l
l
–
c
o
m
b
i
n
e
d
m
a
n
d
a
t
o
r
y
a
n
d
v
o
l
u
n
t
a
r
y
s
c
o
r
e
;
f
o
r
t
h
e
f
o
l
l
o
w
i
n
g
v
a
r
i
a
b
l
e
s
D
i
n
d
i
c
a
t
e
s
c
h
a
n
g
e
f
r
o
m
2
0
0
4
t
o
2
0
0
5
:
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
–
1
i
f
t
h
e
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
l
e
r
i
s
g
o
v
e
r
n
m
e
n
t
(
s
t
a
t
e
)
,
0
o
t
h
e
r
w
i
s
e
;
B
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
B
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
H
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
H
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
s
i
z
e
–
n
a
t
u
r
a
l
l
o
g
a
r
i
t
h
m
o
f
t
o
t
a
l
a
s
s
e
t
s
;
l
e
v
e
r
a
g
e
–
t
o
t
a
l
l
i
a
b
i
l
i
t
i
e
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
N
P
A
T
–
p
r
o
?
t
a
b
i
l
i
t
y
d
e
?
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
n
o
n
-
t
r
a
d
a
b
l
e
–
p
e
r
c
e
n
t
a
g
e
o
f
n
o
n
-
t
r
a
d
a
b
l
e
s
h
a
r
e
s
(
a
s
d
e
s
c
r
i
b
e
d
i
n
f
o
o
t
n
o
t
e
[
1
4
]
i
n
2
0
0
5
t
h
i
s
v
a
r
i
a
b
l
e
m
e
a
s
u
r
e
s
r
e
s
t
r
i
c
t
e
d
t
r
a
d
a
b
l
e
s
h
a
r
e
s
)
;
i
n
s
t
i
t
u
t
i
o
n
a
l
–
p
r
o
p
o
r
t
i
o
n
o
f
t
o
p
2
0
s
h
a
r
e
h
o
l
d
e
r
o
w
n
e
r
s
h
i
p
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
(
i
n
c
l
u
d
e
s
d
o
m
e
s
t
i
c
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
a
n
d
f
o
r
e
i
g
n
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
s
u
c
h
a
s
Q
F
I
I
)
;
f
o
r
e
i
g
n
–
p
e
r
c
e
n
t
a
g
e
f
o
r
e
i
g
n
o
w
n
e
r
s
h
i
p
(
i
n
c
l
u
d
e
s
Q
F
I
I
a
n
d
H
o
n
g
K
o
n
g
i
n
v
e
s
t
o
r
s
)
;
a
u
d
i
t
?
r
m
–
1
i
f
t
h
e
B
i
g
4
a
u
d
i
t
o
r
,
0
o
t
h
e
r
w
i
s
e
;
a
u
d
i
t
c
o
m
m
i
t
t
e
e
–
1
i
f
a
u
d
i
t
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
d
u
a
l
–
e
q
u
a
l
s
1
i
f
t
h
e
c
o
m
p
a
n
y
s
e
p
a
r
a
t
e
d
t
h
e
r
o
l
e
o
f
C
E
O
a
n
d
c
h
a
i
r
m
a
n
b
e
t
w
e
e
n
2
0
0
4
a
n
d
2
0
0
5
,
a
n
d
0
i
f
n
o
c
h
a
n
g
e
,
a
n
d
–
1
i
f
s
e
p
a
r
a
t
i
o
n
i
n
2
0
0
4
w
a
s
r
e
m
o
v
e
d
i
n
2
0
0
5
;
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
–
n
u
m
b
e
r
o
f
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
b
o
a
r
d
s
i
z
e
Table X.
Pearson correlation
matrix between
independent variables
2005
Split equity
reform
39
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
D
R
e
f
o
r
m
e
d
D
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
D
S
i
z
e
D
L
e
v
e
r
a
g
e
D
N
P
A
T
D
N
o
n
-
t
r
a
d
a
b
l
e
D
I
n
s
t
i
t
u
t
i
o
n
a
l
D
F
o
r
e
i
g
n
D
A
u
d
i
t
?
r
m
D
A
u
d
i
t
c
o
m
m
i
t
t
e
e
D
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
D
D
u
a
l
D
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
D
R
e
f
o
r
m
e
d
1
D
U
l
t
i
m
a
t
e
c
o
n
t
r
o
l
0
.
0
4
2
1
D
S
i
z
e
0
.
0
2
9
2
0
.
0
1
1
1
D
L
e
v
e
r
a
g
e
0
.
0
0
2
0
.
0
1
4
0
.
5
9
8
*
*
1
D
N
P
A
T
0
.
1
6
0
2
0
.
0
3
6
0
.
0
9
9
2
0
.
4
1
8
*
*
1
D
N
o
n
-
t
r
a
d
a
b
l
e
2
0
.
7
7
7
*
*
0
.
0
4
9
0
.
0
8
6
0
.
0
6
1
2
0
.
1
5
9
1
D
I
n
s
t
i
t
u
t
i
o
n
a
l
2
0
.
0
1
4
0
.
0
2
2
2
0
.
0
5
9
2
0
.
1
7
4
0
.
1
3
1
2
0
.
0
4
2
1
D
F
o
r
e
i
g
n
2
0
.
1
0
6
0
.
0
3
2
0
.
0
3
0
2
0
.
1
9
8
*
0
.
0
3
7
0
.
0
3
1
0
.
3
6
9
*
*
1
D
A
u
d
i
t
?
r
m
2
0
.
0
9
3
0
.
0
0
0
0
.
0
1
5
2
0
.
0
5
0
0
.
0
0
3
0
.
0
2
0
0
.
0
2
2
0
.
0
1
1
1
D
A
u
d
i
t
c
o
m
m
i
t
t
e
e
0
.
0
0
0
0
.
0
2
4
0
.
1
2
0
0
.
3
0
7
*
*
2
0
.
2
3
2
*
0
.
0
2
1
2
0
.
0
8
2
0
.
0
7
3
0
.
0
0
0
1
D
R
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
0
6
8
0
.
1
9
9
*
0
.
1
3
7
0
.
1
1
9
0
.
0
2
5
0
.
0
9
6
0
.
0
4
6
0
.
1
1
1
0
.
0
0
0
0
.
3
2
1
*
*
1
D
N
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
2
0
.
1
6
3
0
.
0
2
0
0
.
1
1
5
0
.
3
3
5
*
*
2
0
.
0
7
8
0
.
1
3
6
2
0
.
0
7
2
0
.
0
3
9
0
.
0
0
0
0
.
5
6
5
*
*
0
.
3
8
4
*
*
1
D
D
u
a
l
0
.
0
9
7
2
0
.
0
3
6
2
0
.
0
9
3
2
0
.
0
6
2
0
.
0
6
3
2
0
.
0
2
6
0
.
0
7
2
2
0
.
0
5
2
2
0
.
1
7
4
2
0
.
1
2
2
2
0
.
0
4
8
2
0
.
1
5
6
1
D
I
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
2
0
.
0
5
5
0
.
1
1
6
0
.
0
5
9
0
.
0
9
2
2
0
.
0
8
8
0
.
0
5
1
0
.
1
3
4
0
.
1
1
1
0
.
0
4
7
0
.
2
4
8
*
*
0
.
3
3
7
*
*
0
.
2
1
7
*
0
.
0
0
3
1
N
o
t
e
s
:
C
o
r
r
e
l
a
t
i
o
n
i
s
s
i
g
n
i
?
c
a
n
t
a
t
t
h
e
*
0
.
0
5
a
n
d
*
*
0
.
0
1
l
e
v
e
l
s
(
t
w
o
-
t
a
i
l
e
d
)
,
r
e
s
p
e
c
t
i
v
e
l
y
;
m
a
n
d
a
t
o
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
v
o
l
u
n
t
a
r
y
–
c
h
e
c
k
l
i
s
t
i
t
e
m
s
n
o
t
r
e
q
u
i
r
e
d
b
y
C
h
i
n
e
s
e
G
A
A
P
;
o
v
e
r
a
l
l
–
c
o
m
b
i
n
e
d
m
a
n
d
a
t
o
r
y
a
n
d
v
o
l
u
n
t
a
r
y
s
c
o
r
e
;
f
o
r
t
h
e
f
o
l
l
o
w
i
n
g
v
a
r
i
a
b
l
e
s
D
i
n
d
i
c
a
t
e
s
c
h
a
n
g
e
f
r
o
m
2
0
0
4
t
o
2
0
0
5
:
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
–
1
i
f
t
h
e
u
l
t
i
m
a
t
e
c
o
n
t
r
o
l
l
e
r
i
s
g
o
v
e
r
n
m
e
n
t
(
s
t
a
t
e
)
,
0
o
t
h
e
r
w
i
s
e
;
B
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
B
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
H
-
s
h
a
r
e
–
1
i
f
t
h
e
c
o
m
p
a
n
y
h
a
s
H
-
s
h
a
r
e
s
,
0
o
t
h
e
r
w
i
s
e
;
s
i
z
e
–
n
a
t
u
r
a
l
l
o
g
a
r
i
t
h
m
o
f
t
o
t
a
l
a
s
s
e
t
s
;
l
e
v
e
r
a
g
e
–
t
o
t
a
l
l
i
a
b
i
l
i
t
i
e
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
N
P
A
T
–
p
r
o
?
t
a
b
i
l
i
t
y
d
e
?
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
;
n
o
n
-
t
r
a
d
a
b
l
e
–
p
e
r
c
e
n
t
a
g
e
o
f
n
o
n
-
t
r
a
d
a
b
l
e
s
h
a
r
e
s
(
a
s
d
e
s
c
r
i
b
e
d
i
n
f
o
o
t
n
o
t
e
[
1
4
]
i
n
2
0
0
5
t
h
i
s
v
a
r
i
a
b
l
e
m
e
a
s
u
r
e
s
r
e
s
t
r
i
c
t
e
d
t
r
a
d
a
b
l
e
s
h
a
r
e
s
)
;
i
n
s
t
i
t
u
t
i
o
n
a
l
–
p
r
o
p
o
r
t
i
o
n
o
f
t
o
p
2
0
s
h
a
r
e
h
o
l
d
e
r
o
w
n
e
r
s
h
i
p
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
(
i
n
c
l
u
d
e
s
d
o
m
e
s
t
i
c
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
a
n
d
f
o
r
e
i
g
n
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
s
u
c
h
a
s
Q
F
I
I
)
;
f
o
r
e
i
g
n
–
p
e
r
c
e
n
t
a
g
e
f
o
r
e
i
g
n
o
w
n
e
r
s
h
i
p
(
i
n
c
l
u
d
e
s
Q
F
I
I
a
n
d
H
o
n
g
K
o
n
g
i
n
v
e
s
t
o
r
s
)
;
a
u
d
i
t
?
r
m
–
1
i
f
t
h
e
B
i
g
4
a
u
d
i
t
o
r
,
0
o
t
h
e
r
w
i
s
e
;
a
u
d
i
t
c
o
m
m
i
t
t
e
e
–
1
i
f
a
u
d
i
t
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
r
e
m
u
n
e
r
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
–
1
i
f
n
o
m
i
n
a
t
i
o
n
c
o
m
m
i
t
t
e
e
,
0
o
t
h
e
r
w
i
s
e
;
d
u
a
l
–
e
q
u
a
l
s
1
i
f
t
h
e
c
o
m
p
a
n
y
s
e
p
a
r
a
t
e
d
t
h
e
r
o
l
e
o
f
C
E
O
a
n
d
c
h
a
i
r
m
a
n
b
e
t
w
e
e
n
2
0
0
4
a
n
d
2
0
0
5
,
a
n
d
0
i
f
n
o
c
h
a
n
g
e
,
a
n
d
–
1
i
f
s
e
p
a
r
a
t
i
o
n
i
n
2
0
0
4
w
a
s
r
e
m
o
v
e
d
i
n
2
0
0
5
;
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
–
n
u
m
b
e
r
o
f
i
n
d
e
p
e
n
d
e
n
t
d
i
r
e
c
t
o
r
s
d
i
v
i
d
e
d
b
y
t
o
t
a
l
b
o
a
r
d
s
i
z
e
Table XI.
Pearson correlation
matrix between changes
in independent variables
2004-2005
ARJ
23,1
40
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
These correlations are not unexpected, since the institutional variable is the sum of
domestic and foreign investors and the H-share investors who are from Hong Kong are
treated as part of the foreign investors. No two of these highly correlated variables are
included in the same regression. Therefore, multicollinearity is not a concern in our
analysis.
6.1 DiD regression results
The DiD regressions reported in Table XII show that the reform variable is signi?cant
for mandatory, voluntary and overall disclosure scores between 2004 and 2005,
indicating support for hypotheses H1b and H2b. Change in disclosure is also
signi?cantly negatively associated with the 2004 disclosure score, meaning that
companies with lower 2004 disclosure scores tend to have signi?cantly higher 2004-2005
change scores. However, since none of the other control variables are consistently
signi?cant, the model was re-estimated as follows. The variables potentially violating
the equality of the opening position were retained (ultimate control; non-tradable; and
opening disclosure); while the control variables measuring change between 2004 and
2005, found insigni?cant in the model (1) estimation, were dropped. This model is
expressed as model (2):
DTRANSP
i
¼ d
0
þb
1
DReformed
i
þb
2
Ultimate control04
i
þb
3
Non–tradable04
i
þb
4
Disc04 þDu
i
Model ð2Þ
The model (2) DiD regression results are reported in Table XIII. The reform variable is
signi?cant in each regression, indicating that reformed companies had a greater change
in disclosure scores than unreformed companies did. These results support H1b and
H2b that, as a result of the SEreform, mandatory, voluntary and thus overall disclosures
would increase more for reformed companies than for unreformed companies. Given
that the support for H1b has occurred despite the ceiling effect on the mandatory score in
2004. These results suggest that the reform has had a strong effect, given that the full
effect of the reformis not likely to be fully captured in the ?rst year, as the reformperiod
could take up to three years, and given the SEreform’s share escrowconditions outlined
earlier. Since ultimate control and non-tradable for 2004 show some degree of skewness
(Tables II and IV) we reran model (2) without those variables. Our results are unchanged
from those reported in Table XIII.
The DiD reform coef?cient in model (2) shows that the reform improved the
mandatory disclosure score for reformed companies by 0.4 percent from the 2004 base
year. Hence, if reformed companies on average score 96.4 out of 100 for mandatory
disclosure score in 2004, the score has improved to 96.4 þ (96.4
*
0.004) ¼ 96.786 as a
result of the reform. While such an increase may not seem economically large, it is
constrained by the ceiling effect as the increase is based on the high mandatory
disclosure score of 96.4 in 2004. For voluntary disclosure, the reform coef?cient
indicates that voluntary disclosure improved by 3.5 percent from the 2004 base. In
other words, if the average voluntary disclosure for reformed companies is 67.7 out of
100 in 2004, the reform pushes up the voluntary disclosure in 2005 by 67.7
*
0.035 to
70.07 in 2005. While the increase may not seem economically large, the full effect of the
reform would not be expected to be captured until three years after the reform (in 2008).
These results provide support for H3a, that the increase in mandatory disclosures
would be smaller than the increase in voluntary disclosures for reformed companies.
Split equity
reform
41
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
6.2 Sensitivity analyses
A number of additional tests were performed as sensitivity analyses for potential
confounding factors. This section reports the results.
6.2.1 Early and late reformers. Recall that 37 companies completed the reform by
December 31, 2005, while 21 companies announced the reform before December 31,
2005 and completed it by February 2006. H1b and H2b were re-tested to see whether
Change in mandatory
disclosure
Change in voluntary
disclosure
Change in overall
disclosure
B t Sig. B t Sig. B t Sig.
(Constant) 0.588 8.938 0.000 0.303 4.721 0.000 0.592 8.424 0.000
DReformed 0.008 2.341 0.021 0.045 1.928 0.057 0.012 2.246 0.027
DUltimate control 0.001 0.109 0.913 0.032 0.857 0.394 0.006 0.634 0.528
DSize 20.010 21.062 0.291 0.123 1.895 0.061 0.021 1.400 0.156
DLeverage 0.024 1.001 0.319 20.228 21.331 0.186 20.033 20.829 0.409
DNPAT 0.041 1.473 0.144 20.059 20.288 0.774 0.025 0.534 0.595
DNon-tradable 0.033 1.478 0.142 0.084 0.522 0.603 0.043 1.154 0.251
DForeign 0.043 0.498 0.619 20.049 20.079 0.937 0.071 0.495 0.622
DAudit committee 0.002 0.269 0.788 0.007 0.138 0.890 0.003 0.282 0.779
DRemuneration committee 20.004 20.990 0.324 20.034 21.097 0.275 20.010 21.450 0.150
DNomination committee 0.004 0.484 0.629 0.027 0.446 0.657 0.009 0.679 0.499
DDual 0.009 2.386 0.019 0.014 0.508 0.613 0.011 1.759 0.082
Disc_04 20.602 28.953 0.000 20.478 25.984 0.000 20.646 28.566 0.000
Non-tradable 04 0.000 20.065 0.948 20.007 20.150 0.881 20.003 20.241 0.810
Ultimate control 04 0.000 0.025 0.980 0.012 0.598 0.551 0.004 0.945 0.347
F-statistic 8.650 3.703 7.226
Sig. of F 0.000 0.000 0.000
Adjusted R
2
0.482 0.248 0.431
Notes: Mandatory checklist items required by Chinese GAAP; voluntary checklist items not required
by Chinese GAAP; overall combined mandatory and voluntary score; for the following variables D
indicates change from 2004 to 2005; ultimate control 1 if the ultimate controller is government (state), 0
otherwise; DReformed
i
is a variable coded 1 for the reformed group as they change from 0 in 2004 (pre-
reform) to 1 in 2005 (post-reform) while the change in score for the unreformed group is 0 to 0 from
2004-2005; DUltimate control
i
measures the change in the ultimate controller of the company between
2004 and 2005; it is assigned 1 if the ultimate controller changes from privately controlled to
government (state) controlled, and 0 otherwise; DSize
i
measures the change in company size between
2004 and 2005; size is measured by the natural logarithm of total assets of the company; DLeverage
i
measures the change in capital structure between 2004 and 2005; this is proxied by total liabilities
divided by total assets (TL/TA); DNPAT
i
measures the change in pro?tability de?ated by the amount
of total assets between 2004 and 2005; DNon-tradable
i
measures the change in the percentage of non-
tradable shares between 2004 and 2005 (as described in note 15 in 2005 this variable measures
restricted tradable shares); DForeign
i
measures the change in percentage foreign ownership between
2004 and 2005; foreign investors include QFII and Hong Kong investors as reported in annual reports;
DAudit committee
i
equals 1 if the company set up an audit committee between 2004 and 2005, and 0
otherwise; DRemuneration committee
i
equals 1 if the company set up a remuneration committee
between 2004 and 2005, and 0 otherwise; DNomination committee
i
equals 1 if the company set up a
nomination committee between 2004 and 2005, and 0 otherwise; DDual
i
equals 1 if the company
separated the role of CEO and chairman between 2004 and 2005, and 0 if no change, and 21 if
separation in 2004 was removed in 2005; Disc_04 is the disclosure score for 2004; non-tradable 04 is the
percentage of tradable shares in 2004; ultimate control 04 is 1 if the ultimate controller in 2004 is
government (state), 0 otherwise
Table XII.
DiD regressions on
disclosure between 2004
and 2005 – model (1)
ARJ
23,1
42
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the impact of the reform on these two sub-samples is the same. Dividing the sample in
two considerably weakens the power of the tests and the results. For both sub-samples,
the models (1) and (2) regression results for the reform variable are consistent in
direction with the combined sample results for all disclosure measures. However, the
reform variable is only signi?cant in the voluntary disclosures regression for model (2)
of both sub-samples and weakly signi?cant.
6.2.2 Companies selected by the government. Recall that the reformwas completed in
batches, with 11 companies in the ?rst set of reformed companies selected by the
government to pilot the reform, and the remaining 47 companies self-selecting to reform.
A concern is that the ?rst 11 reformed companies might cover some of the “best” stocks
listed on the Shanghai Stock Exchange, and therefore the impact of the reform on these
11 companies could be greater than on the rest of the reformed companies. Thus, these 11
early reformed companies were dropped from the sample (together with their 11
matched controls) to see if the improved disclosure effect of the reform still holds for the
remaining 47 reformed companies. The models (1) and (2) results with the reduced
sample are very similar to those of the full sample.
7. Conclusions
The SE structure was seen as a major obstacle hindering the growth and stability of the
Chinese share market. On April 29, 2005, the Chinese Government launched the SE
reform to address this problem. One of the aims of this reform was to improve corporate
governance and thereby to improve corporate ?nancial transparency. This study
examines the impact of the SE reform on corporate ?nancial transparency in China,
investigating how the reform affected post reform transparency of companies that
commenced the reform (reformed G-share companies) before December 31, 2005.
Mandatory disclosure Voluntary disclosure Overall disclosure
B t Sig. B t Sig. B t Sig.
(Constant) 0.578 9.051 0.000 0.286 4.594 0.000 0.574 8.222 0.000
DReformed 0.004 2.155 0.033 0.035 2.479 0.015 0.008 2.320 0.022
DDual 0.009 2.615 0.010
Disc_04 20.593 29.064 0.000 20.470 26.023 0.000 20.631 28.399 0.000
Non-tradable 04 20.001 20.162 0.872 0.022 0.490 0.625 0.004 0.341 0.734
Ultimate control 04 20.001 20.374 0.709 0.009 0.506 0.614 0.003 0.640 0.524
F-statistic 23.937 11.726 22.411
Sig. of F 0.000 0.000 0.000
Adjusted R
2
0.499 0.272 0.427
Notes: Mandatory checklist items required by Chinese GAAP; voluntary checklist items not required
by Chinese GAAP; overall combined mandatory and voluntary score; for the following variables
D indicates change from 2004 to 2005; ultimate control 1 if the ultimate controller is government (state),
0 otherwise; DReformed
i
is a variable coded 1 for the reformed group as they change from 0 in 2004
(pre-reform) to 1 in 2005 (post-reform) while the change in score for the unreformed group is 0 to 0 from
2004-2005; DDual
i
equals 1 if the company separated the role of CEO and chairman between 2004 and
2005, and 0 if no change, and 21 if separation in 2004 was removed in 2005; Disc_04 is the disclosure
score for 2004; non-tradable 04 is the percentage of tradable shares in 2004; ultimate control 04 is 1 if
the ultimate controller in 2004 is government (state), 0 otherwise
Table XIII.
DiD regressions on
disclosure between 2004
and 2005 – model (2)
Split equity
reform
43
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
It was expected that the SE reform would improve the transparency of the reformed
companies after the reform(in 2005) compared to their pre-reformtransparency (in 2004)
and compared to a control group of unreformed companies in 2005. Transparency was
measured using a 270-item checklist based on IFRS but designed speci?cally for the
Chinese environment. Three transparency scores were derived from the checklist:
(1) a mandatory transparency score, based on the level of disclosure of items
required by Chinese GAAP in 2004 and 2005;
(2) a voluntary transparency score, based on the level of disclosure of items
required only by IFRS in 2004 and 2005; and
(3) an overall transparency score, based on the results of the mandatory and
voluntary transparency scores in 2004 and 2005.
The results of the studysupport the positive impact of the SEreformonthe transparency
of reformed companies. In the post-reform 2005 annual reports, mandatory voluntary
and overall disclosures of reformed companies improved compared to their pre-reform
levels and also compared to those of a matched group of unreformed companies.
Voluntary disclosures increased more than mandatory disclosures, but the mandatory
disclosures still increased signi?cantly despite the ceiling effect on them caused by the
disclosure templates in Chinese GAAP.
A limitation of our study is sample size as we focus on reformed companies listed on
the CSI 300 index. Although we include all such companies plus controls, the resulting
sample is relatively small (n ¼ 116). However, our primary focus is on the effect of the
SE reform, and since we ?nd the reform variable is signi?cant in all regression models,
our sample size is suf?ciently large to detect that effect.
Our study is also limited to companies that commenced the SE reform before
December 31, 2005 and examines the immediate impact of the reform on their corporate
?nancial transparency. Future research could also examine the reform’s long-term
effect on corporate ?nancial transparency in China. The impact of the reform is
expected to be greater over the three years after 2005 especially as the share escrow on
controlling shareholders was lifted by 2008. Therefore, it would be interesting to see if
the practical signi?cance of the reform becomes greater over those latter years on
reformed companies. Moreover, it would be interesting to examine whether the reform
has facilitated other important changes for Chinese companies in areas such as
improved corporate governance, and the attraction of more institutional investors,
especially QFII, to invest in the reformed companies. In addition, consistent with other
disclosure studies, our study only examines changes to the quantity of disclosures and
not the quality of these disclosures. Examining the quality of increased disclosures
would be a fruitful area for future research.
Notes
1. The report by Deloitte Touche Tohmatsu (2007) “A Comparison with current PRC GAAP and
IFRS”, compares the 2005 PRC GAAP, the (then) new2006 Accounting Standards to be applied
fromJanuary 1, 2007 and IFRS: (available at: www.iasplus.com/country/china.htm). There was
no change in the Chinese GAAP between December 31, 2004 and December 31, 2005.
2. The state-related shares also include legal-person shares that are owned by domestic
companies, but these domestic companies are typically other SOEs. Hence, the distinction
between state and legal shares is often just in name.
ARJ
23,1
44
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3. The SE reform attempts in 1999 and 2001 were quickly cancelled due to the negative market
reaction, caused possibly by the fear that a sudden ?oating of a huge number of non-tradable
shares would substantially affect the supplyanddemand balance, leadingto animmediate drop
in share price (Inoue, 2005). Both reform attempts were followed by bear market periods.
4. There is a debate in the corporate ?nancial transparency and corporate governance literature
whether the former is considered a subset of the latter or vice versa (Bai et al., 2004; Bushman
and Smith, 2001). This study treats corporate ?nancial transparency as part of corporate
governance, consistent with previous studies in this region (Kim, 2005; Susilowati, 2008).
5. Note that since these studies are restricted to large companies, the results may not be
representative of all Chinese companies. The current study addresses this through a more
representative sample of companies.
6. The presence of the templates is so “encompassing” their effects also appear in the corporate
governance section and directors’ report in annual reports where each company has the same
format in terms of disclosure and discussion on corporate governance mechanisms and
management discussions. The accounting model then adopted is called “fund accounting”
which is entirely rule-based meaning it has “an entire of code of extremely speci?c
accounting treatments developed to meet nearly every situation like the templates, thus, all
accountants need to do is just to ‘tick in the box’” (Walter and Howie, 2006).
7. Effect size refers to the impact that one variable has on another, and is measured using a
standardized variable such as the Pearson correlation coef?cient. Cohen (1988, 1992)
suggests that X has a large effect on Y if the Pearson correlation between them is at least 0.5.
8. Of the companies, 11 were chosen by the government as the ?rst two pilot reform batches,
the remaining 26 companies self-selected into the reform process. Sensitivity analyses
reported in the results section address potential differences between these two sub-samples.
9. The ?nancial year in China is from January 1 to December 31, with listed companies required
to submit their annual report between March and April in the following year. Therefore, the
second group of companies would still have incentives to improve disclosure in their 2005
annual report. Even so, it is possible that there may be a time lag effect for improved
disclosure after the reform, and if this were the case, the lag could be greater for the group of
companies completing the reform in February 2006 despite announcing the reform plan in
December 2005. Sensitivity analyses reported in the results section address this possibility.
10. These disclosure requirements are more comprehensive than accounting standards and are
applicable to all companies, not just listed companies.
11. Chinese companies issuing H-shares listed in Hong Kong are required to prepare ?nancial
statements using either IFRS or Hong Kong GAAP for their Hong Kong shareholders. Our
study focuses on the ?nancial statements (in Chinese) prepared for domestic Chinese
shareholders. For that purpose, checklist items not required by Chinese GAAP are
considered voluntary. Only six companies (two reformed, four unreformed) in the sample
have H-shares. Thus, the in?uence on our sample of the IFRS reporting requirement for
H-share issuers is minimal.
12. The equation is referred to as a ?rst-differenced equation. The strength of the equation is
that it overcomes “heterogeneity biases” (Wooldridge, 2006), because any omitted variable
problem or unobserved effect a
i
that can impact transparency scores remain constant or
?xed given the same companies are followed in two periods (2004 and 2005 in this study).
By taking the difference across two years, any omitted variable or unobserved effect is
cancelled out. It has been noted that the DiD methodology results in low R
2
(Wooldridge,
2006).
Split equity
reform
45
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
13. Small deviations from parallel assumptions are tolerated since resulting biases would be
small (Wooldridge, 2006).
14. After the 2005 reform, there is no distinction between non-tradable and tradable shares for
the reformed companies. However, since the previously non-tradable shareholders still face
temporary lockup on their shareholdings (share escrow), the non-tradable share category in
2005 is referred to as restricted tradable shares.
15. As discussedinthe results section, when this model was estimated, none of the control variables
measuring change between 2004 and 2005 were found to be signi?cant. Since these variables do
not account for the variation in disclosure between 2004 and 2005, they were dropped from the
model. The ?nal DiD model estimated (model (2)) is reported in the results section.
References
Ang, J.S. and Ma, Y.L. (1999), “Transparency in Chinese stocks: a study of earnings forecasts by
professional analysts”, Paci?c-Basin Finance Journal, Vol. 7 No. 2, pp. 129-55.
Ashbaugh, H. and Pincus, M. (2001), “Domestic accounting standards, international accounting
standards, and predictability of earnings”, Journal of Accounting Research, Vol. 39,
pp. 417-34.
Bai, C.E., Liu, Q., Lu, J., Song, F.M. and Zhang, J. (2004), “Corporate governance and market
valuation in China”, Journal of Comparative Economics, Vol. 32, pp. 599-616.
Botosan, C.A. (1997), “Disclosure level and the cost of equity capital”, The Accounting Review,
Vol. 72 No. 3, pp. 323-49.
Botosan, C.A. and Harris, M. (2000), “Motivations for a change in disclosure frequency and its
consequences: an examination of voluntary quarterly segment disclosures”, Journal of
Accounting Research, Vol. 38 No. 2, pp. 329-53.
Botosan, C.A. and Plumlee, M. (2002), “A re-examination of disclosure level and the expected cost
of equity capital”, Journal of Accounting Research, Vol. 40 No. 1, pp. 21-40.
Bushman, R.M. and Smith, A.J. (2001), “Financial accounting information and corporate
governance”, Journal of Accounting and Economics, Vol. 32, pp. 237-333.
Bushman, R.M. and Smith, A.J. (2003), “Transparency, ?nancial accounting information and
corporate governance”, FBRNY Economic Policy Review, April, pp. 65-87.
Bushman, R.M., Piotroski, J. and Smith, A.J. (2004), “What determines corporate transparency?”,
Journal of Accounting Research, Vol. 42 No. 2, pp. 207-52.
Chau, G.K. and Gray, S.J. (2002), “Ownership structure and corporate voluntary disclosure in
Hong Kong and Singapore”, The International Journal of Accounting, Vol. 37, pp. 247-65.
Chen, J.P., Chen, S.M. and Su, X.J. (2001), “Is accounting information value-relevant in the
emerging Chinese stock market?”, Journal of International Accounting, Auditing and
Taxation, Vol. 10 No. 1, pp. 1-22.
Chow, C.W. and Wong-Boren, A. (1987), “Voluntary ?nancial disclosure by Mexican
corporations”, The Accounting Review, Vol. 62 No. 3, pp. 533-41.
Cohen, J. (1988), Statistical Power Analysis for the Behavioural Sciences, 2nd ed., Academic Press,
New York, NY.
Cohen, J. (1992), “A power primer”, Psychological Bulletin, Vol. 112 No. 1, pp. 155-9.
DeFond, M.L., Wong, T.J. and Li, S. (2000), “The impact of improved auditor independence on
audit market concentration in China”, Journal of Accounting and Economics, Vol. 28,
pp. 269-305.
ARJ
23,1
46
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Deloitte Touche Tohmatsu (2001), “International accounting standards presentation and
disclosure checklist”, available at: www.iasplus.com/fs/iaschk01.pdf
Eng, L.L. and Mak, Y.T. (2003), “Corporate governance and voluntary disclosure”, Journal of
Accounting & Public Policy, Vol. 22 No. 4, pp. 325-45.
Enriques, L. and Volpin, P.F. (2007), “Corporate governance reforms in continental Europe”,
Journal of Economic Perspectives, Vol. 21 No. 1, pp. 117-40.
Feng, L.C. and Xu, W.H. (2007), “Has the reform of non-tradable shares raised prices?”, Emerging
Markets Finance and Trade, Vol. 43 No. 2, pp. 33-62.
Field, A. (2009), Discovering Statistics Using SPSS, 3rd ed., Sage, London.
Haw, I.M., Qi, D. and Wu, W. (2001), “The nature of information in accruals and cash ?ows in an
emerging capital market: the case of China”, The International Journal of Accounting,
Vol. 36, pp. 391-406.
Ho, S.S.M. and Wong, K.S. (2001), “A study of the relationship between corporate governance
structures and the extent of voluntary disclosure”, Journal of International Accounting,
Auditing and Taxation, Vol. 10, pp. 139-56.
Inoue, T. (2005), “Reform of China’s split-share structure takes shape”, Nomura Capital Market
Review, Vol. 8 No. 3, pp. 2-21.
Jian, M. and Wong, T.J. (2003), “Earnings management and tunneling through related party
transactions: evidence from Chinese corporate groups”, working paper, The Hong Kong
University of Science and Technology, Hong Kong.
Kim, T.E. (2005), “Determinants of corporate ?nancial reporting transparency: evidence from
South Korea”, thesis, University of New South Wales, Sydney.
Lin, T. (2001), “Corporatisation and corporate governance in China’s economic transition”,
Economics of Planning, Vol. 34, pp. 5-35.
Liu, Q. and Lu, Z.J. (2003), “Earnings management to tunnel: evidence from China’s listed
companies”, working paper, University of Hong Kong, Hong Kong.
Morris, R.D. and Gray, S.J. (2007), “Corporate transparency differences in the Asian region: the
feasibility of global standardization in practice?”, paper presented at the AFAANZ
Conference, Gold Coast, July.
Morris, R.D., Leung, Q. and Gray, S.J. (2006), “Corporate transparency in China: factors in?uencing
?nancial disclosure levels”, paper presented at the American Accounting Association
International Accounting Section Midyear Meeting, Los Angeles, CA, January 13.
Morris, R.D., Pham, T. and Gray, S.J. (2009), “Corporate governance and transparency in
Malaysia: before and after the Asian ?nancial crisis”, in Baxter, J. and Poullaos, C. (Eds),
Practices, Profession and Pedagogy in Accounting – Essays in Honour of Bill Birkett,
Chapter 5, Sydney University Press, Sydney.
Morris, R.D., Ho, B.U.S., Pham, T. and Gray, S.J. (2004), “Financial reporting practices of
Indonesian companies before and after the Asian ?nancial crisis”, Asia-Paci?c Journal of
Accounting and Economics, Vol. 11 No. 2, pp. 1-27.
Ruland, W., Tung, S. and George, N.E. (1990), “Factors associated with the disclosure of
managers’ forecasts”, The Accounting Review, Vol. 65 No. 3, pp. 710-21.
Spence, M. (1973), “Job marketing signaling: the informational structure of job markets and
related phenomena”, Quarterly Journal of Economics, Vol. 87, pp. 355-74.
Susilowati, I. (2008), International Differences in Corporate Governance and Financial
Transparency: A Comparative Study of In?uential Factors in Australia, India, Indonesia
and Japan, VDM, Saarbru¨cken.
Split equity
reform
47
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Tower, G. and Yan, Z.H. (2006), “Asset impairment disclosure in China”, working paper, Curtin
University of Technology, Perth.
Walter, C.E. and Howie, F.J.T. (2006), Privatizing China: The Stock Markets and their Role in
Corporate Reform, 2nd ed., Wiley, Singapore.
Wang, Q.T., Wong, J. and Xia, L.J. (2007), “State ownership, institutional environment and auditor
choice: evidence fromChina”, working paper, Chinese University of Hong Kong, Hong Kong.
Watts, R. and Zimmerman, J. (1986), Positive Accounting Theory, Prentice-Hall, Englewood Cliffs, NJ.
Wooldridge, J.M. (2006), Introductory Econometrics, 3rd ed., Thomson South Western, Mason, OH.
Xiao, J., Weetman, P. and Sun, M. (2004a), “Political in?uence and coexistence of a uniform
accounting system and accounting standards: recent developments in China”, Abacus,
Vol. 40 No. 2, pp. 193-218.
Xiao, J., Yang, H. and Chow, C. (2004b), “The determinants and characteristics of voluntary
internet-based disclosures by Chinese listed companies”, Journal of Accounting & Public
Policy, Vol. 23 No. 3, pp. 191-225.
Xiao, Z. (1999), “Corporate disclosure made by Chinese listed companies”, The International
Journal of Accounting, Vol. 34 No. 3, pp. 349-73.
About the authors
Wendy Green is a Senior Lecturer at the University of New South Wales who teaches
Auditing and Assurance Services at the undergraduate and postgraduate levels. Wendy Green
researches in the discipline of auditing. Wendy Green’s research to date has focused on factors
affecting audit quality and includes work on audit quali?cations in Australia and in emerging
markets, auditor switching, audit committees, auditor judgment in an analytical procedures
setting, and auditor specialisation. This work has appeared in both Australian and international
refereed journals and has been included in refereed conference proceedings both in Australia and
internationally. More recently, Wendy Green’s research has focused on informing the regulation
of auditing and assurance in the areas of partner rotation and the assurance of carbon emissions.
Wendy Green is the corresponding author and can be contacted at: [email protected]
Richard D. Morris is an Associate Professor at the University of New South Wales. His
research interests are in ?nancial accounting, international accounting, and accounting history.
They include the impact of IFRS adoption in Australia and the EU, international harmonisation
of ?nancial reporting, corporate transparency in the Asia Paci?c region, ?nancial reporting,
and the Asian ?nancial crisis of 1997-1998, the evolution of ?nancial reporting in Britain and
Australia during the nineteenth century, and the economic factors associated with accounting
choices made by ?rms. His work has appeared in international refereed journals, professional
journals, monographs, and chapters in books. Richard D. Morris’s teaching responsibilities are in
?nancial accounting at undergraduate and postgraduate levels.
Haiping Tang was a research student in the School of Accounting, University of New South
Wales, Australia.
ARJ
23,1
48
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Xiaoyan Chen, Xin Ling. 2015. Determinants of Chinese equity financing behaviours: traditional model
and the alternatives. Accounting & Finance n/a-n/a. [CrossRef]
2. Xiaoyan Chen. 2015. Improved corporate governance and Chinese seasoned equity offering announcement
effects. Accounting & Finance n/a-n/a. [CrossRef]
3. Haiyan Jiang, Ahsan Habib. 2012. Split-share reform and earnings management: Evidence from China.
Advances in Accounting 28, 120-127. [CrossRef]
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_916368570.pdf