-
Corporate Ownership Structure and the Informativeness
of Accounting Earnings in East Asia
Joseph P. H. Fan Department of Finance
School of Business and Management The Hong Kong University of
Science and Technology
Clear Water Bay, Hong Kong Phone: 852-2358-8016 Email:
[email protected]
and
T.J. Wong
Department of Accounting School of Business and Management
The Hong Kong University of Science and Technology Clear Water
Bay, Hong Kong
Phone: 852-2358-7574 Email: [email protected]
This version: September 2000
We appreciate helpful comments from Ray Ball, Gary Biddle, Ellen
Engel, Steve Matsunaga, Jevons Lee, Suil Pae, Enrico Perotti, Terry
Shevlin, Sheridan Titman, John Wei, Joanne Wu, Takeshi Yamada,
Jerold Zimmerman, an anonymous referee, and workshop participants
at Hitotsubashi University, The Hong Kong University of Science and
Technology, National Chengchi University of Taiwan and Shanghai
University of Finance and Economics, and conference participants at
the 2000 Conference on Accounting in Transition Economies at the
William Davidson Institute, the 2000 HKUST Accounting Symposium,
the 2000 AAANZ Conference, the 2000 Shanghai APFA conference, and
the 2000 London EFA Conference. T.J. Wong acknowledges the
financial support of Wei Lun Fellowship.
-
2
Corporate Ownership Structure and the Informativeness of
Accounting Earnings in East Asia
Abstract
This paper hypothesizes that the threat of expropriation by
controlling owners in East Asian corporations lowers the
credibility of accounting earnings and hence the stock price
informativeness of those earnings. The complicated share ownership
structure of East Asian corporations, characterized by a voting
control that is highly concentrated in the hands of families, and a
large separation of their voting rights from cash flow rights,
provides controlling owners with both the ability and incentive to
expropriate minority shareholders. We document that the
informativeness of earnings, measured by the earnings-return
relation, decreases with the level of an ultimate owners voting
control and the extent to which the owners voting rights exceed her
cash flow rights. We also find that family control per se does not
lower the informativeness of earnings. Earnings become less
informative when controlling families maintain high voting power
and large separation of voting from cash flow rights. Our results
are generally robust to controls for firm size, market-to-book
assets, leverage, number of industry segments operated by the firm,
and to varying the starting and ending dates of the stock return
window. JEL classification: G32, M41
-
3
1. Introduction
Public corporations in East Asia typically have a low level of
transparency and
disclosure quality. Poor accounting information is often blamed
for high financing costs
in this region. The financing difficulty became evident during
the recent Asian Financial
Crisis. Independent from the Crisis, East Asian firms also face
increasing worldwide
competition, thanks to globalization of trade and opening up of
domestic markets. Poor
accounting information and its associated high capital cost pose
a potential threat to the
competitiveness of East Asian firms.
Some commentators and policy advisors, including the World Bank
and the
International Monetary Fund, believe that a closer adherence to
international disclosure
rules and the adoption of international accounting standards are
essential for improving
corporate transparency in East Asia (World Bank, 1998). Despite
efforts to impose
stricter reporting rules and standards, a recent survey cited by
Asian Wall Street Journal
(November 24, 1999) finds that corporate transparency in this
region is declining. While
the new accounting rules may have increased the quantity of
accounting information,
investors still do not trust the quality of the reported
numbers.1 Thus, it is important for
regulators and policy makers to understand the causes of the low
credibility of reported
financial information in the region.
The objective of this paper is to examine how corporate
ownership structure in
seven East Asian economies affects the credibility of reported
accounting information.
More specifically, we examine how ownership structure affects
the credibility of
1 This view of low information quality was shared among business
professionals at the recent
World Bank Meeting. For example, a local lawyer from Thailand
remarked that the major difference (in
-
4
accounting earnings and hence the informativeness of those
earnings for investors.
Corporate ownership is highly concentrated in East Asia. We
report that the control of
over 70% of East Asian firms as measured by the level of voting
rights is ultimately
concentrated in the hands of families (outside Japan). Under
concentrated ownership,
conflicts of interest arise between controlling and minority
shareholders, and the
controlling shareholders decisions may result in the
expropriation of the minority
shareholders (Shleifer and Vishny, 1997; La Porta et al., 1999;
Johnson et al., 2000).
Moreover, a substantial fraction of these ultimate owners
possess higher voting (control)
rights than cash flow rights.2 The ultimate owners incentives to
expropriate minority
shareholders increase as the separation of their cash flow and
voting control rights
becomes larger. This is because larger voting rights give the
ultimate owners more power
to expropriate their companies, while smaller cash flow rights
reduce the owners share
of losses from the extraction of wealth.3 Also, when an owner
effectively controls a firm,
she gains control of the production of accounting information
and the reporting policies.
As the ultimate owner is perceived to have a strong incentive to
expropriate outside
shareholders, indicated by a large cash-vote divergence, the
market expects that the
owner will report accounting information more for
self-interested purposes rather than to
accounting disclosure) between the past and today is that
statements of accounts now carry more qualifications, not better
information. See the report by Henny Sender (1999).
2 The owner of a corporate share is entitled to three categories
of rights. First, the owner has the right of voting to deploy
corporate assets, i.e., voting (control) rights. Second, she has
the right to earn income, i.e., cash flow rights. Third, the owner
has the right of transferring the share to another party.
3 Claessens, Djankov, Fan, and Lang (1999) report for public
corporations from nine East Asian economies that the concentrated
voting control and the cash-vote divergence in those corporations
diminish firm value, indicating the economic significance of the
agency problem associated with the ownership structures. Consistent
evidence is also found in several other studies. La Porta,
Lopez-de-Silanes, Shleifer, and Vishny (1999) examine over 300
firms from 27 wealthy economies and report higher valuation of
firms with higher cash flow ownership by controlling owners.
Johnson, Boone, Breach, and Friedman (2000) document that levels of
shareholder protection explain the extent of stock market decline
in many emerging markets during the Asian Financial Crisis. They
argue that in countries with weak shareholder
-
5
reflect the firms true underlying economic transactions. This
loss in credibility of
accounting earnings reduces the stock price informativeness of
those earnings.
Therefore, we hypothesize that the informativeness of a firms
accounting earnings to
outside investors decreases with the increase in the ultimate
owners voting rights and the
degree of divergence between the owners voting rights and cash
flow rights.
We expect the relation between share ownership and accounting
information in
East Asia to be different from that in the U.S. The literature
on corporate share ownership
and accounting information is concentrated on U.S. companies
that are typically diffusely
owned. Prior research documents that an increase in managerial
ownership (Warfield et
al., 1995) or institutional ownership (Rajgopal et al., 1999)
would reduce the principal-
agent problem between managers and shareholders, which would in
turn lower the
incentives and opportunities for managers to manage earnings
while raising the price
informativeness of accounting earnings. However, we expect that
this relation between
share ownership and accounting information is not applicable to
East Asian corporations
due to differences in the degree of ownership concentration and
in the associated type of
agency problems.
Since families dominate most East Asian corporations, we also
investigate the role
of family control in the informativeness of earnings. Ever since
the Asian Financial
Crisis, there has been a growing concern that family-dominated
corporations are not
responsive to minority shareholders and are unwilling to reveal
their business plans and
finances to outsiders. South Korea is the first Asian country to
acknowledge the potential
harm of family domination in corporations and has recently
announced reforms to curb
protection, worse economic prospects result in more
expropriation by managers and thus a larger decline in share
prices.
-
6
family ownership of conglomerates (or chaebol). Although some
reformers have begun
to suspect that family control of East Asian companies leads to
inadequate corporate
reporting, there is no empirical evidence indicating the reasons
for or even the existence
of such a link. We start by examining whether family control per
se negatively affects
the informativeness of earnings. We then investigate the voting
rights and cash flow
rights structures and their relation to the informativeness of
earnings.
We find that in East Asia, the ultimate owners voting rights and
the divergence of
her cash flow rights and voting rights have a negative effect on
the informativeness of
earnings as measured by the earnings-return relation. This is
consistent with the
hypothesis that the threat of expropriation by the owner as
proxied by the voting rights
and the cash-vote divergence reduces the credibility of
accounting earnings and the stock
price informativeness of those earnings. Our results also show
that family control in
general does not have a negative impact on the informativeness
of earnings. Only those
firms with controlling owners who maintain high voting rights
and a large cash-vote
divergence report less informative earnings. Our results are
robust to controls for firm
size, market-to-book assets, leverage, number of industry
segments operated by the firm,
and to varying the starting and ending dates of the stock return
window.
This paper makes a contribution to the literature on corporate
ownership and the
properties of earnings by providing results that contrast with
research focusing on U.S.
corporations (Warfield et al., 1995; Rajgopal et al., 1999; and
Bushman et al., 1999) and
U.K. corporations (Pope et al., 1998). Also, compared to this
other body of research, our
results may be more generalizable to other parts of the world
because concentrated
corporate ownership in East Asia as compared to diffuse
corporate ownership in the U.S.
-
7
and the U.K. is a more representative corporate share structure
throughout the world.
Second, several recent accounting studies (Ball, Kothari, and
Robin, 1998; Ball, Robin,
and Wu, 1999; and Ali and Hwang, 2000) have provided evidence
that in addition to
accounting standards, features of the institutional environment
such as corporate
governance, as well as legal and financial systems can also
explain the international
differences in accounting properties. In this paper, we extend
their work by examining
ownership structure as one of the channels through which a
countrys institutional
environment influences each individual firms reporting
credibility. More specifically, we
examine whether corporate governance, measured by the structure
of share ownership at
the firm level, affects the earnings-return relation across
firms. Third, this research may
have implications for East Asian reformers and regulators who
are striving to improve
corporate governance and transparency in their countries.
The paper proceeds as follows. In Section 2, we discuss the
interplay between
ownership structure and the informativeness of earnings, and
present our hypothesis. In
Section 3, we report statistics of the ownership structure of
East Asian firms and present
our empirical analyses. In Section 4, we conclude this paper by
summarizing our findings
and pointing out future research directions.
2. Ownership structure, earnings credibility, and the
informativeness of earnings
As will be detailed later in this paper, the control of East
Asian corporations is
typically concentrated in one owner (or family). The ultimate
control is achieved through
complicated ownership arrangements, i.e., stock pyramids,
cross-shareholdings, and, to a
less extent, multiple classes of stocks. Moreover, these
ownership arrangements often
-
8
create wedges between the ultimate owners cash flows and their
voting rights. That is,
the ultimate owners often have greater voting rights than cash
flow rights.4 In this
section, we discuss how the ownership structure affects the
informativeness of accounting
earnings.
Public corporations, to various extents, are constrained by
agency problems
originating from conflicts of interest among stakeholders. The
type of agency problems
associated with a firm is affected by the structure of the firms
share ownership. When
ownership is diffuse as is typical in the U.S. and the U.K.,
agency problems stem from
the conflicts of interest between managers and shareholders
(Berle and Means, 1932;
Jensen and Meckling, 1976; Roe, 1994). As ownership
concentration increases to a level
where an owner obtains effective control of the firm, the nature
of agency problems shifts
away from the manager-shareholder conflicts to conflicts between
the controlling owner
and minority shareholders (Shleifer and Vishny, 1997). East
Asian corporations, as well
as most other corporations outside the U.S. and the U.K., are
subject to the latter type of
agency problems. More specifically, when a large shareholder
gains effective control of
a corporation, her decisions may result in the expropriation of
minority shareholders.
The conflicts of interest between large and small shareholders
can include outright
expropriation, i.e., the controlling shareholders enrich
themselves through self-dealing
4 An emerging literature documents that a separation between
cash flow rights and voting rights is
common among public corporations around the world. La Porta et
al. (1999) report such evidence for over 600 corporations in 27
rich countries. Claessens et al. (2000) report similar evidence for
public firms from East Asia. A host of papers focus on European
countries and report a typical ownership structure characterized by
separation of cash flow rights and voting rights. See Gugler et al.
(1999) on Austrian companies, Becht and Chapelle (1999) on Belgian
companies, Bloch and Kremp (1999) on French companies, De Jong et
al. (1999) on public companies in the Netherlands, and Bianchi et
al. (1997), Aganin and Volpin (1998), Enriques (1998), and Melis
(1998) on listed companies in Italy.
-
9
transactions in which profits are transferred to other companies
they control.5 They can
also exercise de facto expropriation through the pursuit of
objectives that are not profit-
maximizing in return for personal utilities.6
One key factor that exacerbates the agency conflicts in East
Asia is that
controlling owners often possess more voting control rights than
cash flow rights. This
cash-vote divergence positively affects the controlling owners
incentives to expropriate
other shareholders.7 The incentives of expropriation arise when
these two rights diverge
because the ultimate owners receive the entire benefit but only
bear a fraction of the cost.
We use a simple pyramidal structure to illustrate this point. A
family owns 25% of the
stock of publicly traded Firm A, which in turn has 32% of the
stock of Firm B. As a
modest measure, we say that the family controls 25% of Firm B --
the weakest link in the
chain of voting rights. In contrast, we say that the family owns
about 8% of the cash flow
rights of Firm B, the product of the two ownership stakes along
the chain. Given this
5 Cases of self-dealing transactions are numerous in East Asia.
For example, it was reported that a
tycoon in Hong Kong sold an unprofitable business solely owned
by him at a premium price to a publicly traded company that he
controlled. Scott (1999) studies the role of corporate governance
in four Asian countries that were in financial crisis: Korea,
Indonesia, Malaysia, and Thailand. He concludes by recommending
that strengthening the effective limits on self-dealing
transactions of controlling owners would be the priority task for
these countries.
6Given the threat of expropriation, the question arises as to
why investors still hold minority shares. There are several
potential reasons. First, many of the capital markets in East Asia
are segmented and heavy transaction costs limit investors' choice
of investments. Second, investors can price protect themselves by
paying discounted prices for shares. The third reason is that
investors are attracted to invest in East Asian companies because
their ultimate owners, typically families, maintain business and
political connections that allow their companies to make abnormal
profits. The birds' nests case detailed in Backman (1999, pp.
279-280) is a case in point. Ari Sigit, a grandson of Suharto,
together with other family members controlled Arha Group in
Indonesia. Sigit collaborated with Indonesia's Birds' Nest
Association to seek government backing of monopoly rights on the
export of the nests. The request was granted and Sigit was able to
collect a substantial fortune from the trading of the precious
traditional Chinese delicacies. Fisman (1999) conducts an event
study on the stock price effects of the news announcements of
Suharto's illness. He analyzes the value drops of Suharto connected
firms and reports that the proportion of these firms' share values
that is attributed to Suharto connections is very large -- about a
quarter of each firm's share value. Political connections are
clearly valued by investors in this case.
7 Expropriation as an agency cost of the separation of cash flow
rights from voting rights plays a key role in the theoretical
models of Burkart et al. (1997, 1998), Bebchuk et al. (1999), and
Wolfenzon (1999).
-
10
ownership structure, it only costs the family eight dollars for
every 100 dollars stolen
from Firm B.8
As share ownership structure delineates a firms agency problems,
it also has
impact on the firms financial reporting. Prior research almost
exclusively focuses on
U.S. corporations whose shares are typically diffusely owned.
The relation between
ownership structure and accounting reporting has not been
studied in the context of
concentrated ownership, the dominant organizational form outside
the U.S. When an
owner effectively controls a firm, she also controls the
production of the firm's
accounting information and the reporting policies. As the
controlling owner is perceived
to have strong incentives to expropriate outside shareholders,
as indicated by a large
separation of voting rights and cash flow rights, the
credibility of the firms accounting
information is reduced. That is, outside investors pay less
attention to the reported
accounting numbers, because they expect that the controlling
owner reports accounting
information more for self-interested purposes than to reflect
the firms true underlying
economic transactions. In particular, outside investors may not
trust the firms reported
accounting earnings because the controlling owner may manipulate
earnings for outright
expropriation. In addition, outside investors know that the
controlling owner has an
incentive to avoid reporting detailed and truthful accounting
information that would
attract close monitoring by outside shareholders. This does not
always mean outright
8 We note that there might be rationale for the ownership
structure in East Asia. The concentrated
ownership reduces transaction costs in negotiating and enforcing
corporate contracts with various stakeholders, including managers,
labor, material suppliers, customers, debt-holders, and
governments. Shleifer and Vishny (1997) suggest that the benefits
from concentrated ownership may be relatively larger in countries
that are generally less developed, where property rights are not
well defined and/or protected and enforced by judicial systems.
However, if stock pyramids, cross shareholdings, or dual-class
shares are used to consolidate control, they would also result in a
departure from the one-share-one-vote rule which induces the agency
problem emphasized in this study.
-
11
earnings manipulation for covering up possible earnings effects
of wealth extraction. The
controlling owner may simply bury the wealth effects of her
expropriation activities in
the aggregate earnings number without reporting them as separate
income statement
items. As a result of the loss of earnings credibility, the
stock price informativeness of
the earnings is weakened.9 The effects of earnings credibility
have been found to be
important in prior literature. Teoh and Wong (1993) report that
the market perception of
the quality of accounting earnings, proxied by the firms auditor
size, positively affects
the stock price informativeness of earnings.
The role of accounting in the context of concentrated ownership
contrasts with its
role in the context of diffuse ownership. The accounting
literature contains extensive
research on how the agency problem between owners and managers
affects the role of
accounting in management compensation contracts and how the
reporting incentives of
managers affect accounting information quality in a firm.10
However, Warfield et al.
(1995) carried out one of only a few studies that specifically
examine the relation
between corporate share ownership and the stock price
informativeness of earnings. In a
diffuse ownership context, they document that more managerial
ownership is associated
with greater earnings informativeness. They argue that an
increase in managerial
ownership reduces the conflicts of interest between owners and
managers and thus the
need for accounting-based managerial constraints. The result is
that the informativeness
9 Although the informativeness of earnings is weakened by the
threat of expropriation, it is not lost
entirely. In this paper, we document that accounting earnings
remain positively associated with annual stock returns in these
economies (Table 4 and Appendix 1). Prior research (Alford et al.,
1993; Ball, Robin, and Wu, 1999) also documents evidence consistent
with the view that accounting earnings generally convey relevant
information for stock price valuation among East Asian
corporations.
10 The role of accounting information in management compensation
contracts that are designed to mitigate the agency problem between
owners and managers in a diffuse ownership context is discussed in
Watts and Zimmerman (1986). Past research (Healy, 1985; Gaver et
al., 1995; Holthausen et al., 1995) has examined how bonus
contracts provide managers with incentives to manage earnings.
-
12
of earnings increases because managers have less need to manage
earnings in order to
alleviate constraints. In East Asian corporations, the high
concentration of ownership
nullifies the principal-agent problem between owners and
managers as well as the related
role of accounting-based managerial contracts. We therefore do
not expect this issue
identified by Warfield et al. to be applicable to East Asian
firms.
In examining the relation between the expropriation incentives
of a firms
controlling owner and the informativeness of accounting
earnings, we focus on the joint
effects of the controlling owners level of voting rights and the
degree of separation
between these voting rights and cash flow rights. With a
sufficient level of voting rights
to gain control, the owner has the ability and perhaps the
incentive to expropriate, since
not being the sole owner she only bears part of the loss from
the expropriation. A
divergence between cash flow rights and voting rights would
further increase the
controlling owners incentives to expropriate minority
shareholders. Using these
ownership structure measures as proxies for the risk of
expropriation by the controlling
owner, we expect that the credibility of the firms accounting
earnings and hence their
informativeness to outside investors decrease with an increase
in the voting rights level of
the ultimate owner and the degree to which the level of voting
rights exceeds the
associated level of cash flow rights. Formally, our
(alternative) hypothesis is: The
increase in the ultimate owners level of voting rights and/or
her degree of cash flow
rights and voting rights divergence decreases the
informativeness of her firms earnings.
-
13
3. Empirical analysis
In this section, we introduce the sample and data sources,
describe the ownership
structures of East Asian corporations, and perform statistical
analysis to test the earnings
credibility hypothesis that we developed in the previous
section.
3.1. Data sample
We select our sample firms from seven East Asian economies --
Hong Kong,
Indonesia, South Korea, Malaysia, Singapore, Taiwan and
Thailand. We include firms
that have sufficient ownership, stock return, earnings and other
financial data for
empirical analysis. Below is a description of the data
sources.
3.1.1. Ownership structure
For each firm we need to identify who the ultimate owners are,
and what share of
the cash flow rights and control rights they hold. For these
ownership structures we use
data assembled in Claessens et al. (2000). This ownership
database includes publicly
traded corporations in nine East Asian economies: Hong Kong,
Indonesia, Japan, Korea,
Malaysia, the Philippines, Singapore, Taiwan, and Thailand. The
starting point for the
construction of this ownership database is the Worldscope
database, which generally
provides the names and holdings of the six largest owners of
companies from 49
countries. For the nine East Asian economies, it has over 4,500
publicly traded firms, but
only about 2,000 companies provide detailed ownership
information. The Worldscope
data is supplemented with ownership information from the Asian
Company Handbook
1999, the Japan Company Handbook 1999, Hong Kong Company
Handbook 1997, the
Handbook of Indonesian Companies 1996, the Philippine Stock
Exchange Investments
-
14
Guide 1997, the Securities Exchange of Thailand Companies
Handbook 1997, and the
Singapore Investment Guide. In all cases, the ownership
structure data as of the 1996
fiscal year-end are collected. Information on dual-class shares
is provided in Datastream,
as described in Nenova (1999), and is supplemented by the
country-specific sources
mentioned above for Indonesia, the Philippines, Singapore, and
Thailand, where
Datastream covers a smaller fraction of listed companies.
Various country sources on
business group affiliation as detailed in Claessens et al.
(2000) are employed to study the
pyramid structures and cross-holdings among group-affiliated
firms. The ownership
database ends up having 2,980 companies for which the ultimate
owners can be traced.
3.1.2. Merging ownership and stock return/financial data
We merge the ownership data with the PACAP electronic database,
which is
commercially distributed by the University of Rhode Island.
PACAP contains the
financial and stock return data of publicly traded companies of
the seven East Asian
economies in concern. We exclude Japan from our analysis because
its institutional
environment and its firms ownership structures are quite
different from the other East
Asian economies.11 We select 1991 through 1995 as the period of
analysis and retrieve
the stock return and financial data for that period accordingly.
An exception is Korea, for
which PACAP has data up to only 1994. We do not include 1996
data because PACAP
has yet to update the data for that year. We also exclude
pre-1991 data because we are
concerned that the ownership structure of that time may differ
too much from that of
1996. Although we have ownership data for the Philippines, we do
not include firms
11 Different from the East Asian firms that are typically family
controlled, the dominant ultimate
owners of the Japanese firms are institutions, typically the
main banks of industrial groups. Japanese firms ownership
structures are also quite different from those of the East Asian
firms in both the degree of control and cash-vote divergence
(Claessens et al., 2000).
-
15
from that country because they are not covered by the PACAP
database. The merging of
the 1996 ownership data and the 1991-1995 stock return and
financial data requires us to
assume that the ownership and control structures of the firms
did not change substantially
during that period. This is a reasonable assumption since the
economic and political
conditions were relatively stable during that period. After
merging the data, we obtain a
sample of 1,350 firms with a total of 4,490 firm-years.12
3.2. Measuring ultimate owners cash flow and voting rights
Most prior studies of ownership structure focus on immediate
ownership
common shares directly owned by individuals or institutions.
Immediate ownership is
not sufficient for characterizing the ownership and control
structure of East Asian firms,
as these firms are generally associated with complicated
indirect ownership. Different
from these prior studies, we focus on ultimate ownership. For a
given firm, ultimate
owners and their share of cash flow and voting control rights
are identified. The
procedure of identifying ultimate owners is similar to the one
used in La Porta et al.
(1999). An ultimate owner is defined as the shareholder who has
at least 5% of the
voting rights of the company and who is not controlled by
anybody else. If a company
does not have an ultimate owner, it is classified as widely
held. To economize the data
collection task, we stop tracing for further voting control and
set the ultimate owners
voting rights level equal 50% once it reaches such level. This
ceiling is reasonable since
the ultimate owner unambiguously gains full control once she
secures 50% of voting
rights. To facilitate comparison, ultimate owners are classified
into family and non-
family owners. Family owners are individuals who are usually
managers or family
12 The two extreme percentiles of firm-year observations of
annual stock returns and net earnings
over market value of equity (see section 3.4 for the two
variable definitions) are eliminated from the
-
16
members of managers. Non-family owners include widely held
corporations, widely held
financial institutions, and the state. Although a company can
have more than one
ultimate owner, we focus on the largest ultimate owner. With the
highest level of voting
rights, the largest ultimate owner is more likely the
controlling owner of the firm than the
smaller owners.
As our definition of ownership relies on both cash flow and
voting control rights,
the cash flow rights that support the control by ultimate owners
need to be further
identified. Firm-specific information on pyramid structures,
cross-holdings, and
deviations from one-share-one-vote rules are used to make the
distinction between cash
flow and voting rights. To facilitate the measurement of the
separation of cash flow and
voting rights, we also set the maximum cash flow rights level
associated with any
ultimate owner at 50%.13
Table 1 reports descriptive statistics on the concentration of
the ultimate cash flow
and voting rights of the 1,350 corporations in the hands of the
largest controlling holder.
Broken down by economies, the sample covers 324 Hong Kong firms,
129 Indonesian
firms, 236 Korean firms, 207 Malaysian firms, 173 Singaporean
firms, 120 Taiwan firms,
and 161 Thai firms. The sample covers 40% of all publicly traded
firms in the region.14
From Panel A of the table, East Asian corporations exhibit high
levels of concentration of
control: the mean level of voting rights is 27%. This is in
contrast to U.S. firms studied
by most prior research, which are characterized by diffuse
ownership and control. For a
quarter of the East Asian companies, more than 40% of the voting
rights is in the hands
sample.
13 The minimum level of cash flow rights is allowed to be less
than five percent. 14 As of December 1996, the numbers of listed
firms in these economies were: 583 in Hong Kong,
267 in Indonesia, 760 in Korea, 621 in Malaysia, 266 in
Singapore, 382 in Taiwan, and 454 in Thailand.
-
17
of the largest block-holder. Thai firms display the most
concentrated voting rights, 35%
on average, followed by Indonesian companies (33%), Malaysian
and Hong Kong
companies (28%), Singaporean firms (26%), Taiwanese firms (21%),
and South Korean
firms (18%). The minimum level of voting rights is 5% across the
economies, with the
exception of South Korea, where a level of zero indicates the
existence of widely held
firms. The maximum level of voting rights is 50% across the
seven economies, due to
the 50% ceiling imposed.
Panel B reports the basic statistics for levels of cash flow
rights. The cash flow
rights patterns are similar to the voting rights patterns in
Panel A. The overall average
concentration is 23%. Note particularly that the mean levels of
cash flow rights are lower
than the corresponding level for voting rights in Panel A,
indicating the divergence
between cash flow and voting rights. In Panel C, we report the
basic statistics of the ratio
of cash flow rights over voting rights (CV). The ratio, by
definition, ranges between zero
and one. If a firm is widely held, i.e., zero cash flow and
voting rights, its CV ratio is set
to one. CV indicates the degree of divergence between cash flow
and voting rights. The
closer the ratio gets to zero, the larger the divergence. In
East Asia, the mean CV ratio is
0.86. The mean CV ratios are rather similar across the seven
East Asian economies,
ranging between 0.78 (Indonesia) and 0.95 (Thailand). Over a
quarter of the East Asian
firms display cash-vote divergence (CV
-
18
ownership data and the inability to trace some hidden control
chains would bias our
statistics downward. In addition, small firms tend to have more
concentrated ownership,
but our sample mainly consists of larger firms due to limited
coverage of ownership data
of small firms. Thus, our sample average ownership concentration
is lower than that of
the population. However, we expect that the understatement of
ownership data and the
large firm bias in our sample would weaken but not
systematically bias in favor of our
hypothesis. Notwithstanding the data limitation, it is
sufficient to conclude from Table 1
that the ownership and control structure in East Asia is highly
concentrated, in contrast to
the diffusely owned firms in the U.S. documented in prior
research. The East Asian firms
also differ from U.S. firms in that they are characterized by a
separation of ownership and
control resulting from the controlling owners possession of more
voting power than cash
investment.
Table 2 presents the fraction of firms in our sample that are
associated with family
ultimate owners. Seventy percent of the East Asian firms are
associated with family
owners. South Korea has the highest family control percentage
(79%), followed by
Indonesia (78%), Malaysia (77%), Thailand (74%), Hong Kong
(72%), Taiwan (59%),
and Singapore (50%). Combined with the evidence in Table 1, it
is evident that most of
the family owners have a high degree of voting control over the
firms. These statistics
suggest that families dominate other types of owners in
controlling the East Asian
corporations, as also reported in Claessens et al. (2000).
Table 3 compares the ownership structure of the largest ultimate
owners between
the family controlled and non-family controlled firms. Panel A
presents voting rights
statistics. In East Asia, family controlled firms have higher
levels of voting rights than
-
19
non-family controlled firms. The mean (median) level of voting
rights is 27% (30%) for
family firms and 25% (20%) for non-family firms. These
differences are statistically
significant in terms of both the mean and the median. Economy by
economy, the mean
and median differences in voting rights are generally positive,
with the exception of
Malaysia. However, the differences are significant for only
South Korean and
Singaporean firms.
Panel B presents cash flow rights statistics. Interestingly, the
overall mean level of
cash flow rights in East Asia is statistically significantly
lower for family-controlled firms
(22%) than for non-family controlled firms (24%) at the 5%
level. However, there is no
median difference between the two groups. Economy by economy,
lower family cash
flow rights levels are observed in five of the seven East Asian
economies. The negative
mean and median differences are statistically significant in
Hong Kong, Indonesia, and
Malaysia. Panel C of Table 3 compares the ratio of cash flow
rights over voting rights
between the two groups of firms. Family controlled firms in East
Asia are associated
with a higher degree of cash-vote divergence than are non-family
controlled firms. The
mean CV ratios are 0.84 and 0.93, respectively. The difference
is highly significant, with
difference in means and medians, both at the 1% level.
In summary, the comparison of ultimate ownership structure
reveals that family
controlled firms in East Asia on average have higher voting
rights, lower cash flow
rights, and greater separation of cash flow rights from voting
rights than non-family
controlled firms.
-
20
3.3. Correlation analysis
Theory suggests that the ownership and control structure of the
East Asian
corporations influence their ultimate owners ability and
incentive to expropriate minority
shareholders. We hypothesize that the risk of expropriation
affects the minority
shareholders perception of earnings credibility, and hence the
informativeness of the
firms reported earnings. We expect that the ability to
expropriate is positively associated
with the degree of control that the ultimate owners possess. On
the other hand, we expect
that the incentives to expropriate are positively associated
with the degree of separation
of cash flow rights from voting rights.
As an initial test of the earnings credibility hypothesis, we
compare the
informativness of earnings of two groups of firms with different
degrees of voting control
and cash-vote divergence, respectively. As a simple proxy for
the informativeness of
earnings, we use the Pearson correlation coefficient between
annual earnings scaled by
lagged market value of equity (NI) and cumulative
market-adjusted stock returns (CAR).
The correlation coefficient thus measures the level of the
informativeness of earnings.
CAR is the annual raw returns minus the annual market returns.
The annual returns are
continuously compounded from monthly stock returns starting from
twelve months
before the latest date, as required by law or stock exchange
listing rules, that the firm
discloses its annual report. The legally required deadline of
financial reporting is usually
several months after the firms fiscal year end: three months in
Korea and Thailand, four
months in Taiwan, and five months in Hong Kong, and six months
in Indonesia,
Malaysia, and Singapore. We therefore compute the CAR of each
firm using the twelve-
month window defined by the firms reporting deadline.
-
21
To test whether the level of ultimate owners voting rights,
which proxies for the
ultimate owners ability to expropriate minority shareholders,
reduces earnings credibility
and hence the informativeness of earnings, we divide the samples
by the level of voting
rights. From the basic statistics in Table 1, it is observed
that firms in the different
economies are associated with different levels of ultimate
voting control. This suggests
that the level of a firm's ultimate owner's voting rights is
affected by institutional factors
across economies. As argued by La Porta et al. (1999), a
country's legal system
determines the ownership structure of its firms. To take into
account these potential
cross-economy effects, we use the median level of voting rights
in the economy as a
cutoff to divide the sample firms into high voting control and
low voting control groups.15
Table 4 reports the results of the correlation analysis. All of
the correlation
coefficients are different from zero at one percent level of
significance. As reported in
the third row of Column 3 and 4, there is only a minor
difference in the earnings-return
correlation coefficients between the high and low voting control
groups in East Asia: 0.21
and 0.22, respectively. The argument that high voting control
alone leads to low
informativeness of earnings is only weakly supported.
To examine the effects of cash-vote divergence on earnings
informativeness, we
decompose the sample into two groups. Firms in the first group
are all associated with
cash-vote divergence, i.e., CV
-
22
cash-vote divergence have a lower earnings-return correlation
coefficient (0.15) than
firms with no cash-vote divergence (0.24). This evidence is
consistent with the view that
separation of ownership and control induces expropriation,
lowers earnings credibility,
and hence weakens the informativeness of earnings.
As a further test, we decompose the samples into four groups by
both voting
control and cash-vote divergence. To be consistent with the
earnings credibility
hypothesis, the weakest informativeness of earnings should be
observed in the group of
firms with both high voting control and great cash-vote
divergence because their ultimate
owners have the greatest incentive and ability to expropriate.
This is indeed the case.
From the first and second row of Columns 3 and 4, the smallest
earnings-return
correlation is recorded for the group of firms with high voting
control and large cash-vote
divergence.
We next apply this analysis to investigate the role of family
control in the
informativeness of earnings. It is arguable that family owners
have more ability to
expropriate than non-family owners. This argument is based on
the assumption that
family owners, as natural persons, are more capable of capturing
the benefits from
expropriation than are non-family owners, usually legal persons.
Given this ability to
expropriate, family control would be associated with weak
earnings credibility and hence
poor earnings informativeness. To investigate this possibility,
we divide our samples into
family control and non-family control groups and compare their
earnings-return
correlation coefficients. As reported in Row 3, Columns 5 and 6,
family controlled firms
have a higher correlation coefficient than non-family controlled
firms. Inconsistent with
our conjecture, family controlled firms tend to have higher
earnings informativeness than
-
23
non-family controlled firms. We further cut the samples into
four groups by both family
control status and cash-vote divergence. If family control and
cash-vote divergence
jointly induce expropriation, one would find the weakest
earnings-return correlation in
the high divergence and family control group. As reported in the
first row of Columns 5
and 6, conditional on the existence of cash-vote divergence,
there is little difference in the
earnings-return correlation between family controlled and
non-family controlled firms in
East Asia. In fact, the strongest earnings-return correlation is
found among the family
controlled firms whose ultimate owners have no cash-vote
divergence (CV=1).
To test whether our results are sensitive to the annual windows
used in calculating
CAR, we also calculate CAR using several fixed windows for all
firms in all economies:
nine (six) months before to three (six) months after the current
fiscal year-end. The
results remain the same.
In summary, we find that earnings informativeness, measured by
earnings-return
correlation, is negatively related to the voting control level
and the degree of separation
of cash flow from voting rights of the largest ultimate owners
of East Asian firms. The
evidence is consistent with the view that the ultimate owners'
ability and incentive to
expropriate minority shareholders are factors impacting earnings
credibility and
informativeness in East Asia. On the other hand, we do not find
that family control per
se affects earnings informativeness.
3.4. Regression analysis
We next perform regression analysis to examine the determinants
of earnings
informativeness in East Asia.
-
24
3.4.1. Basic relations between returns and earnings
Before we focus on the role of ownership structure, we perform a
set of ordinary
least squares regressions to determine the basic relations
between returns and earnings in
East Asia:
CARit = a0 + a1 NIit + (Fixed effects) + uit
where, for sample firm i and year t
CARit = cumulative net-of-market twelve-month stock return at
year t;
NIit = net earnings at year t divided by market value of equity
at the beginning of year t;
Fixed effects = dummy variables controlling for fixed effects of
calendar years and/or
economies;
uit = error term at year t.
The regressions are performed year by year, economy by economy,
and pooling all of the
years and economies. The results are reported in Appendix 1.
Because we generally
find heteroskedasticity problems in the regressions, we report
White-adjusted t-statistics
for all the coefficients. Fixed-effects of calendar years and/or
economies, where
appropriate, are included as dummy intercepts in the
regressions, but for simplicity they
are not reported in the table. Consistent with the correlation
analysis in Table 4, the
estimated coefficients of earnings (NI) are positive and
statistically significant across all
the years and economies, suggesting earnings has an information
role in East Asia.
-
25
3.4.2. The effects of ownership structure
We next test the informativeness of earnings conditional on
ownership structure
using the following pooled time-series cross-sectional
regression model:
CARit = a0 + a1 NIit + a2 NIit SIZEit + a3 NIit Qit + a4 NIit
LEVit + a5 NIit SEGi
+ a6 NIit OWNi + (Fixed effects) + uit
where, for sample firm i and year t
CARit = cumulative net-of-market twelve-month stock return at
year t;
NIit = net earnings at year t divided by market value of equity
at the beginning of year t;
SIZEit = natural logarithm of market value of equity in millions
of U.S. dollars at the
beginning of year t;
Qit = market value of equity divided by book value of total
assets at the beginning of year
t;
LEVit = total liability divided by total assets at the beginning
of year t;
SEGi = number of industry segment(s) in which the firm
operates;
OWNi = a set of variables that proxy for ultimate ownership
structure;
Fixed effects = dummy variables controlling for fixed effects of
calendar years and
economies;
uit = error term at year t.
-
26
We include market value of equity to book value of total assets
to control for the
effects of growth on the earnings-return relation.16 Growth
opportunities are likely to be
positively associated with future earnings levels and/or
earnings persistence (Collins and
Kothari, 1989), the higher the market to book assets, the larger
the expected earnings
growth and/or earnings persistence, the stronger the
earnings-returns relation.17 On the
other hand, market to book ratio may also be affected by firm
risk. High growth firms
may be more risky, which weakens the earnings-return relation.
Given these
countervailing effects, the net effect of growth on the
earnings-return relation is therefore
an empirical issue. We incorporate leverage in the regression
because it proxies for the
riskiness of debt or default risk (Dhaliwal et al., 1991).
Highly levered firms are
associated with high risk and hence their earnings-return
relation is weakened. However,
firms with higher leverage may be more subject to monitoring by
their creditors, and
hence have higher earnings-return sensitivity than other firms
with low leverage. In
addition, we include the number of industry segments in which
each sample firm operates
as another control in the anticipation that conglomerate firms,
due to their relatively more
complex earnings-generating process, may have weaker
earnings-return relations than
firms operating in a single industry.18 Finally, we include firm
size as a control for other
missing factors that affect the earnings-returns relation. For
example, prior literature on
the U.S. case (Atiase, 1985; Freeman, 1987) has documented that
the public disclosure
16 The use of market to book value of equity produces
qualitatively similar results in our
regressions. 17 We do not include a separate control for
earnings persistence because the earnings history is
inadequate for its empirical estimation in our sample. 18 The
1996 company segment data were collected from Worldscope and
supplemented with
additional data from the Asian Company Handbook. Since companies
report their segment data at different degrees of detail, we group
the companies segments according to the two-digit Standard Industry
Classification system.
-
27
and private development of non-earnings information are
increasing functions of firm
size.
We employ the ordinary least squares method to regress CAR on
the median
adjusted voting rights level (EV), the degree of separation
between cash flow and voting
rights (CV), and the control variables. Note that each firms EV
is adjusted using the
median voting rights of the economy in which it operates because
firms in different
economies are associated with different levels of ultimate
voting control. However, CV is
not median adjusted because it does not vary significantly
across the East Asian
economies (Table 1). The regressions are performed on the pooled
sample.
Equation (1) of Table 5 presents the regression result. As
before, we report White-
adjusted t-statistics for all the coefficients. We also omit
reporting fixed-effects of
calendar years and economies. Larger firms earnings are more
informative, as indicated
by the significantly positive estimated coefficient of NI*SIZE.
The coefficient of NI*Q
is insignificant in all regressions, suggesting that the risk
and the growth effects are offset
by each other. The estimated coefficient of NI*LEV is
significantly positive, suggesting
a monitoring role of creditors in reducing expropriation by the
ultimate owners and
enhancing earnings informativeness. The coefficient of NI*SEG is
significantly
negative, supporting the view that conglomerate firms report
less informative earnings
than more focused firms. The estimated coefficient of NI is
insignificantly different from
zero, suggesting that only the interaction terms of NI with the
control variables have the
explanatory power. As in regression results in Appendix 1, the
intercept in equation (1)
remains significantly negative. We suspect that the negative
intercepts are caused by the
omitted expected earnings component. When we include lagged
earnings as expected
-
28
earnings by replacing NI with the change in earnings (current
year earnings minus lagged
earnings all divided by lagged market value of equity) in all
our regressions, the
magnitude of the intercepts of the regressions drops by more
than half.
The focus of Table 5 is the role of ownership structure. In
Equation (1), we have
earnings interacting with the median adjusted voting rights (EV)
and the divergence
between cash flow and voting rights (CV). We find that the
estimated coefficient is
negative at the 5% level, suggesting that high voting control
rights has a negative effect
on earnings informativeness. If high voting control by ultimate
owners is associated with
a great ability to expropriate minority shareholders, our
evidence is consistent with the
view that uninformative earnings reported by East Asian firms at
least in part reflect low
credibility in ultimate owners reported earnings.
We now turn to investigate the effects of separation of cash
flow from voting
rights on earnings informativeness. Our hypothesis is that
larger cash-vote divergence
implies higher expropriation incentives, resulting in weaker
earnings credibility, which in
turn leads to lower earnings informativeness. CV, by definition,
is inversely related to
cash-vote divergence. To be consistent with the hypothesis, we
should observe a
significantly positive estimated coefficient of CV. As in
Equation (1), the coefficient of
CV is positive and statistically significant at the 5% level.
This evidence is consistent
with the earnings credibility hypothesis.
We next consider the joint effect of voting control and
cash-vote divergence. We
expect that voting control provides ability and cash-vote
divergence provides incentive to
ultimate owners to expropriate minority shareholders. To be
consistent with the earnings
credibility hypothesis, we should observe the lowest earnings
informativeness when
-
29
ultimate owners have both a high degree of control and cash-vote
divergence. We create
a dummy variable, DEV, which equals one when the voting rights
level is greater or
equal to the median in the economy, and zero otherwise. In the
regression, we include an
interaction term, NI*CV*DEV, in addition to NI*EV and NI*CV. To
be consistent with
the hypothesis, we should observe a significantly positive
estimated coefficient of this
interaction term. From Equation (2) of Table 5, we find this is
indeed the case, with the
White-adjusted t-statistic for NI*CV*DEV significant at the 1%
level. The coefficient of
NI*EV remains significantly negative. The coefficient of NI*CV
becomes insignificant,
suggesting that the effect of CV on earnings informativeness
primarily occurs when the
voting control level is high. The estimated coefficients of
earnings (NI) and the
controlled variables are all of the same signs and significance
levels as reported earlier in
Equation (1).
In summary, we report that cash-vote divergence weakens
earnings
informativeness in East Asia. The effect is magnified if the
voting control level is high.
This evidence confirms the results from the correlation analysis
in Table 4 and lends
support to the earnings credibility hypothesis.
3.4.3. Checks of robustness
Since all voting and cash flow rights that exceed 50% are
capped, the effects of
any variation in voting and cash flow rights of these firms
would not be captured by our
measure. Moreover, if actual voting and cash flow rights both
exceed 50%, their
divergence would not be captured by the CV measure, as it would
be recorded as one
which indicates no divergence. As a sensitivity test of any
possible bias in our results, we
rerun the regression by excluding observations associated with
voting rights equal to or
-
30
more than 50% from our sample. As reported in Equation (3) of
Table 5, the signs of the
coefficients of NI*EV and NI*CV*DEV remain the same, their
magnitudes are reduced
from those estimated from the full sample but remain significant
at the 5 percent level.
Although the exclusion of firms whose controlling owners have
the highest voting power
and hence the highest ability to expropriate weakens the
results, the ownership
coefficients remain statistically significant. We thus confirm
that the 50% ceiling for our
ownership data would not have biased in favor of our
hypothesis.
As further diagnostic checks19, we have estimated the regression
models using
cumulative abnormal returns calculated from two fixed annual
windows: 9 (6) months
prior to 3 (6) months after the current fiscal year end. The
empirical results are not
sensitive to the various annual windows employed to calculate
CARs. We have also used
cumulative raw returns, instead of net-of-market returns, and
used a two-year cumulative
net-of-market return, starting 21 months before to three months
after the fiscal year-end,
as an alternative dependent variable. The two-year return, which
includes both current
and lagged-year returns, attempts to adjust for any differences
in price efficiency in
capturing future earnings between highly concentrated (high EV)
and less concentrated
ownership (low EV) firms (Jacobson and Aaker, 1993; Ali and
Hwang, 2000). Our
results for EV and CV remain qualitatively the same after using
these alternative
dependent variables. Instead of using NI in our regression
model, we have also used NI,
change in earnings (current earnings minus lagged earnings all
divided by lagged market
value of equity). The coefficient of NI*EV remains negative and
statistically
19 The results of the diagnostic checks are not reported in
tables but are available upon request.
-
31
significant, while the coefficient of NI*CV*DEV is positive and
statistically significant.
In addition, the coefficient of NI*CV remains statistically
significant.
We provide a further test of whether or not the effects of
voting control and cash-
vote divergence on East Asian firms cluster in time and/or
economies. Table 6 presents
the results of a set of year-by-year regressions. These
regressions include NI*EV,
NI*CV, and NI*CV*DEV, in addition to the control variables. We
find that the
coefficients of NI*EV, NI*CV, and NI*CV*DEV are mostly of the
expected signs, and
NI*EV and NI*CV*DEV are statistically significant in three
annual samples: 1991,
1993, and 1994. Table 7 presents the results of a set of
economy-by-economy
regressions using the same model. We find ownership effects in
several economies. The
effect of voting control (EV) on earnings informativeness is
negative and significant in
Hong Kong, Indonesia, and Taiwan. The effect of CV is positive
and significant in Hong
Kong, and Thailand. The joint effect of EV and CV is
significantly positive in Indonesia.
From the year-by-year and economy-by-economy results, we do not
find that the effects
of the ownership variables are concentrated in any given year or
economy.
The above diagnostic checks have demonstrated that our empirical
results are
robust to the measurement bias in the ownership variables, and
the various specifications
of cumulative stock returns and earnings. In addition, the
ownership effects are generally
found in our sample, not just in any single year or economy.
3.4.4. The role of family control
Given that family ownership dominates other types of ownership
in the control of
East Asian corporations, we are interested in learning whether
family control affects the
informativeness of accounting earnings. We include in the
regression a dummy variable,
-
32
FAMILY, which equals one if a firm is associated with an
ultimate family owner, and
zero otherwise. In the regression, we allow NI, EV, CV, and
CV*DEV each to interact
with FAMILY. Equation (1) of Table 8 presents the regression
result. The result shows
an insignificant coefficient of NI*FAMILY, suggesting that
family control per se does
not affect earnings informativeness in East Asia. The result is
consistent with the
correlation analysis in Table 4. In the same regression, the
coefficient of
NI*FAMILY*EV is negative and significant at the 1% level. The
coefficient of
NI*FAMILY*CV is positive but insignificant, while the
coefficients of
NI*FAMILY*CV*DEV is positive and significant at the 5% level.
The evidence
suggests that given family ownership, CV is positively
associated with earnings
informativeness, especially when controlling owners possess high
voting control. As a
numerical example, if EV=0 and CV=1, family control has a net
neutral effect (-0.20 - 0
+ 0.24) on earnings informativeness. That is, if an ultimate
family owner avoids
excessive voting control and sets her voting rights equal to her
cash investments, family
control does not reduce earnings informativeness. On the other
hand, family control
hurts earnings informativeness if the controlling family
possesses excessive voting
control (EV>1) and creates divergence between cash flow and
voting rights (CV
-
33
regression results of the family and non-family controlled
sub-samples are reported in
Equations (2) and (3) of Table 8, respectively. From Equation
(2), control rights has a
negative effect on earnings informativeness for family
controlled firms, as indicated by
the negative and significant estimated coefficient of NI*EV.
Separation between cash
flow and voting rights alone does not affect earnings
informativeness, as indicated by the
insignificant coefficient of NI*CV. However, when controlling
families possess high
voting control, separation of cash flow and voting rights reduce
earnings informativeness,
as indicated by the significantly positive coefficient of
NI*CV*DEV. Although we find
ownership structure affects family firms earnings-returns
relations, we fail to find the
same effects among non-family controlled firms. From Equation
(3), none of the
estimated coefficients of NI*EV, NI*CV, and NI*CV*DEV for the
non-family sub-
sample are statistically significant. The overall results
suggest that earnings
informativeness of the family controlled firms is more sensitive
to the ownership
structure than that of non-family controlled firms. One
explanation is that non-family
ultimate owners, such as financial institutions, are unlikely to
consume the expropriation
benefits, hence their earnings credibility and informativeness
are insensitive to ownership
structures. Related to this explanation, it is interesting to
note the effect of leverage.
Recall that we generally find that leverage has a positive
effect on earnings
informativeness. We find here in Table 8 that leverage
positively affects the earnings
informativeness of family controlled firms, but not non-family
controlled firms. This
result further supports the argument that creditors play a
monitoring role that works
against the expropriation by the ultimate owners, which enhances
the informativeness of
accounting earnings of family controlled firms. In Table 8 we
also find that firm size
-
34
positively affects the earnings informativeness of family
controlled but not non-family
controlled firms. This may suggest that large family firms
command a positive effect on
earnings credibility that is absent among small family
firms.
4. Conclusion
The Asian Financial Crisis has caused many East Asian economies
to re-examine
the adequacy of their corporate financial reporting. Poor
accounting information and its
associated high capital cost pose a potential threat to the
competitiveness of East Asian
firms. Despite efforts to improve corporate transparency by
imposing new accounting and
disclosure rules in East Asia, the perception is that the
financial reporting credibility of
corporations remains low. Before prescribing a cure for low
corporate transparency
among East Asian firms, it is important to find out why reported
corporate financial
information has low credibility.
We hypothesize that the high share ownership concentration and
the large
separation of cash flow and voting rights, which are common in
East Asia, weaken the
credibility of reported earnings to outside investors, and hence
lower the informativeness
of the accounting earnings. The earnings credibility is weakened
because minority
shareholders anticipate that the ownership structure gives the
controlling owners both the
ability and incentive to manipulate earnings for outright
expropriation or to report
uninformative earnings to avoid detection of their expropriation
activities. Our empirical
results do show that the ultimate owners voting rights and the
separation of cash flow
and voting rights have a negative effect on earnings
informativeness, as measured by the
earnings-return relation. We also find that family control per
se does not lower earnings
-
35
informativeness. Only firms with controlling families that
maintain high voting rights and
a large voting-control divergence report less informative
earnings. Our results are robust
to controls for firm size, market-to-book asset, leverage,
number of industry segments
operated by the firm, and to varying the starting and ending
dates of the stock return
window.
The analysis of East Asian corporations allows us to study the
subject of earnings
informativeness in a different ownership context from that of
the research on the U.S.
Because of the concentrated share ownership in East Asia,
conflicts of interest in East
Asian corporations are primarily between controlling and
minority shareholders, not
between managers and shareholders as in the case of U.S. firms.
As we have
demonstrated in this paper, the different nature of agency
conflicts creates different
effects on earnings informativeness. Our research results are
also rich in policy
implications. In general, these results support Ball, Kothari,
and Robin (1998) by finding
that policy makers should consider a countrys overall
institutional environment before
prescribing a comprehensive set of rules and regulations for
corporate reporting. Also, it
is important for policy makers and regulators to understand how
the concentrated share
ownership structure in East Asia provides perverse incentives
for managers to reduce
accounting information quality. Lastly, the paper illustrates
that it would be fruitful for
future research to focus on how ownership structures shape
accounting policies in
emerging markets and transition economies.
-
36
References: Aganin, Alexander, and Paolo Volpin, 1998, On the
Origin and Evolution of Pyramidal
Groups: An Empirical Evaluation, Harvard University, mimeo.
Alford, Andrew, Jennifer Jones, Richard Leftwich, and Mark
Zmijewski, 1993, The
Relative Informativeness of Accounting Disclosures in Different
Countries, Journal of Accounting Research, 31, Supplement,
183-223.
Ali, Ashiq, and Lee Seok Hwang, 2000, Country-Specific Factors
Related to Financial
Reporting and the Value Relevance of Accounting Data, Journal of
Accounting Research, forthcoming.
Asian Wall Street Journal, Business Transparency in Region Has
Worsened, Survey
Shows. (November 24, 1999). Atiase, R.K., 1985, Predisclosure
Information, Firm Capitalization and Security Price
Behavior around Earnings Announcements, Journal of Accounting
Research, Spring, 21-36.
Backman, Michael, 1999, Asian Eclipse: Exposing the Dark Side of
Business in Asia,
Singapore: John Wiley & Sons (Asia).
Ball, Ray, S. P. Kothari, and Ashok Robin, 1998, The Effect of
Institutional Factors on Properties of Accounting Earnings:
International Evidence, Working Paper, University of Rochester and
Rochester Institute of Technology.
Ball, Ray, Ashok Robin, and Joanna Wu, 1999, Properties of
Accounting Earnings in
Four East Asian Countries, Working Paper, University of
Rochester and Rochester Institute of Technology.
Bebchuk, Lucian, Reinier Kraakman and George Triantis, 1999,
Stock Pyramids, Cross-
Ownership, and Dual Class Equity: The Creation and Agency Costs
of Separating Control From Cash Flow Rights, Working Paper No.6951,
NBER.
Becht, Marco, and Ariane Chapelle, 1999, Ownership and Control
in Belgium, Solvay
Business School, Brussels, mimeo. Berle, Adolf A., and Gardiner
C. Means, 1932, The Modern Corporation & Private
Property, NJ: Transaction Publishers, 1991, originally published
by Hartcourt, Brace & World.
Bianchi, Marcello, Magda Bianco, and Luca Enriques, 1997,
Ownership, Pyramidal
Groups and Separation between Ownership and Control in Italy,
Working Paper, European Corporate Governance Network.
-
37
Bloch, Laurence, and Elizabeth Kremp, 1999, Ownership and
Control in France, INSEAD Business School, mimeo.
Burkart, Mike, Denis Gromb, and Fausto Panunzi, 1997, Large
Shareholders,
Monitoring, and the Value of the Firm, Quarterly Journal of
Economics, 112: 693-728.
Burkart, Mike, Denis Gromb, and Fausto Panunzi, 1998, Why High
Takeover Premia
Protect Minority Shareholders, Journal of Political Economy,
106, 1: 172-204. Bushman, Robert, Qi Chen, Ellen Engel, and Abbie
Smith, 1999, The Sensitivity of
Corporate Governance Systems to the Timeliness of Accounting
Earnings, Working Paper, University of North Carolina and the
University of Chicago.
Claessens, Stijn, Simeon Djankov, Joseph P.H. Fan, and Larry
H.P. Lang, 1999,
Expropriation of Minority Shareholders in East Asia, Working
Paper, World Bank.
Claessens, Stijn, Simeon Djankov, and Larry H.P. Lang, 2000, The
Separation of
Ownership and Control in East Asian Corporations, Journal of
Financial Economics, forthcoming.
Collins, D. W., and S.P. Kothari, 1989, An Analysis of
Intertemporal and Cross-sectional Determinants of Earnings Response
Coefficient. Journal of Accounting and Economics, 11: 143-81.
De Jong, Abe, Rezaul Kabir, Teye Marra, and Ailsa Roell, 1999,
Ownership and
Control in the Netherlands, Tilburg University, mimeo.
Dhaliwal, Dan, Kyung Lee and Neil Fargher, 1991, The Association
between Unexpected Earnings and Abnormal Security Returns in the
Presence of Financial Leverage, Contemporary Accounting Research,
8: 20-41.
Enriques, Luca, 1998, Pyramid Groups, Intra-Group Operations,
and the Separation
between Ownership and Control in Italy, forthcoming in
Giurizpridenzia Commerciali.
Fisman, Raymond, 1999, Its Not What You Know Estimating the
Value of Political
Connections, Quarterly Journal of Economics, forthcoming.
Freeman, Robert, 1987, The Association between Accounting
Earnings and Security Returns for Large and Small Firms, Journal of
Accounting and Economics, 9: 195-228.
-
38
Gaver, Jennifer, Kenneth Gaver, and Jeffrey Austin, 1995.
Additional Evidence on the Association between Income Management
and Earnings-based Bonus Plans. Journal of Accounting and
Economics, 19: 3-29.
Gugler, Klaus, Alex Stomper, and Josef Zechner, 1999, The
Separation of Ownership
and Control: An Austrian Perspective, University of Vienna,
mimeo. Healy, Paul, 1985, The Effect of Bonus Schemes on Accounting
Choices, Journal of
Accounting and Economics, 7: 85-107. Holthausen, Robert, David
Larcker, and Richard Sloan, 1995. Annual Bonus Schemes
and the Manipulation of Earnings, Journal of Accounting and
Economics, 19: 29-74.
Jacobson, Robert, and David Aaker, 1993, Myopic Management
Behavior with
Efficient, but Imperfect, Financial Markets, Journal of
Accounting and Economics, 16, 383-405.
Jensen, Michael C., and William H. Meckling, 1976, Theory of the
Firm: Managerial
Behavior, Agency Costs and Ownership Structure, Journal of
Financial Economics, 3, 305-360.
Johnson, Simon, Peter Boone, Alasdair Breach, and Eric Friedman,
2000, Corporate
Governance in the Asian Financial Crisis, Journal of Financial
Economics, forthcoming.
Johnson, Simon, Rafael La Porta, Florencio Lopez de Silanes, and
Andrei Shleifer, 2000,
Tunnelling, American Economics Review, forthcoming. La Porta,
Rafael, Florencio Lopez-De-Silanes, and Andrei Shleifer, 1999,
Corporate
Ownership Around the World, Journal of Finance, 54, 471-518. La
Porta, Rafael, Florencio Lopez-De-Silanes, and Andrei Shleifer,
1999, Investor
Protection and Corporate Valuation, NBER Working Paper 7403.
Melis, Andrea, 1998, Corporate Governance in Europe: An empirical
analysis of the
Italian case among non financial listed companies, Working
Paper, Universit di Cagliari.
Nenova, Tatiana, 1999, The Value of a Corporate Vote and Private
Benefits: Cross-
Country Analysis, Working Paper, Harvard University.
Pope, Peter, Ken Peasnell, and Steven Young, 1999, Outside
Directors, Board Effectiveness, and Earnings Management, Working
Paper, University of Lancaster.
-
39
Rajgopal, Shivaram, Mohan Venkatachalam, and James Jiambalvo,
1999, Is Institutional Ownership Associated with Earnings
Management and the Extent to which Stock Prices Reflect Future
Earnings? Working Paper, University and Washington and Stanford
University.
Roe, Mark J., 1994, Strong Managers, Weak Owner The Political
Roots of American
Corporate Finance, Princeton, NJ: Princeton University Press.
Scott, Kenneth, 1999, Corporate Governance and East Asia, Working
Paper No 176,
Stanford Law School.
Sender, Henny, 1999, Smoke and Mirrors, Far Eastern Economic
Review, September 30, 34-38.
Shleifer, Andrei and Robert Vishny, 1997, A Survey of Corporate
Governance, Journal
of Finance, 52: 737-783. Teoh, Siew T. and T.J. Wong, 1993.
Perceived Auditor Quality and the Earnings
Response Coefficients, The Accounting Review, 68: 346-67.
Warfield, T., J.J. Wild, and K. Wild, 1995, Managerial Ownership,
Accounting Choices,
and Informativeness of Earnings, Journal of Accounting and
Economics, 61-91. Watts, Ross and Jerold Zimmerman, 1986, Positive
Theory of Accounting, Prentice-Hall,
Inc. Wolfenzon, Daniel, 1999, A Theory of Pyramidal Structures,
Harvard University,
mimeo, February.
World Bank, 1998, The Road to Recovery: East Asia after the
Crisis, Oxford University Press.
-
40
Table 1
Ultimate ownership structure of the largest block shareholders
of East Asian corporations
Firm No. Mean Std. Err. Q1 Median Q3 Minimum Maximum
Panel A: Voting rights Hong Kong 324 0.28 0.11 0.20 0.20 0.40
0.05 0.50 Indonesia 129 0.33 0.11 0.30 0.30 0.40 0.05 0.50 Korea
(South) 236 0.18 0.10 0.10 0.20 0.30 0.00 0.50 Malaysia 207 0.28
0.11 0.20 0.30 0.30 0.05 0.50 Singapore 173 0.26 0.10 0.20 0.30
0.30 0.05 0.50 Taiwan 120 0.21 0.09 0.10 0.20 0.30 0.05 0.50
Thailand 161 0.35 0.13 0.30 0.40 0.50 0.05 0.50
East Asia 1350 0.27 0.12 0.20 0.30 0.40 0.00 0.50
Panel B: Cash flow rights
Hong Kong 324 0.24 0.11 0.19 0.20 0.30 0.03 0.50 Indonesia 129
0.25 0.11 0.18 0.24 0.32 0.03 0.50 Korea (South) 236 0.16 0.09 0.10
0.14 0.20 0.00 0.50 Malaysia 207 0.24 0.12 0.13 0.20 0.30 0.05 0.50
Singapore 173 0.21 0.11 0.12 0.20 0.30 0.04 0.50 Taiwan 120 0.18
0.09 0.10 0.20 0.22 0.04 0.50 Thailand 161 0.33 0.13 0.20 0.30 0.40
0.04 0.50
East Asia 1350 0.23 0.12 0.12 0.20 0.30 0.00 0.50
Panel C: Cash flow over voting rights
Hong Kong 324 0.88 0.21 0.80 1.00 1.00 0.13 1.00 Indonesia 129
0.78 0.24 0.64 0.86 1.00 0.13 1.00 Korea (South) 236 0.91 0.19 1.00
1.00 1.00 0.13 1.00 Malaysia 207 0.85 0.21 0.73 1.00 1.00 0.16 1.00
Singapore 173 0.79 0.21 0.60 0.80 1.00 0.26 1.00 Taiwan 120 0.85
0.19 0.73 1.00 1.00 0.40 1.00 Thailand 161 0.95 0.15 1.00 1.00 1.00
0.20 1.00
East Asia 1350 0.86 0.21 0.76 1.00 1.00 0.13 1.00
The ownership data is taken as of December 1996 or the fiscal
year-end 1996. The primary data source is Worldscope, supplemented
by other sources as detailed in Claessens, Djankov, and Lang
(2000). An ultimate owner is defined as the shareholder who has at
least 5 percent of voting rights of the firm and who is not
controlled by anybody else. Although a firm can have more than one
ultimate owner, only the one with the largest voting rights is
reported. The voting rights level of an ultimate owner is set at
50% and not traced any further once that level exceeds 50%. The
maximum cash flow rights level associated with any ultimate owner
is also capped at 50%. The minimum level of cash flow rights is
allowed to be less than 5%.
-
41
Table 2
Fraction of East Asian corporations controlled by families No.
of firms Families Non-families
Hong Kong 324 0.72 0.28 Indonesia 129 0.78 0.22 Korea (South)
236 0.79 0.21 Malaysia 207 0.77 0.23 Singapore 173 0.50 0.50 Taiwan
120 0.59 0.41 Thailand 161 0.74 0.26
East Asia 1350 0.70 0.30
The data is taken as of December 1996 or the fiscal year-end
1996. The primary data source is Worldscope, supplemented by other
sources as detailed in Claessens, Djankov, and Lang (2000). A firm
is classified as family controlled if the largest ultimate owner of
the firm is an individual who is a manager or a family member of
the firm's manager. Non-family owners include widely held
corporations, widely held financial institutions, and the
state.
-
42
Table 3
Comparison of ultimate ownership structure of the largest block
shareholders between family and non-family controlled firms in East
Asia
Mean Median Family
controlled Non-family controlled
T-statistics for difference
Family controlled
Non-family controlled
Z-statistics for difference
Panel A: Voting rights
Hong Kong 0.28 0.27 0.58 0.20 0.30 -0.59 Indonesia 0.33 0.33
0.00 0.30 0.30 -0.08 Korea (South) 0.19 0.13 3.32*** 0.20 0.10
4.45*** Malaysia 0.27 0.30 -1.22 0.30 0.30 -1.03 Singapore 0.28
0.25 1.78* 0.30 0.20 1.83* Taiwan 0.21 0.20 0.75 0.20 0.20 1.10
Thailand 0.35 0.33 0.88 0.40 0.30 0.98 East Asia 0.27 0.25 2.12**
0.30 0.20 2.55***
Panel B: Cash flow rights Hong Kong 0.23 0.26 -2.31** 0.20 0.20
-2.32** Indonesia 0.23 0.31 -2.79*** 0.24 0.30 -2.56*** Korea
(South) 0.17 0.13 2.12** 0.19 0.10 3.54*** Malaysia 0.22 0.28
-2.73*** 0.20 0.30 -2.36** Singapore 0.21 0.20 0.41 0.20 0.20 0.52
Taiwan 0.17 0.19 -1.02 0.16 0.20 -0.47 Thailand 0.32 0.33 -0.34
0.30 0.30 -0.28 East Asia 0.22 0.24 -2.01** 0.20 0.20 -1.30
Panel C: Cash flow rights over voting rights Hong Kong 0.84 0.97
-6.89*** 1.00 1.00 -5.10*** Indonesia 0.73 0.96 -6.68*** 0.80 1.00
-5.21*** Korea (South) 0.89 0.95 -2.12** 1.00 1.00 -1.98** Malaysia
0.83 0.95 -4.88*** 1.00 1.00 -4.08*** Singapore 0.75 0.82 -2.29**
0.80 0.90 -2.08** Taiwan 0.79 0.93 -4.10*** 0.80 1.00 -4.35***
Thailand 0.93 1.00 -4.20*** 1.00 1.00 -2.73*** East Asia 0.84 0.93
-8.85*** 1.00 1.00 -7.74*** The data is taken as of December 1996
or the fiscal year-end 1996. The primary data source is Worldscope,
supplemented by other sources as detailed in (Claessens, Djankov,
and Lang, 2000). An ultimate owner is defined as the shareholder
who has at least 5 percent of voting rights of the a firm and who
is not controlled by anybody else. The voting rights level of an
ultimate owner is set at 50% and not traced any further once that
level exceeds 50%. The maximum cash flow rights level associated
with any ultimate owner is also capped at 50%. The minimum cash
flow rights is allowed to be less than 5%. A firm is classified as
family controlled if the largest ultimate owner of the firm is an
individual who is a manager or a family member of the firm's
manager. Asterisks denote significance levels: *** 1%; ** 5%; *
10%.
-
43
Table 4
Correlations between earnings and stock returns: the role of
ultimate voting rights level, family control, and cash-vote
divergence
Full sample V>median in economy V
-
44
Table 5 Regression results of the effects of ultimate voting
rights and cash-vote divergence on
earnings informativeness CARit = a0 + a1NIit + a2NIit*SIZEit +
a3NIit*Qit + a4NIit*LEVit + a5NIit*SEGi + a6NIit*OWNi + (Fixed
effects) +
uit (1) (2) (3) Intercept -0.17*** -0.17*** -0.17*** (-12.15)
(-12.38) (-12.29) NI -0.59 -0.52 -0.63 (-0.24) (-0.16) (-1.36)
NI*SIZE 0.10*** 0.10*** 0.12*** (3.01) (2.94) (3.09) NI*Q -0.07
-0.07 -0.09 (-0.83) (-0.88) (-1.07) NI*LEV 0.59*** 0.61*** 0.58***
(2.80) (2.89) (2.68) NI*SEG -0.09*** -0.09*** -0.11*** (-2.86)
(-3.04) (-3.24) NI*EV -0.94** -2.04*** -1.62** (-2.27) (-3.89)
(-2.03) NI*CV 0.45** 0.10 0.19 (2.40) (0.44) (0.23) NI*CV*DEV
0.50*** 0.45** (2.71) (2.09) Adj-Rsq 0.10 0.10 0.10 Obs. 4490 4490
4022 The sample includes firms from seven East Asian economies:
Hong Kong, Indonesia, South Korea, Malaysia, Singapore, Taiwan, and
Thailand. The ordinary least squares method is employed in the
regressions. CAR is cumulative net-of-market twelve-month stock
return at year t. The annual returns are continuously compounded
from monthly stock returns starting from twelve months before the
latest date, as required by law or listing rules, that the firm
discloses its annual report. NI is net earnings at year t divided
by market value of equity at the beginning of year t. SIZE is
natural logarithm of market value of equity in millions of U.S.
dollar at the beginning of year t. Q is market value of equity
divided by the book value of total assets at the beginning of year
t. LEV is total liability divided by total assets at the beginning
of year t. SEG is the number of industry segment(s). OWN is a set
of ownership variables: EV, CV, and DEV. EV is the voting rights
level adjusted for the median in the economy. CV is the cash flow
rights over voting rights. DEV is a dummy variable equal to one if
EV exceeds zero, or else zero. Fixed-effects of calendar years and
economies are included in the regressions but not reported.
Financial and stock return data are from the PACAP database
covering 1991 through 1995. Ownership and segment data are
primarily from Worldscope as of December 1996 or fiscal year-end
1996. White-adjusted t-statistics are in parentheses. Asterisks
denote significance levels: *** 1%; and ** 5%.
-
45
Table 6 Determinants of East Asian firms' earnings
informativeness: year-by-year regression
results CARit = a0 + a1NIit + a2NIit*SIZEit + a3NIit*Qit +
a4NIit*LEVit + a5NIit*SEGi + a6NIit*OWNi + (Fixed effects) +
uit
1991 1992 1993 1994 1995 Intercept -0.27*** 0.01 -0.25***
-0.24*** -0.12*** (-8.53) (0.18) (-8.54) (-12.36) (-7.38) NI -1.87*
1.75* -2.34** -0.49 0.02 (-1.74) (1.73) (-1.97) (-0.61) (0.03)
NI*SIZE 0.31*** -0.17** 0.30*** 0.07 0.00 (3.87) (-1.99) (3.69)
(1.09) (0.02) NI*Q -0.32* -0.39* -0.19 0.18 0.30* (-1.72) (-1.82)
(-1.20) (1.13) (1.80) NI*LEV 0.21 0.71 -0.25 0.55 1.87*** (0.51)
(1.16) (-0.54) (1.36) (4.31) NI*SEG -0.16** 0.01 -0.16*** -0.02
-0.10* (-2.30) (0.20) (-2.49) (0.35) (-1.83) NI*EV -3.24*** -1.32
-1.56 -3.52*** -0.52 (-2.89) (-1.09) (-1.41) (-3.36) (-1.31) NI*CV
0.05 0.64 0.09 -0.09 0.03 (0.10) (1.22) (0.17) (-0.24) (0.08)
NI*CV*DEV 0.85** 0.29 0.68* 0.86** -0.09 (1.93) (0.69) (1.88)
(2.32) (-0.25) Adj-Rsq 0.21 0.06 0.12 0.18 0.13 Obs. 802 880 948
989 871
The sample includes firms from seven countries: Hong Kong,
Indonesia, South Korea, Malaysia, Singapore, Taiwan, and Thailand.
The ordinary