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: pjfan@ust.hk

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: actjwong@ust.hk

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.

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

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

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

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 principalagent 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.

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.

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

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

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.

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 profitmaximizing 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

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.
6
Given 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).

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.

10

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

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.

11

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.

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.

12

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

13

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).

14

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 nonfamily 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

15

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.
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.
14

16

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<1). The minimum CV ratio is 0.13. The Pearson
correlation coefficients between the voting rights and CV is only -0.07, suggesting that
CV does not simply proxy for voting rights.
The actual degree of ownership concentration in East Asia should be even higher
than the statistics reported here for several reasons. First, the 50% ceiling for the

17

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

18

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.

19

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 twelvemonth window defined by the firms reporting deadline.

20

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<1. Firms in the second group are associated with no cashvote divergence, i.e., CV=1. We then calculate Pearson correlation coefficients for each
of the sub-samples and report their values in the second column of Table 4. Firms with

15

The median adjustment is motivated by the cross-economy variations in company and securities
laws that determine the ownership level needed to exercise effective control. For example, in Korea,
restrictions on voting rights of institutional investors in listed companies and a requirement of 30% of the
vote to file class action suits suggest that a small percentage of voting rights can result in effective control.

21

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

22

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.

23

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.

24

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.

25

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.

26

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 Whiteadjusted 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

27

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

28

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

29

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.

30

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 cashvote 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,

31

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.
NI*FAMILY*CV

is

positive

but

insignificant,

while

The coefficient of

the

NI*FAMILY*CV*DEV is positive and significant at the 5% level.

coefficients

of

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<1).
We next ask a different question: do the effects of EV and CV on East Asian
firms earnings informativeness documented in Table 5 only occur among familycontrolled firms? To address this question, we divide the sample of East Asian firms into
the family controlled and the non-family controlled sub-samples and perform two
different regressions on the two sub-samples separately. The model allows the earnings
variable (NI) to interact with EV, CV, CV*DEV, and the set of controlled variables. The

32

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 subsample 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

33

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

34

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.

35

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, CrossOwnership, 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.

36

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 Crosssectional 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.

37

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: CrossCountry 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.

38

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.

39

Table 1
Ultimate ownership structure of the largest block shareholders of East Asian corporations
Firm No.

Mean

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

324
129
236
207
173
120
161

0.28
0.33
0.18
0.28
0.26
0.21
0.35

East Asia

1350

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

Std. Err.

Q1

Median

Q3

Minimum

Maximum

Panel A: Voting rights

0.11
0.20
0.11
0.30
0.10
0.10
0.11
0.20
0.10
0.20
0.09
0.10
0.13
0.30

0.20
0.30
0.20
0.30
0.30
0.20
0.40

0.40
0.40
0.30
0.30
0.30
0.30
0.50

0.05
0.05
0.00
0.05
0.05
0.05
0.05

0.50
0.50
0.50
0.50
0.50
0.50
0.50

0.27

0.12

0.30

0.40

0.00

0.50

324
129
236
207
173
120
161

0.24
0.25
0.16
0.24
0.21
0.18
0.33

Panel B: Cash flow rights

0.11
0.19
0.20
0.11
0.18
0.24
0.09
0.10
0.14
0.12
0.13
0.20
0.11
0.12
0.20
0.09
0.10
0.20
0.13
0.20
0.30

0.30
0.32
0.20
0.30
0.30
0.22
0.40

0.03
0.03
0.00
0.05
0.04
0.04
0.04

0.50
0.50
0.50
0.50
0.50
0.50
0.50

East Asia

1350

0.23

0.12

0.30

0.00

0.50

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

324
129
236
207
173
120
161

0.88
0.78
0.91
0.85
0.79
0.85
0.95

1.00
1.00
1.00
1.00
1.00
1.00
1.00

0.13
0.13
0.13
0.16
0.26
0.40
0.20

1.00
1.00
1.00
1.00
1.00
1.00
1.00

East Asia

1350

0.86

1.00

0.13

1.00

0.20

0.12

0.20

Panel C: Cash flow over voting rights

0.21
0.80
1.00
0.24
0.64
0.86
0.19
1.00
1.00
0.21
0.73
1.00
0.21
0.60
0.80
0.19
0.73
1.00
0.15
1.00
1.00
0.21

0.76

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%.

40

Table 2
Fraction of East Asian corporations controlled by families
No. of firms

Families

Non-families

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

324
129
236
207
173
120
161

0.72
0.78
0.79
0.77
0.50
0.59
0.74

0.28
0.22
0.21
0.23
0.50
0.41
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.

41

Table 3
Comparison of ultimate ownership structure of the largest block shareholders between
family and non-family controlled firms in East Asia
Family
controlled

Mean
Non-family
controlled

T-statistics for
difference

Family
controlled

Median
Non-family
controlled

Z-statistics for
difference

0.20
0.30
0.20
0.30
0.30
0.20
0.40

0.30
0.30
0.10
0.30
0.20
0.20
0.30

-0.59
-0.08
4.45***
-1.03
1.83*
1.10
0.98

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

0.28
0.33
0.19
0.27
0.28
0.21
0.35

0.27
0.33
0.13
0.30
0.25
0.20
0.33

Panel A: Voting rights

0.58
0.00
3.32***
-1.22
1.78*
0.75
0.88

East Asia

0.27

0.25

2.12**

0.30

0.20

2.55***

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

0.23
0.23
0.17
0.22
0.21
0.17
0.32

0.26
0.31
0.13
0.28
0.20
0.19
0.33

Panel B: Cash flow rights

-2.31**
-2.79***
2.12**
-2.73***
0.41
-1.02
-0.34

0.20
0.24
0.19
0.20
0.20
0.16
0.30

0.20
0.30
0.10
0.30
0.20
0.20
0.30

-2.32**
-2.56***
3.54***
-2.36**
0.52
-0.47
-0.28

East Asia

0.22

0.24

-2.01**

0.20

0.20

-1.30

Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Thailand

0.84
0.73
0.89
0.83
0.75
0.79
0.93

1.00
1.00
1.00
1.00
0.90
1.00
1.00

-5.10***
-5.21***
-1.98**
-4.08***
-2.08**
-4.35***
-2.73***

East Asia

0.84

1.00

-7.74***

Panel C: Cash flow rights over voting rights

0.97
-6.89***
1.00
0.96
-6.68***
0.80
0.95
-2.12**
1.00
0.95
-4.88***
1.00
0.82
-2.29**
0.80
0.93
-4.10***
0.80
1.00
-4.20***
1.00
0.93

-8.85***

1.00

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%.

42

Table 4
Correlations between earnings and stock returns: the role of ultimate voting rights level, family control, and cash-vote divergence

CV<1
CV=1
All

Full sample

V>median in economy

V<median in economy

Family

Non-family

0.15 (1756)
0.24 (3274)

0.14 (1267)
0.24 (2046)

0.18 (489)
0.25 (1228)

0.15 (1454)
0.27 (2080)

0.15 (302)
0.18 (1194)

0.21 (3313)

0.22 (1717)

0.22 (3496)

0.17 (1496)

The sample includes firms from seven East Asian economies: Hong Kong, Indonesia, South Korea, Malaysia, Singapore, Taiwan, and Thailand. The reported numbers are
Pearson correlation coefficients between annual earnings scaled by lagged market value of equity and cumulative market-adjusted stock returns. The market adjusted return is
the annual raw return minus the annual market return. 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. All of the correlation coefficients are different from zero at one percent level of
significance. The earnings and stock return data are taken from the PACAP database covering 1991 through 1995. For each of the two variables, the extreme two percentiles of
the firm-year observations are eliminated from the sample. Numbers in brackets are the numbers of firm-years. 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). V is the voting rights level of the
largest ultimate owner. CV is the ratio of cash flow to voting rights. An ultimate owner is defined as the shareholder who has at least 5% of voting rights of the a firm and who
is not controlled by anybody else. A firm is 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.

43

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***
(-12.15)

-0.17***
(-12.38)

-0.17***
(-12.29)

NI

-0.59
(-0.24)

-0.52
(-0.16)

-0.63
(-1.36)

NI*SIZE

0.10***
(3.01)

0.10***
(2.94)

0.12***
(3.09)

NI*Q

-0.07
(-0.83)

-0.07
(-0.88)

-0.09
(-1.07)

NI*LEV

0.59***
(2.80)

0.61***
(2.89)

0.58***
(2.68)

NI*SEG

-0.09***
(-2.86)

-0.09***
(-3.04)

-0.11***
(-3.24)

NI*EV

-0.94**
(-2.27)

-2.04***
(-3.89)

-1.62**
(-2.03)

NI*CV

0.45**
(2.40)

0.10
(0.44)

0.19
(0.23)

0.50***
(2.71)

0.45**
(2.09)

0.10
4490

0.10
4022

NI*CV*DEV

Adj-Rsq
Obs.

0.10
4490

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 yearend 1996. White-adjusted t-statistics are in parentheses. Asterisks denote significance levels: *** 1%; and **
5%.

44

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

-0.27***
(-8.53)

0.01
(0.18)

-0.25***
(-8.54)

-0.24***
(-12.36)

-0.12***
(-7.38)

NI

-1.87*
(-1.74)

1.75*
(1.73)

-2.34**
(-1.97)

-0.49
(-0.61)

0.02
(0.03)

NI*SIZE

0.31***
(3.87)

-0.17**
(-1.99)

0.30***
(3.69)

0.07
(1.09)

0.00
(0.02)

NI*Q

-0.32*
(-1.72)

-0.39*
(-1.82)

-0.19
(-1.20)

0.18
(1.13)

0.30*
(1.80)

NI*LEV

0.21
(0.51)

0.71
(1.16)

-0.25
(-0.54)

0.55
(1.36)

1.87***
(4.31)

NI*SEG

-0.16**
(-2.30)

0.01
(0.20)

-0.16***
(-2.49)

-0.02
(0.35)

-0.10*
(-1.83)

NI*EV

-3.24***
(-2.89)

-1.32
(-1.09)

-1.56
(-1.41)

-3.52***
(-3.36)

-0.52
(-1.31)

NI*CV

0.05
(0.10)

0.64
(1.22)

0.09
(0.17)

-0.09
(-0.24)

0.03
(0.08)

NI*CV*DEV

0.85**
(1.93)

0.29
(0.69)

0.68*
(1.88)

0.86**
(2.32)

-0.09
(-0.25)

0.21
802

0.06
880

0.12
948

0.18
989

0.13
871

Intercept

Adj-Rsq
Obs.

The sample includes firms from seven countries: 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
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%; ** 5%; and * 10%.

45

Table 7
Determinants of East Asian firms' earnings informativeness: economy-by-economy regression results
CARit = a0 + a1NIit + a2NIit*SIZEit + a3NIit*Qit + a4NIit*LEVit + a5NIit*SEGi + a6NIit*OWNi + (Fixed effects) + uit
Hong Kong
Indonesia
Korea (South)
Malaysia
Singapore
Taiwan
Intercept

Thailand

-0.18***
(-7.77)

-0.11*
(-1.75)

-0.05*
(-1.86)

-0.11***
(-3.60)

-0.03
(-1.08)

0.16***
(5.82)

-0.17***
(-4.19)

NI

-1.04
(-1.43)

-1.14
(-1.26)

3.71**
(2.19)

4.41***
(2.85)

2.08
(1.32)

4.25
(1.06)

-7.70***
(-3.59)

NI*SIZE

0.13**
(2.45)

0.11
(1.32)

-0.11
(-0.93)

-0.34***
(-2.92)

-0.09
(-0.77)

-0.11
(-0.40)

0.40***
(2.57)

NI*Q

0.09
(0.78)

0.49
(1.07)

-0.19
(-0.31)

-0.57***
(-3.09)

0.00
(0.01)

-0.70*
(-1.82)

0.65**
(2.20)

NI*LEV

0.43
(1.41)

1.19*
(1.80)

-0.45
(-1.16)

-0.24
(-0.53)

0.24
(0.37)

-1.94
(-0.96)

4.41***
(4.20)

NI*SEG

-0.04
(-1.22)

0.12
(1.37)

-0.18
(-1.49)

0.01
(0.18)

-0.18**
(-2.00)

-0.30
(-1.25)

-0.05
(-0.25)

NI*EV

-1.60***
(-2.72)

-2.64*
(-1.81)

0.27
(0.10)

0.29
(0.13)

-3.49
(-1.32)

-10.01*
(-1.78)

2.86
(1.02)

NI*CV
y

0.58*
(1.68)

-0.82
(-1.44)

-0.30
(-0.38)

1.02
(1.32)

-0.57
(-0.90)

0.07
(0.07)

3.24***
(2.48)

NI*CV*DEV

-0.14
(-0.51)

0.98*
(1.89)

0.42
(0.79)

-0.12
(-0.20)

0.89
(1.49)

1.59
(1.45)

-0.87
(-1.17)

0.12
1131

0.07
276

0.14
620

0.06
788

0.07
648

0.22
439

0.25
588

Adj-Rsq
Obs.

46

Table 7 (continued)
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 year effects 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. Whiteadjusted t-statistics are in parentheses. Asterisks denote significance levels: *** 1%; ** 5%; and * 10%.

47

Table 8
Regression analysis of the effects of family control on earnings informativeness
CARit = a0 + a1NIit + a2NIit*SIZEit + a3NIit*Qit + a4NIit*LEVit + a5NIit*SEGi + a6NIit*OWNi + (Fixed effects) +
uit
Pooled
Family
Non-family
Intercept

-0.17***
(-12.43)

-0.17***
(-10.16)

-0.17***
(-6.46)

NI

-0.36
(-0.86)

-0.92*
(-1.83)

0.58
(0.56)

NI*SIZE

0.10***
(2.92)

0.14***
(3.34)

-0.01
(-0.12)

NI*Q

-0.06
(-0.74)

-0.04
(-0.39)

-0.11
(-0.84)

NI*LEV

0.61***
(2.87)

0.63***
(2.64)

0.39
(0.86)

NI*SEG

-0.10***
(-3.14)

-0.12***
(-3.39)

-0.03
(0.53)

NI*EV

-1.99***
(-3.24)

-1.38
(-1.32)

NI*CV

0.22
(0.78)

0.34
(0.67)

NI*CV*DEV

0.54**
(2.29)

0.08
(0.23)

0.11
3159

0.08
1331

NI*FAMILY

-0.20
(-0.98)

NI*FAMILY*EV

-2.12***
(-3.46)

NI*FAMILY*CV

0.24
(0.86)

NI*FAMILY*CV*DEV

0.53**
(2.26)

Adj-Rsq
Obs.

0.10
4490

48

Table 8 (continued)
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: FAMILY, EV, CV, and DEV. FAMILY is a dummy variable
equals one if the firm is family controlled, and zero otherwise. 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.
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%; ** 5%; and * 10%.

49

Appendix 1
Results of the simple regressions of stock returns on earnings
CARit = a0 + a1NIit + (Fixed effects) + uit
Intercept
NI
Adj-Rsq

Obs.

1991

-0.25***
(-8.51)

1.46***
(7.71)

0.17

802

1992

-0.04
(-1.35)

0.77***
(4.12)

0.05

880

1993

-0.24***
(-9.34)

1.00***
(6.57)

0.1

948

1994

-0.24***
(-14.20)

1.09***
(8.25)

0.17

989

1995

-0.13***
(-9.69)

0.79***
(6.79)

0.11

871

Hong Kong

-0.17***
(-7.75)

0.77***
(8.10)

0.10

1131

Indonesia

-0.07
(-1.34)

0.71***
(4.08)

0.06

276

Korea (South)

-0.05*
(-1.73)

1.53***
(7.98)

0.14

620

Malaysia

-0.13***
(-5.67)

0.68***
(3.01)

0.05

788

Singapore

-0.02
(-1.18)

0.65***
(3.04)

0.07

648

Taiwan

0.15***
(5.41)

1.37***
(2.60)

0.22

439

Thailand

-0.13***
(-3.41)

2.79***
(8.12)

0.18

588

Pooled

-0.18***
(-12.97)

1.02***
(14.48)

0.08

4490

50

Appendix 1 (continued)
The ordinary least squares method is employed in the regressions. CAR is cumulative net-of-market twelvemonth 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.
Fixed-effects of calendar years and economies are included in the annual and country regressions,
respectively, but not reported. Financial and stock return data are from the PACAP database covering 1991
through 1995. White-adjusted t-statistics are in parentheses. Asterisks denote significance levels: *** 1%;
and * 10%.

51