samaha and dahawy - 2011 - an empirical analysis of cg structures and voluntary corporate disclosure in egyptian

Int. J. Accounting, Auditing and Performance Evaluation, Vol. 7, Nos. 1/2, 2011 61


Copyright © 2011 Inderscience Enterprises Ltd.



An empirical analysis of corporate governance
structures and voluntary corporate disclosure in
volatile capital markets: the Egyptian experience
Khaled Samaha*
Department of Accounting,
School of Business,
The American University in Cairo (AUC),
Room 2058 – BEC Building,
P.O. Box 74,
New Cairo 11825, Egypt
E-mail:
*Corresponding author
Khaled Dahawy
Department of Accounting,
School of Business,
The American University in Cairo (AUC),
Room 2001 – BEC Building,
P.O. Box 74,
New Cairo 11825, Egypt
E-mail:
Abstract: This paper examines the level and determinants (i.e. ownership

structure, board composition and audit committee presence) of voluntary
corporate disclosure in the annual reports of the largest 100 companies listed on
the Egyptian stock exchange (EGX). Our results indicate that overall voluntary
disclosure was low at just 13.43% with a large variation range. This score
places Egypt at a lower level than other emerging capital markets (e.g.
Singapore, Hong Kong and Malaysia). The variances of these results support
the need for individual country level studies and comparative analysis.
Multivariate results show audit committee presence as the most significant
variable influencing voluntary disclosure. Also, companies with a higher ratio
of independent non-executive directors have a higher extent of voluntary
disclosure. It was also evidenced that voluntary disclosure increases with
decreases in block-holder ownership. Results show that two other ownership
aspects – managerial and government – are not related to voluntary disclosure.
Finally, the analysis shows profitability and internationality significantly
impact voluntary disclosure. On the other side, that number of shareholders,
type of auditor, size, liquidity, leverage and industry type of the firm do not
affect the extent of voluntary disclosure.
Keywords: corporate governance; corporate disclosure; voluntary disclosure;
volatile capital markets; Egypt.
Reference to this paper should be made as follows: Samaha, K. and
Dahawy, K. (2011) ‘An empirical analysis of corporate governance structures
and voluntary corporate disclosure in volatile capital markets: the Egyptian



62 K. Samaha and K. Dahawy

experience’, Int. J. Accounting, Auditing and Performance Evaluation, Vol. 7,
Nos. 1/2, pp.61–93.
Biographical notes: Khaled Samaha is an Assistant Professor in the
Department of Accounting at the American University in Cairo (AUC). He has
a PhD (Manchester Business School – UK) and an MSc (Birmingham Business
School – UK). He is a Certified Public Accountant (CPA) from the Egyptian
Society for Accountants and Auditors (ESAA) and is certified by the Egyptian
Accounting Syndicate. He has extensive practical experience in the application
of International Financial Reporting Standards (IFRSs) and he has recently
published three papers about progressing Egypt towards convergence with
IAS/IFRS. His research interests include harmonisation and compliance with
IAS/IFRS, financial reporting and corporate governance mechanisms, audit
procedures and methodologies, financial reporting on the internet and the
implementation of accounting information systems in small and medium size
enterprises (SMEs). Currently, he is serving as a Member on the Editorial
Board of the Afro Asian Journal of Finance and Accounting. Currently, he is
serving as an Audit Consultant on various projects with the Egyptian Ministry
of Transport and the government of the Italian Republic.
Khaled Dahawy is the Director of MBA programs and an Associate Professor
in the Department of Accounting at the American University in Cairo (AUC).
He has a PhD (University of North Texas) and an MBA (Pennsylvania State
University). He is a Certified Public Accountant (CPA) in the State of Illinois,
USA, certified from the Egyptian Society for Accountants and Auditors
(ESAA) and is certified by the Egyptian Accounting Syndicate. He has several
papers and cases published in academic accounting journals and presented at
academic and practitioner conferences. His research interests include corporate
governance disclosure, disclosure in the developing countries, the role of the
audit profession in strengthening transparency and disclosure, the use of
information technology in accounting and the implementation of computerised
accounting information systems. He teaches financial accounting, international

accounting, tax accounting and auditing. He has extensive practical experience
as an expert in the Capital Market Authority (CMA) and has served as a
Consultant in many missions with the World Bank and the National
Democratic Party (NDP).
1 Introduction
The move towards a free market economy characterised by free trade and working of a
market pricing mechanism in the 1990s represented challenges to the Egyptian
government, private sector institutions and the accounting profession. This was the case
because historically Egyptian accounting was not capital-market oriented but followed
the principles of macro accounting, with strong government intervention to control the
economy and was closely connected with tax accounting (Hegazy, 1991). As a result,
the Egyptian government has adopted several far-reaching measures aimed at improving
the local and foreign investment environments in order to cope with the recent changes
that Egypt witnessed. These measures include the introduction of the International
Accounting Standards (IASs). This was parallel with a programme to privatise state-
owned companies as part of a more comprehensive movement towards capitalism, as
well as promotion of economic democracy and widespread of stock ownership.



The Egyptian experience 63




That change in orientation meant that the country had to change its whole
environment including the political, economic, social and legal environments to
accommodate private enterprises (Tesche and Tohamy, 1994). Thus, this was followed by
the Egyptianisation of these standards in 1997, originating the so-called Egyptian
accounting standards that became mandatory starting from 1998. The Egyptian Society of

Accountants and Auditors (ESAA) translated the IASs, organised many conferences
which argued for the introduction of the standards and helped in the regulation of
accounting in Egypt. These accounting changes were coupled with devaluation of the
currency, revision of the tax and customs structure and rates and revision of the
investment law. The adoption of IASs benefited Egypt in several aspects
1 it facilitated Egyptian access to international capital markets
2 it saved time and effort that might have been spent developing the national
accounting standards from scratch
3 ensured the fairness and meaningfulness of financial statements prepared
by Egyptian enterprises for international investors
4 increased the importance of accounting as a profession, due to the introduction of
the professional code and objectives.
Hofstede (1984) cultural variables model shows Egypt as a collectivist society, with large
power distance and strong uncertainty avoidance. According to Gray’s (1988) model
linking Hofstede’s cultural variables with accounting variables, Egyptian accounting
should be characterised by statutory control, uniformity, conservatism and secrecy.
HassabElnaby and Mosebach (2005) argued that these accounting variables are in
conflict with the capitalistic model that IASs is based on. In other terms, the Egyptian
cultural variables may conflict with the application of IASs (Dahawy and Conover,
2007). This argument is confirmed by the findings of Samaha and Stapleton (2008) which
focused on the practice (de facto) compliance with IASs in Egypt and revealed that there
is evidence of very low levels of compliance. To the extent that this evidence is indicative
of poor transparency, and/or the use of creative accounting techniques, it is problematic
for investors. Overall, these results may be seen to confirm the view taken by the World
Bank report (World Bank, 2002) that Egyptian financial reporting lacks reliability.
Egypt’s case presents a classical confrontation between a historically secretive society
and the requirement for high disclosure levels to attract direct foreign investments.
Researchers have, historically, found Egypt’s business society to be highly secretive
(Dahawy et al., 2002). However, the Egyptian government, business world and media
have consistently reported the need for direct foreign investment and hence the increase

in disclosure levels. This has resulted in increased significance of the Egyptian Stock
Exchange (EGX) (Samaha and Stapleton, 2008) as an important venue for attracting
foreign investments and to encourage local residents to invest in shares. Therefore,
Egyptian companies may engage in voluntary disclosure
1
to enhance the value of their
stocks.
This confrontation highlighted above between secretive culture and need for
disclosure presents an interesting motivation to investigate whether the traditional high
secrecy level outweighed the reforming efforts of government to establish a climate of
greater accountability and transparency.



64 K. Samaha and K. Dahawy




Furthermore, understanding why firms disclose information voluntarily is useful to
both preparers and users of accounting information, as well as to accounting policy
makers (Buzby, 1975; Meek et al., 1995). Previous studies on the determinants of
voluntary disclosure have focused mainly on the USA and other developed countries
(e.g. Buzby, 1975; Camfferman, 1997; Cerf, 1961; Choi, 1973; Cooke, 1989; Depoers,
2000; Ferguson et al., 2002; Firth, 1979; Frost and Pownall, 1994; Gray et al., 1995;
Inchausti, 1997; Malone et al., 1993; Meek and Gray, 1989; Meek et al., 1995; Mitchell
et al., 1995; Raffournier, 1995; Turpin and DeZoort, 1998). In contrast, a limited number
of research studies examined disclosure practices of companies in the developing
economies (Needles, 1997).
2

Some studies have examined institutional mechanisms
(i.e. corporate governance) that may influence voluntary disclosure practice. Corporate
governance attributes examined in relation to voluntary disclosure in these studies
include ownership structure, the proportion or existence of independent directors, the
appointment of a non-executive director as chairman and the existence of an audit
committee. However, most of the previous research in the developing countries focused
on the effect of a single corporate governance attribute (see e.g. Chau and Gray, 2002)
and very few examining different governance attributes in a single study (see e.g. Ghazali
and Weetman, 2006). It is worth noting that Okeahalam and Akinboade (2003) reviewed
the corporate governance literature in the African context, and concluded that: “there has
been limited published research on corporate governance in Africa and even less rigorous
academic or empirical research. There is an urgent need to embark on a meaningful
analysis of corporate governance [research] in Africa” (p.28). Therefore, an additional
motivation of this paper is to fill a void of research that explains the impact of corporate
governance on voluntary disclosure in Egypt as an African nation. In light of the above,
our main research questions are:
1 What is the level of voluntary disclosure in the annual reports of listed Egyptian
companies?
2 To what extent are aspects of corporate governance statistically significant in
explaining the level of voluntary disclosure in the annual reports of listed Egyptian
companies?
Therefore, this paper has three aims. Firstly, to examine voluntary disclosure of
companies listed on the EGX. Secondly, to examine whether corporate governance
variables that were found to be significant in explaining voluntary disclosure practices in
developed nations would have the same significance in developing nations. Finally, to
compare the results with other research conducted in other developing nations. This study
adds to the literature on voluntary disclosure in the developing countries and extends that
literature by including corporate governance variables as possible explanatory variables
of voluntary disclosure. Also, the results of this research may be useful for regulators in
Egypt as they continue to deliberate the appropriate corporate governance requirements.

This study should also help investors who are interested in investing in Egypt in
understanding the Egyptian corporate governance and disclosure environment. This study
extends academic research by comparing findings with results from other studies on
developing nations. If the results are the same then we can address developing nations as
a group. If the results of this study are different it will give support for the need to do
research at the individual country level.



The Egyptian experience 65




An overview of the Egyptian economy and the Egyptian corporate governance
framework is provided in Section 2. Section 3 provides a literature review and develops
and formulates the research hypotheses. Data selection and collection, and the research
techniques, are described in Section 4. Results and analysis are presented in Section 5,
with conclusions in Section 6.
2 Overview of the Egyptian economy and corporate governance
framework
Over the past several years, Egypt has actively reviewed and improved its regulatory
frameworks, in particular, corporate governance, transparency and disclosure. However,
the mandatory adoption of International Accounting Standards/International Financial
Reporting Standards (IAS/IFRS) starting from 1997 is not sufficient to resolve the
transparency problem or grantee the quality of disclosure (Samaha and Stapleton, 2008).
Mandatory disclosure rules ensure equal access to basic information (Lev, 1992), but this
information has to be augmented by firms’ voluntary disclosures and information
production (Cheng and Courtenay, 2004). There are major market incentives to disclose
information voluntarily. Managers’ attitudes to voluntary disclosure change according to

the perceived relationship of costs and benefits involved (e.g. see Gray et al., 1990; Healy
and Palepu, 1995).
The Egyptian government reform measures have dramatically improved the outlook
for Egyptian foreign investments, instigated and sustained high levels of growth and
employment creation. The current increase of the number of newly established companies
and the expansions undertaken by companies already in operation, are the results of the
streamlining investment procedures that have been undertaken. Inflows of foreign direct
investment (FDI) have also significantly increased, particularly starting in the financial
year 2003/2004.
3
Net FDI inflows increased from US$509.4 million in FY 2000/2001, to
reach US$6.1 billion in FY 2005/2006 and US$11.1 billion in FY 2006/2007. According
to the World Investment Report, published in 2007 by the United Nations Conference on
Trade and Development (UNCTAD), Egypt has emerged as the lead FDI recipient
country in the African continent. On the monetary side, net international reserves reached
US$31.7 billion in December 2007, achieving an increase nearly by 21.9% compared
with December 2006. In addition, the inflation rate has dropped from 12.4% in 2006 to
6.9% in 2007. The primary capital market reports that the total value of share issuances
has increased from 83.62 billion Egyptian pounds to 102.93 billion Egyptian pounds
which is an increase of 23% in one year. On the secondary market the general
index of the stock market in points has increased from 2,500 to 3,500 during 2007
(www.investment.gov.eg).
The Egyptian Ministry of investment through the Egyptian Institute of Directors
(EIOD)
4
introduced the general framework of corporate governance via an Egyptian
Corporate Governance Code (ECGC) in 2005. The code is based on the Organisation for
Economic Cooperation and Development (OECD) corporate governance principles. The
intention was to enhance the quality of information issued by listed companies, improve
decision making, attract investors and stimulate economic development through increased

competition and enhance the level of confidence of foreign portfolio investors in the
Egyptian capital market (Carana, 2000; MOFT, 2007).



66 K. Samaha and K. Dahawy




This code represents the general framework for corporate governance of Egyptian
enterprises. The corporate governance code was published in Arabic, setting out
principles of best practice for good corporate governance. Compliance with the EIOD
code is not mandatory. The guide is prepared in accordance with the corporate
governance principles issued by the OECD and a number of countries including South
Africa, Malaysia and the Philippines. These rules are primarily applicable to joint stock
companies listed on the stock market, especially those being actively traded. As stated in
guideline (3–4), the board should comprise a majority of non-executive directors with
the technical or analytical skills to benefit the board and the company. All of the
non-executive directors should dedicate the time and attention necessary to fulfil their
obligations to the company and not to accept assignments that could be seen to be a
conflict of interest. In addition, guideline (6–1) stated that the audit committee should
consist of not less than three non-executive members. One member should be a financial
and accounting expert. In case of an insufficient number of non-executives, one or more
members may be appointed from outside the company. A critical aspect of the new EGX
listing rules is that companies must form an audit committee of the board. The audit
committee is responsible for the oversight of internal auditing and control procedures and
reviews annual reports and prospectuses. The audit committee does not propose the
external auditor. The members of the committee should be non-executives, unless
there are no non-executive directors on the board, in which case outsiders can be hired

(EGX, 2006).
Since 2005, the activities of the EGX have indeed increased considerably: market
capitalisation has increased dramatically reaching 46% by the end of 2006 (EGX, 2006);
there were 1.65 million investors in 2005, compared with only 25,000 ten years earlier
(EGX, 2006); and the number of registered companies increased from 218 in 1991 to
1,150 by the end of 2002 (CMA, 2003) and down to 800 companies by the end of 2006
(EGX, 2006). The EGX has raised its international profile and is one of the leading
markets in the Middle East North Africa region. Foreign participants, who enjoy full
market access and suffer no restrictions on capital mobility or convertibility, represented
approximately 17% of value traded in 2001 (Abdel Shahid, 2003) and reaching 31% by
end 2006 (CMA, 2006). This level of growth and the increase in international profile
suggest that the government’s reform process has met with some success, and it has been
argued that the Egyptian Exchange is prepared for the globalisation era (Omran, 2002;
Samaha and Stapleton, 2008). However, as it was seen in the Far East financial crisis in
countries such as Thailand such rapid growth is not necessarily accompanied by increases
in the quality of corporate reporting.
Since best practices in corporate governance and greater disclosure are just being
promoted, there is probably a cross-sectional variance in corporate governance. Hence,
Egyptian firms provide an appropriate sample to examine the issue of corporate
governance and disclosure at this time. We ask whether the subsequent actions of reform
by the Egyptian government increased the awareness of disclosure as a tool of corporate
governance. Examining company annual reports after the introduction of new corporate
governance recommendations in 2005 by the EIOD also gives us the opportunity to
assess transparency by companies that adopted the best practices recommended in the
Egyptian Code of Corporate Governance (ECCG).



The Egyptian experience 67

In Section 3, we review literature which suggests association between voluntary
disclosure and corporate governance based on theories of reporting, including mainly
agency and political costs theories. We then consider the empirical evidence and draw out
six hypotheses for testing on Egyptian companies.
3 Prior literature and development of hypotheses
Corporate governance is aptly defined by Denis and McConnell (2002, pp.1–2) as “…the
set of mechanisms – both institutional and market-based – that induce the self-interested
controllers of a company … to make decisions that maximise the value of the company to
its owners…”. Recent empirical work on the association between disclosure and
corporate governance in the developing countries include Ghazali and Weetman (2006),
Barako et al. (2006), Cheng and Courtenay (2004), Chau and Gray (2002), Eng and Mak
(2003), Haniffa and Cooke (2002) and Ho and Wong (2001). We extend prior work by
examining corporate governance from three aspects. Specifically, this paper examines the
association between the various categories of ownership structure, board composition,
existence of audit committees and voluntary disclosure.
3.1 Ownership structure
Structure of ownership determines the level of monitoring and thereby the level of
disclosure. As in previous research, we introduce ownership structure by including
1 number of shareholders
2 block-holder ownership: the proportion of ordinary shares held by substantial
shareholders which is represented by shareholdings of 5% or more
3 managerial ownership: the proportion of ordinary shares held by the CEO and
executive directors
4 government ownership.
The number of shareholders is used as a measure of dispersion of shareholder control.
Schipper (1981) proposed that monitoring problems that could be solved by issuing
public accounting reports would increase with the number of owners. As the number of

shareholders increases, one would expect disclosure to increase if it can provide a
solution to the additional monitoring problems associated with dispersion in ownership.
Previous studies have found a significant positive association between the number of
shareholders and the extent of financial disclosure (e.g. Chau and Gray, 2002; Cooke,
1989, 1991; Malone et al., 1993).
Block-holder ownership is the percentage of ordinary shares held by substantial
shareholders (i.e. shareholdings of 5% or more). Additional monitoring is required in
cases when share ownership is diffused. Hence, it is expected that voluntary disclosure
increases with decreases in block-holder ownership. Previous research has indicated the
presence of a negative relation between block-holder ownership and disclosure
(McKinnon and Dalimunthe, 1993; Mitchell et al., 1995; Schadewitz and Blevins, 1998).



68 K. Samaha and K. Dahawy




Managerial ownership is the percentage of ordinary shares held by the CEO and
executive directors. Excessive management ownership could be counter-productive to the
firm’s long-term value, as management could effectively wield external threats. This
contention is found in the entrenchment theory (Fan and Wong, 2002; Morck et al., 1988)
which predicts that high management interest leads to lower voluntary disclosure.
Further, the controlling owner of the firm effectively decides “… the accounting
reporting policies” (Fan and Wong, 2002, p.403). It is predicted that this leads to a
low level of disclosure, driven primarily by the controlling owner’s motive to hold
up minority shareholders (Fan and Wong, 2002). In line with this, prior research
(e.g. Eng and Mak, 2003; Ghazali and Weetman, 2006) argues that greater agency
problems exist when managerial ownership is low. In other terms, executives who own a

smaller portion of the company would have high incentives to consume bonuses and low
incentives to maximise job performance. In this instance, outside shareholders may need
to increase monitoring of a manager’s behaviour to reduce the associated increase in
agency costs. The required monitoring by outside shareholders will increase firm costs.
However, these costs can be reduced if managers provide voluntary disclosure.
In this context, voluntary disclosures can act as a substitute for costly monitoring.
Previous studies show managerial ownership to be negatively related to disclosure levels
(e.g. Eng and Mak, 2003; Ghazali and Weetman, 2006; Ruland et al., 1990). Based on the
previous arguments and findings, we expect that voluntary disclosure increases with
decreases in managerial ownership.
There has been no agreement on the impact of governmental ownership on disclosure.
Eng and Mak (2003) argue that agency costs are higher in government-owned companies
because of conflicting objectives between pure profit goals of a commercial enterprise
and goals related to the interests of the nation, resulting in a need for communication with
other shareholders in the form of disclosures. However, Ghazali and Weetman (2006)
argued that companies with government ownership may not need to give extensive
disclosure because of the presence of additional separate monitoring by the government.
In addition, government-controlled companies can obtain cheaper funds from local
banks; therefore, they may not need to attract potential investors. As a result, less
disclosure can be expected in government-controlled companies. It may be expected in a
developing country like Egypt that government-controlled companies are strongly
politically connected and these companies may disclose less information to protect their
political linkages or interests or beneficial owner. Therefore, we expect that companies
with a higher proportion of government ownership will disclose less voluntary
information.
Based on the previous discussion, we introduce the following hypothesis:
H1: There is a positive association between the number of shareholders and the level of
voluntary disclosure.
H2: There is a negative association between block-holder ownership and the level of
voluntary disclosure.

H3: There is a negative association between managerial ownership and the level of
voluntary disclosure.
H4: There is a negative association between government ownership and the level of
voluntary disclosure.



The Egyptian experience 69




3.2 Board composition: independent non-executive directors on the board
Healy and Palepu (2001) argue that the agency problem can be solved by having the
board of directors monitor and discipline management on behalf of the external owners.
The board of directors provides monitoring as it is elected by the shareholders to monitor
their interests. The board of directors is the central internal control mechanism for
monitoring managers. Non-executive directors can be regarded as “professional referees
whose task is to stimulate and oversee the competition among the firm’s top
management” (Fama, 1980, p.294). The perceived importance of the outside directors as
a powerful tool for constraining management behaviour is largely attributed to their being
independent of management (Rosenstein and Wyatt, 1990). The presence of a higher
proportion of non-executive directors on corporate boards should result in more effective
monitoring of managerial opportunism (Chau and Leung, 2006; Fama and Jensen, 1983;
Leftwich et al., 1981; Weir and Laing, 2003). Thus, the firms whose boards are
dominated by outside directors and they are expected to disclose more voluntary
information.
Previous studies report that companies that have more outside directors on the board
have more voluntary disclosures (e.g. Adams and Hossain, 1998; Chen and Jaggi, 2000;
Cheng and Courtenay, 2004; Leung and Horwitz, 2004; Williams, 2002). However, some

researchers, did not find significant relationship between the level of voluntary disclosure
and board independence (Ghazali and Weetman, 2006; Haniffa and Cooke, 2002; Ho and
Wong, 2001). Other researchers report the presence significant negative association
between the level of voluntary disclosure and board independence (Barako et al., 2006;
Eng and Mak, 2003; Gul and Leung, 2004).
In terms of board composition, duality and board of director’s size can be seen as
important factors that impact disclosure. However, the information related to duality is
non-existent in the Egyptian environment and cannot be collected. Given the Egyptian
environment the board size may not be an indicator of monitoring and disclosure.
Appointment of independent non-executive directors is a new governance initiative
recommended in ECCG. Currently, there is no empirical research that has linked board
independence to voluntary disclosure in Egypt. The ECCG guidelines specifically define
an independent director as a person who is not involved in the management of the
company and does not have any direct or indirect interest in the company. Our
expectation is that with the introduction of the ECCG, independent directors will play a
more proactive role in ensuring greater corporate transparency and accountability. This
increased awareness may be partly discharged by more voluntary information disclosure
in annual reports. The expectation is expressed in the following hypothesis:
H5: There is a positive association between the proportion of independent non-executive
directors and the level of voluntary disclosure.
3.3 Existence of an audit committee
Audit committees, among other functions, ensure the quality of financial accounting and
control system (Collier, 1993). Since an audit committee consists mainly of non-
executive directors, it should reduce the amount of withheld information. Agency theory
predicts that audit committees should lower agency costs. Bradbury (1990) views audit
committees as a monitoring mechanism which is formed voluntarily in high agency cost



70 K. Samaha and K. Dahawy

situations to improve the quality of information flow between principal and agents. Audit
committees may improve internal control and thus act as an effective monitoring device
for improving disclosure quality. Forker (1992) found a positive but weak relationship
between the disclosure of the audit committee and the quality of share option disclosure
for UK companies. McMullen (1996) found significant association between the presence
of audit committees and more reliable financial reporting. Barako et al. (2006) also found
strong positive relationship between the presence of an audit committee companies’
voluntary disclosure practices in Kenya.
Both ECCG and listing rules in Egypt require the presence of an audit committee, to
enhance the level of information produced by companies. The quality and meetings of the
audit committee can be seen as factors that impact the monitoring role of the audit
committee. However, given the secretive Egyptian culture and the novice status of the
audit committee requirements, it is impractical, if not impossible, to have the access to
this type of detailed information. Since this is a new requirement. Therefore, it is
hypothesized that:
H6: There is a positive association between the existence of audit committees and the
level of voluntary disclosure.
3.4 Other control variables
Following the practice in prior research
5
we include size, financial leverage, industry
type, type of auditor, profitability, liquidity and internationality as control variables in the
multiple regression models for testing the main hypotheses.
4 Data collection and research design
4.1 Sample selection and data sources
Despite the increased market capitalisation of the Egyptian Stock Market, trades remain

concentrated in few companies that exhibit real strong fundamentals (EGX, 2006). In
2006, the EGX had 800 listed companies of which 30 are the most actively traded
companies as measured by the EGX 30 index. The information on the market
capitalisation of all 800 listed companies is found in the EGX Bulletin that is published
monthly. Therefore, we based our sample selection on the largest 100 firms according to
market capitalisation for the year 2006. This approach is used due to data limitations.
Voluntary disclosure is measured by the amount and detail of non-mandatory
information that is contained in the management discussion and analysis in the annual
report. In Egypt, there is no Data Stream from which the annual reports could be
obtained. However, all the annual reports in addition to board of director’s reports of
listed companies are filed with the Egyptian Company for Information Dissemination
(EGID) which follows the EGX, and are available for a fee. The annual reports and the
Board of Directors reports of the sample companies for the year ending 2006 were
obtained from EGID.
There is no information in the annual reports of listed Egyptian companies about the
number of shareholders, block-holder ownership, managerial ownership, government
ownership, proportion of non-executive independent directors and the existence of audit



The Egyptian experience 71




committees. However, recently the EGX requires listed companies to file this information
with EGID. Therefore, information on these variables for the sample companies for the
year ending 2006 was also obtained from EGID.
4.2 Scoring voluntary disclosure: index construction
The voluntary disclosure checklist was constructed based on the ones used by Chau and

Gray (2002) in Hong Kong and Singapore, and Ghazali and Weetman (2006) in
Malaysia. This provides a useful yardstick for comparison with earlier research. The
major reason for adapting these two checklists is that the original checklist used in these
studies was based on Meek et al. (1995) that was compiled ‘‘based on an analysis of
international trends and observations of standard reporting practice, taking into account
the relevant research studies as well as other comprehensive international surveys of
accounting and reporting’’ (Meek et al., 1995, p.561). The items on the checklist were
checked against the mandatory disclosure requirements of Egypt that are based on the
EASs in order to make sure that there are no mandatory items.
6
The disclosure list (see Appendix) is divided into 12 sub-parts representing the
categories of voluntary disclosure, resulting in a total of 80 disclosure items. These
categories are: general corporate information; corporate strategy; future prospects;
information about directors; review of operations; product/service information; segmental
information; research and development; employee information; social and environmental
reporting; financial review and market-related information.
A separate index is calculated for each sub-part to enable a finer exploration of
changes in the indices, and which improves statistical and descriptive analysis. In
addition, we also calculated three partial indices for the disclosure items in line with
Ghazali and Weetman (2006) and Meek et al. (1995), corresponding to the type of
voluntary information. These are: financial (FIN); strategic (STRAT) and corporate
social reporting (CSR) (see Appendix). This resulted in calculating 15 voluntary
disclosure indices (3 partial and 12 sub-indices). In addition to have partial and
sub-indices, two overall indices for voluntary disclosures (VDIS 1 and VDIS 2) were
calculated in line with Street and Gray (2002). VDIS 1 assigned an equal weighting to
each of the sub-indices. VDIS 2 assigned equal weighting to each item of disclosure
required by the sub-indices.
Each item of disclosure was scored without weighting on a dichotomous basis taking
the commonly used approach of giving the item a score of 1, 0 or not applicable (N/A)
(see e.g. Chau and Gray, 2002; Cooke, 1989, 1991; Ghazali and Weetman, 2006; Haniffa

and Cooke, 2002; Ho and Wong, 2001). To ensure that companies were not penalised for
non-disclosure of irrelevant items each annual report was read in its entirety, following
Cooke (1989, 1991). The voluntary disclosure index was derived by computing the ratio
of actual scores awarded to the maximum possible score attainable for items appropriate
(applicable) to that company.
Following recommendations in the literature (see e.g. Samaha and Stapleton, 2008);
the annual reports of three companies in the sample were randomly drawn and were given
to three independent persons to score. The result of the Spearman rho correlation test
(not reported) suggest that scores obtained by the independent scorers and those of the
present investigators are in substantial agreement; indicating minimal subjectivity in
scoring the disclosure items in the annual reports of the sample companies. In
Section 4.3, we present our empirical findings.



72 K. Samaha and K. Dahawy




4.3 Regression models and definition of variables
A multiple ordinary least squares (OLS) regression model was used to test the association
between the dependent variables of voluntary disclosures (VDIS, FIN, STRAT and CSR)
and the independent variables. Table 1 provides a summary of the operational definition
of independent variables and their sources. Because tests of normality on the continuous
independent and dependent variables suggest non-symmetrical distribution, all
continuous variables were transformed into normal scores based on the Van der Waerden
approach to avoid violating the regression assumptions (a method proposed by Cooke,
1998).
Table 1 Operational definitions of independent variables

Independent variables and
code on SPSS Operational definition Source of information
Test variables
1 Ownership structure
Number of shareholders
(SHARE)
In line with Ghazali and Weetman
(2006), it is calculated as the
number of the owners holding the
total number of shares issued
Ownership structure
information (EGID)
Block-holder ownership
(BLKOWN)
In line with Eng and Mak (2003),
it is calculated as the percentage of
shares owned by the block-holders–
shareholders whose ownership is
equal to or exceeds 5% to the total
number of shares issued
Ownership structure
information (EGID)
Managerial ownership
(MANOWN)
In line with Eng and Mak (2003),
it is calculated as the percentage of
shares owned by the CEO and
executive directors to the total
number of shares issued
Ownership structure

information (EGID)
Government ownership
(GOVOWN)
Percentage of the shares owned
by the public sector (holding
companies, companies, banks,
insurance companies and other
institutions) to the total number
of shares issued
Ownership structure
information (EGID)
2 Board composition
Independent non-executive
directors on the board (INDEP)
In line with Ghazali and Weetman
(2006), it is calculated as the ratio
of the number of non-executive
directors to the total number of the
directors
Board of directors’
report (EGID)
www.egyptwatch.com
3 Existence of audit
committees (ACOM)
Dichotomous, 1 or 0
Board of directors’
report (EGID)
www.egyptwatch.com
Control variables
1 Type of auditor (AUDIT)

Dichotomous, 1 for big 4 or 0
otherwise
Annual report: auditor
report section
2 Industry type (IT1, IT2,
IT3)
Categorised into: manufacturing
IT1, trade/service IT3 and financial
IT2
EGX Bulletin
(December 2006)



The Egyptian experience 73




Table 1 Operational definitions of independent variables (continued)
Independent variables and
code on SPSS Operational definition Source of information
3 Internationality (INTER)
In line with Zarzeski (1996), a firm
is defined as international based on
the following: foreign business
transactions (foreign sales or
exports), or being an MNC
affiliation, trading of firm’s shares
on foreign stock exchanges

(however, this is very rare in listed
Egyptian companies)
Annual report: footnotes
Board of directors’
report (EGID)
4 Leverage (LEVG)
Defined as total debt to total
stockholders’ equity
Annual report: financial
statements
5 Firm size (ASSETS,
SALES)
Log (base 10) of total assets
Log (base 10) of total sales
Annual report: financial
statements
6 Profitability (PROF)
Return on equity defined as net
income before tax to the total
stockholders’ equity
Annual report: financial
statements
7 Liquidity (LIQD)
Is a quick ratio and defined as
current asset – after deducting the
inventory – to the current liabilities
Annual report: financial
statements
Based on the information in Table 1 the following four models are estimated:
VDIS =

E
O +
E
1 SHARE +
E
2 BLKOWN +
E
3 MANOWN +
E
4 GOVOWN
+
E
5 INDEP +
E
6 ACOM +
E
7 AUDIT +
E
8 IT1 +
E
9 IT2 +
E
10 IT3
+
E
11 INTER +
E
12 LEVG +
E
13 ASSETS +

E
14 SALES +
E
15 PROF
+
E
16 LIQD + e
FIN =
E
O +
E
1 SHARE +
E
2 BLKOWN +
E
3 MANOWN +
E
4 GOVOWN
+
E
5 INDEP +
E
6 ACOM +
E
7 AUDIT +
E
8 IT1 +
E
9 IT2 +
E

10 IT3
+
E
11 INTER +
E
12 LEVG +
E
13 ASSETS +
E
14 SALES +
E
15 PROF
+
E
16 LIQD + e
STRAT =
E
O +
E
1 SHARE +
E
2 BLKOWN +
E
3 MANOWN +
E
4 GOVOWN
+
E
5 INDEP +
E

6 ACOM +
E
7 AUDIT +
E
8 IT1 +
E
9 IT2 +
E
10 IT3
+
E
11 INTER +
E
12 LEVG +
E
13 ASSETS +
E
14 SALES +
E
15 PROF
+
E
16 LIQD + e
CSR =
E
O +
E
1 SHARE +
E
2 BLKOWN +

E
3 MANOWN +
E
4 GOVOWN
+
E
5 INDEP +
E
6 ACOM +
E
7 AUDIT +
E
8 IT1 +
E
9 IT2 +
E
10 IT3
+
E
11 INTER +
E
12 LEVG +
E
13 ASSETS +
E
14 SALES +
E
15 PROF
+
E

16 LIQD + e
where VDIS represents overall voluntary disclosure index, FIN represents financial
disclosure partial index, STRAT represents strategic disclosure partial index, CSR
represents corporate social disclosure partial index,
E
0
, },
E
16
represent regression
coefficients, Independent variables are defined in Table 1 and e represents the error term.



74 K. Samaha and K. Dahawy




5 Empirical results and discussion
5.1 Descriptive statistics and correlation analyses
Table 2 presents the descriptive statistics of the explanatory variables.
7
The average firm
size in terms of total assets is 4 billion Egyptian pounds (approximately, 747 million US
dollars) with the maximum total assets being 40 billion Egyptian pounds (approximately
7.4 billion US dollars). The range of the total assets indicates that the sample firms are
widely distributed. Number of shareholders ranges from 3 to 99 shareholders with an
average of 19 shareholders, implying that ownership structure is highly concentrated in
listed Egyptian companies. Only 21 companies (21%) have an audit committee. The

average managerial holding (MANOWN) is very high at 89.6%. The level of block-
holder ownership (BLKOWN) is high with a mean of 59%. On average, slightly more
than one-third of the board of directors (39%) is outside directors (INDEP). The level of
government ownership (GOVOWN) is lower than MANOWN and BLKOWN with a
mean of 29%.
Table 3 (Panel A) shows the distribution of the dependent variable (i.e. extent of
voluntary disclosure measured by VDIS) and the partial categories of information
(FIN, STRAT and CSR). The total voluntary disclosure scores range from 4% to 62%
with a mean score of 13.43% in the case of VDIS 1, and range from 3% to 58% with a
mean score of 10.53% in the case of VDIS 2. Thus, there were large variations in
voluntary disclosure practices among the sample companies in Egypt. The indices and
their ranges suggest that the overall voluntary disclosure level is very low, which is in
line with prior research on disclosure in the developing countries, specifically Egypt
(Dahawy and Conover, 2007; Samaha and Stapleton, 2008). No company obtained an
overall compliance rate above 80%. Consistent with Ho and Wong (2001) in Hong Kong,
the low levels of voluntary disclosure ratio implies that analysts in Egypt may search for
information outside of annual reports (e.g. via investor relations department).
Table 2 Descriptive statistics of the independent variables
Code in SPSS Min. Max. Mean SD
Panel A: continuous explanatory variables
SHARE 3 99 19.15 17.85
BLKOWN 0.00 1.00 0.5912 0.3749
MANOWN 0.00 0.97 0.0896 0.2186
GOVOWN 0.00 0.99 0.2920 0.3301
INDEP 0.00 0.96 0.3913 0.3596
LEVG 0.00 7.66 0.5708 1.2674
ASSET 156,632,696 38,274,231,487 3,770,732,913 6,957,985,258
SALES 555,165 18,730,653,475 1,076,852,202 2,533,005,897
PROF
1.34

1.18 0.2082 0.2775
LIQD 0.06 14.21 1.5223 1.9902
Panel B: categorical explanatory variables
ACOM 0.00 1.00 0.21 0.4093
AUDIT 0.00 1.00 0.49 0.5024
IT1 0.00 1.00 0.55 0.5000
IT2 0.00 1.00 0.13 0.3380
IT3 0.00 1.00 0.32 0.4688
INTER 0.00 1.00 0.79 0.4093



The Egyptian experience 75




Table 3 Descriptive statistics of the voluntary disclosure overall, partial and sub-indices
Panel A: voluntary disclosure overall and partial indices
Total disclosure
VDIS 1 VDIS 2
Financial
information
partial index
Strategic
information
partial index
Corporate social
responsibility
partial index

Mean (%) 13.43 10.53 15.73 10.80 4.17
Min (%) 4 3 4 3 0
Max (%) 62 58 59 72 50
K-S
significance
0.000** 0.000** 0.000** 0.000** 0.000**
VDIS score
(%)
No. of
companies %
No. of
companies %
No. of
companies %
No. of
companies %
Above 80 0 0 0 0 0 0 0 0
60–79.9 1 1 0 0 1 1 0 0
40–59.9 3 3 4 4 4 4 3 3
Below 40 96 96 96 96 95 95 97 97
Total 100 100 100 100 100 100 100
1
00
Panel B: voluntary disclosure sub-indices
Sub-indices Min Max Mean SD
High level of disclosure (80% or more)
No sub-index
Moderate level of disclosure (60–79.9%)
No sub-index
Low level of disclosure (40–59.9%)

General corporate information
sub-index
33 100 43.65 0.19
Very low level of disclosure (below 40%)
Review of operations sub-index 0 100 32.65 0.27
Market-related information
sub-index
0 100 26.60 0.16
Corporate strategy sub-index 0 57 12.87 0.15
Segmental information sub-index 0 50 11.35 0.10
Product/service information
sub-index
0 60 9.00 0.16
Social and environmental
reporting sub-index
0 50 6.25 0.12
Financial review sub-index 0 60 6.00 0.14
Research and development
sub-index
0 100 3.50 0.16
Employee information sub-index 0 50 3.13 0.09
Future prospects sub-index 0 50 2.88 0.09
Information about directors
sub-index
0 100 2.86 0.13
**Significant at the 1% level.



76 K. Samaha and K. Dahawy

The voluntary mean disclosure in 2006 varied from 4.17% in the case of corporate social
reporting information to 15.73% for financial information, with strategic information in
between at 10.8%. Voluntary disclosure that relate to corporate social reporting is lower
than the strategic and financial indices. The range of the STRAT index (0.69) is wider
than the range of the FIN and CSR indices (0.55 and 0.50, respectively). The minimum of
the CSR and STRAT indices (0.00 and 0.03) is lower than the minimum index of the FIN
indices (0.04).
This result is well below the recent results reported by Ghazali and Weetman (2006)
who reported that the voluntary mean disclosure varied from 20.2% in the case of CSR
information to 39.3% for strategic information, with financial information in between at
35.6%. The total voluntary disclosure scores in Malaysia range from 6.3% to 74.0% with
a mean score of 31.4%. However, the results is much higher than Hong Kong and
Singapore (Chau and Gray, 2002) who reported that the voluntary mean disclosure for
Singapore in 1997 varied from 10.68% for financial information to 16.76% for non-
financial information, with strategic information in between at 16.00%. The overall mean
disclosure in 1997 was at 13.83%. For the Hong Kong companies, the voluntary mean
disclosure in 1997 varied from 9.77% in the case of financial information to 18.49% for
strategic information, with non-financial information in between at 10.45%. The overall
mean disclosure in 1997 was 12.23%.
Only four companies (4%) disclosed 40% or more of the 80 items included in the
index. The results are well below the Malaysian results (Ghazali and Weetman, 2006)
which indicated that 26 companies out of 87 (30%) disclosed 40% or more of the items
included in the index. In contrast to the Malaysian study, these results show that even
among the biggest companies on the EGX, there is insignificant variability in the amount
of information voluntarily disclosed in annual reports and only four companies can be
said to be ‘good disclosers’ according to the classification of Samaha and Stapleton

(2008). Table 3 also shows that the scores are not normally distributed as indicated by the
significance of the non-parametric Kolmogrov–Smirnov test (or K–S Lilliefors).
Table 3 (Panel B) shows that the voluntary disclosure sub-indices range from a low of
2.86% in the case of information about directors to a high of 43.65% on general corporate
information disclosures. Table 4 also shows that the mean score of the general corporate
disclosure sub-index is higher than all other sub-indices. Table 3 (Panel B) also shows
that the mean score of seven sub-indices (future prospects, employee, financial review,
products and services, research and development, information about directors and social
and environmental reporting) is below 10% which is indicative of poor voluntary
disclosure in Egypt. The very low disclosure level on Information about directors is
corroborated in the World Bank report (2002);
it is up to company statutes to determine what level of disclosure relating to the
board should be divulged to shareholders at the AGM. For example, the names
of directors and their remuneration is disclosed at the AGM but not published
in the annual report. The remuneration of board members consists of sitting
fees and travel expenses on the one hand and an annual share of profits not
exceeding ten percent of net income (after deducting five percent for legal
reserves and five percent of paid in capital for dividends). Executive
remuneration is not disclosed. (p.13)



The Egyptian experience 77




Table 4 Pearson correlation
*Significant at the 0.05 level.
**Significant at the 0.01 level.

***Significant at the 0.001 level.



78 K. Samaha and K. Dahawy




Table 4 presents the correlation between variables. Disclosure score (VDIS) is positively
correlated with SHARE (r = 0.597), INDEP (r = 0.497), ACOM (r = 0.703), ASSETS
(r = 0.289) and negatively correlated with block-holder ownership (BLKOWN)
(r = 0.590). The univariate analysis supports HI, H5 and H6 that the number of
shareholders, proportion of independent directors and existence of audit committees are
positively correlated with the level of voluntary disclosure. Also, H2 is supported in that
block-holder ownership is negatively correlated with the level of voluntary disclosure.
The correlation between MANOWN, GOVOWN and VDIS is not significant. VDIS is
also correlated with ASSETS (r = 0.289). The results show that MANOWN and
BLKOWN are not correlated with one another. However, MANOWN is negatively
correlated to GOVOWN (r = 0.196). BLKOWN is negatively related to INDEP
(r = 0.446) and SHARE (r = 0.404). Finally, ACOM is positively related to INDEP
(r = 0.494) and SHARE (r = 0.348).
5.2 Regression results: corporate governance attributes and voluntary
disclosure
Table 5 presents the R
2
(coefficient of determination), F-ratio,
E
-coefficients and
t-statistics for four models (one overall (VDIS 2) and three partial (FIN, STRAT and

CSR)) used to test hypotheses H1 to H6.
8.
The regression results are not materially
affected by robustness testing, which included the Kolmogrov–Smirnov (K–S) test, a
visual inspection of the histogram and normal probability plot of the residuals in the four
OLS models. This is to be expected since normal scores are normally distributed,
implying normality of the error term (Camfferman and Cooke, 2002). A further check
was conducted to see if the results are driven by outliers by removing the influential cases
as recommended by Field (2003). The results are not materially affected by any of these
robustness tests.
As shown in Table 5, the multiple regression models
9
for the overall level of
voluntary disclosure (VDIS) have reported F-value of 18.059 (significant at the 0.0001
level), and adjusted coefficients of determination (R
2
) of 40.8. At the 0.01 level of
significance, the hypothesis that all explanatory variable coefficients are simultaneously
equal to zero is rejected.
The explanatory power of this model is higher than Eng and Mak (2003) [20.61%],
Ghazali and Weetman (2006) [36.1%], Ho and Wong (2001) [31.4%] and Cheng and
Courtenay (2004) [7.5%], but lower than Barako et al. (2006) [53.4%], Haniffa
and Cooke (2002) [47.9%] and Chau and Gray (2002) in respect of Hong Kong [42.7%]
and Singapore [72.5%] companies.
However, there were some apparent differences in the explanatory power between the
three partial models (FIN, STRAT and CSR) as shown by the R
2
in Table 5. The amount
of explained variation in voluntary disclosure ranges from 34.6% in the case of FIN to
50.4% for STRAT information and 53.1% for CSR information.

Table 5 shows that only five independent variables entered the equation in the VDIS
regression model. These variables are: block-holder ownership (BLKOWN), ratio of
independent directors on the board (INDEP), existence of audit committees (ACOM),
internationality (INTER) and profitability (PROF). In addition, the directions of the signs
of all significant coefficients in the regression model are in agreement with the
hypotheses.



The Egyptian experience 79




Table 5 Results of cross-sectional OLS regressions
Overall voluntary
disclosure model
Financial
informationmodel
Strategic
information model
Corporate social
responsibility model
Adjusted R
2
40.8 34.6 50.4 53.1
F statistic 18.059 18.451 26.155 38.360
Significance 0.000*** 0.000*** 0.000*** 0.000***
Durbin
Watson test

1.514 1.752 1.633 1.604
Variables Beta t
Sig.
(p-value)
Beta t
Sig.
(p-value)
Beta t
Sig.
(p-value)
Beta t
Sig.
(p-value)
Constant 1.987 0.050 3.415 0.001 0.254 0.800 2.046 0.043
Ownership diffusion
SHARE 0.107 1.208 0.230EX 0.057 0.666 0.507EX 0.118 1.464 0.146EX 0.158 2.748 0.007*IN
BLKOWN 0.211 2.407 0.018*IN 0.162 0.1886 0.062EX 0.188 2.625 0.010**IN 0.120 1.465 0.146EX
MANOWN 0.080 1.016 0.312EX 0.023 0.282 0.779EX 0.066 0.920 0.360EX 0.078 1.136 0259EX
GOVOWN 0.075 0.941 0.349EX 0.099 1.209 0.230EX 0.002 0.023 0.982EX 0.012 0.171 0.864EX
Board composition
INDEP 0.234 2.395 0.019*IN 0.168 1.816 0.073EX 0.175 2.166 0.033*IN 0.225 3.384 0.001***IN
Existence of audit committees
ACOM 0.812 3.843 0.000***IN 1.119 5.951 0.000***IN 0.932 5.481 0.000***IN 0.869 5.850 0.000***IN
Control variables
AUDIT 0.041 0.519 0.605EX 0.143 1.724 0.088EX 0.044 0.618 0.538EX 0.009 0.136 0.892EX
IT1 0.027 0.347 0.729EX 0.080 0.955 0.342EX 0.3192.640 0.010**IN 0.038 0.536 0.593EX
IT2 0.001 0.014 0.989EX 0.080 0.963 0.338EX
0.042 0.532 0.596EX 0.064 0.921 0.359EX
IT3 0.028 0.357 0.722EX 0.027 0.322 0.748EX 0.058 0.532 0.596EX 0.086 1.243 0.217EX
INTER 0.148 1.988 0.049*IN 0.435 2.336 0.022*IN 0.086 1.189 0.237EX 0.118 1.716 0.089EX

LEVG 0.064 0.814 0.418EX 0.051 0.602 0.549EX 0.027 0.367 0.714EX 0.040 0.563 0.575EX
ASSETS 0.074 0.923 0.358EX 0.135 1.651 0.102EX 0.074 0.996 0.322EX 0.006 0.086 0.931EX
SALES 0.010 0.117 0.907EX 0.058 0.617 0.538EX 0.047 0.623 0.535EX 0.003 0.035 0.972EX
PROF 0.160 2.087 0.040*IN 0.198 2.482 0.015*IN 0.097 1.348 0.181EX 0.028 0.403 0.688EX
LIQD 0.097 1.209 0.230EX 0.071 0.859 0.393EX 0.138 1.926 0.057EX 0.007 0.094 0.926EX
*Significant at the 0.05 level.
**Significant at the 0.01 level.
***Significant at the 0.001 level.
Note: IN = variables included by the stepwise regression; EX = variables excluded by
the stepwise regression.
The existence of audit committees is the most important predictor of the extent of overall
voluntary disclosure, with the highest estimated coefficient of 3.843 significant at less
than the 0.0001 level and in the predicted direction. The positive sign for the existence of
audit committees provides support for Hypothesis H6 in that companies with an audit
committee disclose more voluntary information in their annual reports. This is consistent
with Ho and Wong (2001) and Barako et al. (2006) who found the existence of audit
committees to be positively significant. Furthermore, the existence of audit committees is



80 K. Samaha and K. Dahawy




significant at the 0.0001 level in explaining all types of voluntary information disclosure
(FIN, STRAT and CSR) and in the predicted direction. Hence, it may be appropriate for
regulatory authorities (e.g. EGX) to require all listed companies in Egypt to establish an
audit committee in order to secure more corporate transparency.
Next the most significant variables are block-holder ownership and the ratio of

independent directors on the board (with a p-value 0.018 and 0.019, respectively). Both
are in the predicted direction. The negative sign for the block-holder ownership provides
support for H2 in that voluntary disclosure increases with decreases in block-holder
ownership. This is inconsistent with Eng and Mak (2003) who found that the level of
voluntary disclosure is not significantly related with block-holder ownership. However,
block-holder ownership is significant at the 5% level in explaining one type of
information disclosure (START) and in the predicted direction. Information disclosure is
likely to be high in companies with a lower proportion of block-holder ownership. This is
consistent with Gray’s (1988) secrecy hypothesis which argues that where a firm’s shares
are closely held, there is a preference for confidentiality so that disclosure is restricted to
those who are closely involved with the management and financing of the firm.
The positive sign for the ratio of independent directors on the board provides support
for H5 in that companies with a higher proportion of independent directors on the board
disclose more voluntary information in their annual reports. A possible explanation of
this result is that INDEP in Egypt are likely to actively pressing the company to disclose
more non-mandatory information. This result supports the Williamson (1985) framework,
and the intuition that greater board independence is linked to more transparency and
better monitoring. This is inconsistent with Haniffa and Cooke (2002), Ho and Wong
(2001) and Ghazali and Weetman (2006) who found the ratio of independent directors to
total directors on board to be insignificant. Ho and Wong’s disclosure index score
includes the voluntary disclosure items perceived as most important by analysts but the
index score adopted in this study did not take this issue into consideration. This
significance presence of independent non-executive directors may reflect a similarity
between de jure and de facto independence. In contrast, Eng and Mak (2003) found a
significant negative association in Singapore companies. Similarly, Barako et al. (2006)
found a significant negative association in Kenyan companies. The results suggest that
external directors in Egypt play a complementary monitoring role to disclosure whereas
in Eng and Mak (2003) Singapore sample, and in Barako et al. (2006) Kenyan sample
they play a substitute role to disclosure. Again, the proportion of independent directors on
the board is significant at the 5% level in explaining two types of information disclosure

(STRAT and CSR) and in the predicted direction. Information disclosure is likely to be
high in companies with a higher proportion of independent directors on the board.
Number of shareholders is neither significant in explaining overall voluntary
disclosure nor any partial category of disclosure (except CSR), indicating that ownership
diffusion is not influencing voluntary disclosure levels. Thus, H1 that companies with a
larger number of shareholders are likely to have a higher extent of voluntary disclosure is
not supported. This is consistent with Ghazali and Weetman (2006) who found it to be
not statistically significant in explaining any type of information disclosure in Malaysia.
The results show that two aspects of ownership – managerial and government – are
not related to voluntary disclosure. Thus, H3 and H4 that voluntary disclosure increases
with decreases in managerial and government ownership are not supported. The result for
managerial ownership is inconsistent with the findings in Ghazali and Weetman (2006)
who found that information disclosure is likely to be less in owner-managed companies.



The Egyptian experience 81




However, the result for government ownership is consistent with Ghazali and Weetman
(2006) who found that government ownership is not influencing disclosure levels. The
result for both variables is inconsistent with the findings in Eng and Mak (2003) who
found that Singapore firms are less likely to provide voluntary information as managerial
and government ownership increases.
Profitability and internationality were significant in explaining overall voluntary
disclosure (VDIS) and financial (FIN) disclosure. More profitable companies disclose
additional information to signal that the company is being well managed and
professionally run while at the same time screening better-performing companies from

those performing less. International firms competing for foreign resources tend to expand
their financial and accounting disclosure as bonding for resource providers. This
expanded disclosure is assumed to reduce resource providers’ uncertainty about
transactions with the firm and, in turn, enables the firm to obtain resources at lower costs.
The variables representing type of auditor, industry type, size, leverage and liquidity
are not statistically significant in any of the models. The insignificance of the size
variable is inconsistent with prior research (Ghazali and Weetman, 2006; Ho and Wong,
2001) who indicated that larger companies seek to satisfy the information demands of
investors, to attract prospective investors to the company and to compete for international
funding. Large companies are also expected to have better governance structure with
clear separation between owners and managers. This could occur because large firms
need more financing capital than smaller firms. The general lack of influence of size on
voluntary disclosure may be related to the fact that that the EGX 100 companies are the
highest in terms of market capitalisation, so that all sample companies have reacted
similarly.
5.3 Tests on robustness of the model: sensitivity tests
A series of additional tests were conducted to test the models robustness. Firstly, an OLS
regression test was also conducted by dropping all control variables from the model. The
results of this test (not presented in the tables) were qualitatively similar to those of the
model with all control variables (Table 5). Secondly, a General White Test was
performed to examine the distribution of error terms of the OLS regression test conducted
on all firms and the results indicated that heteroskedasticity did not constitute a
significant threat to the validity of our findings.
Thirdly, in corporate governance research the issue of endogeneity is a major
concern. The issue of ownership structure suggests possible double counting particularly
in respect to ownership concentration where the block-holder may also be an insider like
the CEO. In this case, this could lead to collinearity issues. Whilst VIF factors and
correlation do not suggest grave issues this is still a concern. Therefore, we run separate
regressions with one ownership structure variable included but other ownership structure
variables excluded. The results (not reported here) indicated that all experimental

variables (INDEP, ACOM and BLKOWN) remained statistically significant with signs as
predicted in all tests. Thus, we infer from these results that multi-collinearity also did not
provide any threat to the validity of our findings.
Finally, we conducted four reduced models based on the corporate governance
variables and control variables found to be significant in the four full regression models
reported in Table 5. This method is utilised by Haniffa and Cooke (2002) in their study of
corporate governance and voluntary disclosure. The reduced regression model contains



82 K. Samaha and K. Dahawy




all the three test variables in the full VDIS model (INDEP, ACOM and BLKOWN) and
the internationality and profitability control variables, which was significant in two of the
four regression models. The reduced model regressions are all significant (p-value 0.000),
with F-ratios and adjusted R
2
s ranging from 12.547 and 0.368 for FIN to 23.128 and
0.528 for CSR. The results (not reported here) are similar to the full model in that all
experimental variables (INDEP, ACOM and BLKOWN) remained statistically
significant with signs as predicted in all tests.
6 Conclusion
This paper provides evidence in line with Eng and Mak (2003) and Cheng and Courtenay
(2004) in Singapore, Ho and Wong (2001) in Hong Kong and Ghazali and Weetman
(2006) in Malaysia that voluntary disclosure in the developing countries is problematic.
The overall disclosure level is very low at just 13.43%. This evidence about voluntary
disclosure is potentially problematic, as the low disclosure levels may be indicative of

less transparency. These scores place Egypt at a lower level than Singapore, Hong Kong
and Malaysia investigated by Eng and Mak (2003), Cheng and Courtenay (2004), Ho and
Wong (2001) and Ghazali and Weetman (2006), who found results of: 21.75%, 28.91%,
29% and 31%, respectively for the overall voluntary disclosure level. The variances of
these results support the need for individual country level studies and comparative
analysis.
Although, the Egyptian government has made great efforts to set a corporate
governance code for listed companies in recent years, the effect of those regulations or
rules are yet to be seen. In fact, inadequate information is available in the annual reports
of most Egyptian companies for investors. The understandability of voluntary
information disclosed by Egyptian companies is far from acceptable at present, as
insufficient information is available in their financial reports. A probable clarification for
the very low levels of voluntary disclosures by the sample companies may be related to
the poor enforcement process applied by the EGX as argued by Samaha and Stapleton
(2008).
In this paper, we provide evidence on the corporate governance determinants of
voluntary disclosure. Overall, the findings of this research are strongly consistent with the
predictions of agency theory (as measured by existence of audit committees, proportion
of independent directors on the board and block-holder ownership). The analysis shows
that type of auditor, size, leverage, number of shareholders and liquidity of the firm do
not seem to affect the extent of its voluntary disclosure, which limits the support for
agency and signalling theories. Since compliance is not related to manufacturing and
non-manufacturing sectors this limits the support for political process theory. This may
indirectly suggest that the positive accounting perspective may not be entirely applicable
for voluntary disclosure in Egypt.
The new provisions to enhance corporate governance are statistically significant in
explaining voluntary disclosure in annual reports. This implies that the efforts of
regulators to enhance corporate transparency did have some parallel effect in changing
the attitudes of listed companies towards more voluntary information disclosure at the
point of regulatory change. However, the level of disclosure is still minimal.

Consequently, these findings have policy implications regarding the strategies needed to
secure listed firms transparency in the future, and to which future research should be



The Egyptian experience 83




directed. These strategies should include improve corporate governance practices by
enforcing the CG code requirements. EGX should be one of the main driving forces
hand-in-hands with the companies department of the MOFT to regulate the activities of
listed companies. Companies should be required to include a minimum number of
independent directors on board committees dealing with potential conflict of interest
issues. There are no rules in Egypt that govern ‘independence’ of the board. The 2003
listing rules introduced the concept of ‘non-executive director’ for the first time. By most
reports, implementation has been slow, and few companies appoint truly independent
board members outside the banking sector.
Although the EGX listing rules of 2003 require the mandatory creation of audit
committees as a priority, to oversee and work with the companies internal and external
auditors and to secure more corporate transparency, however, it can be seen that just 21
of the top 100 companies on EGX have audit committees. This is supported by the World
Bank report (2004) which revealed that this rule is partially observed. This calls for
greater enforcement of the EGX listing rules. The next revision of the listing rules could
include enforcement for independent nomination and audit committees, and an expansion
of the role of the audit committee.
It is important to recognise that this research must be interpreted in light of some
limitations. Firstly, there is subjectivity inherent in the scoring of annual reports, which
can be managed by detailed rules and cross checking, but it can never be completely

eradicated. Secondly, there has been no attempt to assess the information needs of users,
and each item was weighted equally, assuming that each has the same information
content, whereas, in practice, some information may have higher value to users of
corporate annual reports than others. The choice not to weight the indices may (Naser and
Nuseibeh, 2003) or may not (Chow and Wong-Boren, 1987) affect measured compliance
rates. Finally, the study was cross-sectional in nature and only the reporting practices for
the financial year ended 2006 were examined, which may have been too close to the
change in corporate governance requirements. However, future research to measure
voluntary disclosure longitudinally would allow us to capture the effects of the
requirements after a learning period. Notwithstanding these limitations, the results are
sufficiently robust to consider their implications for Egyptian policy makers and to
warrant an extension to other emerging economies.
Acknowledgements
The authors are grateful to Sara Abdallah from the British University in Egypt and
Nermeen Fathy from the Cairo University for their immense assistance, and participants
in the 32nd European Accounting Association (EAA) Annual Congress – in Tampere,
Finland – for their helpful comments and suggestions. We are grateful to two anonymous
reviewers for constructive comments on this paper and to Prof. J-L.W. Mitchell Van der
Zahn – the guest editor – for guidance in final preparation of this paper.



84 K. Samaha and K. Dahawy




References
Abdel Shahid, S. (2003) ‘The impact of globalization on capital markets: the Egyptian EGX’,
The Arab Bank Review, Vol. 5, No. 2, pp.7–15 (October).

Adams, M. and Hossain, M. (1998) ‘Managerial discretion and voluntary disclosure: Evidence
from the New Zealand life insurance industry’, Journal of Accounting and Public Policy,
Vol. 17, pp.245–281.
Barako, D.G., Hancock, P. and Izan, H.Y. (2006) ‘Relationship between corporate governance
attributes and voluntary disclosures in annual reports: the Kenyan experience’, Financial
Reporting, Regulation and Governance, Vol. 5, No. 1, pp.1–25.
Bradbury, M.E. (1990) ‘The incentives for voluntary audit committee formation’, Journal of
Accounting and Public Policy, Vol. 9, pp.19–36.
Buzby, S.L. (1975) ‘Company size, listed versus unlisted stock, and the extent of financial
disclosure’, Journal of Accounting Research, Vol. 13, No. 1, pp.16–37.
Camfferman, K. (1997) Voluntary Annual Reports Disclosure by Listed Dutch Companies,
1945–1983. New York: Garland Publishing, Inc.
Camfferman, K. and Cooke, T.E. (2002) ‘An analysis of disclosure in the annual reports of UK and
Dutch companies’, Journal of International Accounting Research, Vol. 1, pp.3–30.
Carana (2000) ‘The Corporate Governance Policy Framework in Egypt’, Special Study provided to
the United States Agency for International Development By Carana Corporation under the
USAID Coordinating and Monitoring Services Project Contract # PCE-1-800-97-00014-00,
Task Order 800.
Cerf, A.R. (1961) Corporate Reporting and Investment Decision. Berkeley, CA: University of
California Press.
Chau, G. and Leung, P. (2006) ‘The impact of board composition and family ownership on audit
committee formation: evidence from Hong Kong’, Journal of International Accounting,
Auditing and Taxation, Vol. 15, pp.1–15.
Chau, G.K. and Gray, S.J. (2002) ‘Ownership structure and corporate voluntary disclosure in Hong
Kong and Singapore’, Int. J. Accounting, Vol. 37, pp.247–265.
Chen, C.J.P. and Jaggi, B. (2000) ‘Association between independent non-executive directors,
family control and financial disclosures in Hong Kong’, Journal of Accounting and Public
Policy, Vol. 19, pp.285–310.
Cheng, E.C.M. and Courtenay, S.M. (2004) ‘Board composition, regulatory regime and voluntary
disclosure’, Working Paper, Nanyang Business School, Nanyang Technological University,

Singapore.
Choi, F.D.S. (1973) ‘Financial disclosure and entry to the European capital market’, Journal of
Accounting Research, Vol. 11, No. 2, pp.159–179.
Chow, C.W. and Wong-Boren, A. (1987) ‘Voluntary financial disclosure by Mexican
corporations’, The Accounting Review, Vol. 62, No. 3, July, pp.533–541.
CMA (2003) Economic Bulletin-Stock Exchange. Cairo: Capital Market Authority Information
Center (In Arabic).
Collier, P. (1993) ‘Factors affecting the formation of audit committees in major UK listed
companies’, Accounting and Business Research, Vol. 23, pp.421–430.
Cooke, T.E. (1989) ‘Voluntary corporate disclosure by Swedish companies’, Journal of
International Financial Management and Accounting, Vol. 1, No. 2, pp.171–195.
Cooke, T.E. (1991) ‘An assessment of voluntary disclosure in the annual reports of Japanese
corporations’, Int. J. Accounting, Vol. 26, No. 3, pp.174–189.
Cooke, T.E. (1998) ‘Regression analysis in accounting disclosure studies’, Accounting and
Business Research, Vol. 28, No. 3, pp.209–224.



The Egyptian experience 85




Dahawy, K. and Conover, T. (2007) ‘Accounting disclosure in companies listed on the Egyptian
stock exchange’, Middle Eastern Finance and Economics, Vol. 1, pp.1–20.
Dahawy, K., Merino, B. and Conover, T. (2002) ‘The conflict between IAS disclosure requirements
and the secretive culture in Egypt’, Advances in International Accounting, Vol. 15,
pp.203–228.
Denis, D.K. and McConnell, J.J. (2002) ‘International corporate governance’, Working Paper,
Krannert Graduate School of Management, Purdue University.

Depoers, F. (2000) ‘A cost-benefit analysis of voluntary disclosure: some empirical evidence from
French listed companies’, The European Accounting Review, Vol. 9, No. 2, pp.245–263.
EGX (2006) ‘Cairo and Alexandria Stock Exchange’, Fact Book, December, Cairo, pp.1–61
(in Arabic).
Eng, L.L. and Mak, Y.T. (2003) ‘Corporate governance and voluntary disclosure’, Journal of
Accounting and Public Policy, Vol. 22, pp.325–345.
Fama, E.F. (1980) ‘Agency problems and the theory of the firm’, Journal of Political Economy,
Vol. 88, pp.88–307.
Fama, E.F. and Jensen, M.C. (1983) ‘Separation of ownership and control’, Journal of Law and
Economics, Vol. 26, No. 2, pp.301–326.
Fan, P.H. and Wong, T.J. (2002) ‘Corporate ownership structure and the informativeness
of accounting earnings in East Asia’, Journal of Accounting and Economics, Vol. 33,
pp.401–425.
Ferguson, M.J., Lam, K.C.K. and Lee, G.M. (2002) ‘Voluntary disclosure by state-owned
enterprises listed on the stock exchange of Hong Kong’, Journal of International Financial
Management and Accounting, Vol. 13, No. 2, pp.125–152.
Field, A. (2003) Discovering Statistics Using SPSS for Windows: Advanced Techniques for the
Beginner. London: Sage publications.
Firth, M. (1979) ‘The effect of size, stock market listings, and auditors on voluntary disclosure in
corporate annual reports’, Accounting and Business Research, Vol. 9, pp.273–280.
Forker, J.J. (1992) ‘Corporate governance and disclosure quality’, Accounting and Business
Research, Vol. 22, pp.111–124.
Frost, C.A. and Pownall, G. (1994) ‘Accounting disclosure practices in the United States and the
United Kingdom’, Journal of Accounting Research, Vol. 32, No. 1, pp.75–98.
Ghazali, N.A.M. and Weetman, P. (2006) ‘Perpetuating traditional influences: voluntary disclosure
in Malaysia following the economic crisis’, Journal of International Accounting, Auditing and
Taxation, Vol. 15, pp.226–248.
Gray, S. (1988) ‘Towards a theory of cultural influence on the development of accounting systems
internationally’, Abacus, Vol. 24, No. 1, pp.1–15.
Gray, S.J., Meek, G.K. and Roberts, C.B. (1995) ‘International capital market pressures and

voluntary annual report disclosures by U.S. and U.K. multinationals’, Journal of International
Financial Management and Accounting, Vol. 6, No. 1, pp.43–68.
Gray, S.J., Radebaugh, L.H. and Robert, C.B. (1990) ‘International perceptions of cost constraints
on voluntary information disclosures: a comparative study of UK and US multinationals’,
Journal of International Business Studies, Fourth Quarter, pp.597–622.
Gul, F.A. and Leung, S. (2004) ‘Board leadership, outside directors’ expertise and voluntary
corporate disclosures’, Journal of Accounting and Public Policy, Vol. 23, pp.1–29.
Haniffa, R.M. and Cooke, T.E. (2002) ‘Culture, corporate governance and disclosure in Malaysian
corporations’, Abacus, Vol. 38, pp.317–349.
HassabElnaby, H.R. and Mosebach, M. (2005) ‘Culture’s consequences in controlling agency
costs: Egyptian evidence’, Journal of International Accounting, Auditing and Taxation,
Vol. 14, No. 1, pp.19–32.