achieving high quality, comparable financial reporting:
TRANSCRIPT
An Investigation of Compliance with International Accounting Standards by Listed Companies in the Gulf Co-Operation Council Member States
Abstract
This study investigates the extent of compliance with international accounting standards (IASs) by companies in the Gulf Co-Operation Council (GCC) member states (Bahrain, Oman, Kuwait, Qatar, Saudi Arabia and the United Arab Emirates). Based on a sample of 137 companies (436 company-years) we found that compliance increased over time, from 68% in 1996 to 82% in 2002. There was significant variation in compliance among the six countries and between companies based on size, leverage, internationality and industry. Non-compliance was greater than that observed in developed countries and reflected limited monitoring and enforcement by the bodies responsible for overseeing financial reporting and limited comprehensiveness of audits by external auditors.
Key words: Adoption of International Accounting Standards (IAS), Financial reporting compliance; audit function; enforcement of accounting standards, Gulf Co-Operation Council (GCC) member states.
2
An Investigation of Compliance with International Accounting Standards by
Listed Companies in the Gulf Co-Operation Council Member States
INTRODUCTION
Since 1973 the London based International Accounting Standards Board (IASB), and its
predecessor organisation the International Accounting Standards Committee (IASC), have
been developing international accounting standards (IASs) for use by private sector entities
throughout the world. By 2005, it was estimated that over 90 countries had adopted or planned
to adopt IASB standards in the future (IASB, 2005). Rapid globalisation of financial markets
has given rise to the demand for internationally comparable financial reporting. Harmonisation
of accounting standards, that is, determining common principles and rules to underpin external
financial reporting, has been seen as an important part of achieving more transparent and
consistent reporting at an international level.
However, comparable reporting is unlikely to be achieved by harmonised standards alone.
Enforcement of financial reporting standards has been identified as a key element in the process
of promoting comparable reporting (SEC, 2000; CESR, 2003). Little is known about the extent
of IASs compliance in jurisdictions where use of IASs is mandatory and subject to monitoring
and enforcement. Most previous IASs compliance studies have been in settings where IASs
use is voluntary or not subject to national enforcement (Nobes, 1990; Street, Gray and Bryant,
1999; Tower, Hancock and Taplin, 1999; Street and Bryant, 2000; Street and Gray, 2001;
Glaum and Street, 2003).
3
The aim of this paper is to examine compliance with IASB standards in a group of developing
countries which from 1986 made IASs use mandatory with the stated aim of improving the
quality of financial reporting and promoting capital market development. The countries are the
six Gulf Cooperation Council (GCC) states, namely Bahrain, Oman, Kuwait, Saudi Arabia,
Qatar and the United Arab Emirates (UAE). We investigate factors which are associated with
greater compliance with IASB standards by companies operating in these states, in order to
provide insights about the extent to which compliance, and therefore international
harmonisation, has been achieved in an economically and politically important region.
Using a self-constructed compliance checklist, we measure the extent of listed companies’
compliance with 14 IASs over the period 1996-2002. Our sample includes 137 companies and
436 company-years. Data was collected manually from annual reports for three years for
companies from Oman, Kuwait and Saudi Arabia (1996, 1999 and 2002), for four years for
companies from Bahrain (1996, 1997, 1999 and 2000) and for three years for companies from
Qatar and the UAE (1999, 2000 and 2002), reflecting different national IASs adoption dates in
the region.
For the GCC as a whole, the average level of compliance for all years for both disclosure and
measurement requirements of the 14 IASs was 0.75; that is, on average companies complied
with 75% of the checklist’s disclosure and measurement items. The level of compliance for
disclosure requirements was 0.69 and for measurement requirements it was 0.81. The level of
compliance increased over time, from 0.68 in 1996 to 0.82 in 2002, indicating that compliance
has been improving in the region.
4
The level of disclosure compliance is lower than that observed in developed countries such as
Australia (0.94; Tower, Hancock and Taplin, 1999), Germany (0.81; Glaum and Street, 2003)
and Switzerland (0.74; Street and Gray, 2001). In addition, the level of measurement compliance
is lower than that in Germany (0.86) and Switzerland (0.92; Street and Gray, 2001). This
suggests that incentives for compliance are less in the region than in developed countries and
that there is scope for further improvement in national monitoring and enforcement mechanisms
in the GCC member states.
In addition, the level of compliance for both disclosure and measurement varies among the GCC
member states. This reflects differences in each state’s position over time in relation to the
adoption of IASs, in the effectiveness of its enforcement body and in the nature of the audit. The
highest average compliance level for all years is in the UAE (0.80), followed by Saudi Arabia
(0.78), Kuwait (0.75), Oman (0.74), Bahrain (0.73) and Qatar (0.70). Compliance increased over
time in each state. Based on between-country differences in the financial reporting framework
discussed in this paper, we predicted that highest compliance would be in Kuwait and Oman,
followed by UAE, Saudi Arabia, Bahrain and Qatar in that order. There was only partial support
for this prediction, possibly because the importance of institutional differences were overstated,
or common cultural and other factors already captured in the control variables were too
dominant to yield clear results.
We found compliance varies across companies according to a number of attributes. Industry is
one factor. Companies with a higher level of compliance with IASs are also likely to be larger,
have higher leverage and a greater international presence. The result differs from that found in
5
developed countries (Street and Bryant, 2000; Glaum and Street, 2003) suggesting that company
attributes associated with compliance with IASs may differ between developed and developing
countries. In these six developing countries higher compliance is associated with greater
company size, leverage and foreign shareholders, while Big 4/5 auditor and dual listing on a
foreign stock exchange are among the more significant variables explaining compliance in
developed countries.
The contribution of this study is to extend IASs compliance studies by examining a region
which has not been the focus of previous research. Prior studies have addressed individual
countries (for example Abdelrahim, Hewaidy and Mostafa (1997) and Abdelrahim and Mostafa
(2000) studied companies from Kuwait and Joshi and Al-Mudhahki (2001) included companies
from Bahrain) and considered a relatively small sample (22 – 37 companies) and a limited
number of IASs (between one and three) at one date. This study includes all listed companies
for which data was available in the six countries. It uses a comprehensive compliance index,
based on all IASs applicable in the region at the time and provides a systematic measure of
compliance over the period 1996-2002 when IASs increased in use in the region.
Since IASB standards are used in a range of countries, it is useful to explore adoption in a
region which displays unique institutional and cultural features. Our study provides insights
about the factors influencing adoption and compliance in an important group of developing
countries and suggests the factors may differ between developing and developed countries. In
general, the outcome of adoption of IASs in developing countries is of interest as it affects the
overall level of international harmonisation which can be achieved through use of IASB
standards.
6
The results of this study have important implications for regulators and enforcement bodies in
the GCC region. If their goal is greater compliance with IASs, there is ample scope for further
improvement. While there are existing penalties for those contravening the law, further
enforcement action seems necessary to maximise compliance. The results suggest that in the
face of scarce resources, efforts might be more effective if they focused on monitoring
companies that are smaller, operate locally and borrow proportionately less. Thus, the findings
may be relevant to national regulators and the Gulf Co-Operation Council Accounting and
Auditing Organization (GCCAAO) in their efforts to harmonise financial reporting.
This study also contains important lessons for international organisations interested in the spread
of IASs in developing countries. It provides evidence about factors associated with compliance
and shows how compliance has evolved over time. The GCC experience could be relevant to
regulators interested in improving compliance with accounting standards and the effectiveness
of monitoring and enforcement mechanisms in other countries.
The remainder of the paper is organised as follows. The next section describes the institutional
framework for financial reporting in the GCC member states, reviews relevant literature and
proposes hypotheses. The third section describes sample selection, data collection and statistical
method. Results are presented and robustness checks reported in the fifth section. Conclusions,
limitations and suggestions for further research complete the paper.
BACKGROUND AND HYPOTHESES
7
Prior IASs compliance studies report that differences between companies in their level of
compliance reflect their country of origin (Tower et al., 1999; Street and Bryant, 2000; Street
and Gray, 2001) implying that there are important features in the national financial reporting
framework which affect compliance. A country’s financial reporting framework has a key role
in specifying financial reporting requirements, in establishing a due process for monitoring and
enforcing accounting standards, and in influencing the extent of compliance with those
standards. It is therefore to be expected that the framework (which is proxied by country in
cross-country studies) will influence the level of compliance. Street and Bryant (2000), Street
and Gray (2001) and Glaum and Street (2003) also found that presence of a Big 4/5 auditor and
being cross-listed were associated with greater compliance. Prior studies have not found that
other company level attributes are significant explanatory factors for level of compliance, except
for industry (Street and Gray, 2001). Consequently, we focus on country level differences in the
development of our hypotheses. We describe below the institutional framework of the six
countries included in this study and consider how differences in the framework may affect the
level of compliance.
Country factors affecting compliance
During the period of this study (1996-2002) all six countries had company law administered by
the relevant government department and securities market law by the national stock exchange.
Company law required that financial statements (balance sheet and profit and loss statement) be
prepared and submitted to the shareholders and government (the registrar of companies). It
further required that financial statements give a true and fair view of the company’s operations.
All six countries had laws governing the audit of financial statements and requirements for
auditors to be licensed with the government. All countries required auditors to have work
8
experience, two (Kuwait and Saudi Arabia) had examinations for admission as auditors and
three (Oman, Saudi Arabia and UAE) had professional training requirements. Four countries
(Bahrain, Oman, Kuwait, and the UAE) required the use of International Auditing Standards
(ISA). All countries had penalties which could be applied when auditors breached regulations.1
All listed companies in Oman, Kuwait and Bahrain were required to comply with IASs from
1986, 1991 and 1996 respectively. In Saudi Arabia, Qatar and UAE the central banks required
banks and finance and investment companies to comply with IASs in 1992, 1999 and 1999
respectively.2 According to Abdallah (2001, p. 136), the involvement of foreign ownership in
banks led the Saudi Arabia Monetary Agency (SAMA, which is responsible for registration,
supervision and monitoring of financial institutions’ activities and reporting) to require them to
adopt IASs to provide reliable, understandable and comparable financial statements to local and
foreign investors. The widespread use of the IASs in regard to the financial statements of banks
and other financial institutions in many countries across the world led the UAE Central Bank to
adopt IASs, to increase understanding and comparability of financial reports of banks operating
in the UAE in order to enhance their relative position (UAECB, 1999) . Similarly, in Qatar, the
increasing number of foreign banks that voluntarily used IASs led the Qatar Central Bank
(QCB) to require all banks (foreign and national) to adopt IASs. The reasons given were to unify
accounting standards used by banks, to better observe their performance and to increase the level
of comparability of financial reports, thereby meeting the greater demand for information from
shareholders and the public (QCB, 2004).3
The rapid growth and opening up of capital markets in the GCC member states and pressure
from multinational corporations led the governments to adopt IASs, in the expectation that their
9
adoption will meet greater demands of shareholders and of local and international investors for
more detailed information and for greater comparability in financial reporting (Azzam, 1998;
Hassan, 1998; Shuaib, 1999; Al-Basteki, 2000; Hussain, Islam, Gunasekaran and Maskooki,
2002; Naser and Nuseibeh, 2003a).
By the end of the study period (1996-2002) all countries had an enforcement body in place,
however the extent of their activities varied. Only in Oman and Kuwait did the body check
compliance with IASs (except for UAE where compliance with IAS 30 Disclosures in the
Financial Statements of Banks and Similar Financial Institutions was reviewed). In the other
four countries, the enforcement body relied upon the auditors to check for compliance with
IASs. Member states had laws in place to deal with non-compliance, for example company
officers or directors could be prosecuted (all countries except Kuwait) and fined (all countries)
for non-compliance with regulations (Al-Shammari, Brown and Tarca, 2007).
However, the comprehensiveness of the audit and enforcement function is not the same in all
states and this could be an important explanatory factor for differences in compliance between
countries. The registration requirements of auditors and the existence of an effective supervisory
body could affect the ability of external auditors to detect (and their preparedness to report)
instances of non-compliance. The effectiveness of the enforcement body would be reflected in
its willingness to investigate any allegation against an auditor and to impose a penalty on an
auditor who did not report an instance of non-compliance. For this reason we might expect to
observe between-state differences in the extent of non-compliance with IASs. As noted above,
prior studies in developed countries have found greater compliance among companies audited
by Big 4/5 auditors and where International Auditing Standards are used (Street and Bryant,
10
2000; Street and Gray, 2001; Glaum and Street, 2003). Nevertheless, cases where auditors assert
that financial statements comply with IASs when the accounting policies and notes indicate
otherwise have been observed (Cairns, 1997, p. 2).
Arrangements for monitoring companies’ compliance with IASs differ among the member
states.4 The enforcement bodies in Qatar, Saudi Arabia, the UAE and Bahrain rely primarily on
the independent audit report. Those of Kuwait and Oman undertake their own monitoring.
Generally, there is little if any proactive monitoring, the exceptions being Oman and Kuwait;
however, the enforcement body in the UAE does investigate some disclosures. Correspondingly,
provision is made for reactive monitoring if enough shareholders complain. Reactive monitoring
is conducted by appointing another external auditor in all states except Oman.
It seems that the provisions concerning compliance monitoring are less stringent in Qatar,
Bahrain, the UAE and Saudi Arabia. Potential reasons for the comparative absence of effective
monitoring by these enforcement bodies are a lack of professional training, salaries insufficient
to attract technically qualified staff and a lack of funding in that surveillance departments have
other responsibilities as well. To date, no audit firm, board member or manager has been taken
to court over a non-compliance matter except in Oman and Kuwait.
Thus compliance with IASs and other regulations is not enforced uniformly across the member
states. This is another reason to expect that the level of compliance varies by state. In Qatar,
there is no effective enforcement body; whereas, in Saudi Arabia and the UAE, the enforcement
bodies monitor at least some regulations. Therefore, it can be expected that the level of
compliance in the UAE and Saudi Arabia by banking and investment companies is higher than
11
in Qatar. Assuming, as seems to be the case, that the enforcement bodies in Kuwait and Oman
are more active than in Bahrain, it can be expected that the level of compliance with IASs in
Oman and Kuwait is higher than in Bahrain. Furthermore, the average level of compliance with
IASs in Kuwait and Oman is expected to be higher than in Bahrain, Qatar, the UAE and Saudi
Arabia.
In general, the level of compliance with IASs is expected to have increased over time in the
GCC region if only because enforcement activities have increased in all member states.
Recently, at least two external auditors have been required to audit companies’ accounts in Qatar
and Saudi Arabia. In the UAE, although there is no legal requirement for more than one external
auditor to be appointed, common practice is for UAE banking and investment companies to
have their accounts audited by two and sometimes three external auditors. In Bahrain,
shareholders have recently begun to question instances of a possible violation of IASs or other
regulations. This suggests that they may have become more aware of and more willing to pursue
their statutory rights. In Oman and Kuwait, the enforcement bodies have become more active in
monitoring companies’ compliance with IASs and other regulations in recent years.
We predict that companies from countries that adopted IASs earlier (Oman, Kuwait and Saudi
Arabia) are more likely to have a higher level of compliance than companies from countries that
adopted IASs later (Bahrain, Qatar and UAE). Additionally, companies from countries with
stronger enforcement practices (Oman and Kuwait) are more likely to have a higher level of
compliance than companies from countries with relatively weaker enforcement practices.
Similarly, companies from countries where there are entrance examinations or professional
training requirements (Kuwait, Saudi Arabia, Oman, UAE) are more likely to have a higher
12
level of compliance than companies from countries without qualification requirements.
Accordingly, it is hypothesized that:
H1: The level of compliance with IASs in Kuwait and Oman is higher than in Saudi
Arabia, Qatar, Bahrain and UAE.
H2: The level of compliance with IASs in the UAE is higher than in Saudi Arabia,
Bahrain and Qatar.
H3: The level of compliance with IASs in Saudi Arabia is higher than in Bahrain and
Qatar.
H4: The level of compliance with IASs in Bahrain is higher than in Qatar.
Collectively, these four hypotheses predict that compliance in Kuwait and Oman exceeds that in
the UAE, which exceeds that in Saudi Arabia, which exceeds that in Bahrain, which exceeds
that in Qatar, all other things being equal. However, there are several other factors which may
affect compliance and these are discussed below.
Company factors affecting compliance
Prior studies have examined the relationship between compliance and a number of company
attributes including size, age, internationality, profitability, leverage, ownership diffusion,
liquidity and industry as well as type of auditor (discussed above). Attributes other than type
of auditor, internationality (foreign stock exchange listing) and industry have not been found
to be related to compliance (Tower et al., 1999; Street and Bryant, 2000; Street and Gray, 2001;
Glaum and Street, 2003). Variables representing company attributes are included in this study
as control variables. The focus of prior studies has been mainly on developed countries so it is
13
important to include these variables to determine whether they have explanatory power for the
companies from developing countries included in this study. There are several reasons why
this may be the case.
Larger companies are more visible and therefore may be more likely to comply with
accounting standards. Watts and Zimmerman (1978) and Holthausen and Leftwich (1983) have
argued that larger companies act to protect their reputation and avoid government intervention.
While the authors based their ideas on developed markets, they also apply in the GCC
countries where large companies are politically visible and economically important. Recent
privatisations of large state owned companies mean that they are a focus of government and
investor attention. In addition, larger companies have more resources to spend on compliance
and are less likely to be affected by disclosure of proprietary information than smaller
companies. Larger companies may be older with more established reporting systems, meaning
that compliance is less costly for them.
Further, larger companies are likely to be more international, that is, to have more foreign
investors, foreign sales or to have foreign stock exchange listings. Street and Bryant (2000),
Street and Gray (2001) and Glaum and Street (2003) have shown that companies which are
cross-listed have higher levels of compliance. In the GCC setting companies are not listed on
foreign stock exchanges, however they do seek foreign investors. This may provide an
incentive for greater compliance, to make financial reporting more transparent and comparable
and to increase the company’s credibility.
14
More profitable companies may comply more with IASs to signal their success and strength to
potential foreign investors and market participants. Dumontier and Raffournier (1998) argued
that compliance with IASs by profitable companies is one way to signal their superior
performance to the market. Although Dumontier and Raffournier do not mention it, another
signalling reason is that if IASs lead to more conservative income, companies with larger profits
that can ‘afford’ to comply with IASs might do so to signal their greater earnings ‘quality’.
Agency theory (Jensen and Meckling, 1976) suggests that agents will increase disclosure to
their principals to reduce information asymmetry and thus agency costs. Companies with
higher leverage can be expected to disclose more information to reduce agency costs by
reassuring debtholders that their interests are protected. A strategy of providing greater
disclosure can be expected to lead to more compliance with IASs. In the GCC setting, banks
play a dominant financial role in the economy (Azzam, 1998; Oweiss, 2000; Al-Shimmiri, 2003)
and have substantial and enduring financial relationships with companies. While banks may not
need to rely on public information to the same extent as in, say, the USA, it can still be argued
that companies with a higher level of leverage may be more likely to disclose additional
information, in order to reduce agency costs and information asymmetry with shareholders, than
companies with a lower level of leverage.5
Companies with higher leverage have, by definition, less equity and probably, in turn, relatively
fewer shareholders. Consequently, they are more likely to be subject to higher equity risk than
companies with a lower level of leverage and, therefore, be subjected to greater shareholder
demand for information to assess both the probability their company will meet its debt
obligations and the riskiness of future cash flows arising from their investments.
15
In the GCC member states, three shareholder groups typically have substantial equity ownership
in companies listed on the GCC stock exchanges. These groups are the government and its
agencies, dominant families and institutional investors. These groups may influence the level
and quality of disclosure and the level of compliance with IASs. In the GCC member states,
these groups are considered insiders because they usually have representatives on the
companies’ boards of directors; thereby, they have better access to internal information.6
Therefore, we expect that companies with more insiders (that is, more closely held ownership)
have less motivation to comply with IASs than companies with widely held share ownership.
It can also be argued that companies with lower liquidity ratios are more likely to comply with
IASs than companies with higher liquidity ratios. Craswell and Taylor (1992) pointed out that
while debtholders may negotiate with companies for release of additional information,
shareholders of listed companies depend on public disclosure. Likewise, demand for company
information by the GCC countries’ Ministries of Commerce and shareholders is expected to
increase when liquidity is lower. As a result, a company with a lower liquidity ratio is more
likely to disclose additional information through compliance with IASs, in order to reduce
agency costs and assure shareholders and the Ministries of Commerce that its financial position
is sufficient to meet short-term commitments.
A final factor which may be relevant is industry membership. Industry type may capture
sensitivity to political costs not captured by other proxies that differ by industry (Ball and Foster,
1982; Bazley, Brown and Izan, 1985; Watts and Zimmerman, 1986, p. 239). In this connection,
Ball and Foster (1982) argued that industry type can be a more appropriate proxy for political
16
cost sensitivity than size. Industry type may also proxy for differences in the proprietary costs of
disclosure, which have been found to be correlated with accounting method choice (Malone,
Fries and Jones, 1993). Ferguson, Lam and Lee (2002), for instance, argued that companies in
identifiable, highly competitive industries may disclose less information to avoid loss from the
leakage of proprietary information. Accounting and disclosure practices are often observed to
reflect industry commonalities. Malone et al. (1993) and Wallace, Naser and Mora (1994), for
example, argued that the adoption of industry-related practices may lead to differential levels of
disclosure on similar items in financial reports published by companies in different industries.
As a consequence, one can expect that companies in a given industry may comply more closely
with a particular IASs that is more applicable to their activities, such as the banking industry
with IAS 30. Accordingly, different industry groups may have a different compliance level with
IASs because of differences in their sensitivity to political costs not captured fully by other
proxies, to proprietary costs and to other unspecified industry commonalities.
Proxies for company attributes are included in the regression models to investigate whether
they are associated with compliance in the manner predicted by the theories and research
outlined above. In summary, compliance is predicted to differ between countries and by
industry and to increase with appointment of an audit firm affiliated with one of the large
international audit firms. It may also increase with company size, leverage, profitability,
internationality, ownership diffusion and the company’s maturity and decrease with greater
liquidity.
DATA AND METHOD
17
Selection of countries and companies
The six GCC member states (Bahrain, Oman, Kuwait, Saudi Arabia, Qatar and the UAE) were
selected for study because they are a group of developing countries that made IASs mandatory
for some or all listed companies during the period 1986 to 1999. The turning point for the
increased acceptance of IASs in the region was 1999, when Qatar and the UAE adopted IASs.
For Oman, Kuwait and Saudi Arabia the study covers three years (1996, 1999 and 2002)
because IASs became mandatory prior to 1996 for all types of company in Oman and Kuwait,
and for banks and investment companies in Saudi Arabia. For Bahrain the study covers four
years (1996, 1997, 1999 and 2002) because all types of company in Bahrain were required to
adopt IASs in 1996. Since companies in Bahrain might not have had adequate time to adopt
IASs in 1996, the study also covers 1997. For Qatar and the UAE the study covers three years
(1999, 2000 and 2002). Banks and investment companies in Qatar and the UAE were required to
adopt IASs in 1999; and because those companies might not have had enough time to shift to
IASs by 1999, the study covers 2000 as well.
A total of 200 companies were listed over the full period. Given the small size of the population,
the study aimed to include all relevant listed companies. There were nine (4.5%) Islamic
institutions listed on the GCC stock exchanges. They were excluded because they follow
specific disclosure requirements and are exempted from following IASs. A further 28 companies
(14%) were excluded because they were delisted for various reasons (liquidations, litigations,
mergers and other reasons not provided) during the period of the study, and their annual reports
were either not issued or were not available. A further seven companies (3.5%) were excluded to
avoid double-counting, since these companies are cross-listed on the Kuwait Stock Exchange
(KSE) and their home stock exchange (one of the other GCC stock exchanges). Consequently,
18
the number of sample companies was reduced from 200 to 156. Annual reports were requested
by mail from these companies with two follow-up requests to non-respondents. A full set of
annual reports (three or four years) was collected for 137 companies, giving 436 company-years
(Table 1).
TABLE 1 ABOUT HERE
Table 2 provides descriptive statistics for the sample companies. Mean size was US$1,732
million, ranging from US$3 million to US$27,691 million. On average, companies in Saudi
Arabia were the largest in the sample (US$13,159 million) and in Oman the smallest (US$203
million) reflecting the concentration of large banking sector companies in Saudi Arabia, the
UAE and Qatar. Leverage ranged from zero to 0.98, with a mean of 0.45. The figure of zero
indicates that some companies effectively had no debt, while a ratio of 0.98 implies that the
company had little equity. Qatar, Saudi Arabia and the UAE companies had the highest
leverage because their sample was limited to banks and investment companies. Liquidity for
the full sample ranged from 0.13 to 336, with a mean of 4.36. Profitability ranged from –0.97
to 0.44, with a mean of 0.08.7 The proportion of foreign ownership ranged from zero to 0.59,
with a mean of 0.04. This small mean indicates that the GCC member states were in the early
stages of attracting foreign investors. Saudi Arabian companies had the highest proportion of
foreign ownership (a mean of 0.25) and the lowest mean was Qatar (zero) as Qatar had not yet
opened its share market to direct ownership by foreign investors.8 Institutional ownership ranged
from zero to 0.80, with a mean of 0.26 for the whole sample. Kuwait had the highest
institutional ownership (mean 0.36) and Saudi Arabia the lowest (0.11). Company age ranged
from 1 to 50 years, with a mean of 21.59.
19
TABLE 2 ABOUT HERE
Of the 436 annual reports, 390 were audited by local audit firms with an international
affiliation (Big Five); 46 were not. Of the 390, 109 were audited by affiliates of Ernst &
Young, followed by KPMG 108, PricewaterhouseCoopers 89, Arthur Andersen 52, and
Deloitte Touche Tohmatsu 32. The market for auditing listed companies in the GCC member
states was obviously dominated by local audit firms with an international affiliation. The
industry classification of the sample was 58 banking and investment companies (171 company-
year observations), eight insurance companies (27 observations), 38 manufacturing companies
(130 observations) and 33 services companies (108 observations). These industry classifications
reflect the strong representation of financial services in the GCC member states.
Data collection
The level of mandatory compliance with IASs was measured by a self-constructed compliance
index consistent with prior compliance studies (Tower et al., 1999; Street and Bryant, 2000;
Street and Gray, 2001; Glaum and Street, 2003). The checklist was based on 14 standards;
namely, IAS 1, 10, 14, 16, 18, 21, 23, 24, 27, 28, 30, 32, 33 and 37. The 14 selected IASs were
the ones most applicable to GCC companies and have been found to be among the most
important and controversial standards in prior studies (Street and Bryant, 2000; Street and Gray,
2001). IASs not included were those not applicable to GCC companies (IAS 12, 15, 19, 26, 29,
34, 11, 17, 20, 22, 31, 35, 38), not applicable in the study period (IAS 39) or those where there
was little or no within-sample variation (IAS 2 and 7) or little or no disclosure (IAS 8).9
20
The checklist was developed based on the requirements shown in IASs published by the IASC
and IASB. It distinguishes between disclosure and measurement requirements because
compliance may differ between them (Street and Gray, 2001). Table 3 shows the number of
disclosure and measurement items in the checklist in total and for each year of the study. The
checklist (available from the authors) was validated by comparison with checklists used by Ernst
& Young and KMPG in Kuwait. In addition, its validity and comprehensiveness was confirmed
by an external auditor from Kuwait.
TABLE 3 ABOUT HERE
Since several different reporting years for each member state were covered in this study, a range
of standards were applicable in each member state. Each standard’s applicability to companies’
annual reports was determined and the relevant portions of the checklist were used in data
collection. For example, IAS 1 Disclosure of Accounting Policies was effective from 1 January
1975 until its replacement by IAS 1 Presentation of Financial Statements from 1 July 1998.
Therefore, for the purposes of this study, companies’ annual reports were checked for
compliance with the original IAS 1 in 1996 and 1997, while reports for 1999, 2000 and 2002
were checked against its replacement. The scoring procedure takes account of such differences
in requirements over time, by expressing the compliance index in ratio form.
Data for the index were extracted from the annual reports. The full annual report was read and
data hand collected by one researcher to ensure consistency in coding. Each disclosure item on
the checklist was assigned a value of one if it was disclosed and zero if the item obviously
applied but was not disclosed. Items obviously not applicable to the report were coded as “NA”.
21
Items where not enough information was given to discern their applicability were coded as
“DK” for “do not know”. The primary index is an overall compliance measure, excluding NA
and DK items, and comprising a disclosure and a measurement compliance measure. Robustness
tests were completed to determine the sensitivity of the results to the treatment of the potentially
ambiguous NA and DK items. Items in the checklist were not weighted because this process can
introduce subjectivity and bias (Cooke, 1989 and 1991; Wallace and Nasser, 1995).
Regression model
Multivariate analysis was used to explore state-by-state effects, time trends and relationships
between the level of compliance and company attributes. The basic model is shown below.
Since the dependent variable (the score for the compliance index) lies between zero and one, it
was transformed by taking the logarithm of the “odds ratio” (Hanushek and Jackson, 1977, pp.
187-189; Fox, 1997, p. 151). If the overall level of compliance with IASs for a company is given
by P, the logarithm of the odds ratio Y, is given by:
Y = log ⎟⎠⎞
⎜⎝⎛
− PP
1
This transformation has been used in prior disclosure studies (Ahmed and Nicholls, 1994;
Ahmed, 1996; Inchausti, 1997; Makhija and Patton, 2004). The transformed compliance index
was then regressed on company attributes and year-by-year dummy variables in order to assess
the effect of each attribute on compliance levels and any time trends. Specifically, the following
regression model for the full model is estimated. (Company-year subscripts are omitted for
convenience.)
εββ ++= ∑ =
=
20
10i
i ii XY ,
where:
22
Y = compliance score
0β = constant term
1X = Company size
2X = Leverage
3X = Profitability
4X = Liquidity
5X = Internationality
6X = Age
7X = Ownership diffusion
8X = Auditor
9X = Industry 1
10X = Industry 2
11X = Industry 3
12X = Bahrain
13X = Oman
14X = Kuwait
15X = Saudi Arabia
16X = Qatar
17X = 1996
18X = 1997
19X = 1999
20X = 2000
ε = Error term.
Ten company attributes and year-by-year dummy variables were examined for their association
with the level of compliance to discover if the level of compliance with IASs was influenced by
company attributes and time trends as well as institutional differences (proxied by country
variables). Data for four of them (company size, leverage, internationality and ownership
diffusion) were obtained from companies’ annual reports or from annual companies guides
published by the stock exchanges. Data for the remaining six (country of origin, auditing firm,
company age, industry, liquidity and profitability) were obtained from the companies’ annual
reports.
23
RESULTS
Level of compliance
The level of mandatory compliance (measurement and disclosure) with the 14 standards,
averaged over all companies and all years, was 0.75. The mean level of disclosure compliance
was 0.69 and measurement compliance was 0.81. The level of compliance averaged over all
companies increased over time, from 0.68 in 1996 to 0.82 in 2002 (Table 4). This indicated that
compliance with IASs has been improving in the region; however, no company in any year
within the period studied fully complied with all relevant IASs.
TABLE 4 ABOUT HERE
The level of compliance with IASs differed between the GCC member states. The highest
average compliance level over all years sampled was found in the UAE (0.80). This was
followed by Saudi Arabia (0.78), Kuwait (0.75), Oman (0.74), Bahrain (0.73) and Qatar (0.70)
(Table 4). Significant differences in median compliance are shown in Table 5, with Bahrain,
Kuwait and UAE greater than Qatar, Bahrain greater than Oman and Oman greater than Saudi
Arabia.
TABLE 5 ABOUT HERE
With respect to disclosure compliance, the UAE and Saudi Arabia both averaged 0.75, followed
by Kuwait (0.72), Qatar (0.69), and Bahrain and Oman (both 0.65). The highest level of
24
measurement compliance was again in the UAE (0.85), Saudi Arabia (0.83), Bahrain and Oman
(both 0.82), Kuwait (0.78) and Qatar (0.71). Overall, measurement compliance was better than
disclosure compliance for all years and the increase in compliance was relatively greater for
disclosure items. Compliance levels of 80% for disclosure and 84% for measurement were
achieved in 2002 (Table 4).
The level of compliance varied across standards. The highest average level of compliance for all
years was 0.89 for IAS 18, and the lowest was 0.43 for IAS 37. The compliance level was
highest for IAS 1, 16, 18, 23, 24, 27 and 30 (averaging more than 0.80); whereas, it was lowest
for IAS 14 and 37 (less than 0.50). Moderate compliance was found for IAS 10, 21, 28, 32 and
33 (more than 0.50 and less than 0.80). In general, the level of compliance increased from 1996
to 2002 for all standards, indicating that the GCC states, collectively, were progressing towards
achieving greater de facto harmonisation with IASs. A possible explanation for the higher level
of compliance with IAS 1, 16, 18, 23, 24, 27 and 30 is that they are less difficult to implement
when compared with IASs with requirements to disclose more proprietary information or with
more complicated requirements, such as IAS 14 and 37.
Regression results
Full sample, all years
Model I of Table 6 reports the results for the overall compliance index excluding NA (not
applicable) and DK (do not know) items. The model was significant overall (F = 12.059, p <
0.001) with R2 (adj.) of 0.356. The adjusted R2 is consistent with prior compliance studies (such
as the 0.297 found by Glaum and Street, 2003). The results reveal that, as predicted, the level of
mandatory compliance with IASs exhibits a home-state effect and increases with a company’s
25
size, leverage and internationality. Also, there were significant differences in the level of
compliance across years with compliance in 1996, 1997 and 1999 all being less than 2002.
TABLE 6 ABOUT HERE
In Model I average compliance in the UAE was reflected in the constant term, showing that
compliance in Kuwait and Oman was significantly higher than in the UAE and compliance in
Qatar was significantly less, giving partial support for H1 and H2. The results of models where
country dummy variables are rotated (to maximise the number of direct comparisons between
countries) are shown in Table 7 (regressions not reported in detail). In summary, the predicted
relationships are supported by significant results in the multivariate analyses as follows: Kuwait
and Oman greater than Qatar and UAE (H1); and UEA, Saudi Arabia and Bahrain greater than
Qatar (H2, H3 and H4). The predictions not supported were: Kuwait and Oman greater than
Saudi Arabia (H1) and UAE greater than Saudi Arabia (H2), as these variables were not
significantly different in regression models. The magnitude of the coefficient was in the
direction predicted for the first country sub-group (Kuwait and Oman greater than Saudi Arabia)
but not for the second (UAE greater than Saudi Arabia).
The hypothesized relationship between Bahrain and the other countries was not supported. The
coefficient for Bahrain was positive and significant in all regressions, suggesting being a
Bahraini company was a factor explaining more compliance rather than less as predicted in H1,
H2 and H3. (The results for H4 are as predicted: Bahrain significantly greater than Qatar.) Table
7 shows that nine of the 15 unique paired comparisons were statistically significant at the 0.05
confidence level or better (one-tailed t-test) and in the manner predicted. Thus, taken as a
26
whole, the multivariate results lend substantial support for hypotheses which proposed that
differences in compliance levels would be associated with country differences in enforcement
activities of auditors and surveillance departments of stock exchanges, central banks and
government departments.
TABLE 7 ABOUT HERE
Robustness tests demonstrated that the regression results are not sensitive to alternative
specifications of the dependent variable, that is, to the treatment of potentially ambiguous items
in the checklist. Several measures of the dependent variable were investigated (regressions not
reported in detail). First, the overall compliance index was calculated including NA and DK
items, but recording NA and DK as non-compliance, thus giving a lower compliance score
because all items were considered applicable and companies were penalised for not disclosing
an item, irrespective of its applicability. Second, the overall compliance index was calculated
including DK items, but recording DK as non-compliance and excluding NA items. Only
companies with DK items received a lower score. Overall, the results reported in Model I (Table
6) were not especially sensitive to the alternative transformed compliance indices, despite some
differences with respect to the significance of country and industry dummy variables. We also
fitted Model I using the untransformed dependent variables. The results were essentially
unaffected, although explanatory power of the transformed dependent variable models was
slightly (1–3%) higher.
Subsamples
27
The GCC member states were divided into two groups, according to the extent to which each
country adopted IASs. Sub-sample 1 included countries requiring compliance by companies in
all industries (Bahrain, Kuwait and Oman). Sub-sample 2 included countries requiring
compliance only by banks and finance and investment companies (Saudi Arabia, Qatar and the
UAE). Model II for sub-sample 1 was significant (F = 12.793, p < 0.001) with R2 (adj.) of 0.341
(Table 6). The results exhibit a within-group effect and confirm that compliance with IASs has
increased over time and increased with a company’s size, leverage, age and internationality.
This result was similar to that for the full sample (Model I), presumably because of the very
large overlap in the data.10
Model III for sub-sample 2 was significant overall (F = 4.269, p < 0.001) with R2 (adj.) of 0.477.
The findings confirm that compliance with IASs has increased over time and increases with
company size and declines with liquidity and the concentration of shares held by institutional
investors. 11 In each sub-sample, there were significant differences between years, suggesting
other factors were changing over time, such as the effectiveness of the enforcement body and the
comprehensiveness of the audit function.
Country
There was some variation in the relative importance of the company attributes across the GCC
member states (Table 8). Size was the most influential factor in every state except Bahrain,
while a company’s age contributed significant explanatory power in Bahrain, Oman and Kuwait.
Internationality and leverage were significant in Oman and Kuwait, while industry effects were
found only in Kuwait. Profitability was positively related to the extent of compliance in Qatar,
and liquidity was negatively associated with the level of compliance in Saudi Arabia. The
28
remaining variables made no significant contribution in any state. These results suggest that the
degree of influence of company attributes that are associated with compliance can differ
between countries, even when these countries are from one region and have substantial
economic and cultural ties.
TABLE 8 ABOUT HERE
There are significant differences between years for each country except Qatar. This result is
consistent with the increasing influence of enforcement bodies and more comprehensive
auditing over time. It is also consistent with growth in companies’ incentives to comply for other
commercial reasons.
The significant differences observed in the level of compliance between the GCC member states
are consistent with Tower et al. (1999), who found significant differences among six Asia–
Pacific countries: Australia, Hong Kong, Malaysia, the Philippines, Singapore and Thailand. It is
also consistent with Street and Gray (2001), who found significant differences among seven
groups of countries (Germany, France, Switzerland, other Western Europe, Africa, China, and
the Middle East, former Soviet Block and “other countries”).
The findings from this study are consistent with Street and Gray (2001), who reported that the
level of voluntary disclosure compliance differed by industry. However, they conflict with
Tower et al. (1999), who found no association between voluntary compliance and size, leverage
and industry. They also conflict with Street and Bryant (2000), who found size and industry
were not associated with voluntary compliance for a sample of 17 mainly developed countries.
29
The results are inconsistent, too, with Street and Gray (2001), who found size was not
significant; whereas, appointment of a Big Five auditor was. Another inconsistency is with
Glaum and Street (2003) who found, in Germany, that size was not significant, but a Big Five
auditor was. Overall, our findings are consistent with prior studies which conclude that country
is a primary explanatory factor for level of compliance. We identify some additional factors
which are not significant in prior research, possibly related to the sample of countries included in
this study and thus further reflecting country differences.
CONCLUSIONS
The aim of this study was to investigate empirically the extent of mandatory compliance with
international accounting standards (IASs) by companies in the Gulf Co-Operation Council
(GCC) member states — namely, Bahrain, Oman, Kuwait, Qatar, Saudi Arabia and the United
Arab Emirates (UAE) — between 1996 and 2002, and to explain why some companies
comply more than others. A self-constructed compliance index based on 14 IAS applicable to
GCC companies during the period was used to measure of compliance with both recognition and
disclosure requirements for 137 listed companies (436 company-years).
The average level of compliance for all companies and all years was 75% of the items in the
index. No company within the examined time period fully complied with all requirements. The
average level of compliance increased over time, though, from 68% in 1996 to 82% in 2002.
There was significant variation in the level of compliance across the six GCC member states as
well. The highest average level of compliance was in Saudi Arabia, where it reached 88% in the
last year of the study.
30
We report only partial support for the hypotheses that compliance is highest in Kuwait and
Oman, followed by the UAE, Saudi Arabia, Bahrain and Qatar, possibly because we have
overstated the potential effect of institutional differences. Alternately, common cultural and
other factors already captured in the control variables may have been too dominant to yield a
clear result on all of the cross-country differences because the sample size was not large enough.
That is, the number of companies in some states might be too small to detect a strong country
effect. The level of compliance increased over time in every GCC member state, suggesting that
each state individually was moving towards greater accounting harmonisation with IASs. These
averages should be treated with caution, however, because they do not allow for differences in
the composition of the sample for each state.
The degree of non-compliance with IASs across the GCC member states was partially
attributable to limited monitoring and enforcement by the bodies responsible for overseeing
financial reporting, and to the limited comprehensiveness of audits by external auditors. In
addition to the role of enforcement bodies and external auditors, several company attributes
helped explain the level of compliance with IASs. Compliance variation increased with a
company’s size, leverage and internationality. The level of compliance varied by industry too;
however, company profitability, liquidity, ownership diffusion and whether the audit was
conducted by a major international audit firm were found not to be significant factors.
A closer look at changes in the level of compliance over time in each GCC member state
indicated that compliance improved substantially in Kuwait and Oman from 1996 to 1999, more
so than in the other GCC member states. This could, at least partially, have resulted from the
31
improved effectiveness of monitoring and enforcement in Kuwait and Oman, beginning in 1999.
The UAE had a higher level of compliance than Saudi Arabia and Qatar, probably because its
enforcement body is more active in compliance monitoring.
Despite the level of non-compliance and the responsibilities of auditors under the law, no action
has been taken against an audit firm or member of the board of directors, or a manager, for
violating an accounting regulation, except for two cases in Oman and one in Kuwait. This study
also found that external auditors in the GCC member states reported a client company fully
complied with all IASs even where that was clearly not the case. This must raise questions about
the quality of auditing in the region.
Only the enforcement bodies in Kuwait and Oman conducted proactive monitoring, as specified
in their respective company laws. However, the level of mandatory compliance with IASs in
Kuwait and Oman for all years is less than in the UAE and Saudi Arabia but higher than in Qatar
and Bahrain. A possible explanation for this could be related to the fact that the level of
compliance in the UAE and Saudi Arabia was measured for one sector only (banks and
investment companies), while in Kuwait and Oman it was measured for other industries as well.
Additionally, it is common practice in the UAE for banks to have three external auditors, and it
is a requirement in Saudi Arabia to have at least two. The relatively high level of compliance in
the UAE and Saudi Arabia also could be influenced by the higher level of auditing or training
because the Ministry of Commerce in the UAE and the Saudi Organization for Certified Public
Accountants require auditors to undertake professional training programs.
32
The study adds to literature which has investigated IASs compliance by considering a group of
developing countries of economic and political importance. The findings reveal that non-
compliance was widespread and probably higher than in developed countries. The level of
mandatory compliance with IAS disclosure and measurement requirements in each GCC
member state for each year and for all years was lower than the level of even voluntary
compliance with disclosure and measurement requirements in Australia (0.94; Tower et al.,
1999). The level of mandatory disclosure compliance in the GCC member states for all years
was also lower than companies listed on the Germany’s “Neuer Markt” (0.81; Glaum and Street,
2003), although it was higher in Kuwait, Saudi Arabia and the UAE in 2002. Mandatory
disclosure compliance for all years in Bahrain, Oman, Kuwait and Qatar was lower than
voluntary disclosure compliance in Switzerland (0.74; Street and Gray, 2001). Mandatory
measurement compliance in each GCC member state for all years was also lower than voluntary
measurement compliance in Switzerland (0.92; Street and Gray, 2001). These findings imply
that the company attributes that explain within-sample variance in compliance with IASs can
differ between developed and developing countries. In general, the level of compliance in
developing countries is associated with size, leverage and internationality; whereas, auditor and
cross-listing are the most important explanatory variables in developed countries.
The study explores the impact of enforcement activities on IASs compliance following
mandatory IASs adoption. The findings may be relevant to regulators in countries and regions
which have more recently adopted IASs and to stock exchanges which permit companies to
submit financial statements based on IASs. Our results suggest that although a set of
mechanisms to promote compliance among GCC companies are in place, activities of
enforcement bodies have been inadequate. Possible reasons for this inadequacy are a lack of
33
professional training and salaries insufficient to attract highly qualified accountants. Also, the
enforcement bodies have other responsibilities and there may be a lack of commitment from
governments, at a policy level, to enforcement.
There are several future research opportunities in this area. The study of compliance levels
among GCC companies could be repeated at a future date to determine whether compliance
levels have continued to increase while stock markets have become more open to foreign
investors. This would provide important feedback to regulators in the region. In addition,
research could consider the relationship between corporate governance and compliance, which
has not been investigated for GCC companies. Our results raise questions about factors
affecting, first, the quality and independence of auditors in GCC member states and, second,
their role in promoting compliance with IASs and other regulations. Both issues could be
examined in future research.
Compliance studies in other developing countries would also be of interest, as IASB standards
are intended for use throughout the world. Identifying factors which promote compliance and
therefore harmonisation across countries with different cultural backgrounds and levels of
economic development would be useful to regulators and standard setters. As more countries
adopt IASs, the role of enforcement in promoting cross-country comparability becomes a key
topic. Studies which identify effective enforcement strategies and can therefore provide input
to policy decisions about enforcement would be beneficial to national and international capital
market development.
34
REFERENCES Abdallah, W. M. (2001). Managing multinationals in the Middle East: Accounting and tax
issues. Connecticut: Quorum Books.
Abdelrahim, A. M., A. M. Hewaidy, and G. N. Mostafa. (1997). The relevance of international accounting standards to the evaluation of fixed assets in Kuwait corporations. Arab Journal of Administrative Sciences, 5 (1): 53-96.
Abdelrahim, A. M. and G. N. Mostafa. (2000). Accounting for investment for listed companies on the KSE. Al-Muhasiboon (The Accountants), 6 (16): 30-45.
Abdulla, J. (1998). Corporate financing policy in Oman: Perceptions of executives. Asian Review of Accounting, 6 (2): 117-142.
Ahmed, K. and D. Nicholls. (1994). The impact of non-financial company characteristics on mandatory disclosure compliance in developing countries: The case of Bangladesh. International Journal of Accounting, 19 (3): 62-77.
Ahmed, K. (1996). Disclosure policy choice and corporate characteristics: A study of Bangladesh. Asian-Pacific Journal of Accounting, 3 (1): 183-203.
Al-Bastaki, H. (1997). Extent of disclosure in the annual reports of banks operating in Bahrain. Arab Journal of Administrative Sciences, 5 (1): 241-268.
Al-Basteki, H. (2000). Strategies for enhancing the accounting profession and practices in Bahrain: Accounting experts' perceptions. Research in Accounting in Emerging Economies, 4: 165-189.
Al-Shammari, B., P. Brown and A. Tarca. (2007). Development of enforcement mechanisms following adoption of International Accounting Standards in the Gulf Cooperation Council Member States, Working paper, University of Western Australia.
Al-Shimmiri, T. B. (2003). Capital structure and performance of non taxpaying entities. Arab Journal of Administrative Sciences, 10 (3): 363-385.
Azzam, H. (1998). Financial deepening in the Arabia Gulf region. In Devlin, J., (eds), Gulf Economies: Strategies for Growth in the 21st Century. UAE: The Emirates Center for Strategic Studies and Research Publications.
Ball, R. and G. Foster. (1982). Corporate financial reporting: A methodological review of empirical research. Journal of Accounting Research, 20 (Supplement): 161-234.
Bazley, M., P. Brown, and H. Y. Izan. (1985). An analysis of lease disclosure by Australian companies. Abacus, 21 (1): 44-62.
Cairns, D. (1997). IFAC: 20 years on. World Accounting Report (October).
35
Committee of European Securities Regulators (CESR), (2003) Standard No. 1 on financial information: Enforcement of standards on financial information in Europe. CESR/03-073. March. Available at http://www.europefesco.org/v2/default.asp.
Cooke, T. E. (1989). Voluntary corporate disclosure by Swedish companies. Journal of International Financial Management and Accounting, 1 (2): 171-195.
Cooke, T. E. (1991). An assessment of voluntary disclosure in the annual reports of Japanese corporations. International Journal of Accounting, 26 (3): 174-189.
Craswell, A. T. and S. L. Taylor. (1992). Discretionary disclosure of reserve by oil and gas companies. Journal of Business Finance and Accounting, 19 (2): 295-308.
Doha Securities Market (DSM). (2005). Amiri Decree amending Law No. 13 of 2000 permitting foreigners to trade in Doha Securities Market (DSM): http://www.dsm.com.qr/regulations.
Dumontier, P. and B. Raffournier. (1998). Why firms comply voluntarily with IAS: An empirical analysis with Swiss data. Journal of International Financial Management and Accounting, 9 (3): 216-245.
Ferguson, M. J., K. C. Lam, and G. M. Lee. (2002). Voluntary disclosure by state-owned enterprises listed on the stock exchange of Hong Kong. Journal of International Financial Management and Accounting, 13 (2): 125-152.
Fox, J. (1997). Applied regression analysis, linear models, and related methods. California: Sage Publications.
Glaum, M. and D. L. Street. (2003). Compliance with the disclosure requirements of Germany's New Market: IAS versus US GAAP. Journal of International Financial Management and Accounting, 14 (1): 64-1000.
Hanushek, E. A. and J. E. Jackson. (1977). Statistical methods for social scientists. New York: Academic Press.
Hassan, N. A. (1998). The impact of socioeconomic and political environment on accounting system preferences in developing economies. Advances in International Accounting, Supplement 1: 43-88.
Hussain, M., M. M. Islam, A. Gunasekaran, and K. Maskooki. (2002). Accounting standards and practices of financial institutions in GCC member countries. Managerial Auditing Journal, 17 (7): 350.
Inchausti, B. G. (1997). The influence of company characteristics and accounting regulation on information disclosed by Spanish firms. European Accounting Review, 6 (1): 45-68.
36
International Accounting Standards Board (IASB). (2005). Countries claimed to comply with IASs: http://www.iasb.org/about/faq/asp?showpageContent=no&xml=18_18_24_17122003.htm.
Jensen, M. and W. Meckling. (1976). Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3 (3): 305-360.
Joshi, L. and H. Al-Bastaki. (2000). Judgment based accounting: A survey of accountants' practices in Bahrain. Paper presented at the 23rd Annual European Accounting Association Congress, Munich.
Joshi, L. and J. Al-Mudhahki. (2001). Empirical study of compliance with international accounting standards (IAS 1) by stock exchange listed companies in Bahrain. Journal of Financial Management and Analysis, 14 (2): 43-54.
Makhija, A. and J. Patton. (2004). The impact of firm ownership structure on voluntary disclosure: Empirical evidence from Czech annual reports. Journal of Business, 77 (3): 457-491.
Malone, D., C. Fries, and T. Jones. (1993). An empirical investigation of the extent of corporate financial disclosure in the oil and gas industry. Journal of Accounting, Auditing and Finance, 8 (3): 249-273.
Naser, K. and R. Nuseibeh. (2003a). Quality of financial reporting: Evidence from the listed Saudi nonfinancial companies. International Journal of Accounting, 38 (1): 41-69.
Naser, K. and R. Nuseibeh. (2003b). User's perception of corporate reporting: Evidence from Saudi Arabia. British Accounting Review, 35 (2): 129-153.
Naser, K., R. Nuseibeh, and A. Al-Hussaini. (2003). Users' perception of various aspects of Kuwaiti corporate reporting. Managerial Auditing Journal, 18 (6): 599-617.
Nobes, C. W. (1990). Compliance by US corporations with IASC standards. British Accounting Review, 22 (1): 41-49.
Oweiss, I. (2000). The Arab Gulf economies: Challenges and prospects. The Emirates Center for Strategic Studies and Research, the Emirates Lecture Series No. 26.
Qatar Central Bank (QCB). (2004). Annual report: http://www.qcb.gov.qa/annualreport.
Securities and Exchange Commission (SEC). (2000). ‘The concept release on international accounting standards’, Available at http://www.sec.gov/rules/concept/34-42430.htm.
Shuaib, A. 1999. Financial reporting in the GCC member countries. Paper presented at the 11th Asian-Pacific Conference on International Accounting Issues, Melbourne.
37
Street, D. L., S. J. Gray, and S. M. Bryant. (1999). Acceptance and observance of international accounting standards: An empirical study of companies claiming to comply with IASs. International Journal of Accounting, 34 (1): 11-48.
Street, D. L. and S. J. Gray. (2001). Observance of international accounting standards: Factors explaining non-compliance by companies referring to the use of IAS. London: ACCA Research Monograph No. 74.
Street, D.L. and S. M. Bryant. (2000). Disclosure level and compliance with IASs: A comparison of compliance with and without U.S. listings and filings. International Journal of Accounting, 35 (3): 305-329.
Tower, G., P. Hancock, and R. Taplin. (1999). A regional study of listed companies' compliance with international accounting standards. Accounting Forum, 23 (3): 293-305.
United Arab Emirates Central Bank (UAECB). (1999). Central Bank of the United Arab Emirates Circular No. 20 of 1999: Adoption of IASs by the UAE banking and investment sector. Abu Dhabi: UAE Government Printing Press.
Wallace, R. S. O. and K. Naser. (1995). Firm-specific determinants of the comprehensiveness of mandatory disclosure in the corporate annual reports of firms listed on the stock exchange of Hong Kong. Journal of Accounting and Public Policy, 14 (4): 311-368.
Watts, R. and J. Zimmerman. (1978). Towards a positive theory of the determination of accounting standards. Accounting Review, 53 (1): 112-134.
Watts, R. and J. Zimmerman. (1986). Positive accounting theory. Englewood Cliffs: Prentice-Hall.
38
Table 1. Sample selection
Bahrain Oman Kuwait Saudi Arabia
Qatar UAE Total % of the total population
Companies listed on GCC stock exchanges on 31 Dec, 1996
36 79 60 9* 6* 10* 200 100%
Exclusions: Islamic institutions (2) 0 (2) (2) (2) (1) (9) 4.5% Delisted companies during the period of the study (1996-2002)**
(8) (19) (1) 0 0 0 (28) 14%
Cross-listed in GCC stock exchanges
0 0 (7) 0 0 0 (7) 3.5%
Companies where a letter was sent
26 60 50 7 4 9 156 78%
Companies where their data not available (excluded)***
(1) (17) 0 (1) 0 0 (19) 9.5%
Companies with usable data included in the study
25 43 50 6 4 9 137 68.5%
Companies’ annual reports collected (1996-2002)
100 129 150 18 12 27 436
* This number represents only companies in the banking and investment industries because these are the only companies in Saudi Arabia, Qatar and UAE that must adopt IASs.
** Liquidations, mergers, and other reasons not provided. *** Companies did not respond to two requests for their annual reports.
39
Table 2. Descriptive statistics
Mean Median Min Max Std. Dev. All Countries (N = 436) Company size (US$ million)* 1732.00 142.35 3.02 27691.32 4151.41 Leverage 0.45 0.44 0.00 0.98 0.28 Liquidity 4.36 1.32 0.13 336.46 20.19 Profitability 0.08 0.11 -0.97 0.44 0.54 Internationality 0.04 0.01 0.00 0.59 0.11 Ownership diffusion 0.26 0.20 0.00 0.80 0.22 Age 21.59 9.38 1.00 50.00 9.36 Bahrain (N = 100) Company size (US$ million)* 1484.11 73.57 8.51 27691.40 4614.30 Leverage 0.40 0.22 0.00 0.98 0.31 Liquidity 9.70 1.40 0.24 336.46 40.59 Profitability 0.07 0.09 -0.27 0.35 0.10 Internationality 0.08 0.01 0.00 0.59 0.17 Ownership diffusion 0.16 0.09 0.00 0.63 0.17 Age 21.59 21.00 2.00 45.00 8.32 Oman (N = 129) Company size (US$ million)* 203.51 30.85 3.00 4064.16 566.48 Leverage 0.48 0.49 0.04 0.94 0.23 Liquidity 2.00 1.18 0.15 26.62 3.06 Profitability 0.09 0.12 -0.97 0.40 0.16 Internationality 0.02 0.01 0.00 0.36 0.06 Ownership diffusion 0.29 0.27 0.00 0.79 0.24 Age 14.85 15.00 1.00 29.00 6.91 Kuwait (N = 150) Company size (US$ million)* 1163.33 240.77 17.37 20746.55 2705.29 Leverage 0.35 0.29 0.02 0.89 0.25 Liquidity 3.90 1.81 0.13 33.35 4.95 Profitability 0.11 0.11 -.44 0.44 0.10 Internationality 0.01 0.01 0.00 0.02 0.02 Ownership diffusion 0.36 0.36 0.00 0.80 0.22 Age 26.30 25.00 9.00 50.00 8.72 Saudi Arabia (N = 18) Company size (US$ million)* 13159.46 12687.38 2730.42 26442.03 6696.04 Leverage 0.78 0.78 0.69 0.86 0.05 Liquidity 0.72 0.72 0.43 1.09 0.17 Profitability 0.16 0.16 0.10 0.23 0.04 Internationality 0.25 0.30 0.00 0.40 0.15 Ownership diffusion 0.11 0.10 0.00 0.19 0.04 Age 24.50 22.00 16.00 45.00 8.54 Qatar (N = 12) Company size (US$ million)* 2959.01 1531.13 704.18 9525.29 2959.17 Leverage 0.80 0.80 0.74 0.87 0.04 Liquidity 1.48 0.99 0.64 6.42 1.60 Profitability 0.17 0.13 -0.55 1.07 0.35 Internationality 0.00 0.00 0.00 0.00 0.00 Ownership diffusion 0.19 0.20 0.10 0.24 0.04 Age 25.08 23.50 16.00 38.00 7.51
40
Table 2. (Continued) Descriptive statistics
Mean Median Min Max Std. Dev. UAE (N = 27) Company size ($US million)* 4984.28 6084.63 99.02 11785.31 3793.41 Leverage 0.66 0.77 0.07 0.88 0.24 Liquidity 1.50 0.97 0.28 5.23 1.58 Profitability 0.10 0.12 -0.11 0.18 0.07 Internationality 0.04 0.01 0.00 0.02 0.01 Ownership diffusion 0.13 0.07 0.01 0.82 0.21 Age 24.11 23.00 4.00 39.00 9.50
Company size was measured by market value of equity plus book value of debt; Leverage was the ratio of book value of total debt as a proportion of book value of total debt plus market value of equity; Liquidity was the ratio of current assets to current liabilities; Profitability was the ratio of earnings before taxes, Zakat and interest to total shareholders’ equity; Internationality was the percentage of foreign ownership (number of shares owned by foreign investors divided by total shares at year-end); Ownership diffusion was the percentage of institutional investors (number of shares owned by institutional investors divided by total shares issued at year-end); and Age was the number of years passed since foundation.
* In the regressions, company size was operationalised as the natural logarithm of market value of equity plus book value of debt.
41
Table 3. Total number of disclosure and measurement items for each standard
Standards (effective year) Total items Disclosure items Measurement items IAS 1 (1975) IAS 1 (1998)
2 35
2 35
- -
IAS 10 (1980) IAS 10 (2000)
11 7
7 5
4 2
IAS 14 (1983) IAS 14 (1998)
14 23
14 23
- -
IAS 16 (1995) 24 16 8 IAS 18 (1995) 3 3 - IAS 21 (1995) 20 9 11 IAS 23 (1995) 5 3 2 IAS 24 (1986) 5 5 IAS 27 (1990) 12 7 5 IAS 28 (1990) 9 7 2 IAS 30 (1991) 45 45 - IAS 32 (1996) 10 10 - IAS 33 (1998) 8 5 3 IAS 37 (1999) 14 12 2 Maximum (1996) 160 128 32 Maximum (1997) 160 128 32 Maximum (1999) 224 187 37 Maximum (2000) 220 185 35 Maximum (2002) 220 185 35
Note: IAS 30 applies only to banks and financial institutions.
42
Table 4. Results for compliance indices by country and year Country Total
compliance
Disclosure
compliance
Measurement compliance
Country Total compliance
Disclosure compliance
Measurement compliance
Mean Mean Mean Mean Mean Mean
Total (N = 436)
1996 0.68 0.56 0.80
1997 0.71 0.63 0.80
1999 0.74 0.69 0.80
2000 0.76 0.73 0.80
2002 0.82 0.80 0.84
All years 0.75 0.69 0.81
Bahrain (N = 25) Saudi Arabia (N = 6)
1996 0.69 0.56 0.81 1996 0.72 0.65 0.80
1997 0.71 0.63 0.79 1999 0.76 0.74 0.80
1999 0.74 0.70 0.82 2002 0.88 0.85 0.91
2002 0.80 0.74 0.88 All years 0.78 0.75 0.83
All years 0.73 0.65 0.82
Oman (N = 43) Qatar (N = 4)
1996 0.65 0.50 0.80 1999 0.64 0.62 0.67
1999 0.73 0.64 0.82 2000 0.67 0.68 0.65
2002 0.83 0.80 0.85 2002 0.81 0.79 0.83
All years 0.74 0.65 0.82 All years 0.70 0.69 0.71
Kuwait (N = 50) UAE (N = 9)
1996 0.68 0.60 0.75 1999 0.74 0.65 0.82
1999 0.76 0.74 0.79 2000 0.80 0.75 0.86
2002 0.81 0.81 0.81 2002 0.86 0.84 0.91
All years 0.75 0.72 0.78 All years 0.80 0.75 0.85
43
Table 5. Differences in median overall compliance between countries
Bahrain Oman Kuwait Saudi Arabia Qatar
Bahrain
Oman < 0.065*
Kuwait 0.496 0.432
Saudi Arabia 0.609 < 0.010* 0.618
Qatar < 0.072* 0.137 < 0.030** 0.264
UAE 0.486 0.290 0.510 0.803 > 0.090* This table shows significant differences (based on t-tests) between the overall all-years median compliance score reported in Table 5 for each pair of countries. The < and > symbols indicate that compliance was lower/higher in the country identified at the start of each row. ** Significant at the 0.05 level (two-tailed). * Significant at the 0.10 level (two-tailed).
44
Table 6. Regression results: compliance with IASs
Independent variables Coefficients (Expected sign) Model I Model II Model III Company size (+) 0.087 +++ 0.084 +++ 0.284 ++
Leverage (+) 0.146 +++ 0.149 ++ -0.326 (a)
Liquidity (-) 0.001 0.001 -0.032 +
Profitability (+) 0.045 0.018 0.215
Internationality (+) 0.138 + 0.155 ++ -0.523 (a)
Ownership diffusion (-) 0.019 0.029 -0.384 +++
Age (+) 0.001 0.002 + 0.011
Auditor (+) 0.017 0.013
Industry 1 (Bank and investment) -0.024 -0.022
Industry 2 (Insurance) 0.045 0.044
Industry 3 (Manufacturing) 0.034 0.033
Bahrain (-) 0.212 (a) 0.108 ***
Oman (+) 0.158 ++ 0.059
Kuwait (+) 0.103 +
Saudi Arabia (-) 0.032 -0.129 *
Qatar (-) -0.140 ++ -0.162 *
Year (1996) (-) -0.336 +++ -0.321 +++ -0.463 +++
Year (1997) (-) -0.359 +++ -0.343 +++
Year (1999) (-) -0.199 +++ -0.179 +++ -0.382 +++
Year (2000) (-) -0.171 ++ -0.311 ++
Constant -0.294 * -0.183 ** -0.101 *
Adjusted R2 0.356 0.341 0.477 F 12.059 12.793 4.269 Prob. (F) < 0.001 < 0.001 < 0.001 No. of companies 137 118 19 No. of observations 409 365 44 This table presents the results of regression models that examine the relationship between level of compliance and independent variables for the full sample based on all countries and years (Model I) and two subsamples. Model II includes companies from Bahrain, Oman and Kuwait, where all companies adopted IASs. Model III includes companies from Saudi Arabia, Qatar and UAE which adopted IASs for banks and finance and investment companies only. Company size = log of market value of equity plus book value of debt; Leverage = ratio of book value of total debt/book value of total debt plus market value of equity; Liquidity = current assets/current liabilities; Profitability = earnings before taxes, Zakat and interest/total shareholders’ equity; Internationality = number of shares owned by foreign investors/total shares issued at year-end; Ownership diffusion = number of shares owned by institutional investors/total shares issued at year-end; Age = number of years passed since foundation; Auditor was a dummy variable = 1 for companies audited by local audit firms with an international affiliation (Big Five) and 0 for companies audited by local audit firms without an international affiliation (non Big Five). Industry 1 = Banking and investment; Industry 2 = Insurance; Industry 3 = manufacturing. In Models I and III the omitted country dummy variable is the UAE; in Model II it is Kuwait. The omitted dummy variable is Year 2002 and Industry 4 (Services) in all models. +++ t-test (one-tailed) significant p < 0.01 ++ t-test (one-tailed) significant p < 0.05 + t-test (one-tailed) significant p < 0.10 *** t-test (two-tailed) significant p < 0.01. **t-test (two-tailed) significant p < 0.05. *t-test (two-tailed) significant p < 0.10. (a) Coefficient was statistically significant but had the opposite sign to that hypothesized.
45
Table 7. Summary of the multivariate results for paired comparisons of countries
Bahrain Oman Kuwait Saudi Arabia
Qatar UAE
Bahrain 0.054* 0.108*** 0.180*** 0.353*** 0.212***
Oman -0.054* 0.054 0.125** 0.298*** 0.158**
Kuwait -0.108*** -0.054 0.071 0.244*** 0.103*
Saudi Arabia -0.180*** -0.125** -0.071 0.172** 0.032
Qatar -0.352*** -0.298*** -0.244*** -0.173** -0.140**
UAE -0.212*** -0.158*** -0.103* -0.032 0.140**
Relationships consistent with hypotheses
1 4 3 2 5 3
This table summarises the results for Model I when country dummy variables were rotated in sequence show the effect of country in the regression models after controlling for other factors. The entry in each cell indicates the coefficient of the dummy variable for the country at the start of the row, where the country effect of the country identified at the top of each column was captured in the constant term. *** Significant at the 0.01 level (two-tailed) ** Significant at the 0.05 level (two-tailed) * Significant at the 0.10 level (two-tailed).
46
Table 8. Regression results: compliance with IASs by country
Coefficients Independent variables (Expected sign)
Bahrain Oman Kuwait Saudi Arabia
Qatar UAE
Company size (+) -0.008 0.159 +++ 0.091 ++ 0.366 +++ 1.168 + 0.097 ++
Leverage (+) 0.219 0.103 + 0.131 + -3.935 (a) 0.537 3.058
Liquidity (-) 0.001 0.002 0.003 -0.196 + -0.048 -0.166
Profitability (+) -0.078 (a) 0.071 -0.184 2.921 0.002 ++ -0.462
Internationality (+) 0.148 0.275 ++ 3.758 +++ -0.613 (a) 2.683
Ownership diffusion (-)
0.050 0.218 (a) 0.164 (a) 1.189 5.675 (a) -0.350 ++
Age (+) 0.001 + 0.009 +++ 0.001 + 0.018 0.045 -0.007
Auditor (+) 0.105 -0.023 -0.021
Industry 1 0.097 0.016 -0.168 ***
Industry 2 0.122 0.043 -0.081
Industry 3 -0.052 0.055 0.019
Year (1996) (-) -0.432 +++ -0.275 +++ -0.194 +++ -0.316 +++
Year (1997) (-) -0.394 +++
Year (1999) (-) -0.239 +++ -0.146 +++ -0.053 ++ -0.399 +++ -0.039 -0.439 +++
Year (2000) (-) -0.239 -0.343 +
Constant 0.642 ** -0.102 ** -0.122 1.650 * -1.164 -0.270 Adjusted R2 0.579 0.474 0.298 0.791 0.772 0.778 F 10.744 9.844 3.981 9.919 3.472 6.083 Prob. (F) < 0.001 < 0.001 < 0.001 < 0.008 < 0.097 0.049 No. of companies 25 43 50 6 4 9 No. of observations 100 129 136 18 12 17 This table presents the results of regression models that examine the relationship between the level of compliance and independent variables representing company attributes for each country. Company size = log of market value of equity plus book value of debt; Leverage = ratio of book value of total debt/book value of total debt plus market value of equity; Liquidity = current assets/current liabilities; Profitability = earnings before taxes, Zakat and interest/total shareholders’ equity; Internationality = number of shares owned by foreign investors/total shares issued at year-end; Ownership diffusion = number of shares owned by institutional investors/total shares issued at year-end; Age = number of years passed since foundation; Auditor was a dummy variable = 1 for companies audited by local audit firms with an international affiliation (Big Five) and 0 for companies audited by local audit firms without an international affiliation (non Big Five). Industry 1 = Banking and investment; Industry 2 = Insurance; Industry 3 = manufacturing. The omitted dummy variable is Year 2002 and Industry 4 (Services) in all models. +++ t-test (one-tailed) significant p < 0.01 ++ t-test (one-tailed) significant p < 0.05 + t-test (one-tailed) significant p < 0.10 *** t-test (two-tailed) significant p < 0.01. **t-test (two-tailed) significant p < 0.05. *t-test (two-tailed) significant p < 0.10. (a) Coefficient was statistically significant but had the opposite sign to that hypothesized.
47
NOTES 1 See Bahrain Company Law No. 28 of 1975 and replaced by Law No. 21 of 2001, Bahrain Securities
and Stock Exchange Law No. 4 of 1987; Oman Company Law No. 4 of 1974 (as amended) and
Muscat Securities Market Law No. 53 of 1988 replaced by Capital Market Law No. 80 of 1998;
Kuwait Company Law No. 15 of 1960 (as amended) and Kuwait Stock Exchange Law of 13/8/1983;
Saudi Arabia Company Law No. 6 of 1965 (as amended) and Saudi Share Registration Company
(SSRC) Law No. 8/1230 of 1984; Qatar Company Law No. 11 of 1981 and Doha Securities Market
Law No. 14 of 1995; UAE Company Law No. 8 of 1984 (as amended) and its implementing
Ministerial Resolutions and UAE Securities Market Law No. 4 of 2000.
2 In all GCC member states, Islamic institutions listed on the stock exchange must meet a separate
set of disclosure and measurement requirements and are exempt from complying with IAS (Abdel
Karim, 2001).
3 Seven foreign banks were operating in Qatar as of June 2005 (QCB, 2005).
4 The following information was collected during 14 interviews conducted with Ministry of
Commerce and Stock Exchange officials in each of the GCC countries over the period September –
October 2003. See Al-Shammari, Brown and Tarca (2007).
5 Evidence that banks often have access to information directly from companies has been provided
for Saudi Arabia (Naser and Nuseibeh, 2003b), Kuwait (Al-Shimmiri, 2003; Naser, Nuseibeh and
Al-Hussaini, 2003), Bahrain (Joshi and Al-Bastaki, 2000) and Oman (Abdulla, 1998).
6 Evidence of institutional investors’ access to information directly from companies is observed in
Saudi Arabia (Naser and Nuseibeh, 2003b), in Kuwait (Al-Shimmiri, 2003) and in Bahrain (Al-
Bastaki, 1997).
48
7 The figure of –0.97 implies that the entire equity of the company (it was in Oman) was eroded in a
single year (1999); nonetheless, it survived.
8 In Qatar, foreign investors have been able to invest indirectly in national companies listed on the
Doha Securities Market (DSM), via an investment fund, since 2001. Since April 2005, foreigners
have been permitted to trade unconditionally on the DSM (DSM, 2005).
9 This information was gathered from a pilot study of 43 companies from the six countries.
10 An exception is that a company’s age is weakly significant in Model I, Table 6.
11 For sub-sample 1, three dummy variables were used to represent the first three years (1996, 1997
and 2000) because the sample for Bahrain included annual reports for four years; whereas, there were
three years for Oman and Kuwait. The results for 2002 were captured by the constant term. Similarly,
for sub-sample 2, three dummy variables were used to represent 1996, 1999 and 2000. The sample
from Saudi Arabia included annual reports for 1996, 1999 and 2002; whereas, for Qatar and the UAE
it contained annual reports for 1999, 2000 and 2002.
49