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