The Impact Of Culture And Governance On Corporate Social Reporting

Journal of Accounting and Public Policy 24 (2005) 391–430 www.elsevier.com/locate/jaccpubpol

The impact of culture and governance on corporate social reporting R.M. Haniffa a, T.E. Cooke a

b,*

School of Management, Bradford University, Emm Lane, Bradford BD9 4JL, UK School of Business and Economics, University of Exeter, Exeter EX4 4PU, UK

b

Abstract Our aim is to increase understanding of the potential effects of culture and corporate governance on social disclosures. The ethnic background of directors and shareholders is used as a proxy for culture. Corporate governance characteristics include board composition, multiple directorships and type of shareholders. The dependent variable, disclosure in annual reports of Malaysian corporations, is measured by an index score as well as in terms of number of words. Our results indicate a significant relationship between corporate social disclosure and boards dominated by Malay directors, boards dominated by executive directors, chair with multiple directorships and foreign share ownership. Four of the control variables (size, profitability, multiple listing and type of industry) were significantly related to corporate social disclosure with the exception of gearing. This study has public policy implications for Malaysia as well as a number of other countries in the Asia–Pacific region. Ó 2005 Elsevier Inc. All rights reserved. Keywords: Corporate social reporting; Culture; Corporate governance; Malaysia; Disclosure

*

Corresponding author. Tel.: +44 1392 263201; fax: +44 1392 263210. E-mail address: [email protected] (T.E. Cooke).

0278-4254/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jaccpubpol.2005.06.001

392 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

1. Introduction Corporate social reporting has been the subject of substantial academic research for more than two decades.1 Some of the main research questions that have been tackled or are currently being researched include: what companies are reporting;2 can social and environmental disclosure practices be linked to attributes of economic performance or to factors such as size, industry membership, risk, market reaction, external influences, firm reputation, country of origin or proximity to individual consumers;3 and what motivates companies to make particular social and environmental disclosures.4 However, most of the factors and proxies that have been considered in previous studies are company-specific. Since disclosure is an ‘‘. . . accounting activity involving both human and non-human resources or techniques as well as the interaction between the two’’ (Perera, 1994, p. 268), studies in this area would benefit if both cultural and corporate governance factors are considered. The inclusion of ethnicity (a proxy for culture) of decision makers within and without an organisation is important in some countries because the traditions of a nation are instilled in its people and might help explain why things are as they are. As such, how a firm operates and reports will be influenced by the social values of the relevant publics within which it exists (Lehman, 1995; Deegan and Rankin, 1996). The mind of Malaysian managers is influenced by ethnicity, education and type of organisation they work for (Chuah, 1995). Thus, operationally, ethnicity acts as a suitable surrogate for culture in this study but the researchers acknowledge that the measure is partial. Ethnicity is chosen because it is a significant marker of class relations and provides a principle ‘‘... according to which conflicts over wealth and state power take place’’ (van Fossen, 1998, p. 89). Furthermore, in multiracial countries, the prevailing societal values may not reflect the values of the nation as a whole especially if each racial group has chosen to maintain its own ethnic identity and values. Differences between the groups are greater if there exists a history of conflict or in which

1 See Mathews (1997) for further discussion on the development of social reporting over that period. 2 See, for example, Ernst and Ernst, 1976; Teoh and Thong, 1984; Andrew et al., 1989; Guthrie and Parker, 1990; Harte and Owen, 1991; Adams et al., 1995; Deegan and Gordon, 1996; Newson and Deegan, 2002. 3 See, for example, Ingram and Frazier, 1980; Anderson and Frankle, 1980; Trotman and Bradley, 1981; Freedman and Jaggi, 1982; Ullman, 1985; Cowen et al., 1987; Roberts, 1992; Herremans et al., 1993; Tilt, 1994; Clarke and Gibson-Sweet, 1999; Newson and Deegan, 2002. 4 See, for example, Guthrie and Parker, 1989; Patten, 1992; Roberts, 1992; Donaldson and Preston, 1995; Deegan and Gordon, 1996; Deegan and Rankin, 1997; Adams et al., 1998; Neu et al., 1998; Deegan, 2000.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 393

racial differences coincide with national or socio-economic differences (Pettigrew, 1979). Similarly, the extent of shared values and degree of co-operation is determined by organisational co-ordinating activities and formality in the system (Birnberg and Snodgrass, 1988). As such, it is important to acknowledge that values may differ between groups even within a nation (Specter and Solomon, 1990). As well as culture, it is important when discussing corporate social disclosure (CSD) practices to consider the values, motives and choices of those involved in formulating policy and taking decisions in the organisation (Maclagan, 1999). Hence, consideration of corporate governance, including ownership structure and constituents of boards that exist in an organisation (proxy for leadership style and control of decision makers), are important because it is top management who oversee information disclosure in annual reports (Gibbins et al., 1990). Longitudinal studies of corporate social disclosure indicate increases over time, both in terms of the number of companies making disclosures and in the extent of information being reported (Harte and Owen, 1991; Deegan and Gordon, 1996).5 Theories explaining social disclosure patterns include the following: Ôsocial contracting theoryÕ which suggests that companies have a social contract with society to perform certain tasks within the bounds of justice; Ôlegitimacy theoryÕ which extends social contracting theory and involves companies responding to demands of divergent interest groups by legitimising their actions; Ôaccountability theoryÕ which also extends social contracting theory and considers companiesÕ compliance with the law; and lastly, Ôdecision usefulness theoryÕ which incorporates users other than investors (Tilt, 1994). The objective of this paper is twofold. First, to explore CSD practices in two different time periods (1996 and 2002) to determine whether differences exist in the extent and variety of social disclosure over time. Our second objective is to explain variability in CSD, and specifically to assess whether culture (ethnicity) and governance structures, besides size, profitability, gearing, listing status and industry type, are good explanatory variables. We adopted legitimacy theory as the premise of our theoretical framework in this paper because we seek to understand what factors may cause variability in corporate social disclosure and to what extent our variables of interest viz. ethnicity and governance structure within an organisation may influence organisational actions in responding to various stakeholder groups especially in situations where there is

5 However, longitudinal studies of disclosure practices in the annual reports of a particular organisation indicate fluctuations in the variety and extent of CSD (Guthrie and Parker, 1989; Deegan et al., 2002).

394 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

government ÔinterventionÕ via certain socio-economic policies.6 Legitimacy needs to be placed in its environmental context to avoid the problem of ethnocentricity which can occur when a country such as Malaysia is evaluated by the norms prevalent in an Anglo-Saxon culture that are taken to be absolutes. For example, in an Anglo-Saxon society it may well be that demand for legitimacy comes from groups such as consumers whereas in Malaysia that demand comes from government and certain elite groups. We analysed the 1996 and 2002 annual reports of 139 non-financial companies7 listed on the Kuala Lumpur Stock Exchange (KLSE). The earlier year reflects an environment when Malaysia enjoyed remarkable economic growth providing a reasonably high standard of living (Jayasankaran and Hiebert, 1997). An aspect of this growth was the strong collusion (patron–client networks) between the government and certain businessmen who benefited from the governmentÕs National Economic Policy (NEP) (Jayasankaran, 1997; Far Eastern Economic Review, 1998). We recognise that in certain societies there exists a close relationship between business and government. Witness, for example, the chaebol in South Korea and keiretsu in Japan. However, Malaysia differs in the sense that the co-operation (collusion) is often specific to an individual rather than to business in general. In contrast to 1996 we selected 2002, a year in which companies were recovering from the 1997 financial crisis and some business reforms had been implemented. We considered ethnicity (proxy for culture) in our study because it is an important explanatory factor in the context of countries where there exists institutionalised positive discrimination, particularly in terms of job offers and concessions such as grants and trade. Such policies have important implications for corporate behaviour including disclosure and governance practices. Investigating CSD practices in Malaysian annual reports over two time periods and identifying culture and governance structure as possible explanatory factors makes a contribution to the literature. Our findings may have implications for other developing countries where the government takes special interest in racial and business policies such as Fiji, Hong Kong, India, Indonesia, Japan, Sri Lanka and Singapore. The remainder of the paper is organised as follows: the next section reviews the literature on organisational legitimacy theory and how CSD in annual reports forms part of the strategic legitimation process. The development of hypotheses are outlined in Section 3, followed by details of the research method 6 Although political–economy theory has been used by others to explain this phenomenon, we felt that legitimacy theory would be more appropriate when considering voluntary rather than mandatory social disclosure. 7 All banks, insurance and unit trusts companies were excluded, primarily because of the different statutory requirements that apply. In addition, companies in these sectors rarely disclose social responsibility information, unlike entities in developed countries.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 395

in Section 4. Section 5 discusses the results followed by the summary and concluding remarks including avenues for further research.

2. Theoretical framework: organisational legitimacy Legitimacy theory is defined as ‘‘a generalised perception or assumption that the actions of any entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs and definitions’’ (Suchman, 1995, p. 574). The entity, through its top management, seeks congruency between organisational actions and the values of its general and relevant publics (Dowling and Pfeffer, 1975; Lindblom, 1994) or its stakeholders. Sethi (1979) argues that if an actual or potential disparity exists between organisational and social values, then organisational legitimacy will be jeopardised giving rise to a Ôlegitimacy gapÕ. In essence, seeking organisational legitimacy is deemed important in demonstrating social worthiness (Oliver, 1991), ensuring continued inflows of capital, labour and customers (Pfeffer and Salancik, 1978), as well as demonstrating that the firm is in tune with societal concerns (Clarke and Gibson-Sweet, 1999) and values to help close any perceived legitimacy gaps. Actions by corporate management to convince wider society that the organisation is socially responsible are part of the legitimation process (Gray et al., 1995a). Lindblom (1994) identifies four broad legitimation strategies that firms may use to secure organisational legitimacy: informing stakeholders about intended improvements in performance; seeking to change stakeholdersÕ perceptions of an event; distracting attention away from an issue; and changing external expectations about its performance. Such legitimation strategies include gaining, maintaining or repairing legitimacy (Suchman, 1995). Dowling and Pfeffer (1975) suggest three modes of action that firms can take to enhance legitimacy: adapt outputs, goals and methods of operation to conform to prevailing definitions of legitimacy; attempt through communication to alter the definition of social legitimacy to conform to present practices, output and values; and finally, attempt through communication to become identified with symbols, values or institutions which have a strong base for social legitimacy. Gray et al. (1995a) link the strategies suggested by Lindblom (1994) and the actions proposed by Dowling and Pfeffer (1975) within the framework of legitimacy theory, a relationship illustrated in Fig. 1. The adoption of an appropriate strategy depends on how best management feel they can close the legitimacy gap. Legitimacy theory is also the basis for our analysis because ‘‘. . . it is difficult to separate the notion of legitimacy from the idea of crisis, in fact, as this is often the only time that constituents in a power system will consciously assert where they believe authority should be concentrated, and how it should be

396 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

Legitimacy theory Strategy

Educate society

Alter society’s

Alter society’s

Divert society’s

of changes in

perceptions of

expectations of

attention away

organisation’s

the organisation

the organisation

from issues of

actions

Action

concern

Adapt outputs,

Through

Through

Through

goals and

communication,

communication,

communication,

methods of

alter definitions

alter definitions

try to become

operations to

of social

of social

identified with

conform to

legitimacy so as

legitimacy so as

symbols, values

prevailing

to conform to

to conform to

or institutions

definitions of

present

present

which have

legitimacy

practices,

practices,

strong base of

outputs and

outputs and

social legitimacy

values

values

Source: adapted from Gray et al. (1995a)

Fig. 1. Link between LindblomÕs legitimation strategies and Dowling and PfefferÕs legitimation actions.

used’’ (Sutton, 1993, p. 2). The financial crisis in the Far East in 1997 raised questions of corporate legitimacy and its governance structures and the relationship between the corporation and the social context governing the functioning of such entities. Empirical studies in social accounting have employed a number of proxies to test aspects of legitimacy theory although content analysis has been most commonly used. Proxies, that help explain environmental disclosures include ownership structure i.e. public vs private company (Cormier and Gordon, 2001); industry sensitivity (Deegan and Gordon, 1996); major corporate events (Guthrie and Parker, 1989; Patten, 1992); and type of users of annual reports (Deegan and Rankin, 1997). Proxies used to explain community involvement disclosure include a corporationÕs public profile i.e. proximity of a company to individual consumers (Clarke and Gibson-Sweet, 1999). More recent studies have focused on direct questioning of managers to test their motivations for CSD. For example, the study by OÕDonovan (2002) supports legitimacy theory and provides insight into management disclosure behaviour based on scenarios that have different impacts. Where the perceived threat is minimal, disclosure is

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 397

deemed unnecessary and if it is threatening, the disclosure reactions are deemed necessary and aimed at either gaining, maintaining or repairing legitimacy. OÕDwyer (2002) suggests that the aim of companies in disclosing environmental and social information is to influence opinion: either in a defensive way to repair perceived loss of legitimacy or in a proactive way to be seen as having social conscience and enlightened self-interest. However, his results questioned the explanatory ability of legitimacy theory in the context of Ireland. Where a legitimacy gap exists and is recognised by senior management then the company has to consider a response. One response is to ignore the gap, presumably on the basis that no adverse consequences will arise. Alternatively, senior management responds by disclosing information helpful in reducing the legitimacy gap, and this is the focus of this study. Management perceives that a gap exists and wishes to reduce it through additional disclosure. We do this by considering social disclosure practice in the 1996 and 2002 annual reports of Malaysian companies.

3. Determinants of CSD and hypothesis development 3.1. CSD over time Previous studies on developed countries have shown that CSD in annual reports has increased over time in response to a number of factors. Some of the reasons may be attributed to increases in legislation, risk, activities of pressure groups, ethical investors, specific events, awards, economic activities, media interest, societal awareness, and politics. Over the past decade, especially the early 1990s, Malaysia experienced substantial economic growth and tremendous social change before being hit by the economic crisis in 1997. The rapid change in business environment was attributed to the conducive environment for both local and foreign businesses including political stability, excellent infrastructure, indigenous supply of natural gas and petroleum, well trained and versatile workforce, and the wide use and recognition of English as an international language (Seetharaman, 1993). At the same time, the massive expansion in the economy resulted in a shortage of workers in certain sectors, higher consumer spending on goods and services, and an increase in awareness of environmental issues as a result of the boom in the construction and property market. However, the economic crisis in 1997 opened the PandoraÕs Box revealing strong collusion between the ruling political group and business. A survey published in the Far Eastern Economic Review (1998) indicated that 86.2% of businessmen felt that the government is too close to business to the point that the ‘‘. . . private sector can dictate terms, telling government what to do based on their links to leadership, who got overambitious and the private sector who

398 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

got greedy’’ (Jayasankaran and Hiebert, 1998, p. 15). There were also worries of rising corruption following the governmentÕs emphasis on economic growth at any price (Jayasankaran, 1997). The Securities Commission and the KLSE addressed some of these weaknesses through reforms in corporate governance. There has also been greater emphasis on better disclosure and greater transparency of information. Given the scenario of events in the chosen time periods, our directional hypothesis is as follows: H1: There is a significant increase in the extent and variety of CSD in the annual reports of Malaysian companies between 1996 and 2002.

3.2. Culture Culture may be represented by perceptions of loyalty to an ethnic group in which the group is a collection of people who share patterns of normative behaviour (Cohen, 1974) and embrace notions of kinship and common origin, however mythical (Harowitz, 1985). It is important to acknowledge that values may differ between groups even within a nation (Specter and Solomon, 1990), especially when various groups choose to maintain their ethnic identity (Sendut, 1991). Malaysia is such a multiracial country. Divergence in cultural values within the nation results from ethnic polarisation, partly as a result of the governmentÕs NEP which is an affirmative action plan to eliminate the identification of race with economic function (Jomo, 1986).8 In reality, the NEP is a form of institutionalised positive discrimination in favour of the bumiputras9 by offering them various concessions including business contracts. The governmentÕs NEP, which discriminates on the basis of ethnicity, affects corporate behaviour including its social disclosure practice. During the economic boom of the mid-1990s there was substantial collusion between the government and certain Malay businessmen (Jayasankaran, 1997; Far Eastern Economic Review, 1998). Since government is one of the relevant publics of the firm, developing a corporate reputation as being socially responsible through performing and disclosing social responsibility activities may be seen as part of a strategic plan in managing stakeholder relationships (Roberts, 1992). Hence, it may be expected that firms managed by the governmentÕs favoured ethnic group, the Malays, choose to disclose more social responsibility 8 NEP was introduced in 1970 following race riots in 1969 due to the economic dominance of the Chinese amidst the poverty of the Malays (Jayasankaran and Hiebert, 1997). See Haniffa (1999) for a discussion of NEP and its implications on the business environment. 9 Bumiputra refers not only to Malaysians of Malay but other indigenous ethnic groups (Malaysia, 1991). However, for the purposes of this study, bumiputra refers to the Malay group since they form the majority.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 399

information as a legitimisation strategy aimed at changing the perceptions of other stakeholders on the business capability of Malay directors. An additional objective is to divert the attention of other stakeholders away from the close relationship between Malay directors and government, so that such favoured groups can continue to have an influential voice at both governmental and institutional levels. Similarly, since the finance director is normally involved in the preparation of annual reports and disclosure policies, the ethnicity of the finance director would influence corporate social responsibility disclosure policy and practice.10 In the same vein, the ethnicity of the majority of shareholders of a company would influence disclosure strategies of the company. Companies with Malay dominated shareholders would adopt legitimation strategies aimed at attracting or diverting their attention to or from issues of concern to such a group. For example, with regards to religion, Malay shareholders would be concerned to ensure that business activities are halal (permissible), and also whether employees are provided with a place to worship in the workplace. Thus, the intention of CSD by firms dominated by a particular ethnic group whose values are specifically supported by government, is corporate legitimation. Three proxies of culture11 based on ethnicity (firms with boards dominated by Malay directors; Malay finance director; and Malay shareholders) are selected to test the following hypotheses: H2: Ceteris paribus, the extent of CSD is greater for companies with boards dominated by Malay directors. H3: Ceteris paribus, the extent of CSD is greater for companies with a Malay Finance Director. H4: Ceteris paribus, the extent of CSD is greater for companies dominated by Malay shareholders. 3.3. Corporate governance Corporate governance may be broadly defined as ‘‘the manner in which companies are controlled and in which those responsible for the direction of companies are accountable to the stakeholders of these companies’’ (Dahya et al., 1996, p. 71). According to Selznick (1992, p. 290), ‘‘to govern [as opposed 10

Interviews conducted in Malaysia reveal that top management, including the finance director, in conjunction with PR specialists, makes decisions on information disclosure. In this study, finance director refers to the individual who has the greatest power in financial decision making as well as disclosure matters regardless of whether the individual sits on the board or not. It is also common in Malaysia to find the CEO to be the finance director as well. 11 We acknowledge that culture is a complex issue that cannot be resolved by a simple assertion of proxies. Nevertheless, the proxy chosen is thought appropriate to support the theoretical framework of this study.

400 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

to manage] is to accept responsibility for the whole life of the institution. . . . Governance takes account of all the interests that affect the viability, competence and moral character of an enterprise’’. Issues of governance include the legitimacy of corporate power, corporate accountability, to whom and for what the corporation is responsible, and by what standards it is to be governed and by whom (Worthy and Neuschel, 1983). The CSD process can be seen as a strategy aimed at closing a perceived legitimacy gap between management and shareholders (especially foreign shareholders) via non-executive directors. Non-executive directors are seen as the check and balance mechanism, not only in ensuring that companies act in the best interests of owners, but also other stakeholders; advising on the public presentation of the companyÕs activities and performance; and providing Ôadditional windows on the worldÕ (Tricker, 1984, p. 171). Furthermore, they are likely to respond to concerns about honour and obligations and would generally be more interested in satisfying the social responsibilities of the firm (Zahra and Stanton, 1988) as this may enhance their prestige and honour in society. Thus, non-executive directors can put pressure on companies to engage in CSD in ensuring congruence between organisational actions and societal values or organisational legitimacy. Therefore, boards dominated by non-executive directors are expected to have more influence on CSD as they are supposed to represent the interests of other stakeholders. Therefore, H5: Ceteris paribus, the extent of CSD is greater for companies with boards dominated by non-executive directors. The chairman of the board may have greater power and influence than other board members although, de jure, all directors share equal power (ignoring the issue of share ownership). Influence may extend to disclosure practices of the firm. Since CSD is largely voluntary in nature, awareness of issues of concern raised and discussed in other companies by chairmen who hold multiple directorships12 would influence disclosure practice of the company to be in tune with others as part of its reputation management and legitimacy strategy (DiMaggio and Powell, 1983). Multiple directorships refer to the situation where directors sit on more than one board.13 This aspect is often discussed 12 Before the introduction of the Malaysian Code of Corporate Governance (MCCG) in 2000 which among others requires directors to undergo training to equip themselves to undertake the role, the general perception is that non-executive directors sitting on the boards of Malaysian companies lack skills and experience to contribute to performance of companies and merely plays a Ôrubber stampÕ function except where they sit on more than one board. 13 We have not come across any literature that directly discusses multiple directorships. However, it has been discussed indirectly under directorship interlocks. It is quite difficult to segregate crossdirectorships/interlocks and multiple directorships because, by creating interlocks, the incidence of directors holding more than one directorship will be higher.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 401

in the literature under directorship interlocks (see Haunschild, 1993; Zajac and Westpal, 1996). The informational effects of interlocks have been widely discussed in the literature. For instance, Useem (1984) suggests that interlocks function as a channel of information about business practices while Lorsch and MacIver (1989) assert that the interlocking of CEOs of other firms on boards is desirable because of their experience and credibility as peers. Furthermore, ‘‘serving on a board is a way to see how somebody else is doing the same thing youÕre doing’’ (Lorsch and MacIver, 1989, p. 27). Thus, CEOs join other boards and thereby create interlocking relationships specifically to ÔembedÕ what they are doing (Davis, 1996). Another argument in favour of interlocking directors is that members of other boards can offer insights or comparisons derived from personal knowledge of other organisations (Dahya et al., 1996). Thus, decisions at one board become part of the raw material for decisions at other boards. There was also evidence that board members who have participated in various strategic/structural changes on other boards may bring those beliefs to advocate changes in another board (Bettenhausen and Murnighan, 1985; Mizruchi, 1992; Haunschild, 1993).14 Hence, H6: Ceteris paribus, the extent of CSD is greater for companies with a chairman having multiple directorships. The ownership structure of the company may give rise to legitimacy gaps. Different shareholders may demand different disclosures and the demand is greater when foreigners, due to the separation between management and owners geographically, hold a high proportion of shares (Schipper, 1981; Bradbury, 1991; Craswell and Taylor, 1992). Since a substantial fund in the Malaysian capital market comes from foreign investors, higher disclosure of information, including social and environmental information to aid decisionmaking may be expected. Thus, the hypothesis is, H7: Ceteris paribus, the extent of CSD is greater for companies dominated by foreign shareholders. 3.4. Control variables (firm-specific characteristics) 3.4.1. Size Previous studies have indicated a positive relationship between the extent of CSD and company size. One explanation for the association is that large companies undertake more activities and have greater impact on society (Trotman 14

For a discussion of the impact of interlocking directorships in Japan, see Cooke (1996).

402 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

and Bradley, 1981; Teoh and Thong, 1984; Andrew et al., 1989). Larger companies are also subject to greater scrutiny by various groups in society and therefore would be under greater pressure to disclose their social activities to legitimise their business (Cowen et al., 1987). Thus, H8: Ceteris paribus, the extent of CSD is greater for larger companies. 3.4.2. Profitability Unlike size, the relationship between profitability and CSD is inconclusive (see Mangos and Lewis, 1995; Patten, 1991; Roberts, 1992). A possible explanation for a positive association between CSD and profitability is that management has the freedom and flexibility to undertake and reveal more extensive social responsibility programmes to shareholders. Profitable companies disclose social information to demonstrate their contribution to societyÕs well being (legitimise their existence) and therefore the hypothesis is, H9: Ceteris paribus, the extent of CSD is greater for highly profitable companies. 3.4.3. Gearing In a highly geared company, management needs to legitimise its actions to creditors as well as shareholders. Gearing has been found to be an important explanatory variable by Belkaoui and Kahl (1978), Malone et al. (1993) and Wallace et al. (1994). Highly geared companies disclose more information to assure creditors that shareholders and management are less likely to bypass their covenant claims (Myers, 1977; Schipper, 1981) as well as to meet some of the needs of lenders (Cooke, 1996). Hence, the next hypothesis is, H10: Ceteris paribus, the extent of CSD is greater for highly geared companies. 3.4.4. Listing status Stakeholders in foreign countries have diverse interests and power and may, therefore, exert different pressures on companies. For example, in developing countries, there are few consumer and interest groups that are powerful and articulate enough to put pressure on companies to be socially responsible (Andrew et al., 1989). As such, companies listed on the domestic capital market in a developing country would not disclose a great deal of social information due to the absence of rules and regulations as well as public awareness. In contrast, companies that are listed in developed countries may have to adhere to rules and regulations pertaining to such information or adopt good practice by com-

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 403

peting companies and respond to the pressures of various interest groups. This also suggests that pressure to legitimise a corporation is greater in developed rather than developing countries. Hence, H11: Ceteris paribus, the extent of CSD is greater for companies with multiple listing status. 3.4.5. Industry type Companies tend to provide information that is in line with the peculiarities of their industries (Dye and Sridhar, 1995). For instance, labour intensive industries such as manufacturing will choose to disclose more information on employees compared to companies in the extractive and chemical industries that are likely to disclose more environmental information to reflect sensitivity to their particular problems. Similarly, consumer-oriented industries can be expected to exhibit more social disclosure to enhance their corporate image among market consumers, which in turn influences the amount of sales generated (Cowen et al., 1987). Thus, the influence of industry type on CSD practice depends on how critical the effects of their economic activities impacts on society. As such, hypothesis 12 is, H12: Ceteris paribus, the extent of CSD is associated with the type of industry.

4. Research method 4.1. Sample design and data collection The sample was drawn from non-financial companies listed on the main board of the KLSE in 1996. Companies were selected at random on a proportional allocation basis to ensure a representative sample from all other industrial sectors. Letters were sent to 160 companies requesting their English version annual reports and the response rate was 83%.15

15

The 1996 annual reports were collected in 1997. A sample size of 167 was selected based on the table developed by Krejcie and Morgan (1970) as a guideline for sample size decisions. The response rate in 1997 was 83% (139 companies) and due to time constraint after sending reminder letters twice, additional searches were not undertaken. Based on the same list of companies that responded in 1997, letters were sent in August 2003 asking for their 2002 annual reports to enable the undertaking of this longitudinal study. The English version of annual reports were selected to ease analysis and the researchers do not foresee any problems related to inconsistency in different language versions of the report as Malaysians are well versed in different languages.

404 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

Based on the pedagogical note suggested by Wallace and Mellor (1988), the third type of check on non-response bias, which involved assessing whether responses received validly represent the entire population based on certain selected characteristics, was undertaken. This is important because information relating to the whole population from which the sample is selected may enable the researcher to assess whether the responses are free of bias (Holmes et al., 1991).16 Although companies may use other media of communication to demonstrate social responsibility disclosures (see Zeghal and Ahmed, 1990), this study focuses on annual reports because of the high degree of credibility attached to information disclosed in this way (Tilt, 1994). Additionally, annual reports are used by a number of stakeholders as the sole source of certain information (Deegan and Rankin, 1997); have greater potential to influence due to widespread distribution (Adams and Harte, 1998); offer a snapshot of managementÕs mindset in a particular period (Neimark, 1992); and are more accessible for research purposes (Woodward, 1998). The same 139 companies were used for 2002 and our response rate was 100%. 4.2. Measurement of variables 4.2.1. Dependent variable—corporate social disclosure Content analysis, a method of codifying the text (or content) of a piece of writing into various groups (or categories) depending on selected criteria (Weber, 1988), was used to collect the necessary data. An essential element of content analysis is the selection and development of categories into which content units can be classified. The categories and items were drawn from previous research in the area (Ernst and Ernst, 1976; Cowen et al., 1987; Guthrie and Parker, 1989, 1990; Gray et al., 1995a) and applicability to the Malaysian environment (Hossain et al., 1994; Haniffa and Cooke, 2002). A research instrument (see Appendix A) was designed covering items relating to five themes (environmental, employee, community, product and value-added). Two types of measure (number of CSD items expressed as an index [CSDI] and the length of CSD items expressed in terms of number of

16 Non-response bias was tested using the parametric t-test based on means of two critical variables viz. total assets and return on equity (proxies for company size and profitability respectively) in testing the sets of respondents and non-respondents. The test for non-response bias indicated that the sample is soundly based i.e. no significant difference in the critical variables between the two groups. We also tested for non-response bias based on means of board composition and share ownership. The results also indicated that the sample is soundly based i.e. no significant difference based on culture and governance structure between the two groups.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 405

words [CSDL])17 were used to capture the nature of disclosure made in each of the five themes. The former measurement captures ÔvarietyÕ of disclosure while the latter captures the ÔextentÕ of disclosure. One of the reasons for adopting both measures is because the latter cannot capture pictures and graphics, which are potentially powerful and highly effective methods of communication (Beattie and Jones, 1992, 1994; Preston et al., 1996) and excluding them may be considered a limitation. We measure pictures and graphics through the disclosure index i.e. a company scores one for disclosure of any number of pictures or graphics under each item in a theme, hence indicating/capturing ÔvarietyÕ of disclosure rather than ÔextentÕ. Since our focus is on the nature of disclosure rather than the importance of items disclosed, we consider our approach to be reasonable. The preliminary research instrument was pilot tested on 20 companies selected at random from the 1996 and 2002 samples to ensure that items that were unique or important to the Malaysian environment were added and those not relevant omitted. This also ensured that items peculiar to a particular industry were taken into account. The final checklist instrument consisted of 41 corporate disclosure items. To ensure consistency, only one of the authors coded all the annual reports and a set of basic coding rules was constructed to ensure reliability and validity. 4.2.1.1. Corporate social disclosure index. The approach to scoring items is essentially dichotomous in that an item in the research instrument scores one if disclosed and zero if it is not18 although no penalty is imposed if the item is considered irrelevant. To ensure that judgement of relevance is not biased, the entire annual report is read before any decision is made (Cooke, 1992, 1996). The scores for each item were then added to derive a final score for 17 Different volume measurement has been employed in previous studies and each has its advantages and limitations. Number of pages (Patten, 1992; Deegan and Rankin, 1996) and proportion of page (Guthrie and Parker, 1990; Gray et al., 1995a) reflect the amount of total space given to a topic and by inference, the importance of that topic (Krippendorff, 1980). However, such measurements may be affected by font size, margins and treatment of blank parts of a page. The use of number of words (Zeghal and Ahmed, 1990; Deegan and Rankin, 1996; Deegan and Gordon, 1996) is more practical and easily categorised but may be affected by concise and verbose styles of writing (Hackston and Milne, 1996). Number of sentences (Tsang, 1998; Hackston and Milne, 1996) has the advantages of being more easily identifiable, less subjective to interjudge variations, and avoids problems of allocations based on proportion of page and standardising number of word but is more suitable in inferring meanings (Krippendorff, 1980). In this study, both word and sentence counts were considered but due to a high correlation between the two, only word count will be reported in the paper. 18 Use of dichotomous procedure is considered a limitation because it treats disclosure of one item as equal to a company that makes 50 disclosures and does not indicate how much emphasis is given to a particular content category but the advantage is that it gives coders less choice (Hackston and Milne, 1996, p. 88). The use of word count partly overcomes this problem.

406 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

the firm. The approach to scoring is additive and equally weighted19 and was calculated as follows to give the final CSDI (index): Pnj X ij CSDIj ¼ t¼1 nj where CSDIj = corporate social disclosure index, nj = number of items expected for jth firm, nj 6 41, Xij = 1 if ith item disclosed, 0 if ith item not disclosed so that

0 6 Ij 6 1.

4.2.1.2. Extent of corporate social disclosure. Using the same research instrument, the number of words in every sentence related to each item in the checklist is counted. Items relating to graphical presentation in the checklist were excluded from the word count. The number of words related to each item under the five themes was added together to compute the CSDL (length). 4.2.2. Independent variables The independent variables representing the constructs are given in Table 1. 4.3. Data analysis To test whether there is a significant difference in the extent and variety of disclosures in the two selected periods, both the parametric paired sample t-test and the non-parametric Wilcoxon test were conducted. We used regression analysis to test the interrelationship between the various independent variables and the two measures of overall CSD. The assumptions underlying the regression model were tested for multicollinearity based on the correlation matrix as well as the variance inflation factor (VIF). In addition, an analysis of residuals, plots of the studentised residuals against predicted values as well as the Q–Q plot, were conducted to test for homoscedasticity, linearity and normality assumptions. Normality tests based on skewness, kurtosis and Kolmogorov– Smirnov Lilliefors (KSL) were also conducted. The regression models based on the two measures are as follows:

19 Wiseman (1982) used a weighted index to measure disclosure but in this study the items were not weighted because of potential scoring bias and scaling problems and is consistent with Cooke (1989) and Ahmed and Nichols (1994).

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 407 Table 1 Constructs of the independent variables Explanatory variables

Measurement

Malay dominated board of directors (MALDIR) A Malay finance director (MALFD) Malay dominated shareholders (MALOWN) Composition of non-executive directors (COMPNED) Chairperson with multiple directorships (CHMD) Ownership by foreign shareholders (FOROWN) Size (STA) Profitability (ROE) Gearing (DTE) Multiple listing (LIST) Industry type (IT)

Proportion of Malay directors to total directors on the board 1 = Malay finance director and 0 = others Proportion of Malay shareholders to total shareholders Proportion of non-executive directors to total directors on the board 1 = chairperson with multiple directorships and 0 = otherwise Proportion of foreign shareholders to total shareholders Proxy used is total assets Return on equity = Earnings after tax/Total equity Debt to equity = Long-term debt/Total equity 1 = multiple listings and 0 = single listing KLSE classification: IT1 = consumer, IT2 = construction/property, IT3 = trading/services, IT4 = plantations/mining, IT5 = industrial

CSDI ¼ b0 þ b1 MALDIR þ b2 MALFD þ b3 MALOWN þ b4 COMPNED þ b5 CHMD þ b6 FOROWN þ b7 STA þ b8 ROE þ b9 DTE þ b10 LIST þ b11 IT1 þ b12 IT2 þ b13 IT3 þ b14 IT4 þ b15 IT5 þ it CSDL ¼ b0 þ b1 MALDIR þ b2 MALFD þ b3 MALOWN þ b4 COMPNED þ b5 CHMD þ b6 FOROWN þ b7 STA þ b8 ROE þ b9 DTE þ b10 LIST þ b11 IT1 þ b12 IT2 þ b13 IT3 þ b14 IT4 þ b15 IT5 þ it where CSDI = corporate social disclosure index, CSDL = corporate social disclosure length, MALDIR = proportion of Malay directors to total directors on the board, MALFD = Malay finance director, MALOWN = proportion of Malay shareholders to total shareholders, COMPNED = proportion of non-executive directors to total directors on the board, CHMD = chairperson with multiple directorships, FOROWN = proportion of foreign shareholders to total shareholders,

408 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

STA = size based on total assets, ROE = profitability based on return on equity, DTE = gearing based on debt to equity, LIST = multiple listing, IT1 = consumer, IT2 = construction/property, IT3 = trading/services, IT4 = plantations/mining, IT5 = industrial (the excluded dummy variable), b0, . . . , b15 = coefficients to be estimated, it = disturbance term. One other issue involved in the construction of the dependent variable relates to equal weighting since the directional magnitude may not be clearcut. Even though the scores and scoring instrument are connected to a numerical continuum associated with the dependent variable, it might not be appropriate to treat raw scores as interval measures since underlying characteristics are more akin to ordinal data. In addition, the direction of some of the relationships discussed earlier between the dependent and independent variables is not clear although assumed to be monotonic. One method to deal with this issue is to transform the data to normal scores (Cooke, 1998) and this is the approach adopted here. Normal scores are used since the resulting tests have exact statistical properties because significance levels can be determined, the F and t-tests are meaningful, the power of the F and t-tests can be used, and the regression coefficients derived are meaningful. In addition, normal scores offer a means whereby a non-normal dependent variable can be transformed into normality, thus implying that errors are normally distributed by the assumptions of OLS. To further verify the results of the regression analysis and in drawing conclusions based on legitimacy theory, telephone interviews20 were conducted with ten respondents involved in the corporate disclosure process. Initial contacts were made by telephone seeking their cooperation to participate, giving an approximate idea of how long the interview would last and setting up a mutually convenient time when the telephone conversation would not be interrupted. A semi-structured questionnaire (see Appendix B) was used to obtain insights into reasons for engaging/not engaging in CSD and the suitability of the proxies used in this study to reflect pressures from within and without the organisation. 20 Telephone interviews were used, as it is easy to contact different people outside the country in a relatively short period of time. E-mail was not used because respondents indicated preference to respond via short informal telephone conversations rather than having to read and answer the questionnaire sent by e-mail.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 409

5. Discussion of results 5.1. Descriptive analysis of CSD Table 2 presents the results of the descriptive analysis of CSD by themes for both years. The table summarises: the number of companies making at least one disclosure for both years (columns 1 and 2); incidence figures i.e. the number of companies making at least one disclosure as a percentage of the total sample for both years (columns 3 and 4); the means and standard deviation of disclosure based on the number of items disclosed under each theme for both years (columns 5–8); extent of disclosure as measured by the word count for both years (columns 9 and 10); the extent of disclosure as measured by the word count to total words for all disclosures in the sample (columns 11 and 12) for both years; and the means and standard deviations of disclosure based on the number of words disclosed under each theme for both years (columns 13–16). Relying on incidence rates alone may not give a complete picture as the approach treats companies that disclose one or more items in each theme as equal. For example, based on incidence rates, the percentage of companies (11%) disclosing at least one item under the environmental theme in 1996 is higher than the number of companies (9%) disclosing at least one item under the community involvement theme but in terms of the percentage of words disclosed by the former group (5% of total disclosures), it is less than the latter group (9% of total disclosure). This may partly be due to the exclusion of graphics and pictures when using word count. A total of 37,243 words of CSD was provided in the 1996 annual reports for the 139 companies examined, representing an average of 268 words per annual report. In the 2002 annual reports, a total of 38,293 words of CSD was recorded, an average of 275 words per annual report. Table 3 presents the results of the test on the hypothesis that there is a significant difference in the extent and variety of CSD in the annual reports of Malaysian companies for the two periods under study. Tests based on both Wilcoxon and paired sample t-test based on number of items indicate significant differences in four themes of CSD (environmental, employees, community and product) while words measurement do not show significant differences21 in two of those themes viz. community and product. Based on the descriptive analysis and statistical tests, we can accept the directional hypothesis (H1) i.e. CSD in annual reports of Malaysian companies has significantly increased (based on items measured) and the extent and variety of CSD differs between the two years for four of the five themes. This finding is consistent with other longitudinal studies.

21

Test for the theme Ôvalue-addedÕ cannot be computed due to zero standard deviation.

Themes

Disclosing companies (making at least one disclosure)

Disclosing companies as a percentage of total sample (incidence)

Mean

1996

2002

1996

2002

1996

Environment Employees Community Products Value-added

15 59 13 22 3

24 60 29 24 4

11 42 9 16 2

17 43 21 17 3

0.29 2.45 0.53 2.82 0.14

Total Note: Total sample of companies for each year—139.

Std. Dev.

Number of disclosed words (amount)

Disclosed words as a percentage of all disclosed words

Mean

Std. Dev.

2002

1996

2002

1996

2002

1996

2002

1996

2002

1996

2002

0.35 2.71 0.68 3.00 0.14

0.74 2.08 1.03 1.66 0.49

0.75 2.13 1.03 1.54 0.49

1909 11,867 3419 18,687 1361

2543 12,518 3542 18,329 1278

5 32 9 50 4

7 33 9 48 3

13.73 134.44 24.60 134.4 9.80

18.30 90.06 25.48 131.86 8.64

33.59 72.85 47.11 70.37 35.0

40.63 70.07 43.64 63.44 32.0

37,243

38,293

100

100

410 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

Table 2 Descriptive statistics for social disclosure measures in Malaysian listed companies

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 411 Table 3 Paired-sample t-test and Wilcoxon test for social disclosure measures in Malaysian listed companies Disclosures

Mean difference

t-value

Sig.

Mean

Z

Sig.

Overall Items

0.6403

8.028

0.000

6.597

0.000

Words

7.554

1.975

0.052

1996—6.230 2002—6.871 1996—267.94 2002—275.49

1.483

0.138

Environmental Items

0.065

3.091

0.002

3.000

0.003

Words

4.561

3.486

0.001

3.181

0.001

Employee Items

0.252

4.923

0.000

4.550

0.000

Words

4.684

2.753

0.007

2.955

0.003

Community Items

0.144

4.144

0.000

3.911

0.000

Words

0.885

0.584

0.560

0.903

0.366

Product Items

0.179

3.257

0.001

3.350

0.001

Words

2.576

0.896

0.372

1.481

0.139

N/A

N/A

N/A

N/A

N/A

Value-added Items Words

1996—0.288 2002—0.353 1996—13.74 2002—18.30 1996—2.453 2002—2.701 1996—85.37 2002—90.06 1996—0.532 2002—0.676 1996—24.60 2002—25.48 1996—2.820 2002—3.000 1996—134.4 2002—131.9 1996—0.1367 2002—0.1367 1996—9.7914 2002—9.832

Although CSD in Malaysia is voluntary in nature, companies that have disclosed in the past continue to disclose at least the same extent of information over time. The difference in extent and variety may be attributed to a number of factors such as the Best Corporate Social Reporting Award, the Hibiscus Award and the Environmental Reporting Awards.22 These awards recognise 22 This is one of the special category awards under NACRA (National Annual Corporate Report Awards) organised by the KLSE (now known as Bursa Malaysia), the Malaysian Institute of Certified Public Accountants (MICPA), Malaysian Institute of Management (MIM) and Malaysian Institute of Accountants (MIA) to honour excellent corporate reporting among public listed companies.

412 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

exemplary commitment by companies to the natural and physical environment; greater awareness and interests of relevant publics about corporate social activities; governmentÕs interest in the environmental effects of business activities; and interest aimed at demonstrating good corporate behaviour and thereby to continue to benefit from government support and attract funds from ethical investors. 5.2. Multivariate analysis 5.2.1. Descriptive statistics Table 4 provides descriptive statistics of the voluntary disclosure index and the continuous independent variables. It is noticeable that the means of the CSDI in 1996 and 2002 were 16.28% and 17.13% respectively. The range in Table 4 Descriptive statistics for the dependent and continuous independent variables Variables 1996 CSDI CSDL MALDIR MALOWN COMPNED FOROWN STA (RM million) ROE DTE 2002 CSDI CSDL MALDIR MALOWN COMPNED FOROWN STA (RM million) ROE DTE

Mean

Std. Dev.

Min.

Max.

Skewness Kurtosis K–S (Lilliefors)

16.28 11.11 2.63 52.63 5.22 267.94 178.83 15.00 857.00 4.55 46.05 24.46 0.00 100.00 2.20 36.85 22.84 6.01 97.30 4.42 43.34 11.54 20.00 82.00 3.81 16.19 12.60 0.00 48.85 3.54 883030.7 1519155 3986 11867584 21.45

2.44 1.29
1.38
0.38 1.97
0.39 59.44

0.125** 0.102** 0.119** 0.151** 0.129** 0.099** 0.281**

0.36 1.27

2.14
0.31

0.069 0.064

17.13 10.47 2.50 56.10 6.06 275.49 168.52 16.00 801.00 4.66 47.19 23.70 0.000 100.00 1.95 38.73 23.20 10.40 94.00 3.52 49.87 10.14 25.00 74.00
1.20 14.90 10.12 0.00 43.13 2.16 3681429 9519842 8529 81859055 27.96

4.36 1.27
1.14
1.43
1.40
1.14 95.37

0.149** 0.110** 0.116** 0.132** 0.140** 0.070 0.350**

9.81 38.62

12.14 14.5

8.20
18.17 20.67 1.07

7.35 4.82


5.30 0.20

31.79 89.56

35.33 46.13

1.79 3.86

0.06 1.50

0.088 0.038

Notes: CSDI = corporate social disclosure index; CSDL = length of corporate social disclosure; MALDIR = proportion of Malay directors on the boards; MALOWN = proportion of Malay shareholders; COMPNED = proportion of non-executive directors on the board; FOROWN = proportion of foreign shareholders; STA = size of total assets; ROE = return on equity (proxy for profitability); and DTE = debt to equity ratio (proxy for gearing). ** K-S (Lilliefors) with significance <.05, hence data not normally distributed.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 413 Table 5 Correlation matrix CSDL

CSDI

BPDIR BPOWN NED

FOOWN SIZETA ROE DTE

1996 CSDL CSDI MALDIR MALOWN COMPNED FOROWN STA ROE DTE

1 .933** 1 .219** .259** 1 .148 .135 .232** 1
.182*
.192* .126
.047 .075 .172*
.176* .016 .441** .419** .138 .056 .310** .268**
.207*
.013
.042
.041 .015
.067

1
.162 1 .062 .145
.182* .157
.003 .015

2002 CSDL CSDI MALDIR MALOWN COMPNED FOROWN STA ROE DTE

1 .940** 1 .216* .243** 1 .264** .248** .231** 1
.241**
.186* .066
.048 .193* .258**
.144 .158 .195* .210* .102 .066 .333** .356**
.109
.024
.063
.043 .023 .106

1
.122 1 .001 .076
.164 .162 .173*
.138

1 .042 .078

1 .031 1

1
.046 .137

1
.094 1

* Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

1996 was from 2.6% to 52.6% while in 2002, the range was between 2.5% and 56.1%. The mean of the CSDL in 1996 was 267 words and ranged between 15 and 857 words while the mean of the CSDL in 2002 was 275 words with a range between 16 and 801 words. Both measures of dependent variables were not normally distributed as indicated by standard tests on skewness and kurtosis as well as the non-parametric Kolmogorov–Smirnov normality test (or K–S Lilliefors).23 Similarly, the continuous independent variables were found not to be normally distributed except for profitability and gearing in 1996 and foreign ownership and gearing in 2002. As such, the dependent and continuous independent variables were transformed to normal scores before conducting the regression analysis. Besides testing for normality, it is important to check for multicollinearity, homoscedasticity and linearity. Table 5 presents the correlation matrix for the dependent and continuous independent variables for both years. It can be seen

23 K–S Lilliefors with significance of >.05 indicates normality and small significance value indicates reason to doubt the normality assumption (see Norusis, 1995).

414 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

that multicollinearity is not a problem.24 Similarly, an analysis of residuals25 indicates no problems of heteroscedasticity and linearity. Table 6 summarises the results for both years of the multivariate regression models using normal scores26 for the CSDI i.e. based on ÔvarietyÕ of disclosure items expressed in the form of an index. The adjusted R2 for 1996 was 0.389 and for 2002 was 0.453. Regression results for both years indicate a positive relationship at the 1% and 5% significance level, as predicted, between CSDI and companies with Malay dominated boards and companies dominated by foreign shareholders, respectively. Interestingly, companies with boards dominated by non-executive directors were also significantly related at the 1% significance level to CSDI, but in the direction opposite to that predicted. In 1996, results indicate a significant positive relationship at the 5% level between CSDI and chairman with multiple directorships, and in the direction predicted. Out of the five firm-specific control variables, multiple listings and profitability were found to be significant at the 1% level while size was found to be significant at the 5% level for both years. Gearing and industry membership were not significant for both years. Table 7 presents the regression results for both years using normal scores for the CSDL i.e. based on ÔextentÕ of disclosure (word count). The adjusted R2 for 1996 was 0.485 and for 2002 was 0.438. Regression results for both years indicate a positive relationship at the 1% significance level, as predicted, between CSDL and companies with Malay dominated boards but in the direction opposite to that predicted for boards dominated by non-executive directors. The regression model also indicates a significant relationship, in the direction predicted, between chairman with multiple directorships and CSDL at the 1% and 5% levels for 1996 and 2002 respectively. Malay share ownership was found to be significant at the 5% level for only 2002. Three of the control variables (size, profitability and multiple listing) were again found to be significant for both years at less than the 1% level. In 2002, none of the industry variables were significant at the 5% level but in 1996, two industrial sectors (trading and services and construction and property) were found to disclose significantly more than the other sectors. Based on the two models for 1996 and 2002 we find that Malay dominated boards are positively related to CSD, thus providing support for H2.

24

The rule of thumb for checking problems of multicollinearity is when the correlation is >0.800 (Gujarati, 1995). Based on the VIF, multicollinearity is a problem if the factor exceeds 10 (Neter et al., 1983; Kennedy, 1992). 25 Residuals are what are left over after the model is fit and they are also the difference between the observed value of the dependent variable and the value predicted by the regression line (Norusis, 1995, p. 447). 26 Four different transformations of the dependent and independent variables were conducted and the normal scores approach produced the best fit.

1996

2002

Variables

Predicted sign

Coefficient value

t-statistic

Sig t

VIF

Coefficient value

t-statistic

Sig t

VIF

Intercept MALDIR MALOWN MALFD COMPNED CHMD FOROWN STA ROE DTE LIST IT1 IT2 IT3 IT4

+ + + + + + + + + + ± ± ± ±

0.371 0.021
0.112
0.221 0.140 0.166 0.193 0.215
0.058 0.238 0.059 0.126 0.146
0.036


1.640 5.061 0.300
1.619
3.147 2.133 2.148 2.511 2.781
0.743 3.238 0.745 1.494 1.647
0.369

0.104 0.000** 0.765 0.108 0.002** 0.035* 0.034* 0.013* 0.006** 0.459 0.002** 0.457 0.138 0.102 0.713

1.214 1.119 1.086 1.114 1.292 1.353 1.339 1.348 1.371 1.222 1.410 1.617 1.780 2.199

0.359 0.101
0.105
0.205 0.087 0.153 0.188 0.278
0.017 0.288 0.043 0.124 0.068
0.104


1.119 5.005 1.491
1.577
3.085 1.336 2.197 2.584 3.759
0.230 4.115 0.580 1.542 0.803
1.113

0.234 0.000** 0.138 0.117 0.003** 0.184 0.021* 0.011* 0.000** 0.818 0.000** 0.563 0.126 0.424 0.268

1.301 1.148 1.117 1.117 1.082 1.233 1.338 1.376 1.365 1.238 1.408 1.636 1.806 2.212

Adjusted R2 = 0.389, F-statistic = 7.287, p = 0.000

Adjusted R2 = 0.453, F-statistic = 9.164, p = 0.000

Note: IT1 = consumer; IT2 = construction/property, IT3 = trading/services; IT4 = plantations/mining; IT5 = industrial and the excluded industry group is industrial. * Significant at 5%; ** significant at 1%.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 415

Table 6 Multiple regression results using index (CSDI) as the dependent variable

1996

2002

Variables

Predicted sign

Coefficient value

t-statistic

Sig t

VIF

Coefficient value

t-statistic

Sig t

VIF

Intercept MALDIR MALOWN MALFD COMPNED CHMD FOROWN STA ROE DTE LIST IT1 IT2 IT3 IT4

+ + + + + + + + + + ± ± ± ±

0.313 0.030
0.060
0.217 0.173 0.108 0.210 0.219
0.013 0.307 0.068 0.182 0.288 0.033


3.292 4.591 0.453
0.933
3.319 2.441 1.506 2.933 3.054
0.183 4.485 0.930 2.309 3.484 0.363

0.001** 0.000** 0.651 0.352 0.001** 0.004** 0.135 0.004** 0.003** 0.855 0.000** 0.354 0.023* 0.001** 0.717

1.214 1.119 1.086 1.114 1.250 1.353 1.339 1.348 1.371 1.222 1.410 1.617 1.780 2.199

0.273 0.138
0.080
0.193 0.139 0.088 0.232 0.211 0.013 0.305 0.084 0.076 0.169
0.018


2.147 3.745 2.018
1.192
2.867 2.036 1.192 3.141 2.817 0.180 4.298 1.110 0.927 1.971
1.194

0.034* 0.000** 0.046 0.236 0.005** 0.045* 0.236 0.002** 0.006** 0.857 0.000** 0.269 0.356 0.051 0.846

1.301 1.148 1.117 1.117 1.082 1.344 1.338 1.376 1.365 1.238 1.408 1.636 1.806 2.212

Adjusted R2 = 0.485, F-statistic = 11.007, p = 0.000

Adjusted R2 = 0.438, F-statistic = 8.688, p = 0.000

Note: IT1 = consumer; IT2 = construction/property; IT3 = trading/services; IT4 = plantations/mining; IT5 = industrial and the excluded industry group is industrial. * Significant at 5%; ** significant at 1%.

416 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

Table 7 Multiple regression results using words (CSDL) as the dependent variable

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 417

Respondents in fact identified ethnicity of boards as a determinant of CSD partly due to the ÔfeminineÕ cultural value of the Malays (see, for example, Hoecklin, 1994, p. 38) and also because of the close link to government (patron–client network). The existence of a policy of positive discrimination in business opportunities based on ethnicity affects corporate behaviour including its disclosure practice. Being the governmentÕs favoured ethnic group, boards dominated by Malays adopt a reactive legitimation strategy to change perceptions and divert the attention of its various stakeholders away from the close relationship they enjoy with government by increasing social responsibility disclosures. The influence of ethnicity of finance directors, the second cultural variable, was not significantly related to CSD. This implies that Malay finance directors in Malaysian listed companies do not seem to pay a great deal of attention to legitimation strategies on CSD when preparing annual reports. As for the third cultural variable, the proportion of Malay shareholders, it is noticeable that there is a positive relationship at the 5% level only in 2002. This suggests that companies dominated by Malay shareholders do tend to disclose more in 2002, perhaps to address some perceived legitimacy gap following the financial crisis event. In terms of corporate governance variables, the regression models for both years indicate that the composition of non-executive directors is statistically related to CSD but in the opposite direction to that predicted. This implies that Malaysian companies with boards dominated by non-executive directors play a limited role in influencing CSD policy and practice. The telephone interviews conducted, confirmed this and was explained in terms of the relative lack of experience and knowledge of non-executive directors and also, in some situations, because of indifference towards societal concerns. However, chairs with multiple directorships were significantly related to CSD in the direction predicted for both years using CSDL as the dependent variable and in 1996, for the CSDI dependent variable, but not in 2002. This seems to confirm that non-executive directors as a group do not have much influence but a chairman with experience gained by sitting on more than one board may be able to influence disclosure. Interview respondents agreed that sometimes the chairman may express a desire to disclose certain issues based on knowledge of events in other companies in which he/she is a director or simply, to ensure congruence between organisationsÕ actions and societal concerns to enhance his/her prestige and honour in society (Zahra and Stanton, 1988). The results also support the informational effect organisational theory as proposed by Useem (1984), Lorsch and MacIver (1989) and Davis (1996). The third corporate governance variable, foreign share ownership, was found to be statistically significant at the 5% level based on CSDI in 1996 and 2002 but not when the dependent variable is CSDL. Firms dominated by foreign shareholders engage in CSD to attract funds from a wide range of sources, an explanation confirmed with the telephone respondents.

418 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

The two models also show that size, profitability and multiple listings were all statistically related to CSD, results consistent with Foo and Tan (1988) and Andrew et al. (1989) on Malaysian companies. Large companies make more social disclosures for reasons of accountability and visibility as outlined in legitimacy theory (Cormier and Gordon, 2001). The significance of profitability was consistent with Roberts (1992) but inconsistent with Cowen et al. (1987) and Patten (1991). This indicates that Malaysian companies use annual reports as an avenue to publicise their image and legitimise their activities. The significance of multiple listings was consistent with Hackston and Milne (1996) and indicates that in the absence of rules and regulations on social disclosures in Malaysia, companies with listings on overseas stock exchanges adopt legitimation strategies to reflect societal concerns in the global market. The industry– social disclosure relationship seems to be less significant with the interaction of other variables, suggesting that Malaysian companies do not adopt legitimation strategies to address specific concerns relating to core economic activities in their industry grouping. Similarly, gearing as proxy for risk, does not seem to impact on CSD.

6. Conclusion 6.1. Overview of findings This study has examined whether the extent of CSD in the annual reports of Malaysian listed companies changes over time and whether there is an association with three groups of variables: culture, corporate governance and firmspecific (control) variables. Consistent with previous studies, content analysis is adopted to achieve the objectives. Descriptive analysis of our longitudinal study and the results of both the parametric paired sample t-test and non-parametric Wilcoxon test indicates significant differences in the extent and variety of CSD for the two years, despite minimal legislative guidance for such disclosures. Some of the reasons identified from the interviews of companies choosing to engage in CSD include getting awards, enhancing corporate image, receiving government support, obtaining funds and a bandwagon effect. Regression analysis is used to explain variability in the dependent variable with the explanatory variables being culture and corporate governance with company-specific factors acting as control variables. Two different dependent variables are used in the regression models: CSDI and CSDL. We find two cultural, three corporate governance and four of the control variables to be significant regressors that help explain variability in CSD practice of Malaysian companies in both 1996 and 2002. The significant relationship between Malay directors and Malay shareholders with CSD practice in the annual reports of Malaysian companies suggests

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 419

that disclosure cannot be culture free and is attributed primarily to government policy. Hence, CSD is used as a reactive legitimation strategy aimed at diverting attention from questionable business practices, cronyism, nepotism and close affiliation with the government as well as a proactive legitimation strategy to ensure a continued influential voice at both governmental and institutional levels. The significant relationship between executive directors and chair with multiple directorships with CSD indicates that those who are aware of the business environment make disclosure decisions for a purpose. The values of executive directors may not be congruent with other stakeholders and CSD can be used as a legitimation strategy to appease some of the concerns of the relevant publics. Similarly, the chairman may want companies to engage in CSD as a legitimation strategy to demonstrate social worthiness and that the firm is in tune with societal concerns. The significant relationship between CSD and foreign shareholders indicates that Malaysian companies use CSD as a proactive legitimation strategy to obtain continued inflows of capital and to please ethical investors. Our results have implications for other countries in the Asia–Pacific region, an area where power tends to be concentrated within an elite group thereby maintaining the traditional order of patron–client networks (Maidment and Mackerras, 1998). Countries in the Asia–Pacific region where racial, ethnic and linguistic discrimination and inequality exists includes Fiji, India, Indonesia, Hong Kong and Japan. Distinctions between ethnic groups in these countries may be institutionalised, even to the extent of being enshrined in legal systems. For example, in Fiji electors may only vote for candidates from their own ethnic group (Maidment and Mackerras, 1998). However, racial, ethnic and linguistic distinctions are much less important in the Anglo-Saxon influenced countries (areas) of Australia, New Zealand and Hawaii and therefore our research has further implications for these countries. 6.2. Limitations and future research The findings of this research are subject to several limitations. First, this study examined the disclosure practice of companies listed only on the main board of the KLSE for the years 1996 and 2002 and as such, may not be generalised to other periods. A longitudinal study on a yearly basis that can trace the disclosure practice of a particular company or a particular industry is a potential avenue of research as it may help to provide insights into the relationship between strategic changes in the company or industry over the years and its CSD practices. Similarly, it will help trace the trend of disclosure and the impact of culture and corporate governance against the backdrop of social and economic development in the country. Since this study considered the influence of culture and corporate governance factors on Malaysian listed

420 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

companies, studies in the future could test the influence of these variables on CSD in other countries. Secondly, the study focused on only disclosures in corporate annual reports although it is known that management utilise other mass communication mechanisms. Hence, future research may consider disclosures in other media such as newspapers, the internet, and in-house magazines. Thirdly, developing accurate proxies for culture and corporate governance dimensions in the CSD models and selection of variables to be included in the models are constrained by data availability. Future research may consider other cultural proxies such as religious values and culture dimensions proposed by Hofstede (1991) and corporate governance dimensions such as age, qualification and share ownership of board of directors. Finally, given the exceedingly complex nature of the business environment as well as time and geographical constraint, there are inherent limits in the ability of positive empirical research to capture all dimensions that influence CSD policy and practice. Hence, further survey work involving more detailed interviews may help our understanding of these issues. Appendix A. Research instrument used in the study NAME OF COMPANY: INDUSTRY: I. COMMUNITY INVOLVEMENT 1. General philanthropy 2. Participation in government social campaigns 3. Community programs (health & education) II. ENVIRONMENTAL 1. Environmental policies 2. Raw materials conservation & recycling 3. Environmental protection programme 4. Awards for environmental protection 5. Support for public/private action designed to protect the environment III. EMPLOYEE INFORMATION 1. Employees appreciation 2. Recruitment problems

ITEMS

WORDS

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 421

Appendix A (continued) NAME OF COMPANY: INDUSTRY:

ITEMS

WORDS

3. Discussion of ways to overcome recruitment problems 4. Picture of employees welfare 5. Discussion of employees welfare 6. Profit sharing schemes policy 7. Number of employees 8. Breakdown of employees by line of business 9. Breakdown of employees by geographic area 10. Categories of employees by functions 11. Categories of employees by race 12. Categories of employees by age 13. Number of employees for 2 or more yrs 14. Reasons for changes in employee number 15. General redundancy/ retrenchment information 16. Information on accidents 17. Cost of safety measures 18. Health & safety standards 19. Corporate policy on employee training 20. Nature of training 21. Number of employees trained 22. Amount spent on employees training 23. Categories of employees trained IV. PRODUCT OR SERVICE INFORMATION 1. Discussion of major types of products 2. Pictures of major types of products (continued on next page)

422 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

Appendix A (continued) NAME OF COMPANY: INDUSTRY:

ITEMS

WORDS

3. Improvement in product quality 4. Improvement in customer services 5. Distribution of mktg network for finished products—domestic market 6. Distribution of mkg network for finished products—foreign market 7. Customer awards/ratings received V. VALUE-ADDED INFORMATION 1. Value-added statement 2. Qualitative value-added statement 3. Value-added data/ratios Total Index score

Appendix B. Semi-structured questionnaire Section A 1. Current job status/position in the organisation 2. Number of years worked in the organisation (a) Less than 1 (b) 1–2 (c) 3–5 (d) 6–10 (e) Over 10 3. Number of other organisations worked for before joining this organisation (a) None (b) One (c) Two (d) Three (e) Four or more

h h h h h

h h h h h

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 423

Appendix B (continued) Section B 1. Who is responsible for preparing and publishing the companyÕs annual report in your company? 2. Which other parties may influence decisions as to accounting policies and formats and the level of disclosure? 3. Who has the most influence in the disclosure decision making process in your company? (by rank) Finance Director/Chief Accountant Chairman of the board of directors Managing Director Public relations consultant External Auditor Audit Committee Other 4. Is your company engaged in CSD? Yes No 5. Why does your company engaged in CSD? Enhance corporate image/Good corporate citizenship/marketing/PR Win awards Bandwagon/fashionable to do so Obligations to community/accountability Public awareness and concerns on CSD issues Improve morale of employees Appease ethical investors/seek credit for good deeds Obtain funds from wider sources Pressures from stakeholders Lower political pressures Competitors in the industry Receive government support Directors desire to engage in CSD Media attention Stability and improvement in share prices Other 6. Why doesnÕt your company engaged in CSD? Cost of data collection and processing/ auditing/publication/technical problems Competitive disadvantage/Disadvantages in terms of bargaining power Possibility of intervention by government agencies (continued on next page)

424 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430

Appendix B (continued) Section B

7.

8.

9.

Possibility of claims from political or consumer groups/increase demands/heighten suspicion Not to set precedence/make an issue Other What factors may influence companies in Malaysia to engage in CSD? Enhance corporate image/Good corporate citizenship/marketing/PR Win awards Bandwagon/fashionable to do so Obligations to community/accountability Public awareness and concerns on CSD issues Improve morale of employees Appease ethical investors/seek credit for good deeds Obtain funds from wider sources Pressures from stakeholders Lower political pressures Competitors in the industry Receive government support Directors desire to engage in CSD Media attention Stability and improvement in share prices Other What factors may influence companies in Malaysia not to engage in CSD? Cost of data collection and processing/ auditing/publication/technical problems Competitive disadvantage/Disadvantages in terms of bargaining power Possibility of intervention by government agencies Possibility of claims from political or consumer groups/increase demands/heighten suspicion Not to set precedence/make an issue Other Which of the following characteristics of companies may influence decisions to engage in CSD? Size Type of industry

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 425

Appendix B (continued) Section B

10.

11.

12.

Listing status Listing age Profitability Other Which of the following personal characteristics of directors may influence decisions to engage in CSD? Race Social status Academic background Other Which of the following characteristics of boards may influence decisions to engage in CSD? Cross-holdings of directorships Board composition (executive vs non-executive) Finance director on the board Other Any other matters/issues?

Note: Some of the questions are adapted from OÕDwyer (2002).

References Adams, C.A., Harte, G., 1998. The changing portrayal of the employment of women in British banksÕ and retail companiesÕ corporate annual reports. Accounting, Organizations and Society 23 (80), 781–812. Ahmed, K., Nichols, D., 1994. The impact of non-financial company characteristics on mandatory disclosure compliance in developing countries: The case of Bangladesh. International Journal of Accounting 29 (1), 62–77. Adams, C., Hill, W.Y., Roberts, C.B., 1995. Environmental, employee and ethical reporting in Europe. ACCA, London. Adams, C., Hill, W.Y., Roberts, C.B., 1998. Corporate social reporting practices in Western Europe: Legitimating corporate behaviour. British Accounting Review 30, 1–21. Anderson, J.C., Frankle, A.W., 1980. Voluntary social reporting: An iso-beta portfolio analysis. The Accounting Review 55 (3), 467–479. Andrew, B.H., Gul, F.A., Guthrie, J.E, Teoh, H.Y., 1989. A note on corporate social disclosure practices in developing countries: The case of Malaysia and Singapore. British Accounting Review 21, 371–376. Beattie, V., Jones, M.J., 1992. The use and abuse of graphs in annual reports: Theoretical framework and empirical study. Accounting and Business Research 22 (88), 291–303. Beattie, V., Jones, M.J., 1994. An empirical study of graphical format choices in charity annual reports. Financial Accountability and Management 10 (3), 215–236. Belkaoui, A.R., Kahl, A., 1978. Corporate financial disclosure in Canada, Research Monograph No. 1 of Canadian Certified General Accountants Association, Vancouver.

426 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 Bettenhausen, K., Murnighan, J.K., 1985. The emergence of norms in competitive decision-making groups. Administrative Science Quarterly 30, 350–372. Birnberg, J., Snodgrass, C., 1988. Culture and control: A field study. Accounting, Organisations and Society 13, 447–465. Bradbury, M.E., 1991. Characteristics of firms and voluntary interim earnings disclosure: New Zealand evidence. Pacific Accounting Review 3, 37–62. Chuah, B.H., 1995. The unique breed of Malaysian managers, Management Times. New Straits Times Press, Malaysia, March 7, 6. Clarke, J., Gibson-Sweet, M., 1999. The use of corporate social disclosures in the management of reputation and legitimacy. Business Ethics: A European Review 8 (1), 5–13. Cohen, A., 1974. Two-Dimensional Man. Routledge and Kegan Paul, London. Cooke, T.E., 1989. Disclosure in the corporate reports of Swedish companies. Accounting and Business Research 19 (74), 113–124. Cooke, T.E., 1992. The impact of size, stock market listing and industry type on disclosure in the annual reports of Japanese listed companies. Accounting and Business Research 22 (87), 229– 237. Cooke, T.E., 1996. The influence of the Keiretsu on Japanese corporate disclosure. Journal of International Financial Management and Accounting 7 (3), 191–215. Cooke, T.E., 1998. Regression analysis in accounting disclosure studies. Accounting and Business Research 28 (3), 209–224. Cormier, D., Gordon, I.M., 2001. An examination of social and environmental reporting strategies. Accounting, Auditing and Accountability Journal 14 (5), 587–616. Cowen, S.S., Ferreri, L.B., Parker, L.D., 1987. The impact of corporate characteristics on social responsibility disclosure: A typology and frequency-based analysis. Accounting, Organizations and Society 12 (2), 111–122. Craswell, A., Taylor, S., 1992. Discretionary disclosure of reserves by oil and gas companies: An economic analysis. Journal of Business Finance and Accounting 19, 295–308. Dahya, J., Lonie, A.A., Power, D.M., 1996. The case for separating the roles of Chairman and CEO: An analysis of stock market and accounting data. Corporate Governance—An International Review 4 (1), 71–77. Davis, G.F., 1996. The significance of board interlocks for corporate governance. Corporate Governance 4, 154–159. Deegan, C., 2000. Financial Accounting Theory. McGraw Hill Book Company, Sydney. Deegan, C., Gordon, B., 1996. A study of the environmental disclosure practices of Australian companies. Accounting and Business Research 26 (3), 187–199. Deegan, C., Rankin, M., 1996. Do Australian companies report environmental news objectively? An analysis of environmental disclosures by firms prosecuted successfully by the Environmental Protection Authority. Accounting, Auditing and Accountability Journal 9 (2), 52–69. Deegan, C., Rankin, M., 1997. The materiality of environmental information to users of annual reports. Accounting, Auditing and Accountability Journal 10 (4), 562–583. Deegan, C., Rankin, M., Tobin, J., 2002. An examination of the corporate social and environmental disclosures of BHP from 1983–1997: A test of legitimacy theory. Accounting, Auditing and Accountability Journal 15 (3), 312–343. DiMaggio, P.J., Powell, W.W., 1983. The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review 48, 147–160. Donaldson, T., Preston, L., 1995. The stakeholder theory of the corporation—concepts, evidence, and implications. Academy of Management Review 20 (1), 65–92. Dowling, J., Pfeffer, J., 1975. Organizational legitimacy: Social values and organizational behaviour. Pacific Sociological Review 18 (1), 122–126. Dye, R.A., Sridhar, S.S., 1995. Industry-wide disclosure dynamics. Journal of Accounting Research 33 (1), 157–174.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 427 Ernst and Ernst, 1976. Social Responsibility Disclosure. Cleveland, OH, USA. Far Eastern Economic Review, 1998. Survey Results. June 18, p. 36. Foo, S.L., Tan, M.S., 1988. A comparative study of social responsibility reporting in Malaysia and Singapore. Singapore Accountant. August 12–15. Freedman, M., Jaggi, B., 1982. Pollution disclosures, pollution performance and economic performance. The International Journal of Management Science 10 (1), 167–176. Gibbins, M., Richardson, A., Waterhouse, J., 1990. The management of corporate financial disclosure: Opportunism, ritualism, policies and processes. Journal of Accounting Research 28 (1), 121–143. Gray, R., Kouhy, R., Lavers, S., 1995a. Corporate social and environmental reporting: A review of the literature and a longitudinal study of UK disclosure. Accounting, Auditing and Accountability Journal 8 (2), 47–77. Gujarati, D., 1995. Basic Econometrics. McGraw Hill, NY. Guthrie, J.E., Parker, L.D., 1989. Corporate social reporting: A rebuttal of legitimacy theory. Accounting and Business Research 19 (76), 343–352. Guthrie, J.E., Parker, L.D., 1990. Corporate social disclosure practice: A comparative international analysis. Advances in Public Interest Accounting 3 (2), 159–176. Hackston, D., Milne, M.J., 1996. Some determinants of social and environmental disclosures in New Zealand companies. Accounting, Auditing and Accountability Journal 9 (1), 77–108. Haniffa, R.M., 1999. Culture, corporate governance and disclosure in Malaysian corporation. Unpublished thesis, University of Exeter. Haniffa, R.M., Cooke, T.E., 2002. Culture, corporate governance and disclosure in Malaysian corporations. Abacus 38 (3), 317–349. Harowitz, D.L, 1985. Ethnic Groups in Conflict. University of California Press, Berkeley. Harte, G., Owen, D., 1991. Environmental disclosure in the annual reports of British companies: A research note. Accounting, Auditing and Accountability Journal 4 (3), 51–61. Haunschild, P.R., 1993. Interorganisational imitation: the impact of interlocks on corporate acquisitions activities. Administrative Science Quarterly 38, 564–592. Herremans, I.M., Akathaporn, P., McInnes, M., 1993. An investigation of corporate social responsibility reputation and economic performance. Accounting, Organizations and Society 18 (7/8), 587–604. Hoecklin, L., 1994. Managing Cultural Differences. Addison-Wesley, Wokingham. Hofstede, G.H., 1991. Management in a multicultural society. Malaysian Management Review 26 (1), 3–12. Holmes, S., Kent, P., Downey, G., 1991. The Australian differential debate: A survey of practioners. Accounting and Business Research 21 (82), 125–132. Hossain, M., Tan, M.L, Adams, M.B., 1994. Voluntary disclosure in an emerging capital market: Some empirical evidence from companies listed on the KLSE. The International Journal of Accounting 29 (4), 334–351. Ingram, R.W., Frazier, K.B., 1980. Environmental performance and corporate disclosure. Journal of Accounting Research 18 (2), 614–622. Jayasankaran, S., 1997. Hit the brakes. Far Eastern Economic Review (Dec 18), 14–15. Jayasankaran, S., Hiebert, M., 1997. Malaysian dilemmas. Far Eastern Economic Review (Sept 4), 18–20. Jayasankaran, S., Hiebert, M., 1998. Calling doctor Daim. Far Eastern Economic Review (Feb 19), 14–17. Jomo, S.K., 1986. A Question of Class: Capital, the State, and Uneven Development in Malaya. Singapore University Press. Kennedy, P., 1992. A Guide to Econometrics. MIT Press, MA. Krejcie, R., Morgan, D., 1970. Determining sample size for research activities. Educational and Psychological Measurement 30, 607–610.

428 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 Krippendorff, K., 1980. Content Analysis: An Introduction to Its Methodology. Sage, London. Lehman, G., 1995. A legitimate concern for environmental accounting. Critical Perspectives on Accounting 6, 393–412. Lindblom, C.K., 1994. The implications of organizational legitimacy for corporate social performance and disclosure. Working Paper for the American Accounting Association Public Interest Section, USA. Lorsch, J., MacIver, E., 1989. Pawns or Potentates: The Reality of AmericaÕs Corporate Boards. Harvard University Press, Cambridge, MA. Maclagan, P., 1999. Corporate social responsibility as a participative process. Business Ethics: A European Review 8 (1), 43–49. Maidment, R., Mackerras, C., 1998. Culture and Society in the Asia–Pacific. Routledge, London. Malaysia, 1991. Second Outline Perspective Plan 1991–2000. National Printing Department, Kuala Lumpur. Malone, D., Fries, C., Jones, T., 1993. An empirical investigation of the extent of corporate financial disclosure in the oil and gas industry. Journal of Accounting, Auditing and Finance 8, 249–273. Mangos, N.C., Lewis, N.R., 1995. A socio-economic paradigm for analysing managersÕ accounting choice behaviour. Accounting, Auditing and Accountability Journal 8 (1), 38–62. Mathews, M.R., 1997. Twenty-five years of social and environmental accounting research: Is there a silver jubilee to celebrate? Accounting, Auditing and Accountability Journal 10 (4), 481– 531. Mizruchi, M.S., 1992. The Structure of Corporate Political Action: Interfirm Relationships and Their Consequences. Harvard University Press, Cambridge, MA. Myers, S.C., 1977. Determinants of corporate borrowing. Journal of Financial Economics 5, 147– 175. Neimark, M.D., 1992. The Hidden Dimensions of Annual Reports. Paul Chapman, London. Neter, J., Wasserman, W., Kutner, M., 1983. Applied Linear Regression Models. Richard D. Irwin, Illinois. Neu, D., Warsame, H., Pedwell, K., 1998. Managing public impressions: Environmental disclosures in annual reports. Accounting, Organisations and Society 23 (3), 265–282. Newson, M., Deegan, C., 2002. An exploration of the association between global expectations and corporate social disclosure practices in the Asia–Pacific region. The International Journal of Accounting 37 (2), 183–213. Norusis, M.J., 1995. SPSS 6.1 Guide to Data Analysis. Prentice Hall, NJ. OÕDonovan, G., 2002. Environmental disclosures in the annual report: Extending the applicability and predictive power of legitimacy theory. Accounting, Auditing and Accountability Journal 15 (3), 344–371. OÕDwyer, B., 2002. Managerial perceptions of corporate social disclosure: An Irish story. Accounting, Auditing and Accountability Journal 15 (3), 406–436. Oliver, C., 1991. Strategic responses to institutional processes. Academy of Management Review 15, 145–179. Patten, D.M., 1991. Exposure, legitimacy, and social disclosure. Journal of Accounting and Public Policy 10 (4), 297–308. Patten, D.M., 1992. Intra-industry environmental disclosures in response to the Alaskan oil spill: A note on legitimacy theory. Accounting, Organisations and Society 17 (5), 471–475. Perera, M.H.B., 1994. Culture and international accounting: Some thoughts on research issues and prospects. Advances in International Accounting 7. Pettigrew, A.M., 1979. On studying organisational cultures. Administrative Science Quarterly 24, 570–581. Pfeffer, J., Salancik, G.R., 1978. The External Control of Organisations: A Resource Dependence Perspective. Harper and Row.

R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 429 Preston, A., Wright, C., Young, J., 1996. Imag(in)ing annual reports. Accounting, Organizations and Society 21 (1), 113–137. Roberts, R.W., 1992. Determinants of corporate responsibility disclosure. Accounting, Organisations and Society 17 (6), 595–612. Schipper, K., 1981. Discussion of voluntary corporate disclosure: the case of interim reporting. Journal of Accounting Research (Supp) 19, 85–88. Seetharaman, A., 1993. Changing role of financial management in Vision 2020. The Malaysian Accountant, 20–25. Selznick, P., 1992. The Moral Commonwealth: Social Theory and the Promise of Community. University of California Press, Berkeley. Sendut, H., 1991. Managing in a multicultural society: The Malaysian experience. Malaysian Management Review 26 (1), 61–69. Sethi, S.P., 1979. A conceptual framework for environmental analysis of social issues and evaluation of business response patterns. Academy of Management Review 4 (1), 63–74. Specter, C.N., Solomon, J.S., 1990. The human resource factor in Chinese management and reform: Comparing the attitudes and motivations of future managers in Shanghai, China; Baltimore, Maryland; and Miami, Florida. International Studies of Management and Organisations 20, 69–83. Suchman, M.C., 1995. Managing legitimacy: Strategic and institutional approaches. Academy of Management Review 20 (3), 571–610. Sutton, B., 1993. The Legitimate Corporation. Blackwell, England. Teoh, H.Y., Thong, G., 1984. Another look at corporate social responsibility and reporting: An empirical study in a developing country. Accounting, Organizations and Society 9 (2), 189–206. Tilt, C.A., 1994. The influence of external pressure groups on corporate social disclosure: Some empirical evidence. Accounting, Auditing and Accountability Journal 7 (4), 56–71. Tricker, R.I., 1984. Corporate Governance—Practices, Procedures and Powers in British Companies and Their Boards of Directors. Gower Publishing Company Ltd., England. Trotman, K.T., Bradley, G.W., 1981. Association between social responsibility disclosure and characteristics of companies. Accounting, Organizations and Society 6 (4), 355–362. Tsang, W.K., 1998. A longitudinal study of corporate social reporting in Singapore: The case of the banking, food and beverages and hotel industries. Accounting, Auditing and Accountability Journal 11 (5), 624–635. Ullman, A.E., 1985. Data in search of a theory: A critical examination of the relationships among social performance, social disclosure and economic performance of US firms. Academy of Management Review 10 (3), 540–557. Useem, M., 1984. The Inner Circle. Oxford University Press, NY. van Fossen, A., 1998. Race, ethnicity and language. In: Maidment, R., Mackerras, C. (Eds.), Culture and Society in the Asia–Pacific, Routledge. Wallace, R.S.O., Mellor, C.J., 1988. Nonresponse bias in mail accounting surveys: A pedagogical note. British Accounting Review 20, 131–139. Wallace, R.S.O., Naser, K., Mora, A., 1994. The relationship between the comprehensiveness of companies annual reports and firms characteristics in Spain. Accounting and Business Research 25, 41–53. Weber, R.P., 1988. Basic content analysis. Sage University Paper Series on Quantitative applications in the social sciences, Series No. 07-049, Sage, Beverly Hills, CA and London. Wiseman, J., 1982. An evaluation of environmental disclosures made in corporate annual reports. Accounting, Organizations and Society 7 (1), 53–63. Woodward, D.G., 1998. Specification of a content-based approach for use in corporate social reporting analysis. Southampton Institute working paper. Worthy, J.C., Neuschel, R.P., 1983. Emerging Issues in Corporate Governance. Chicago Northwestern University Press.

430 R.M. Haniffa, T.E. Cooke / Journal of Accounting and Public Policy 24 (2005) 391–430 Zahra, S.A, Stanton, W.W., 1988. The implications of Board of DirectorsÕ composition on corporate strategy and performance. International Journal of Management 5 (2), 229–236. Zajac, E.J., Westpal, D.J., 1996. Director reputation, CEO-Board Power, and the Dynamics of Board Interlocks. Academy of Management Journal 39, 64–90. Zeghal, D., Ahmed, S.A., 1990. Comparison of social responsibility information disclosure media used by Canadian firms. Accounting, Auditing and Accountability Journal 3 (1), 38–53.