In an era when the world is facing numerous economic, social and environmental crises, many questions arise about the validity and sustainability of the existing business and development models. The term sustainable development was coined in the paper Our Common Future, by the Brundtland Commission (1987, p. 43) which defines sustainable development “as development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” At its core, the principle of sustainable development aims to find a balance between short-term profitability and long-term sustainability. Sustainable Development is however not possible without the participation of firms as key players of societies. Corporate Social Responsibility (CSR) can be defined very simply as firms’ commitment to a more sustainable development. The World Business Council for Sustainable Development’s (1999, p.6) definition of CSR is “the continuing commitment by businesses to behave ethically and contribute to economic development while improving the quality of life of the workforce and their families as well as of the local community and society at large”. The concept of CSR thus requires companies to take into account social, environmental and economic considerations related to their activities and their interactions with their stakeholders on a voluntary basis (European Commission, 2001).
On the other hand, another phenomenon known as ‘globalisation’ has also been revolutionising the way organisations operate since a few decades. The removal of trade barriers and free flow of capital and resources across national boundaries has increasingly led to the emergence and growth of multinational enterprises (MNEs). Globalisation has also played a significant role as a catalyst for corporate social responsibility as the majority of MNEs have headquarters in developed economies, where organisations are pressured by government regulation, mass media, and citizen groups such as NGOs to be socially responsible. In addition, business operations have become much more transparent with the advancement of modern information and communication technologies. This heightened visibility ensures that unethical corporate actions are increasingly subject to public scrutiny and criticism. MNEs also need to pursue the goal of social responsibility to gain acceptability in the different markets that they operate in, in line with the legitimacy theory. Claims that legitimacy has become one of the most critical issue for companies in the twenty-first century (Marais, 2012) are now widely accepted.
Legitimacy theory is based on the idea that a “social contract” exists between business and society. Society is considered to allow companies to exist and have rights, and in return expecting them to fulfil its expectations about how their operations should be conducted. Therefore, in order to survive a company must ensure that the activities it undertakes actually are or are perceived as being in accordance with the values and norms of society. When society’s expectations are not fulfilled, that is, a company’s actual or perceived behaviour is not in accordance with social values and norms, a breach of contract exists and a legitimacy gap may develop. Recently, legitimacy and the social contract have been discussed under the construct of the “social license to operate” (Demuijnck and Fasterling, 2016).
Companies are thus expected not only to have activities which are aligned with societal values but also to communicate about such activities. Legitimacy theory studies find industry affiliation to be related to social responsibility disclosures; with certain industries having high visibility or a potentially more important environmental impact in general being more involved in the disclosure of social responsibility information (Patten and Crampton, 2004).
There has been growing interest for CSR of MNEs as ‘good corporate citizens’ (Kolk and Van Tulder, 2006; Fortanier and Kolk, 2007) particularly in the context of emerging economies (Vives, 2006) as well as developing countries (Luken and Stares, 2005), though the African continent remains much less represented than other regions (Kolk & Lenfant, 2013).
Research in the developing countries has however generally found that the extent of CSR disclosures in annual reports is lower than in the developed countries. This study thus investigates the level and the reasons for CSR disclosure of MNEs in the hospitality sector in Mauritius through a review of their annual reports using a content analysis procedure. The aim is to explore the possibility of using the legitimacy theory as a plausible explanation for CSR reporting practices by multinational hotel groups in the context of a developing country.
Mauritius is a small island state in the Indian Ocean and is one of Africa’s success stories. The choice of the Mauritian tourism industry is motivated by several reasons. An important contributor to the ‘Mauritian miracle’ has been the expansion of the luxury tourism sector. The country is also actively trying to attract foreign investment in terms of FDI and multinational enterprises (MNEs). Furthermore, the country is one of the first in the world to have made it mandatory for profit making entities to devote 2% of their book profits to CSR activities since 2009 with a view to promote a more sustainable development. With the rapid development of tourism industry, issues such as over-development of the coastal region, increasing pollution and disregard for natural resources are becoming important concerns. The need for players in the tourism industry to adopt a socially and environmentally responsible behaviour is increasingly turning out to be important.
Findings from this study could thus shed more light on the motivations of multinationals hotel groups to engage in and report certain types CSR initiatives and provide a basis for the development of public policy concerning CSR reporting for small island states and developing countries. The present study therefore contributes to the literature from the perspective of an open small island state, largely dependent on tourism as well as from an African and developing country’s perspective.
This paper first provides a brief overview of the theoretical and empirical literature surrounding CSR and CSR reporting with particular emphasis on MNEs and the hospitality industry. The research methodology is then described. Next, the research findings are presented and discussed. The study concludes with recommendations as well as limitations of the study and directions for future research.
Corporate social responsibility is known by various appellations such as social responsibility (SR), corporate citizenship and corporate sustainability especially in the tourism industry. CSR has been defined in numerous and even contradictory ways since CSR is a multidimensional concept and there is a lack of consensus over what social responsibility entails. One of the more cited definitions of CSR in the literature by Carroll (1979) asserts that business organisations that practice social responsibility fulfil “economic, legal, ethical, and discretionary (philanthropic) expectations that society has of organisations at a given point in time” (Carroll, 1979, p. 499).
Davis (2005) argues that CSR represents a strategic opportunities as well as obligations. Effective strategic management of CSR can reduce risks (Husted, 2005); and bring significant benefits to the firm (McWilliams and Siegel, 2001). These benefits go beyond mere reputation-building to the development of valuable organisational capabilities (Sharma and Vredenburg, 1998) as a source of differentiating innovation (Kanter, 1999) and provide a competitive advantage (Porter and Kramer, 2002); while the failure to manage CSR strategically can have serious economic consequences for the firm.
Although, there are different theoretical approaches to social responsibility including instrumental theories, political theories, integrative theories and ethical theories (Garriga and Mele´, 2004), this study focuses on the legitimacy theory as a plausible explanation for CSR reporting practices.
CSR and legitimacy theory
The legitimacy theory is based on the fact that community support is vital for the survival, growth and image of companies. However, in order to acquire this support, corporations constantly try to ensure that they operate within the bounds and norms of society (Deegan, 2002) and must disclose specific information to convince society that their activities are legitimate and beneficial (Alawi and Rahman, 2011). In this context, Bowman and Haire (1976, p. 13) defined corporate social responsibility (CSR) broadly as “including the concern for the impact of all of the corporation’s activities on the total welfare of society.” The legitimacy theory indicates that when there is a discrepancy between corporate actions and societal expectations, management employs such disclosure media as annual reports in order to help to alleviate community concerns or, more accurately, what they perceive to be community concerns. The inclusion of CSR information in these reports is intended to alleviate public concern and show that the corporation is meeting community and societal expectations (Deegan et al., 2002).
The legitimacy theory is closely linked to the stakeholder theory which is also commonly used in the CSR literature. Stakeholders consist of groups of people or individuals who are involved in some capacity with the organisation. The concept of stakeholder specifies the firm’s societal responsibilities by defining specific groups or persons that the business should consider in its CSR orientation (Carroll, 1979).
In terms of legitimacy theory, companies in some industries are socially more visible and are more exposed to the public scrutiny. They are therefore considered to face greater social and political pressure to act in a more socially desirable manner and to provide information in certain areas of social responsibility Legitimacy theory studies suggest that companies in industries with a high visibility are expected to exhibit greater concern to improve the corporate image as this is susceptible of influencing sales and may be considered more likely to make social responsibility disclosure. Companies in industries with larger potential environmental impact are more likely to provide environmental information and companies in industries with high visibility among final consumers (such as banks) are more likely to consider important issues of community involvement and disclose information related to such involvement (Clarke and Gibson-Sweet, 1999, p. 6).
In terms of the multinational firm, some scholars posit that MNEs should respond to pressures for global integration and local responsiveness with respect to CSR issues similarly as their organisational strategies respond to pressures of integration and responsiveness in their product markets (Grant and Gnyawali, 1996). There is however no explicit theory that distinguishes between global and local CSR except for some discussions of the issue in the literature (Donaldson and Dunfee, 1994). In some cases, CSR responsibilities and stakeholder demands require MNEs to respond to both global issues and local issues; the underlying argument, being that diverse stakeholders and conflicting value systems require complex CSR strategy responses (Logsdon and Wood, 2005). However, Logsdon and Wood (2005) claim that global MNEs often fail to respond effectively to issues of importance in their host countries.
The need for CSR is actually more important in the developing countries than in developed countries because development projects and other social infrastructure are lacking in most of these countries and they may not be provided by the government (Amadi & Abdullah, 2011). Under these circumstances, MNEs tend to come under increasing requirements and expectations to bridge those gaps (Baughn et al., 2007). For instance, Lodge and Wilson (2006) argue that the MNEs involvement in CSR is vital to global poverty reduction, especially in the developing countries, because poverty reduction requires systemic change and MNEs can be proficient drivers of change in the countries they are found in.
It has however been argued that the CSR agenda is largely shaped by actors from developed countries, with a clear lack of participation/integration of developing country actors (Blowfield, 2003). In their analysis of CSR in Argentina, Newell and Muro (2006, p. 64) conclude that: “(Multinational Corporations) MNCs are the key drivers and respondents (of CSR). Often responsive to their home countries and receptive to shareholder and activist pressures overseas, their strategies reflect European or North American priorities”. Research also indicates that CSR-related perceptions tend to be context-bound, reflecting generic differences in the way CSR is conceptualised and prioritised in different milieus. Blowfield and Frynas (2005) refer to a survey carried out by the World Business Council for Sustainable Development WBCSD), asking respondents from different countries what CSR meant to them. It turned out that there were considerable differences. For example, Thai and Ghanaian respondents stressed environmental issues and community empowerment respectively.
Companies’ CSR practices and related perceptions regarding their roles and responsibilities in society to a certain extent reflect respective expectations of the contexts in which they operate (DiMaggio and Powell, 1983; Hahn, 2005). However, not all stakeholders of a company can be expected to impact and shape a company’s actions in an equal manner. Instead, the company’s primary stakeholders, i.e. those stakeholders “without whose continuing participation the corporation cannot survive as a going concern” (Clarkson, 1995, p. 106), are likely to be most influential.
Since the creation of the World Business Council for Sustainable Development (WBCSD) at the Rio Earth Summit in 1992, the business community has acknowledged its responsibility to contribute to the discussion concerning economic, environmental and social issues through systematic public reporting on their environmental and social performance, together with economic performance (ISO 2004). Effective communication with stakeholders towards economic prosperity, environmental quality and social justice has been recognised as a defining characteristic of corporate responsibility in the twenty-first century (Wheeler and Elkington, 2001). The UK even passed the operating and financial review (OFR) regulation, which requires companies to generate reports of their social and environmental activities. The last two decades thus witnessed a move from core financial reporting to the adoption of triple bottom line. The triple bottom line principle entails reporting of both economic performance of company and most importantly reporting on the environmental and social performance i.e., reporting of CSR activities.
Social responsibility disclosure has been broadly defined as the “process of communicating the social and environmental effects of organisations’ economic actions to particular interest groups within society and to society at large” (Gray et al., 1996, p.3). Thus, it seeks to reflect several social and environmental aspects upon which companies’ activities may have an impact: employee related issues, community involvement, environmental concerns, and other ethical issues. Social responsibility disclosure therefore refers to the disclosure of information about companies’ interactions with society.
A range of corporate guidelines such as the UN Global Compact principles, the Indicators of the Global Reporting Initiative (GRI) or Article 30 of Agenda 21 prescribe comprehensive lists of corporate roles and responsibilities covering a wide economic, social and environmental aspects. The GRI Guidelines is one of the most commonly used frameworks around the world. These guidelines can be used by corporations of any size and any type for reporting on environmental, social and economic performance (GRI, 2014). The UN Global Compact is a multi-stakeholder initiative with the aim to promote corporate citizenship. It has become the largest and arguably most visible of its kind, currently comprising more than 5000 corporate participants and around 1000 non-corporate stakeholders. Nevertheless, despite this very broad conceptualisation of corporate responsibility towards society, actual corporate practice tends to focus on a narrower set of issues when compared to these normative targets (Barkemeyer, 2009).
A number of studies have attempted to empirically test the link between CSR disclosure and the public concern arising from corporate behaviour that is inconsistent with community expectations, as posited by legitimacy theory. Companies disclose social information using several medium such as annual reports, websites, newspapers and promotional material (Hamid and Atan, 2011). Guthrie and Parker (1989) were among the first to make such an attempt as a test of legitimacy theory, although, as their results failed to show an empirical link between such public concern and CSR disclosures. On the other hand, Deegan and Rankin (1996) found that corporations increase their reporting of favourable environmental information surrounding periods of increased media attention toward environmental issues, concluding that this finding is consistent with a legitimating motive. Various authors such as Gray et al. (1995) and Branco and Rodrigues (2006) have also used legitimacy theory to explain social disclosure practices. Overall, the literature provides somewhat mixed results.
Some scholars, argue that these reporting are primarily carried out for public image building purposes. A key concern is therefore whether firms’ reporting practices reflect substantive or symbolic behaviour, the latter often referred to as “green washing” (Delmas and Burbano, 2011). Perez-Batres et al. (2012) examine the question of what determines a firm’s strategic choice toward more symbolic or substantive CSR action. They empirically investigate two voluntary self-regulatory codes: the United Nations Global Compact and the Global Reporting Initiative (GRI). First, the empirical results show that not all stakeholders are equally relevant when firms decide to engage in self-regulatory codes. Instead, it is important to consider the degree of power of individual stakeholder groups. Second, the authors show that high stakeholder scrutiny has a positive effect on firm decisions to engage substantively, not only symbolically. Substantive involvement reflects true commitment to sustainability whereas symbolic engagement means the firm’s CSR actions are decoupled from actual implementation within the organisation. Third, the authors find even though a firm has slack resources, it still may consider the symbolic option. Fourth, the results suggest that firms from perceived “dirtier” industries are more likely to join substantive self-regulatory codes than firms from perceived “cleaner” industries.
CSR reporting has become a global trend. Covering 4100 firms across 41 countries, the 2013 KPMG Survey of Corporate Responsibility Reporting suggested that the Americas has become the leading CSR reporting region (KPMG International, 2013). To report CSR practices, companies use different terms, such as CSR Report, Corporate Responsibility (CR) Report or Sustainability Report. As of year 2013, 76% of companies from the Americas, 73% from Europe, and 71% from Asia Pacific reported their CSR activities. Besides, 93% of the largest 250 corporations in the world reported their CSR activities (KPMG International, 2013). According to the Ernst & Young Value of Sustainability Reporting, 39% of U.S. companies reported their CSR activities compared to 61% of companies from the rest of the world (EY, 2013).
Research in developing countries has generally found that the extent of CSR disclosures in annual reports is lower than in the developed countries. Savage (1994) examined the CSR disclosure practices of 115 companies in South Africa. He found that as many as 50% of the companies had some CSR disclosures, with human resource disclosures being the most popular. This is closely followed by disclosures on community involvement and environmental performance. The extent of disclosure, however, is low. Imam (2000) and Belal (2001) surveyed CSR disclosure practices in Bangladesh. Imam found that the level of such disclosures was very poor and inadequate and emphasised ‘good news’. The findings appear to mirror the results of prior studies done in other developing countries (Savage, 1994). Abu Baker and Naser (2000) utilised content analysis to examine the annual reports of 143 listed Jordanian companies. They examined the content-category themes, methods and location of CSR disclosures within annual reports. The predominant themes for CSR disclosure amongst companies were human resources and community involvement.
Legislated sustainability reporting
United Nations Environment Programme (UNEP), KPMG Advisory N.V., Global Reporting Initiative (GRI) and Unit for Corporate Governance in Africa (2010) carried out a survey on voluntary or obligatory sustainability reporting approaches in thirty countries, and the following has been identified: 142 national laws and / or standards referring to sustainable development reporting guidelines or obligations; 16 international standards referring to sustainable development reporting and 14 standards for assurance engagements and certification of disclosed information.
However, it should be noted that there is still controversy over the mandatory or voluntary aspect of CSR Reporting. According to Bowen (1953, p. 139), “it is perhaps not necessary, or even desirable, that the obligations of businessmen be codified fully in detail. One of the advantages of informal moral rules is that they are more flexible and adaptable to changing circumstances”. In other words, if CSR reporting will be considered as mandatory, the standardisation will exclude a key CSR driver which is maximising the competitiveness of companies through raised brand value.
CSR in the hospitality sector
Agenda 21 put forward by the World Travel and Tourism Council (WTTC), the World Tourism Organization (WTO), and the Earth Council set international guidelines relative to sustainable tourism. The WTO went on to establish the Global Code of Ethics for Tourism (GCET), which is a “comprehensive set of ten principles whose purpose is to guide stakeholders in tourism development” (World Tourism Organization, 2005). These ten principles are not legally binding but serve as guidelines to central and local governments, local communities, tourism operators, and tourists concerning preservation of the environment.
Kalisch (2002) argues that the trend in the hospitality and tourism industry has been to focus on environmental issues, use of new technology and efficient use of energy. Initiatives such as that of “Green Hotels” have also gained popularity among the lodging operators. “Green Hotels” focuses on programs that are designed to save water and energy and reduce solid wastes (Green Hotels Association, 2005). Program components such as towel hanging and sheet-changing cards have achieved positive responses by hotel management staff as well as guests for over a decade now (Green Hotels Association, 2005).
Several research articles (Jenkins and Yakovelva, 2005; Alnajjar, 2000; Esrock and Leichty, 1998) identified five broad categories in terms of the activities that hotel companies report. The categories are community, environment, marketplace, vision and values and workforce. Medrado and Jackson (2016) examined non- financial disclosures by hospitality and tourism firms on corporate social responsibility and sustainability dimensions. The study utilised content analysis and found that in general, firms utilised the Global Reporting Initiative as the standard guideline for reporting. Findings also suggested that there is a vast difference in the types of information reported by firms in the industry’s various sectors. Lodging firms disclosed the most information on the CSR/sustainability dimensions than firms operating in the food and beverage and cruise line sectors. These findings suggested that CSR/sustainability reporting in the hospitality and tourism industry is in its infancy. Overall, the most frequently disclosed information related to performance on indicators associated with water usage, energy conservation, and waste generation. Community involvement activities were also frequently disclosed by sampled firms.
CSR reporting framework in Mauritius and empirical evidence
The Code of best practice on corporate governance in Mauritius was published in October 2003. Its aim is to provide indicative guidelines for Mauritian companies to improve on their disclosure practices and recognise the importance of their practices on the community in which they operate. Compliance with the code was required as from the reporting year ending 30 June 2005. Section 1 of The Code of Corporate governance for which deals with compliance and enforcement states that the Code applies to the business enterprises listed on the Stock Exchange of Mauritius, banks and non-banking financial institutions, large public companies, state owned enterprises including statutory corporations and parastatal bodies as well as large private companies.
Section 7 of the Code deals with “integrated sustainability reporting” and requires companies above to report (within the annual report) to stakeholders on issues linked to: Environment, Ethics, Health and Safety and Social issues.
The sampled companies in this study form of the category of large public companies. Although the code of practice for disclosing CSR information exists, the prescriptions are not very stringent and thus companies have full discretion as to how to report CSR disclosure in their annual reports. Therefore, this lack of standards could very well give rise to CSR disclosures being very much public-relations oriented.
Furthermore, the Income Tax Act (1995) amended in 2009 requires all profitable companies to contribute 2% of their book profit towards CSR activities. These activities were restricted to the following areas: socio-economic development, health, education and training, leisure and sports, environment, catastrophic interventions support (MEF, 2011). However, as from 2015 (during the course of the project), the activities considered as CSR have been opened up. However, this study was initiated before the change in the legislation during the course of 2015 and uses 2014 annual reports. Therefore, how companies decide to spend the CSR levy could constitute an area for future research.
Despite the attention given to CSR since 2009 in Mauritius in the wake of the introduction of the CSR legislation, academic research on the topic is still very scanty. Ragodoo et al. (2009) and Gokulsing (2011) have mainly concentrated on the contribution of CSR towards poverty alleviation. Kassean and Juwaheer (2010) surveyed the practice of CSR practices in private and public sector companies by means of face to face interviews. The only studies related to the disclosure of CSR was carried by Mahadeo et al. (2011) which used content analysis to measure the level of CSR disclosures of listed companies to investigate the impact of profitability, size, leverage and industry affiliation on CSR reporting; Ramdhony and Hanuman (2012) investigated the motivations for CSR reporting from the perspective of accounting professionals by means of questionnaires. More recently, Ramdhony (2015) examines CSR disclosures by commercial banks operating in Mauritius. Annual reports for the year 2011 are analysed using content analysis. He finds that “Human resources” is a prominent theme of CSR disclosure among Banks in Mauritius. However, in Mauritius mandatory, CSR guidelines do not allow companies to spend their CSR levy on employees. Ramdhony (2015) argues that this can be interpreted as an attempt to reassure employees that they are important stakeholders.
Research design and methodology
Many studies, Alawi and Rahman (2011), Rizk et al. (2008) and Ahmad et al. (2003), have focused on CSR disclosures in annual reports. The annual report is considered to be the main corporate communication tool. This statement is supported by Gray et al., (1995), who argue that annual report is probably the most important document in terms of the way the organisation constructs its own social imagery to all stakeholders. Annual reports are considered to be more credible (Neu et al. 1998) in the sense that, the annual report has full narrative materials and they are verified by auditors. Although CSR information can be disseminated in a number of ways such as the Internet and media, this study considers only disclosures made in the corporate annual reports. This is motivated by the fact that the annual report is a statutory document and can be considered as being more reliable than other means of disclosure.
In particular, the objectives of this study are as follows:
To examine the CSR category disclosure practices of a sample of multinational hotel groups operating the 4 and 5 star categories in Mauritius;
To examine the methods used for CSR disclosures, i.e., monetary, non-monetary-quantitative or declarative;
To examine the news-type of CSR disclosures, i.e., ‘good news’, ‘bad news’ or neutral
To assess if legitimacy theory can be used to explain the CSR disclosures.
The study’s target population comprises of all the groups owning and managing 4 and 5 star hotels in Mauritius. 10 such hotel groups exist. The list of hotel groups was catalogued from the list of registered hotels from the Association of Hoteliers and Restaurants (ARHIM) in Mauritius. It should be noted that there was no official star rating of hotel in Mauritius until 2016. The star ratings used in this study are those that are self-reported by the hotel groups themselves. Of those, 3 are private groups whose headquarters are not found in Mauritius and their annual reports could not be acquired or accessed. The remaining are 7 publicly listed hotel groups but one was eliminated because it operates solely in Mauritius and thus does not meet the criteria of being a multinational. The remaining 6 multinational hotel groups were then divided into outgoing multinationals with headquarters located in Mauritius and having operations in other countries in addition to Mauritius, and incoming multinationals having their headquarters in other countries and having operations in Mauritius. The sample is thus made up of 3 outgoing hotel groups and 3 incoming ones. The three incoming multinationals have headquarters located in Asia (India), the United States, and Europe (France) respectively. Table 1 in Appendix 1 lists the names of hotel groups in the sample as well as their country of origin.
The latest annual report for the year 2014 was downloaded from the groups’ corporate websites. All groups that were analysed for this research project maintained company websites and provided information regarding CSR issues via downloadable PDF format documents of their annual reports. The 2014 annual report for one of the groups was yet not available due to a merger having taken place. Company annual reports for the year 2014 have been used as they were the latest available for the largest dataset of companies as the closure of accounts for the different hotel groups originating from different countries do not coincide.
Content analysis procedure
This study uses a content analysis as a technique to identify and describe patterns in the annual reports of the identified hotel groups. Neuman and Kreuger (2003, p. 219) defines content analysis as “a technique for gathering and analysing the content of text. The content refers to words, meanings, pictures, symbols, ideas, themes, or any message that can be communicated.” Content analysis has been widely employed in previous studies on corporate disclosures (Guthrie et al., 2004), particularly the disclosures of CSR information (Deegan and Gordon, 1996; de Villiers and van Staden, 2006).
CSR is made up of many dimensions from the environment to the responsible treatment of employees. This fact means that different companies often label similar socially responsible behaviour in a different manner. For the content analysis process, themes have to be identified to be applied to the data analysed. According to Milne and Adler (1999), the construction of a categorisation scheme is an essential stage in content analysis research. This involves the selection and development of categories into which content units can be classified. The Global Reporting Initiative (GRI) categories have been used in this study. GRI is the one of the most commonly used guideline for sustainability reporting. The objective of GRI is to mainstream disclosure on environmental, social and governance performance. The GRI categories include Economic, Environmental, Labour Practices and Decent Work, Human Rights, Society and Product Responsibility. The economic aspect will not be considered in this study, the objective being an assessment of the social and environmental reporting. Each company has to be coded for the presence of behaviours relating to the different categories. The measurement instrument used in this study is adapted from Hackston and Milne’s (1996) and Nik Ahmad et al.’s (2003) work (see Appendices 1, 2 and 3). It should be noted that tallies within each category represent the frequency of the observed item and is not an assessment of quality of each instance. The measurement instrument contains different categories with four testable dimensions. These are summarised below:
Themes: according To GRI categorisation
Evidence: monetary, non-monetary-quantitative and declarative (Ernst and Ernst, 1978)
News type: good, bad and neutral
Number of sentences
To ensure that the coding process is reliable, keywords of what comprise each category of CSR disclosures were detailed out in a separate sheet. These were drawn from a review of the literature in particular, the works of Hackston and Milne (1996) and Nik Ahmad et al.’s (2003) as well as the Global Reporting Initiative (GRI)‘s category guidelines (Table 2 in Appendix 2). In addition, the same coding rules used by Hackston and Milne (1996) were adopted in the present study (Appendix 3). These helped make the coding process simpler and also reduce much of the subjectivity in deciding what actually constituted CSR disclosures and how to categorise them. The use of these key terms, categories and decision rules improves the reliability of the coding process (Milne and Adler, 1999). Consistent with other studies, the unit of analysis for the present study is the sentence.
Unerman (2000) argues that the method of measurement is one crucial factor in content analysis research. Unerman favours the use of number of sentences to measure the volume of disclosure as he claims that this may be done with less use of subjectivity. Ingram and Frazier (1980), Milne and Adler (1999) and Tsang (1998) support the use of sentences for several reasons. They argue that a sentence is easily identifiable, less subject to inter-coder variations and has been preferred as an appropriate measurement unit in similar studies. Hackston and Milne (1996) also advocate the reliability of the sentence as a unit of analysis. They argue that the use of sentences avoids the problems linked to apportionment of a page and removes the need to standardise the number of words.
The general procedure for performing the content analysis of annual reports is outlined below: the principal investigator and the research assistant performed the content analysis. Prior to the coding process, the research assistant was given a detailed briefing on what constitutes a CSR disclosure, the use of the coding instrument and an explanation of the different categories and types of disclosure. During the briefing session, the principal investigator used several different annual reports that were not part of the actual sample used for the study to highlight different types of disclosures. At the end of the session, the research assistant was then handed a few annual reports to attempt the coding. Upon completion of this preliminary round of coding, the principal investigator reviewed the coding and categorisations performed by the research assistant. The principal investigator addressed specific issues in coding raised by the research assistant. Following this initial round, the research assistant and the principal investigator both were involved in the coding process of the annual reports and discussed any ambiguity of the items of disclosure, which were reviewed and resolved. This was carried out to ensure consistency and reliability in the content analysis procedure.
The research findings are presented below. It is important to highlight that this study analyses the reporting of CSR activities and does not measure or verify those activities that might have carried out but which for some reason were not publicly reported. Esrock and Leichty (1998) assert that there may be little relationship between a reporting of a company’s CSR and their actual CSR performance.