i UNIVERSITY OF GHANA CORPORATE SOCIAL RESPONSIBILITY DISCLOSURES AND ITS CONSEQUENCES ON FIRM VALUE IN A DEVELOPING ECONOMY BY ALHASSAN MUSAH A THESIS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FUFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL IN ACCOUNTING DEGREE JULY, 2015 University of Ghana http://ugspace.ug.edu.gh ii DECLARATION I do hereby declare that this work is the result of my own research and has not been presented by anyone for any academic award in this or any other university. All references used in the work have been fully acknowledged. I bear sole responsibility for any shortcomings …………………………………….. …………………………. ALHASSAN MUSAH DATE (10274372) University of Ghana http://ugspace.ug.edu.gh iii CERTIFICATION We hereby certify that this thesis was supervised in accordance with procedures laid down by the university. ……………………………………………. …………………….. DR. FRANCIS K. ABOAGYE-OTCHERE DATE 29/01/2016 ………………………………………. …………………… DR. WILLIAM COFFIE DATE University of Ghana http://ugspace.ug.edu.gh iv DEDICATION This work is dedicated to my family. University of Ghana http://ugspace.ug.edu.gh v ACKNOWLEDGEMENT I am indebted to my supervisors; Dr. F.K. Aboagye-Otchere and Dr. W. Coffie for their timeless dedication during the supervision of this work. I am also very grateful to all faculty members for their comments and contributions during the seminar presentations and my colleague students who also assisted me during data collection as well as provided useful comments. Finally, I will like to express my profound gratitude to Mr. Akomeah Evans and Mr. Erasmus Gakpetor for their support and advice. I am very grateful to you all. University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENT Contents DECLARATION ....................................................................................................................... ii CERTIFICATION ................................................................................................................... iii DEDICATION .......................................................................................................................... iv ACKNOWLEDGEMENT ......................................................................................................... v TABLE OF CONTENT ............................................................................................................ vi LIST OF TABLES ..................................................................................................................... x LIST OF ABBREVIATIONS .................................................................................................. xii ABSTRACT ........................................................................................................................... xiii CHAPTER ONE INTRODUCTION 1.1Background of the study ....................................................................................................... 1 1.2 Motivations for the study ..................................................................................................... 4 1.3 Objective of the study .......................................................................................................... 7 1.4 Significance of the study ...................................................................................................... 7 1.5 Scope and limitations of the study ....................................................................................... 9 1.6 Chapter disposition .............................................................................................................. 9 CHAPTER TWO THEORETICAL LITERATURE REVIEW 2.0 Introduction ........................................................................................................................ 11 2.1 Legitimacy theory .............................................................................................................. 11 2.2 Framework of CSRD Determinants ................................................................................... 15 CHAPTER THREE EMPIRICAL LITERATURE REVIEW 3.0 Introduction ........................................................................................................................ 18 University of Ghana http://ugspace.ug.edu.gh vii 3.1 The nature of CSRD........................................................................................................... 19 3.2 CSRD and its categories .................................................................................................... 23 3.3 The quantity of CSRD ....................................................................................................... 25 3.4The quality of CSRD in annual reports............................................................................... 28 3.5 Determinants of CSRD ...................................................................................................... 30 3.6 Corporate characteristics .................................................................................................... 32 3.7 Degree of multinational activity ........................................................................................ 39 3.8 Corporate governance variables ......................................................................................... 40 3.9 Consequence of CSRD ...................................................................................................... 46 3.10 Hypothesis development .................................................................................................. 53 2.10.1 Corporate size ............................................................................................................... 53 3.10.2 Industry type ................................................................................................................. 54 3.10.5 Degree of multinational activity ................................................................................... 55 3.10.6 Board size...................................................................................................................... 56 3.10.7 Board Composition ....................................................................................................... 57 3.10.8 Presence of Corporate social committee ....................................................................... 58 3.10.9 Corporate ownership diffusion ..................................................................................... 59 3.10.10 Consequence of CSRD on market value ..................................................................... 60 CHAPTER FOUR RESEARCH METHODOLOGY 4.0 Introduction ........................................................................................................................ 61 4.1 Research philosophy and design ........................................................................................ 61 4.2 Data and source .................................................................................................................. 63 4.3 Sample and selection criteria ............................................................................................. 65 4.4 Research approach and execution strategy ........................................................................ 66 4.5 Empirical model estimation ............................................................................................... 68 University of Ghana http://ugspace.ug.edu.gh viii 4.6 Methodological issues ........................................................................................................ 69 4.6.1 Normality of residuals..................................................................................................... 70 4.6.2 Heteroskedasticity ........................................................................................................... 70 4.6. 3 Multicollinearity ............................................................................................................ 71 CHAPTER FIVE ANALYSIS AND DISCUSSION OF FINDINGS 5.0 Introduction ........................................................................................................................ 73 5.1 Descriptive statistics of CSRD........................................................................................... 73 5.2 Analysis of CSRD in annual reports .................................................................................. 75 5.3. The incidence of CSRD categories ................................................................................... 78 5.4 Disclosure location in Annual reports................................................................................ 80 5.5 CSRD in different sectors .................................................................................................. 82 5.6 CSRD categories in different economic sectors ................................................................ 85 5.7 Relationship between key disclosure variables.................................................................. 88 5.8 Descriptive statistics .......................................................................................................... 89 5.8 Correlation results for CSRD variables ............................................................................. 91 5.9 Panel data analysis ............................................................................................................. 94 5.9.1 The Quantity of CSRD disclosure .................................................................................. 95 5.9.4 The quality of CSRD disclosure ................................................................................... 103 5.9.5 Discussion of findings on determinants of the quality of CSRD .................................. 104 5.10 consequences of CSRD .................................................................................................. 110 CHAPTER SIX SUMMARY, CONCLUSION AND RECOMMENDATIONS 6.0 Introduction ...................................................................................................................... 115 6.1 Summary of findings........................................................................................................ 116 6.2 Conclusions ...................................................................................................................... 118 University of Ghana http://ugspace.ug.edu.gh ix 6.3 Recommendations ............................................................................................................ 119 References .............................................................................................................................. 121 Appendix: CSRD disclosure index ........................................................................................ 132 University of Ghana http://ugspace.ug.edu.gh x LIST OF TABLES Table 4.6.4: VIF of independent variables 71 Table 4.6.4: Results of the Hausman test 72 Table 5.1 Descriptive statistics on CSD 73 Table 5.2 Categories of CSD in annual report 76 Table 5.3 Incidence of CSD categories 78 Table 5.6 The incidence of disclosure categories in different economic sector 86 Table 5.8: Descriptive statistics of CSD variables 89 Table 5.8.1 Correlation results for quantity of CSD determinants 91 Table 5.8.2 Correlation results for quality of CSD determinants 92 Table 5.9.1 Panel regression results for quantity of CSD 95 Table 5.9.2 The impact of different variables on the quantity of CSD 97 Table 5.9.4 Panel regression results for quality of CSD 103 Table 5.10.5 The impact of different variables on the quality of CSD 104 Table 5.10.6 Correlation results for the consequences of CSD 110 Table 5.10.2 Panel regression results for the consequence of CSD 111 University of Ghana http://ugspace.ug.edu.gh xi LIST OF FIGURES Figure 3.1: Conceptual framework of determinants of CSD 60 Figure 5.1: CSD over the study period 75 Figure 5.3: Disclosing companies CSD for each category 80 Figure 5.4: Disclosure location in the annual report 81 Figure 5.5: analysis of CSD in different sectors 84 Figure 5.6: Modified determinants framework of CSD 109 University of Ghana http://ugspace.ug.edu.gh xii LIST OF ABBREVIATIONS CSRD Corporate Social Responsibility Disclosures SEC-G Securities and Exchange Commission- Ghana GSE Ghana Stock Exchange UK United Kingdom USA United States of America University of Ghana http://ugspace.ug.edu.gh xiii ABSTRACT The impact of corporate organizations activities on society is a growing global concern. As a result, the expectations of stakeholders on the role of businesses in society have increased over the last decade. To this end, greater attention has been given to CSR activities and its subsequent disclosures (Corporate Social Responsibility Disclosure). The disclosure of corporate social responsibility activities of corporations provides information to the public regarding corporate activities that relate to society, such as about effort to reducing environmental impact, improving waste management, compliance with environmental regulations, and efforts to protect employees. The study presented a framework based on legitimacy theory that explained the determinants of CSD in terms of both quantity and quality of CSRD of listed firms on the Ghana stock exchange and also examined the consequence of CSRD on firm value. The study sampled 33 firms listed on the Ghana stock exchange over a six year period. The study adapted a disclosure index by Hackstone & Milne (1996) as adapted by Deegan et al. (2002) and Hassan (2014) and used content analysis to extract social information from the firm’s annual report. The results of the panel regression showed that, quantity of CSRD, and to lesser extent quality of CSRD, are determined by the following variables: corporate size, type of activity, foreign ownership, degree of multinational activity, proportion of non-executive directors, the presence of social responsibility committee on the board and ownership diffusion. Profitability was significantly related to the quality of CSRD but not the quantity of CSRD. With regards to the consequence of CSRD, the empirical results show that CSRD has no significant relationship with firm value. The implication of the results is that, CSRD has no direct financial benefit to firms listed on the Ghana Stock Exchange. University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE INTRODUCTORY CHAPTER 1.1Background of the study The impact of corporate organizations activities on society is a growing global concern. The expectations of various stakeholders such as consumers, employees, investors, and local communities on the role of businesses in society have increased over the last few decades (Hassan, 2014). The concept of corporate Social Responsibility (CSR) is a well- known practices among companies the world over. CSR has over the years become the best alternative of conceiving and doing business that encompasses economic, social, and environmental operations for achieving competitive advantage (Akano, Jamiu, Olaniran & Oluwalogbon, 2013). Although the underlying foundations of CSR are not new, the practice of CSR has been growing in importance, significance and acceptance in the last decades within the corporate world as well as in the academic setting (Dobers & Halme, 2009). CSR can be defined as a responsible and ethical performance about economic, social, and environmental aspects that can improve the quality of life of different stakeholders such as employees, suppliers, consumers, government, community, society, etc. (Gomez, 2011). CSR usually describes the relation that exists between companies and society and the interrelationship between economic, social and environmental features (Lina & Gokiene, 2011). According to Morsing and Schultz (2008), organizations that promote social and environmental initiatives have the power to evoke positive reactions between stakeholders. At the same time, CSR also brings beneficial opportunities for companies in the form of stakeholder engagement and corporate image (Du et al. 2010). But doing well by being good University of Ghana http://ugspace.ug.edu.gh 2 is worth it not only for achieving company performance and success, but also for improving the planet and people’s life. The nature of CSR reporting and disclosure however is still in disputes with different scholars given different aspect of social information that should be contained in a CSR report (Lina & Gokiene, 2011). For instance, Carroll (1979) identified four categories of information to be reported as corporate social responsibilities: economic, legal, ethical, and discretionary (or philanthropic). The economic responsibilities are concerned with business’s financial performance in terms of profitability and share value and the provision of goods and services. Legal responsibilities on the other hand are concerned with compliance with societal laws and regulations. Ethical responsibilities relate to following societal norms and codes of conduct, and discretionary responsibilities relate to voluntary involvement and support of wider societal activities. The benefits that a company derives from operating in a socially responsible manner vary from one researcher to another (Daub, 2007). There are many people that seem to agree that corporations have the responsibility to behave responsively especially in the wake of calls for sustainable environments. Despite the fact that CSR as a perspective is growing, there has always been critics. One of the most influential critics is Nobel Prize winner in economics Milton Friedman, who claims that companies’ sole purpose is to maximize profit for their stockholders. Furthermore, he claims that CSR is a waste of the stockholders’ money in his famous statement “the business of business in business” (Friedman, 1962). However, several relatively new research and reports argue that CSR increases financial performance in the long run (Murray et al., 2006; Said et al., 2009; Yeganeh et al. 2013). ). As a result of this, University of Ghana http://ugspace.ug.edu.gh 3 greater attention has been given to CSR activities and its subsequent disclosures (Corporate Social Responsibility Disclosure (CSRD). The disclosure of corporate social responsibility activities of corporations provides information to the public regarding corporate activities that relate to the society, such as about reducing environmental impact, improving waste management, compliance with environmental regulations, and efforts to protect employees (Yeganeh et al. 2013). By the 1980s, a large volume of literature had developed on environmental accounting building on diverse views from social accounting (Hassan, 2010). In the 1980s, literature relating to the disclosure of social and environmental information in the form of what is commonly known as social and ethical accounting and auditing and reporting (SEAAR) was developed (Hassan, 2010). The emergence and increasing interest in social and environmental accounting and disclosure reflects the increasing demand for transparency driven by a number of factors, including:  Societal interest in social and environmental issues has become increasingly mainstreamed.  The development of the CSR field generally and of CSR standards in particular.  Increasing demand for new and better information to meet the information needs of different stakeholders.  Advances in communication technology.  Investor interest in CSR performance as a business issue.  Declines in the levels of trust afforded to some institutions (Hassan, 2010). University of Ghana http://ugspace.ug.edu.gh 4 CSD provides companies an opportunity to use CSR as a powerful marketing tool that can be used to engage with stakeholders and also to enhance firm value and brand image (Nikolaeva & Bicho, 2011; Sweeney & Coughlan, 2008). According to Hoeffler & Keller (2002), when CSR initiatives have been reported and disclosed, it have favourable impact on how stakeholders think, feel and act toward the company concerned and its brand. Whiles CSR reporting have been acknowledged by marketers as relevant and hence gained popularity within the business community, scholars in the area of accounting have shown far less interest in the topic (Chen &Bouvain, 2009; Ziek, 2009). 1.2 Motivations for the study CSRD is a voluntary activity and as such it is important to understand which companies disclose information about social and environmental activities; which companies disclose more information than others and what are the benefits of CSRD for companies. For about four decades now, corporate social responsibility reporting and disclosure by businesses has been the subject of empirical research (Fifka, 2013). These studies had mostly focused on developed and western economies but the last few years have witness some changes to this trend. Even though, there have been some studies on the subject in emerging and developing economies in recent times, Hassan (2014) argues that there is far more research on corporate social responsibility (CSR) in developed countries than there is in developing countries even though the need for CSR is more pronounced in developing countries than developed countries. This is because there exist more gab in social provisions and governance in developing countries than there is in developed economies (Fifka, 2013). In short, International surveys and comparative studies indicate a significant gap between developed and developing countries in corporate social responsibility disclosure (CSRD) University of Ghana http://ugspace.ug.edu.gh 5 practices, as CSRD is considered a western phenomenon (Baughn, Bodie & McIntosh, 2007; Dobers &.Halme, 2009; Fifka, 2013). There have also been some few research on CRS disclosures in some developing economies but have mostly focused on a specific industry (Akano et al., 2013; Pretten & Mashat, 2009; Khan, Halabi & Samy, 2009). For instance, Akano et al. (2013) studied CSR disclosure activities by commercial banks in Nigeria and found that the value of total assets have positive relationship and statistically significant with the level of corporate social responsibility activities disclosure. Khan et al. (2009) conducted similar study on Bangladesh banks but focused on perceptions and not determinants. In Ghana, there have been some recent research in corporate social responsibility activities but these studies have not focused on disclosures (Hinson, Boateng & Madiche, 2010; Ofori and Hinson & Ndziba, 2009; Rahaman, 2000; Rahaman et al. 2004). For instance, Hinson et al.(2010) looked at corporate social responsibility reportage on banks websites in Ghana and found that a bank that have won an award for CSR had the poorest reportage than some of the banks that have not won anything but had some structured form of website reporting on CSR. Rahaman (2000) conducted a small exploratory study on 12 companies in Ghana and found that pressure from international bodies like IMF and government regulations as external factors, and management attitudes and desire to obtain listing on international stock market as internal factors driving CSRD in Ghana. The only known study on determinants of CSR disclosure in Ghana is a thesis by Henry (2007) which sampled 20 companies in Ghana. The study however concentrated on corporate characteristics as determinants ignoring other critical determinants such as the degree of multinational activities and corporate governance. The extant literature generally, supports the argument that corporate governance mechanisms, if managed and implemented appropriately, are positively associated with the levels of University of Ghana http://ugspace.ug.edu.gh 6 corporate disclosure. This is because, the association between a firm’s governance structure and its disclosure policies is based on the premise that well-governed firms use increased disclosure as a means of mitigating agency problems between managers and shareholders (Goodwin et al., 2007). However, a firm’s governance structure and disclosure policies are to some extent influenced by regulation. Regulation attempts to overcome information asymmetry between managers and shareholders and protect each party’s contractual rights. Schwartz and Scott (2003) suggest that regulations smoothens the progress of the efforts of contracting parties to maximize the joint gains (the contractual surplus). Prior research on disclosures show low levels of disclosures by firms in Ghana and the reasons assigned for that was weak regulations (Rahaman, 2004) Also, most studies on determinants of CSRD focus on quantity of CSRD and ignore quality of CSRD which this study seeks to examine. In addition to the above, Rizk (2006) as cited by Hassan (2010) reviewed literature on CSD research over the years and argues that social accountability issues date back more than 30 years and a fundamental problem in the field of CSR research is that there is no standard, universal definition to provide a framework or model for the systematic collection, organisation and analysis of corporate data. The need to create a framework that can explain CSRD in terms of both determinants and consequences remains an important concern. To explain why particular disclosures are made, or in describing how companies should make particular disclosures, reference is made to a particular theoretical perspective. In the area of social and environmental disclosure practices, recent research has tended to rely upon the approaches of legitimacy theory, and to a lesser extent, stakeholder theory, both of which have their roots in political economy University of Ghana http://ugspace.ug.edu.gh 7 theory. In line with previous literature, the study seeks to suggest a framework based on legitimacy theory as the theoretical background. Even though there is recognition of CSR disclosure and reporting as a means to operationalize CSR, whether and how CSR reporting impacts on corporate social and financial performance is still open to debate (Perrini, Russo & Vurro, 2011). Hassan (2010) argues that the consequence of CSRD has received little attention in literature. It is obvious from the forgoing that, there is little study on CSR disclosure is developing countries and Ghana for that matter. Again, there is little studies that have tried to comprehensive examine determinants of CSRD and its consequence on firm performance in one study. There is also little study that has examined determinants of CSRD in terms of both quality and quantity. This study therefore seeks to cure these deficiencies by examining CSRD determinants on both quality and quantity in the context of a developing economy and its consequence on firm value using the legitimacy theory. 1.3 Objective of the study 1. To examine the nature of corporate social responsibility information disclosed by listed firms in Ghana. 2. Examine the determinants of CSR disclosures of these listed firms in Ghana 3. Examine the relationship between CSR disclosures and firm financial value of firms listed on GSE. 1.4 Significance of the study The study adds to CSRD-related literature in various ways: University of Ghana http://ugspace.ug.edu.gh 8 The study developed a theoretical framework that is used to explain CSRD in terms of determinants for both quality and quantity in an integrated framework. Despite this framework being based on previous literature, its presents a different approach to legitimacy theory by presenting a new dimension on how companies respond to social pressure. Within the framework, the study throws more lights on the relationship between CSRD and variables that did not receive attention in previous studies such as the degree of multinational activities, corporate governance practices and corporate market value. The study also contributes to knowledge on corporate social responsibility disclosure because it examines the level of disclosure of four themes under CSRD and their determinants in a unique setting of a developing country in West Africa, Ghana. The study also presents two new variables to CSRD literature: the presence of corporate social responsibility committee on the board and the Tobin’s q. Finally on the contribution to knowledge, the study adds to literature by not examining only the quantity of CSRD as determinants but the quality of CSRD as well. In the area of practice, the findings of the study is relevant to managers of these firms as its gives them an idea as to the potential benefits of being socially responsible and disclosing same information in their annual report. The findings could also be useful to policy makers in the area of improving voluntary disclosures in Ghana and providing a framework for reporting same. University of Ghana http://ugspace.ug.edu.gh 9 1.5 Scope and limitations of the study The population for the study included all companies that were mandated to adopt IFRS by 2008. However, the study was limited in scope to firms listed on the Ghana Stock Exchange because of the non-availability of data from non-listed firms. The study to focused mainly on corporate social responsibility disclosures and not the act of corporate social responsibility and as such only examines those CSR activities that have been disclosed in the annual reports. The study also relied on one theory and not multiple of theories as suggested by current researchers in the area and as such will rely on the legitimacy theory and its underpinning theory to explain the determinants of CSRD and its consequences. On the consequence of CSRD, the study focused on the consequence on firm value and did not include consequence of firm reputation. The study also focused on listed companies on the Ghana stock exchange irrespective of their line of business. On the limitations of the study, like in all accounting studies, the results should be interpreted in the context of the proxies used to measure the different variables, the sample and the time period of the study. The study also looked at CSRD in total and did not break it down into voluntary and mandatory disclosures. 1.6 Chapter disposition The study is organised in six chapters Chapter one of the study began with a background on corporate social responsibility, its reporting, disclosures and significance. It also contains the motivations for the study which University of Ghana http://ugspace.ug.edu.gh 10 highlights the research gaps identified. This is followed by the research objectives and questions and the contributions of the study as well. Chapter two of the study provides theoretical literature in relation to CSRD and focused mainly on the legitimacy theory which is the theory underpinning this study. The chapter concluded by presenting a suggested framework based on legitimacy on the determinants on CSRD. Chapter three is the first section of literature review and focused mainly on literature relating to the nature of CSRD, its definitions and historical background and other related concept like CSR among others. It also includes a review of empirical literature relating to the disclosure media of CSRD and the methods of measuring the quality and quantity of CSRD. The second aspect of the literature review focused mainly of literature about the determinants of CSD and its consequences, the framework that sought to explain CSRD in terms of both determinants and consequences. Chapter four provided detailed methodology of the study including the research strategy, population, sample, variables and their measurement as well as the sources of data. The fifth chapter presented and analysis of results and presentation of findings. The sixth chapter which is the last chapter of this thesis provides a summary of the findings of this research as well as a discussion of the limitations of the research and recommendations for further research. University of Ghana http://ugspace.ug.edu.gh 11 CHAPTER TWO THEORETICAL REVIEW AND CONCEPTUAL FRAMEWORK 2.0 Introduction This chapter presents a review of the theory underpinning this research. According to Hassan (2014), literature on CSRD has relied mostly on legitimacy theory and to some extent stakeholder theory. The two theories to some extent are related as Hooghiemstra (2000) stated that firm’s uses corporate social activities disclosure in the annual report to legitimize company behaviour by providing information intended to influence stakeholders and eventually society’s perceptions about the company. Legitimacy gap arise when there is no congruent between society’s expectations and perception of a corporation’s activities and actions (Akano et al., 2013). This study relies on the legitimacy theory as its theoretical framework underpinning this study. 2.1 Legitimacy theory According to Liu et al. (2006) previous studies uses legitimacy theory to explain incentives for corporate voluntary disclosure for which social responsibility disclosure is an important component. In line with legitimacy theory, CSRD aims to legitimise a company behaviour by providing information intended to influence stakeholders’ and eventually society’s perceptions about the company (Hooghiemstra, 2000).Deegan et al. (2000) posits that the increase in social disclosures is a strategy by corporations to alter the perceptions about the legitimization of an organization in order to achieve the continuing mandate of the society. Branco & Rodriguez (2008) also argue that CSRD is a critical way for firms to communicate with society, to convince the public that they are meeting their University of Ghana http://ugspace.ug.edu.gh 12 social expectations. According to Akano et al. (2013) legitimacy theory agues for companies to carry out their activities within the framework (bounds and norms) of the society they are operating. They further argue that these bounds and norms are not fixed hence the need for organizations to watch out for changes and respond to them appropriately. This means that companies use CSRD as a means of addressing exposure to public policy (Hassan, 2014). According to Wang et al. (2013), legitimacy is fundamental for an organization to survive and exist in the society in the long term, which cannot come from either the making of profit or the mere observing of legal requirements, but the continuing mandate of society at large. Patten (2002), argues that the differences in the level of social responsibility disclosures are a systematic function of differences in public policy pressure facing individual companies. In the majority of cases, individuals are not aware that a particular course of action is the best way to achieve a certain objective, or that one objective is better than another (Hassan, 2010). In the face of such uncertainty, a social system develops as rules, norms, values, and models come to be accepted by social actors as legitimate (acceptable, desirable, appropriate). According to Zimmerman & Zeitz (2002) when companies are faced with uncertain decisions, social actors refer back to these rules, norms, values, and models in order to proceed. Corporate organizations are expected to act with society’s rules to gain acceptance (legitimacy) from other areas of society. According to Zimmerman & Zeitz (2002) legitimacy is not directly observable and has to be conceived as a social assessment or appraisal of acceptance, appropriateness and/or desirability. Aerts & Cormier ( 2006) argues that the concept of legitimacy has several dimensions and as such can be described as multi-faceted. To this end, Tyler (2006) argues that legitimation refers to the characteristic of being legitimised by being placed within a framework through University of Ghana http://ugspace.ug.edu.gh 13 which something that is viewed as right and proper. Legitimacy is considered to be an intangible asset that determines the ability of a corporation to accumulate capital and personnel, which may influence its survival (Hassan, 2014). Zimmerman & Zeitz, (2002) argue that there are strategic actions which can lead to legitimacy. These actions generates other resources and makes possible survival and growth (Zimmerman & Zeitz, 2002). In addition, legitimacy has an important role in securing support for the corporation’s activities. It shapes a reservoir of support. The companies which are viewed as legitimate are more highly insulated from unsystematic variations in their stock prices (Aerts & Cormier, 2006). Patten (2007) argues that social legitimacy is distinct from economic legitimacy but is not monitored in the marketplace but rather through public policy process. Scott (1995, cited in Aerts & Cormier, 2006) distinguishes between regulative, normative and cognitive legitimacy. According to Aerts and Cormier (2006) legitimacy that is treated in corporate environmental reporting literature is mainly of a normative legitimacy, which incorporates a significant regulatory component. Laws, regulations and formal rules constitute objective reference points and provide formalised and objective assessment parameters. The closer the company’s behaviour aligns with the relevant norms and values, the higher its normative legitimacy (Aerts & Cormier, 2006). Tilling (2004) posits that there are two major classes of legitimacy theory: institutional legitimacy which refers to the type of organisational structure (capitalism or government) that has gained acceptance from society as a whole; and organisational legitimacy (strategic legitimacy) which refers to companies seeking legitimation by approval or avoidance of sanction from groups in society (Hassan, 2010) . Organisational legitimacy theory suggests that a company may occupy one of four phases in relation to its University of Ghana http://ugspace.ug.edu.gh 14 legitimacy: establishing legitimacy; maintaining legitimacy; extending legitimacy; and defending legitimacy (Hassan, 2014). Tilling (2004) suggests that to develop legitimacy theory, the possibility that company may not successfully be able to defend its legitimacy and therefore start to lose legitimacy should be added to the model . This loss phase is most likely to be preceded by sustained media, NGO scrutiny, and government regulation and monitoring. During the loss phase, the company is most likely to increase its social and environmental disclosure in order to meet the specific threat. Suchman (1995) posits that legitimacy theory from an institutional viewpoint refers to companies, managers, performance measures and audience needs which are components within a larger institutionalised cultural framework that produce demands on each other. Legitimacy theory is closely related to stakeholder theory. The demand for enhanced disclosure has been fuelled by the increasing popularity of the stakeholder theory that has resulted in an increased recognition that the interactions of a company are not limited to shareholders (Boesso & Kumar, 2009). Legitimacy theory comprises two basic ideas: companies need to legitimise their activities; and this legitimacy process provides some benefits for companies. The first element is consistent with the argument that CSRD is linked to the presence of social pressure. In this context, it can be argued that the need for legitimacy is not equal for all companies due to the differences in both the degree of social pressure facing companies and the level of the response to this pressure. There are a number of factors which determine the degree of social pressure facing companies and the responses to these pressures. These factors are the potential determinants of CSRD. The second element refers to the fact that the companies can expect to gain benefits from achieving legitimacy through CSRD. Wang et al. (2013) suggests that social information in University of Ghana http://ugspace.ug.edu.gh 15 annual reports can influence public policies, directly by addressing public and/or legislative concerns, or indirectly by projecting an image of the company’s social awareness. It is argued that the greater the likelihood of adverse shifts in public policy, the greater the need to influence the process through social disclosure. On the other hand, researchers have developed the political cost hypothesis to argue that firms employ social responsibility activities to reduce the risk of governmental intrusions, such as regulation, that may adversely affect firm value (Watts and Zimmerman, 1978). Governments control critical resources, and have the power to affect wealth transfers between different groups, while the business enterprises are vulnerable to these wealth redistributions. This has compelled corporations to employ a lot of devices, such as corporate social responsibility disclosure, to counter the potential government intrusions (Wang et al. 2013). In the context of social responsibility disclosure research, legitimacy theory is rooted in the voluntary context (Hassan, 2014) while political cost perspective is associated with government regulation. The study relied on the argument on the voluntary nature of CSRD as social responsibility reporting in Ghana is voluntary. 2.2 Framework of CSRD Determinants Based on the legitimacy theory the determinants of CSRD are those factors which are related to social pressure. In this regard, two important questions come to mind: Are all companies facing the same degree of social pressure? Are all companies responding to this pressure to the same degree? Literature has argued that both the degree of social pressure and the degree of companies’ responsiveness differ across companies. Two types of factors determine these variations. The first type is factors that can determine the degree of social pressure facing companies. These factors include corporate characteristics (size, University of Ghana http://ugspace.ug.edu.gh 16 industry and degree of multinationality). It can be argued that large industrial companies which incorporate a substantial degree of multinational activities, and which face widespread media pressure garner greater attention from the broader society and consequently face the highest degree of social pressure. Therefore, according to legitimacy theory, these companies will need to legitimise their activities more strongly. Thus, a positive relationship between these factors and CSRD is expected. The second type of factor is the degree of company response to social pressure. These factors include corporate ownership and corporate governance. It can be argued that well-governed companies and those with a more dispersed ownership tend to respond positively to social pressure and provide greater levels of disclosure. From the forgoing, the framework below can be suggested to explain these variables properly. University of Ghana http://ugspace.ug.edu.gh 17 Figure 3.1: Conceptual framework of determinants of CSRD Adapted from Hassan (2010) The degree of social pressure facing companies The degree of companies’ response to social pressure CSRD Corporate Ownership Corporate Governance Size Industry Multinational Activity University of Ghana http://ugspace.ug.edu.gh 18 CHAPTER THREE EMPIRICAL LITERATURE REVIEW 3.0 Introduction This chapter begins with a general overview of CSRD, its measurement in terms of both quantity and quality. The measurement of these variables according to literature is very controversial and as such much literature will be spent of clarifying the issue. This will be followed by an empirical review on determinants and consequence of CSRD. The literature in this area has mainly focused on three main questions; which, why and what. The first question is: What companies seem more interested in providing information on their social responsibilities? To answer this question, aspect of the literature provides empirical analysis of the factors (determinants) affecting CSRD. The second question is: why are companies interested in social and environmental disclosures? Part of the review will also focus on the motivations of companies to disclose social and environmental information. The third question is: what are the benefits (consequence) of CSRD to the company? In this context, the review focused on the financial consequence of social and environmental disclosures. The importance of this chapter is that it represents a basis on which to recognise a gap in previous literature and develop research questions. Limitations of previous studies and also provide a basis for defining some concerns that should be taken into account in the development of this study. University of Ghana http://ugspace.ug.edu.gh 19 3.1 The nature of CSRD In a study by Hackson & Milne (1996) on some determinants of social and environmental disclosures in New Zealand, they defined CSRD as the provision of financial and non- financial information about a company’s interaction with its physical and social environment as stated in its annual reports or any other corporate report. Gray et al. (2001) in a similar study in the United Kingdom argued that social and environmental disclosures in its simplest form typically involves providing information relating to an organizations’ activities, aspirations and public image with regard to environment, community, employee and consumer issues. Said et al. (2009) in their study about the relationship between CSRD and corporate governance characteristics in Malaysia argued that CSRD can provide either positive information which presents the firm as operating in harmony with the environment, such as stating that the company have training programmes for employees or that waste management policies are being adhered to, or negative information which presents the company as operating to the detriment of the environment, such as the inability to control or reduce pollution or failure to solve a social problem. The unique feature of this definition of CSRD is the fact that, it combines both positive and negative information on the company’s social and environmental performance. As stated by Lina & Gokiene (2011), the essential aim of social reporting and disclosures is to create social transparency system for responsible decision making. The media for such disclosures have also attracted academic attention. In previous studies, various tittles have been given to CSR disclosure documents. Notable among them include; social progress report, social responsibility reports, social sustainability reports, sustainability University of Ghana http://ugspace.ug.edu.gh 20 reports etc (Hassan, 2010). Razek (2014) argues in the light of all these differences that CSR report is the total reports that covers sustainability, environmental, environmental and social reports. CSRD activities is voluntary in nature and as such Gray and Milne (2002) argues that the voluntary approach highlights two aspects; one, the natural environment and social justice are safe in the hands of businesses and this arises because; two, such ideas are mere extension of good business practice. However, Fukukawa et al. (2007) posits that the voluntary nature of social and environmental information has been criticised as bias and an attempt by corporations to manipulate public perceptions, and as such some developed economies like the United States and the United Kingdom have intensified efforts to increase mandatory disclosures. Rizk et al. (2008) states that there is public pressure on stakeholders to develop corporate social policies mainly as a result of the failure of voluntary disclosures. This pressure therefore acts as a driver for governments, as well as other stakeholders to provide different accreditation mechanisms, guidelines and standards for CSR practices and reporting (Hassan, 2010). It must however be emphasized that these efforts do not attempt to make CSRD mandatory, but seek to find a middle way that seeks to hold companies accountable for their actions. A study by Van der Lann et al. (2005) on differences in social disclosures internationally among Danish/ Norwegian and United States firms indicates an emerging trend for CSRD, which they called the solicited disclosure in which companies are increasingly asked to account in various ways their interactions with and impacts on society. Stakeholders like Non-governmental organizations (NGOs), rating agencies, regulatory agencies, ethical or socially responsible fund managers and other interested agencies are requesting social information from companies. According to Hassan (2010), the difference between voluntary University of Ghana http://ugspace.ug.edu.gh 21 and solicited disclosures is the motive to provide the information. He argues that the motive to provide voluntary CSRD comes from management of companies, either in response to a threat to legitimacy or a need to account, while the motive for solicited CSRD is a direct request from the identified stakeholders. In addition to the above arguments on the nature of CSRD, The whole idea of social and environmental disclosures is considered to be a western phenomenon; as a result, research in the area has concentrated primarily on developed countries (Fifka, 2013). There are some challenges facing CSRD in developing countries where the institutions, standards and appeals system, which support CSRD in western countries are relatively weak (Kemp, 2001). In addition to the above argument of CSRD been a western phenomenon, Chambers et al. (2003) suggest a number of arguments which can explain the low levels of CSRD in developing countries: First, corporate social responsibility is a function of wealth. Second, civil society is developed economies stimulates CSR by exerting greater societal demands and expectations of business responsibility and finally, western corporate governance is more developed than in developing countries and thus encourages greater corporate social responsibility. Hope (2003) in a study on disclosure practices, enforcement of accounting standards and analyst forecast across some selected European countries suggested a measure of enforcement based on five country-level factors: audit spending; insider trading laws; judicial efficiency; rule of law; and shareholder protection. Aerts et al. (2008) applied this measure in order to explain the international differences in the demand for environmental disclosures across a sample of countries. The research found that the United States of America has the strongest enforcement level, followed by Canada, the Netherlands, France, Belgium and Germany. It has been argued that, in high enforcement countries, the mandatory disclosure University of Ghana http://ugspace.ug.edu.gh 22 was found to be highly regulated, resulting in higher levels of disclosure. Thus North American companies which operate in a more regulated context compared with continental Europe companies disclose more environmental and social information related to expenditure and risk, abatement and remediation. Also, Cormier & Gordon (2001) argues that publicly owned companies disclose more social and environmental information than the privately owned companies. On the reporting framework of CSRD, Solomon (2000) conducted a survey to identify aspects of environmental reporting and comparing these with financial reporting. The results provide that the implicit conceptual framework for environmental reporting mirrors the explicit financial reporting conceptual framework in the UK in the following ways: First, the users of financial reporting are also users of environmental reporting; the difference being in the greater emphasis attached to some users (employees, legislators and regulators) and less emphasis attached to others (shareholders). Second, all qualitative characteristics relevant in financial reporting are also relevant in environmental reporting. Third, verification is necessary for environmental reporting as with financial reporting. Fourth, companies bear the cost of environmental reporting as with financial reporting. Fifth, the most appropriate presentation of environmental reporting is within annual reports as financial reporting. However, a number of differences exist between environmental reporting and financial reporting, notable among them includes: University of Ghana http://ugspace.ug.edu.gh 23 First, the elements of environmental reporting are totally different from those of financial reporting. Second, there is no consensus on who should perform verification for environmental reporting. 3.2 CSRD and its categories The measurement of CSRD in firm’s annual reports requires a clear definition of CSRD and its categories as applied in research. Some studies define CSRD based on an early survey by Ernst & Ernst (1979). This survey categorizes CSRD information as follows;  Environment (pollution control, prevention or repair of environmental damage, conservation of natural resources, and other environmental disclosures);  Energy (conservation, energy efficiency of products, and other energy-related disclosures);  Fair business practices (employment of minorities, advancement of minorities, employment of women, advancement of women, employment of other special interest groups, support for minority businesses, socially responsible practices abroad, and other statements on fair business practices);  Human resources (employee health and safety, employee training, and other human resource disclosures);  Community involvement (community activities, health and related activities, education and arts, and other community activity disclosures); University of Ghana http://ugspace.ug.edu.gh 24  Products (safety, reducing pollution from product use, and other product-related disclosures); and  Other social responsibilities disclosed (other disclosures, and additional information). The most cited work according to Hassan (2010) that comprehensively defines CSRD and its categories is the work by Gray et al. (1995). This study identified four themes for corporate social responsibility, namely; natural environment, employees, community and customers. However, recent studies (Yagemeh, 2013; Hassan, 2014) have identified some problems with this categorization and raise the following concerns; First, there are some elements which are socially responsible activities but fall outside these categories. Examples, value added statements, product development and research. Second, these categories did not divide them into mandatory and voluntary. The most comprehensive index according to Hassan (2014) is the one developed by Hackston &Milne (1996) and adopted by Deegan et al. (2002) and applied in a study in New Zealand. This study will adapt that index to measure CSRD in Ghana. Surveys of CSRD practices in the western developed countries reveal that companies place the greatest emphasis on the disclosure of human resources (Gray et al., 2001), including employee numbers and remuneration, equal opportunities, employee share ownership, disability policy and employee training. Little disclosure exists in sensitive areas such as trade union activities, pay awards and redundancy schemes and costs (Adams et al., 1998). The vast majority of disclosures are qualitative in nature. It is also important to add that no CSRD index have been developed for any developing country and Ghana for that matter. The few studies in other developing countries like Nigeria, South Africa, and Egypt etc adopted University of Ghana http://ugspace.ug.edu.gh 25 CSRD index used in developed countries. This study in line with previous literature adapted Hackstone % Milne (1996) index of CSRD. 3.3 The quantity of CSRD This section discusses issues relating to the measuring of the quantity of CSRD in annual reports. The accurate measurement of this disclosure is critical to ensuring the accuracy of the results of the study. The research method that has been commonly used in literature to measure the quantity of CSRD is content analysis (Gray et al., 2001; Yagemeh, 2013; Hassan, 2014). According to Krippendorf (2007), content analysis can be defined as “a research technique for making replicable and valid inferences from data according to their context”. The study further argues that, the main characteristic of content analysis is that, data are coded and measured in a reliable and systematic manner. Abbott (1979) defines content analysis as a technique for gathering data that consist of codifying qualitative information in anecdotal and literacy form into categories in order to derive quantitative scales of varying levels of complexity. According to Hassan (2010), content analysis technique involves determining the documents used in analysis (annual reports), define CSRD and its categories, unit of analysis and reliability of the content analysis. The document used in this study is the annual report and the justification is provided in the methodology of the study. CSRD and its categories have also been dealt with in the previous section. With regards to the unit of analysis, the most commonly used method for quantifying social disclosures include the number of characters, words, pages and the proportion of volume of CSRD to total disclosure (Fifka, 2013). According to Hassan (2014), literature in the area does not provide any theoretical justification for the choice of any of the method listed above. University of Ghana http://ugspace.ug.edu.gh 26 However, Deegan & Gorgon (1996) justified the use of words to quantify disclosures by stating that the volume of disclosure can be recorded in greater detail using number of words. In addition, Zeghal & Ahmed (1990) also argues that words are the smallest unit of measurement and can be expected to provide the maximum strength to the study in assessing disclosure quantity. On the contrary, other studies have criticised the use of words to quantity disclosures. For instance, Hackston & Milne (1996) argued that the use of words as a measure of CSRD is ambiguous as it leaves the researcher pondering which individual word is a CSRD and which is not; hence the possibility of disagreement between different coders is quite high. Hassan (2010) supported this reasoning by stating that words have no meaning to provide a sound basis for coding CSRD without a sentence or sentence for context. On the use of sentence to measure CSRD, Hackston & Milne (1996) provided the following reasons for using sentences to code CSRD. First, sentences can be counted with more accuracy than words. Second, sentences are used to convey meaning rather than discerning individual words which could be problematic. Third, sentences overcome the problem of allocation of portions of pages and remove the need to account for the number of words. Forth, sentences are more natural unit of written English to count than words. Darrel & Schwartz (1997) also stated that a sentence consider a conventional unit of speech and writing, while portion of pages does not. The use of sentences as a unit of measurement seems to ignore that the differences in use of grammar may result in conveying the same message by using a number of words and similar amount of space but using different number of sentences. Milne &Adler (1999) also supported the use of sentence when they argued that measurement in terms of sentences give results to measurement of volume in terms of proportion of the page. University of Ghana http://ugspace.ug.edu.gh 27 On the use of pages to measure CSRD, Ng (1985) argues that print sizes, column sizes and pages sizes may differ from annual report to annual reports. Milne & Adler (1999) posits that using pages to measure CSRD adds unnecessary unreliability. Despite these limitations, some studies adopted this technique in measuring CSRD. Gray et al. (1995) and Unerman (2000) measured CSRD using portion of pages used a grid with 25 rows of equal height and four columns of equal width was laid across each CSR disclosure, with volume being counted as the number of cells on the grid taken up by disclosure and making allowance for any blank parts of a page. A study by Newson & Deegan (2002) measure CSRD to the nearest hundredth of a page using a transparent plastic A4 sheet divided into a grid of 100 rectangles, each side is divided into 10 after allowing for a standard margin of approximately 25mm. In all, Unerman (2000) criticized the over reliance on only words, sentences or characters alone to the neglect of non-narrative disclosures like photographs and charts which is argued to be another effective way of disclosures. The paper further argued that sometimes photographs are more effective tools of disclosure than narrative forms especially to stakeholders who do not have time to read every word in the annual report. For the purpose of this study, the measurement of the quantity of CSRD in annual reports was done using the number of sentences as well as graphs and photographs that represent CSRD disclosures. This objective of this choice is to reduce the weaknesses in relying on only written words to the neglect of photos which is argued to be an effective way of CSRD. Also, the study did not use the number of pages because CSRD reports in annual reports are not always structured and presented alone. Sometime they are contained in the directors’ reports or other reports which cannot be isolated in terms of number of pages (Hassan, 2014). University of Ghana http://ugspace.ug.edu.gh 28 3.4The quality of CSRD in annual reports In an attempt to understand a phenomenon, it is not enough to measure the volume of information as a means to measurement. To fully understand CSRD, the study also seeks to evaluate the quality of CSRD. The measurement of CSRD quality is believed in literature to be difficult because of the challenge in getting an objective measure of information quality. To fully understand corporate social disclosure in Ghana, the study want to compliment the quantity of disclosure with quality to present a better understanding of the level of social disclosure in Ghana. This concept of disclosure quality according to Nasir (2010) is a controversial subject in academic literature. The definition of quality in business studies is that the quality of product or service indicates the perception of the degree to which the product or service meets users’ expectations. ISO 9000 defines quality as “degree to which a set of inherent characteristics fulfils requirements”. Brammer & Pavelin (2006) argued that disclosure quality refers to whether the disclosure reports specific actions, quantifies environmental impacts, sets formal targets and is subject to external audit. The quality of environmental disclosure varies widely across companies since the disclosure content is not strictly regulated (Aerts et al , 2004). It can be deduced from the above definition that disclosure quality can simply be defined as the extent to which information disclosed meets user needs. It has also been well acknowledged in literature that the measurement of disclosure quality is a disputed subject. Beattie et al. (2004) indicate that there are two principle ways to measure quality of disclosure: use subjective analyst disclosure quality ranking; and use researcher-constructed disclosure indices, in which the amount of disclosure is used as a University of Ghana http://ugspace.ug.edu.gh 29 proxy for disclosure quality. Hammond &Miles (2004) argues that the measure of disclosure quality can be determined through the following attributes: quantitative disclosure; third party verification; establishment of appropriate targets and reporting progress against targets; warts and all reporting; the adoption of reporting guidelines and standards; the ability to accurately assess performance from disclosure; clear statement of vision from chief executive; good coverage of significant issues; wide access; reporting of normalised data; and awards/accolades (Hammond &Miles, 2004). According to Hassan (2010), the common method used in literature to measure disclosure quality is to rank disclosure information according to its type or predetermined factors. To this end, various ranking methods have been used in literature to achieve this objective. The distinctive characteristic of these ranking methods is that, they present different scale point for ranking the quality of information disclosure in annual reports. For instance, Robertson & Nicholson (1996) and Cormier et al. (2005) suggested a 3-point scale system, Gamble et al. (1995) and Raar (2002) used a 7-point scale and Van Staden & Hooks (2007) developed a 5-point scale to assess the quality of environmental disclosure which is a major component. On the reliability of the ranking systems stated above, Hassan (2014) argued that using a ranking system that consists of many points could reduce reliability in the measurement, as an increasing number of points leads to a greater opportunity for the existence of a subjective judgement of measurement. It could therefore be stated that using a ranking scale with lower point could improve the reliability of the results. Al-Tuwajiri et al. (2004) proposed qualitative disclosure measures which used weights to represent different disclosure items, based on the perceived importance of each item to various user categories: weight 3 to quantitative disclosure, weight 2 to non-quantitative University of Ghana http://ugspace.ug.edu.gh 30 disclosure that reports specific activities, and weight 1 to common qualitative disclosure. Cormier et al. (2005) also used a three scale rating score as follows: 3 for items described in quantitative terms, 2 when an item is described specifically and 1 for an item discussed in general terms. According to Toms (2002), quantified disclosure items are more likely to represent actual activities and could as well represent a distinguishing feature among competitors. Hammond & Miles (2004) also conquered with this argument when they stated that quantitative disclosures is one of the key attributes used to measure quality of social disclosures. Toms (2002) surveyed a pilot questionnaire on fund managers and analyst on rating for social disclosures. The results showed that low rating was given to non-quantified information while high rating was given to externally-monitored environmental report. 3.5 Determinants of CSRD Analysis of disclosure determinants is a major consideration in accounting literature (Hassan, 2010). Literature on CSRD has looked at both internal determinants and external determinants (Fifka, 2013). These studies have attempted to answer two major questions. First, what attitudes do companies adopt toward accounting disclosure, either general or specific? This question leads to analysis of the disclosure level. Second, why do some companies disclose more, or less, information than others? This question leads to analysis of disclosure determinants. Because CSRD is a voluntary activity, it is important to understand which companies disclose information about social and environmental activities, or which companies disclose such information, more so than others. Adams (2002) indicated that an understanding of the factors which influence disclosure is necessary for improving accountability. University of Ghana http://ugspace.ug.edu.gh 31 Determinants of CSRD are different from determinants of corporate financial reporting, in the sense CSRD addresses the social accountability of companies and focuses on a broader audience (stakeholders), than corporate financial reporting, with its primary focus is on the information needs of investors and creditors (Smith et al., 2005). Various variables have been examined in several studies as determinants of CSRD. Adams (2002) categorized the determinants of CSRD into three categories as in the following table; Category Variables Corporate characteristics Size, industry membership, corporate age, profit General contextual factors The country of origin of the company, social and political context, economic context, period of time, specific events, media pressure, stakeholder power. Internal context Corporate chair, presence of CSR committee Adopted from Hassan (2010) Lee & Hutchison (2005) categorized the forces affecting the decision to disclose environmental information, in line with the results of prior studies, into; laws and regulations, firm/industry characteristics, rational cost-benefit analysis, cultural forces and attitudes, and legitimacy, public pressure, and publicity. Sylvie et al (2003) reviewed literature and indicated that the evidence from studies suggests that voluntary environmental disclosure increases with;  Corporate size and membership in environment-sensitive industries such as oil and gas, chemicals, forest and paper products or utilities  The extent to which a company is widely-owned University of Ghana http://ugspace.ug.edu.gh 32  A company’s exposure to environmental -related legal proceedings or fines related to the environment  A company’s media exposure of its environmental activities  The probability of being involved in accidents in the future  Environmental lobby groups’ concerns about company’s environmental performance To explain the differences of CSRD among different companies, previous studies mainly analysed the impact of corporate characteristics on CSRD and some studies have focused on studying the impact of media pressure. These studies have been primarily concerned with the impact of corporate characteristics and to a lesser extent, the impact of media pressure, while less attention has been paid to the impact of the degree of multinational activities and corporate governance mechanisms. 3.6 Corporate characteristics A lot of studies on CSRD in accounting have focused largely on corporate characteristics as potential determinants especially at the firm level (Fifka, 2013). Ahmed & Courtis (1999) in their study about the relationship between corporate characteristics and corporate disclosure levels in annual reports using a sample of over 2000 annual reports argued that accounting literature has been interested in the association between corporate characteristics and corporate annual report disclosure since 1961. These studies that related to CSRD were particularly interested in examining the impact of corporate characteristics on it. For instance, Hassan (2010) cited Cowen et al. (1987) study where they examined the influence of corporate characteristics on the categories of CSRD in annual reports. The study used Ernst & Whinney (1978) survey to categorize CSRD into; University of Ghana http://ugspace.ug.edu.gh 33 environment, energy, fair business practices, human resources, community involvement, and products. The corporate characteristics variables that were used in the study are; size, industry, profitability, and presence of social responsibility committee (which introduced as a corporate characteristics for the first time) (Hassan, 2010). The results based on a sample of 134 US firms, showed that corporate size has a significant association with disclosure elements; environmental, energy, fair business practice, community involvement, but no influence over human resources or product disclosures. The study also found that industry type influenced some types of disclosures such as energy and community involvement but most disclosure types like fair business practice, human resources, products, others are not significantly affected by industry type. Human resources information appears to be related to the presence of social responsibility committee. Hackstone &Milne (1996) examined the determinants of social and environmental disclosures in New Zealand using corporate characteristics such as size, industry type, corporate profitability and country of reporting and country of origin. The empirical results show that corporate size and industry type are significantly associated with the level of CSRD while profitability is not. In the context of developing countries, Hossain et al. (2006) examined the relationship between social and environmental disclosures and corporate characteristics in Bangladesh. Using a sample of 50 companies and variables such as size, profitability, industry, subsidiaries of multinational company, and audit firm, and the study found that industry and profitability were associated with the level of social and environmental disclosures. In a related study, Ahmad et al. (2003) examined factors influencing environmental disclosure in annual reports for Malaysian listed companies. Their study focused on some corporate characteristics such as corporate size, financial leverage, profitability, industry University of Ghana http://ugspace.ug.edu.gh 34 membership, auditor type, and effective tax rate. Based on a sample of 299 companies, the empirical results revealed that environmental disclosure is significantly associated with both auditor type and financial leverage. Ho & Taylor (2007) examined the impact of corporate characteristics such as; size, profitability, industry membership, leverage, and liquidity) and country on triple bottom- line reporting (TBL) in annual report, stand-alone report, and special website reports, in both US and Japan. The empirical results based on sample of 50 of the largest US and Japanese companies, show a positive and significant association between both corporate size and country and TBL disclosure, negative association between corporate profitability, liquidity, and industry membership and TBL reporting, while there is no significant association between leverage and TBL reporting. Despite this study extending the analysis of CSRD to other disclosure media over annual reports, it does not present a clear framework for the variables that were examined. In addition to small sample size, the study examined between determinants at the company level and at the country level. Garcia-Sanchez (2008) study using Spanish firms also showed that both corporate size and industry membership are associated with CSRD, while there is no association between profitability and CSRD, based on a sample of 35 companies. Reverte (2009) examined the influence of corporate characteristics, industry type as well as media pressure on CSRD of Spanish listed firms using legitimacy theory. The results show that firm size, industry type as well as media pressures were associated with the level of CSRD. On the contrary, profitability and leverage were not associated with the level of CSRD. The study implied that legitimacy theory is the most relevant theory to explain the CSRD practices of Spanish companies. University of Ghana http://ugspace.ug.edu.gh 35 Monteiro & Aibar-Guzman (2009) examined determinants of environmental disclosure in the annual reports for large Portuguese companies. The study examined the following variables; firm size, industry membership, profitability, quotation on the stock market, foreign ownership, and environmental certification. The empirical results based on a sample of 109 companies show that firm size and the fact that a company is listed on the stock market are positively associated with environmental disclosure. In addition to concentration on environmental disclosure in annual reports, this study did not provide a theoretical background for selected determinants. Jinfeng & Huifeng (2009) on their part examined the factors influencing the level of environmental protection information disclosure in annual reports of Chinese firms. The study examined some corporate characteristics such as; corporate size, profitability, industrial nature, financial leverage, and type of accounting firm. The empirical results based on a sample of 248 Chinese companies, show that corporate size, industry nature, and type of accounting firm are significantly associated with environmental disclosure in annual reports. Akano et al. (2013) examined the determinants of corporate social responsibility disclosure by commercial banks of Nigeria using a sample of 13 banks and variables such as size and number of branches. The empirical results show that the value of total assets which was used to measure size have positive relationship and statistically significant with the level of corporate social responsibility activities disclosure. The limitation of this study is that the sample size was too small and it’s examined very few variables and also fails to examine the quality of CSRD. A number of studies have also focused on the influence of corporate characteristics, with other variables on CSRD. In this regard, Brammer & Pavelin (2008) used a sample of University of Ghana http://ugspace.ug.edu.gh 36 134 of the largest UK companies, examined the relation between disclosure and corporate characteristics (size and industry) and they found that disclosure strategy is influenced by industry type and it is positively related to corporate size. Also, Gao et al (2005) examined the influence of corporate size and industry on CSRD in Hong Kong companies, and they found that size and industry influence the level of CSRD. Branco & Rodrigues (2008) examined the factors influencing CSRD in both annual reports and websites for Portuguese listed companies. They argued that, based on both legitimacy theory and a resource-based perspective, CSRD is used by companies as a legitimacy tool to improve corporate image. They examined the following factors; degree of international activity, company size, industry, consumer proximity (a company’s proximity to final consumer), environmental sensitivity, and media pressure. The empirical results, based on a sample of 49 companies, show that only company size and media pressure are significantly associated with CSRD, while other variables are not associated. Although this study used a clear theoretical framework, it did not provide a clear link between factors selected as determinants of CSRD and this theoretical framework. Also, this study added to literature by examining CSRD in both annual reports and internet disclosure, but it suffered from a small sample size. In summary, previous studies have focused on corporate characteristics as determinants of CSRD, and the variables that have been used to represent company characteristics are corporate size, profitability and industry type. The following points can be noted: First, Corporate size is a prevalent variable that has been examined as a determinant of CSRD in previous studies. There are various indicators used in previous studies to measure company size (sales, total assets, Market capitalization, Fortune rank, Turnover and interest received and receivable for banks), but there is no theoretical reason for University of Ghana http://ugspace.ug.edu.gh 37 using a particular measure of size (Hackston & Milne, 1996). The empirical results seem to provide evidence of a significant impact of corporate size on CSRD. Second, Profitability is also a common variable which represents the financial condition of the company. The empirical results appear to provide inconclusive evidence about the impact of profitability on CSRD. There are some methods used to measure corporate profitability such as ROA, ROE, and five year average ROE, and one year lagged ROA. Third, Regarding industry type, there is almost total agreement among studies on the use of a binary code (high profile and low profile) as a measure of industry type. Brammer & Pavelin (2004) defines high-profile industries as those industries with high consumer visibility, a high level of political risk, or are characterised by concentrated, intense competition. Commonly perceived high-profile industries are petroleum, chemical, mining and extractive, forest, paper automobile and airline. Low-profile industries are food, health and personal products, hotel , and appliance and household products (Robert, 1992) as cited by Hassan (2010). The empirical results provide evidence that industry type is positively associated with CSRD. Forth, it has been well articulated in literature the influence of foreign institutions and agencies influence on social and environmental reporting (Dobers & Halme, 2009, Fifka, 2013). Literature however has not explored this variable as determinants of CSRD. However, the few studies on social reporting in developing countries like Ghana have acknowledged the role that multinationals as well as international agencies have helped shaped business practice (Rahaman et al., 2004). Multinational companies operating in Ghana are more likely to disclose more social information because based on legitimacy theory, they have larger exposure and pressure and hence the need to legitimize their activities. For instance, Rahaman (2000) in his exploratory study on environmental reporting in Ghana using some University of Ghana http://ugspace.ug.edu.gh 38 selected firms found that pressure from international lenders like the IMF and governmental regulation as external factors, and management attitudes and the desire to obtain listing on international stock markets as internal factors led to more reporting of social information. Gray et al. (2001) presented an analysis of the relationship between CSRD and corporate characteristics as the following: First, the relationship between CSRD and size has produced somewhat consistent results, as the majority of studies conclude that there is a relation, but Singh & Ahuja, (1983) find no relation, and Cowen, et al. (1987) find the relation only holds for certain areas of disclosure. Second, the relationship between CSRD and industry membership has produced less than consistent results, Hackston & Milne,(1996) find strong relation, Ness & Mirza, (1991) find that this relation holds specifically for the oil industry, while, some other studies find that specific areas of disclosure are related to industry. Third, the relationship between CSRD and profit has produced inconsistent results, some studies find some relation, while some other studies find no relation or an inverse relation, and on the other hand some studies find a positive relation (Gray, et al. ,2001). And from the results of their empirical study, Gray, et al. (2001) believe that in the UK, CSRD is related to corporate characteristics. This conclusion cannot be said about African companies and for the matter Ghana because few studies have been examined in this context. The detailed functional models of the relationships between different measures of CSRD and the corporate characteristics vary with both variables chosen and the time period selected. So, we are unable to claim that there is any unique and/or stable relationship between any measure of disclosure and any corporate characteristics. And University of Ghana http://ugspace.ug.edu.gh 39 they stated that “whilst we can confidently claim that corporate characteristics are highly indicative of a predisposition to disclose, researchers need to make their claims about the relationship between disclosure and characteristics with more care” 3.7 Degree of multinational activity According to Hassan (2014), in contrast to wide attention that has been paid for corporate characteristics; size, industry membership, and profitability, the degree of multinational activities as one of corporate characteristics did not received adequate attention. Toms et al. (2007) and Toms and Hasseldine (2008) stated that although a large literature on the determinants of CSRD has evolved; there is a little evidence on the relation between international activity and CSRD. In their view the level of CSRD in multinational corporations depend on the political and environmental risk in the countries in which they operate and not just the number of foreign countries in which they operate. Their study therefore examined the impact of number of foreign countries, political risk in foreign countries, and environmental risk in foreign countries on CSRD in the sample of multinational companies from environmentally sensitive industries. The empirical results show that both political and environmental risks in countries in which the companies operate are associated with CSRD, while the number of countries is not associated. The limitations of these studies are the concentration on some environmentally sensitive industries (oil, gas, and chemicals) when examining social disclosure, the small sample, and the confusion between determinants at the country level and on the company level. University of Ghana http://ugspace.ug.edu.gh 40 Branco & Rodrigues (2008) in their study on factors influencing social responsibility disclosures examined the impact of multinationality, with some other corporate characteristics variables, where the degree of multinational activity was measured by the ratio of foreign sales, on CSRD by Portuguese listed companies through both annual reports and web sites. The empirical results from the study based on 47 companies showed non-significant association between multinationality and CSRD in both media. Stanny & Ely (2008) examined factors influencing environmental disclosure and it effects on climate change using US firms where they examined the degree of multinational activity on environmental disclosures measures as the ratio of foreign sales. The empirical results based on 490 firms revealed that foreign sales are significantly associated with disclosure. Hassan (2014) examined the impact of corporate governance, degree of multinational activity on social responsibility disclosures using UK firms. The study measured the degree of multinational activity in two ways, first, the ratio of foreign sales and second, the number of foreign countries the firms operates. The empirical results based on 327 firms revealed that the degree of multinationality is not significantly associated with the quantity and quality of CSRD in annual reports. It appears that previous studies have used two indicators to measure multinationality: ratio of foreign sales, and number of foreign countries, and they provide inconsistent results concerning the relationship between multinationality and CSRD. 3.8 Corporate governance variables Halme & Huse, (1997) in their study on the influence of corporate governance, industry and country factors in environmental reporting examined the relation between environmental University of Ghana http://ugspace.ug.edu.gh 41 disclosure and corporate governance variables such as; ownership concentration and board size), industry, and country variables in four European countries (Finland, Norway, Spain, and Sweden). The empirical results based on a sample of 40 companies in four countries, show no significant association between both ownership concentration and the number of board members with environmental disclosure, while industry appear to be the most important factor in explaining environmental disclosure. Chau and Gray (2002) examined the association of ownership structure with the voluntary disclosures including environmental disclosures of listed companies in Asian settings of Hong Kong and Singapore. Using a sample of 60 companies in Hong Kong and 62 companies from Singapore, the study revealed that the extent of outside ownership is positively associated with voluntary disclosures including environmental disclosures. Gul and Leung (2004) examined the linkages between outside directors on the board and voluntary corporate disclosures. The empirical results based on a sample of 385 firms in Hong Kong revealed that the extent to which managers will disclose more corporate information is likely to be affected by the composition and quality of the board of directors. Haniffa and Cooke (2005) examined the extent to which Corporate Social Responsibility Disclosure (CSRD) in the annual reports of Malaysian listed company’s changes over time and whether there is an association with three groups of variables: cultural, corporate governance and firm specific (control) variables. The study measured culture value based on the ethnic background of directors and shareholders. The governance mechanisms that were studied are the proportion of non-executive directors on the board, multiple directorships for chairman, and proportion of foreign shareholders. The results based on a sample of 167 non-financial companies show that CSRD is significantly associated with culture, University of Ghana http://ugspace.ug.edu.gh 42 multiple directorship, and foreign ownership. However, Non-executive directors are negatively associated with CSRD. In the African context, Barako et al. (2006) investigated the extent to which corporate governance attributes such as ownership structure and company characteristics influence voluntary disclosure practices including environmental disclosure in Kenya. Using a sample of 54 firms listed on the Nairobi Stock Exchange over a ten year period, the empirical results show that voluntary disclosure including environmental disclosure is significantly associated with corporate governance attributes , ownership structure and company characteristics. In a related study, Barako and Brown (2008) examined the influence of gender and board representation on communication of corporate social reporting by Kenyan banks. The results of the study based on a sample of 20 firms indicate that board representation can fundamentally improve corporate communication. A higher level of women representation and independent directors greatly improves disclosure. Ghazali (2007) argued in his study that no studies have been done on the association between corporate ownership structure and CSRD, so his study examined the influence of ownership structure on CSRD. The variables that were examined are: ownership concentration; director ownership; government ownership; company size; profitability, and industry. The empirical findings, based on sample of 86 Malaysian companies, indicated that two ownership variables, director ownership and government ownership, are significant influence of CSRD in annual report, while, third ownership variable, ownership concentration, is not statistically significant in explaining the level of CSRD. Profitability is not a significant factor in explaining CSRD, while industry was also significant factor influencing CSRD. University of Ghana http://ugspace.ug.edu.gh 43 Abdullah et al. (2011) examined the influence of board independence and ownership on the decision to disclosure social responsibility disclosures. The empirical results based on the sample of 100 largest firms in Malaysia revealed that boards of family owned firms are negatively associated with the level and the quality of CSR disclosure. In addition to the above, some studies examined certain variables of corporate governance with other variables. Naser, et al. (2006) examined factors influencing CSRD in developing country using Qatar as a case study. The variables that were examined are corporate growth, market capitalisation, profitability, leverage, and ownership variables (governmental ownership, institutional ownership, and major shareholders). The results based on a sample of 21 Qatari companies show positive association between extent of CSRD and corporate size, leverage, and corporate growth. Ownership variables were not associated with CSRD. Brammer & Pavelin (2006) examined the influence of corporate ownership and board composition (with some other variables) on environmental disclosure. They distinguish between the decision to make a voluntary environmental disclosure and decisions concerning the quality of such disclosure. Brammer & Pavelin (2008) examined the factors that influence the quality of environmental disclosure. The factors that were examined as determinants of quality of environmental disclosure are; type of activities, environmental performance, corporate size, media pressure, corporate ownership, profitability and leverage, and number of non-executive directors on the board. The empirical results based on a sample of 447 companies in the UK, show that both corporate size and type of activities is significantly associated with the quality of corporate environmental disclosure. Despite this study being characterized by the large size of the sample, and evaluation of environmental disclosure in annual reports and other statements, it is limited by its concentration on the quality University of Ghana http://ugspace.ug.edu.gh 44 of environmental disclosure and it does not present a theoretical background for selected determinants. Also, Prado- Lorenzo, et al. (2009 a) examined the impact of a number of independent directors on CSRD in Spanish companies, and their results indicated no relation between the two variables. The study also test a stakeholder theory approach to analysing CSRD, and examined shareholder power and ownership dispersion on CSRD. The variables which have been examined are; the presence of financial institution in the corporate ownership structure, the presence of a physical person that represents a dominant shareholder and a number of independent directors. The empirical results, based on a sample of 99 Spanish companies, reveal only a limited association between the presence of a physical person that represents a dominant shareholder and CSRD. Liu & Anbumozhi (2008) examined the factors affecting the disclosure level of corporate environmental information in Chinese listed companies on the bases of stakeholder theory. The empirical results based on sample of 175 companies showed that government power (the environmental sensitivity of industry) is positively related with the level of environmental disclosure, while shareholder power (percent of floating stock possessed by the top 10 shareholders) and creditor power (debt/asset) are not associated with the level of disclosure. Parsa & Kouhy (2008) noted that the literature was extremely focused on the disclosure of social information by large companies, so they examined the determinants of the disclosure of social information by small -and medium-sized companies in UK. Based on a sample of 100 UK companies, they used a correlation test to examine the relation between CSRD and some variables. The empirical results show that the corporate age is not associated with CSRD, while industrial background, corporate University of Ghana http://ugspace.ug.edu.gh 45 size, and gearing, are associated with it. These results imply that small-and medium- sized companies are similar to large companies in the impact of both corporate size and industry membership on CSRD. Despite widespread academic and business interest in CSRD, the theoretical framework of the underlying determinants of CSRD is still elusive (Cormier, et al ., 2005), and each study has concentrated on examining some variables as determinants of CSRD. Previous literature on determinants of CSRD has been primarily concerned with the impact of corporate characteristics, following mainstream disclosure literature, and some studies have examined the impact of media pressure on CSRD. By contrast, little attention has been paid to the impact of other variables, such as corporate governance and degree of multinational activities, on CSRD. Regarding this concern, Haniffa & Cooke (2005) argued that since disclosure is an accounting activity involving both human and non-human resources, studies in this area would benefit if both cultural and corporate governance factors are considered (Haniffa & Cooke, 2005). Consideration of corporate governance is important because it is top management that oversees disclosure in annual reports (Gibbins, et al.,1990). It can therefore be argued that previous studies related to determinants of CSRD display two basic limitations. One limitation is related to CSRD itself , as these studies, in most cases, have provided an incomplete picture of CSRD through concentration on quantity of environmental disclosure in annual reports. These studies seem to be motivated by growing global concerns concerning environmental issues and increasing attention from the public with regard to the impact on the environment of companies’ activities. In this aspect, some studies indicated that environmental information is not the prevalent social disclosure category in annual reports; Hackston & Milne (1996), indicated that University of Ghana http://ugspace.ug.edu.gh 46 human resources and community information are the most important categories in New Zealand companies. Other studies have also supported this view (Rizk et al. 2008; Sobhani et al. 2009; Hassan, 2014). The second limitation is related to determinants, as there is no theoretical background to explain the selected variables. 3.9 Consequence of CSRD It is clear that general awareness and concern in society for matters such as environmental degradation, habitat destruction, global climate change, human rights, and stakeholder involvement, continue to increase. It can be argued that the number of potential areas in which social or environmental activity can have relatively direct financial consequences must increase. According to Murray et al (2006) these consequences can be of a cost-saving nature, cost or liability avoidance, revenue-generating, or even simple signals of best-in-class management practices. A number of studies have paid attention to studying the market reaction to CSRD (Dhaliwal, 2009; Orlens, 2010;., Khlif & Souissi, 2015). Parsa & Kouhy, (2001) examined the relation between CSRD and different aspects of performance. The different aspect of performances that were measured includes accounting ratios in the areas of profitability, managerial efficiency, gearing and liquidity. The empirical results show that profitability and gearing were found to be positively associated with CSRD, concerning of managerial efficiency only stock ratio was associated with CSRD, while no association between liquidity and CSRD. Ingram(1978) in a study on information content of social disclosure, examined the impact of corporate social disclosure on security returns using two related tests; the first test analysed returns from a broad spectrum of the market, and the second test investigated the return performance of specific market segments. The results indicated that in first test, University of Ghana http://ugspace.ug.edu.gh 47 the social disclosure has no information content, while the second test suggest that the information content of social disclosure varies between companies. These results suggest that the information content of corporate social disclosure is conditional upon the market segment with which the company is identified. Anderson & Frankle, (1980) study the influence of voluntary social disclosure on the beta of a portfolio of stocks. The study compared the returns to portfolios composed of securities of socially disclosing companies with the returns to portfolios of equivalent risk composed of securities of non-disclosing companies. The mean returns of the portfolios were compared over two six-month periods (pre- and post-fiscal year), if social performance information has no information content. Then the mean returns for the disclosing and non-disclosing portfolios should be equal. The results indicated that social disclosure has information content and that the market values this disclosure positively. Chan & Milne (1999) study on investors reactions to environmental saints and sinners examined the investors’ reaction to two states of corporate environmental performance; first, when the company discloses about its bad environmental performance(bad news), and second when the company discloses it is a leader in environmental management (good news). The results indicated that the investors react strongly and negatively to the poor environmental performance, while there is no significant reaction to the good environmental performance. Epstein & Freedman (1994) examined whether individual investors demand social information in annual reports. Using a survey of 300 individual investors, the results revealed that non-institutional shareholders are interested in having their corporations report on some information concerning social activities. It appears that there is a strong University of Ghana http://ugspace.ug.edu.gh 48 demand for information about product safety and quality and about company’s environmental activities. According to Hassan (2010), Belkaoui (1972) and Longstreth & Rosenbloom (1973) were the two first studies to examine the relevance of the social information for investors. The first study surveyed groups of bank officers, practicing accountants, and students, and the results indicated that the disclosure of pollution costs had an impact on investment decision. The second study surveyed institutional investors and found that approximately 57% of the respondents stated that they did consider social factors in addition to economic factors (Ingram, 1978). Balabanis et al. (1998) investigated the relationship between CSR (performance and disclosure) and economic performance (including financial performance measured by return on equity and gross profit to sales ratio, and market performance measured by systematic risk and excess market valuation) based on sample of large 56 UK companies. The empirical results show that CSRD is associated with concurrent financial performance. The results of the study also showed that gross profit to sales ratio was found to influence social disclosure positively. A combination of high CSP (Corporate social performance) and good CSRD was found to have positive effects on overall company’s profitability. A combination of low CSP and good CSRD or high CSP and poor CSRD was found to be not-economically rewarding strategies. On the reaction of the market, they found negative market reaction in the subsequent period to companies with high CSRD. Richardson & Welker (2001) in their study examined the relation between social and financial disclosure and the cost of equity capital based on a sample of Canadian companies. They argued that social disclosure could play a role similar to financial University of Ghana http://ugspace.ug.edu.gh 49 disclosure and reduce the cost of equity capital by reducing transactions costs and/or reducing estimation error. Also, social disclosure could influence the cost of equity capital directly through investors preference effects if investors are willing to accept a lower expected returns on investment that fulfil social objectives. The basic hypothesis is that there is a negative relation between social disclosure, as financial disclosure, and cost of equity capital . The results indicated, in contrast to the hypothesis, that there is a statistically significant positive relation between the level of social disclosure and cost of equity capital. Murray et al. (2006) examined the concerns of capital market over social and environmental disclosures in the UK. The study argued that financial markets are variously seen as offering the biggest single impediment or the greatest possible opportunity for international capitalism to re-invent itself in a new form that is compatible with the exigencies of sustainability. In the absence of an apparent proper regulation, stakeholders, rely on the incentive, and persuasion to encourage markets to act in a manner less incompatible with the social and environmental aims of sustainability. A potential major factor in achieving this ambitious re-direction must inevitably be information and , in particular, information about corporation social and environmental activities. They further argued that with significant growth in ethical investment funds, it is apparent that investors become less obsessed by financial return. Consequently, social and environmental disclosures may well offer an important source of direct input to these ethical investors’ decisions. The results indicated that over a period of time, total social and environmental disclosure is significantly related to market returns even after adjusting for the size effect. University of Ghana http://ugspace.ug.edu.gh 50 Cormier & Magnan (2007) examined the influence of voluntary environmental reporting (in both annual reports and environmental reports) on the stock market valuation of a company’ earnings from two perspectives; first, the North American financial reporting context (Canada), and second, continental European financial reporting context (France and Germany). The results indicated that environmental disclosure has moderating impact on the stock market valuation of German company earnings, and does not influence on stock market valuation of Canadian and French company earnings. They presented three potential reasons for the absence of an impact of environmental disclosure on market valuation of company earnings. First, investors use other sources to be aware about environmental performance. Second, investors may not care about environmental disclosure. And finally, what is being disclosed is too thin to draw any conclusion. Dhaliwal et al (2009) examined the influence of voluntary non-financial disclosure with emphasis on social responsibility disclosure on cost of equity capital. The study adopted various types of analyses to answer some questions; first, whether companies produce social and environmental reports in an attempt to reduce cost of capital. The empirical results in relation to the above question indicate that companies with higher cost of equity are more likely to produce social responsibility reports reflecting positive answer to the question. The second question the study examined whether the producing of social responsibility reports leads to lower cost of equity capital. The empirical results indicated that publishing of social responsibility reports is not associated with lower cost of equity, while social performance indicators are significantly associated with lower cost of equity. This result shows that investors evaluate corporate social performance. The third question the study examined was the mechanism that causes social responsibility reports to be associated with lower cost of equity capital? The results indicate that reporting companies with superior University of Ghana http://ugspace.ug.edu.gh 51 social performance attract institutional investors and have lower forecasts errors and dispersion. The last question the study examined was whether reporting companies exploit the benefit of a reduction in the cost of equity capital by issuing equity capital after CSR reporting. The results indicate that reporting companies are more likely to conduct seasoned equity offerings reflecting a positive answer to the question. The results of this study suggest that investor are more likely to be interested in social performance, and at the same time, it appears that investors are more interested in the firm’s ability to generate adequate returns. Vurro & Perrini (2011) examined a three-year disclosure experience for a sample of Fortune 100 companies in an attempt to test a model that relates the structure of corporate social responsibility disclosures to corporate social performance. With a combination of content analysis and corporate social performance data based on a sample of 38 firms the empirical results show that not only the importance of structuring the report in a comprehensive way, and extending coverage to multiple stakeholders and related issues, but also the need for balance between informative needs, thus avoiding concentrated structures. To this end, companies that report on more themes, presenting a balanced and comprehensive product, develop a better ability to manage their stakeholder network gain higher corporate social performance. Crisostomo et al. (2011) examined the relationship between corporate social responsibility and firm performance taking into consideration firm value and financial accounting performance in an emerging economy of Brazil. The empirical results based on a sample of 234 firms show that CSRD is negatively related to firm value. The relationship between CSRD and accounting performance produced neutral effect. University of Ghana http://ugspace.ug.edu.gh 52 Mosaid & Bouti (2012) examine the relationship between social responsibility disclosures and performance indices of Islamic banks. The empirical results based on a sample of 8 banks showed no relation statistically significant between performance indices (ROA, ROE) and CSRD index. Khlif et al. (2015) examined the influence of corporate social and environmental disclosures on corporate performance in South Africa and Morocco. Using the Tobin’s Q to measure corporate performance and based on a sample of 28 firms in both countries; the empirical results revealed that social and environmental disclosure has a significant positive effect on corporate performance only in South Africa. Previous studies on the consequence of CSRD suffer some limitations based on the review above. The studies that are interested in the analysis of market reaction to CSRD, suffer from similar limitations concerning CSRD as stated in previous chapters, as these studies have concentrated on quantity of disclosure in annual reports to the neglect of Quality of CSRD. While there are many studies that examine the relationship between corporate social performance and corporate financial performance, there are few studies that were concerned with the relationship between CSRD and corporate financial performance and, in particular, corporate market value, which is considered the final result for company management (Hassan, 2010).It is also important to add that these studies have been done in developed economies with few on emerging economies like Brazil and Malaysia. University of Ghana http://ugspace.ug.edu.gh 53 3.10 Hypothesis development It can be argued from the review that the determinants of CSRD are those factors that are related to social pressure. These factors can be differentiated on two levels; First, factors which can determine the degree of social pressure facing companies. These factors include corporate characteristics (size, industry and degree of multinationality, foreign ownership). It can be argued that large industrial companies which have a high degree of multinational activities, and face wide media coverage, have the broader society ’s attention and consequently face the highest degree of social pressure. Therefore, according to legitimacy theory , these companies are more likely to need to legitimise their activities to a greater extent, so a positive relationship between these factors and CSRD is expected. Secondly , factors which can determine the degree of companies’ response to social pressure. These factors include corporate ownership and corporate governance. It can be argued that well-governed companies, with more dispersed ownership, tend to respond positively to social pressure and provide more disclosure. The individual factors and their respective hypothesis are presented below. 2.10.1 Corporate size According to Fifka (2013), corporate size is one of the most examined internal determinants of CSRD across the globe. Literature of CSRD seems to be consistent on the positive significant association between firm size and CSD (Hassan, 2014). Depoers (2000) argues that the significant association between firm size and voluntary disclosure, is based on the fact that larger companies can afford increasing costs for voluntary disclosure, and as University of Ghana http://ugspace.ug.edu.gh 54 such tend to employ highly skilled individuals and sophisticated management-reporting systems. Also, there may be greater demands on larger companies to provide information to analysts and the public. A positive association between corporate size and corporate disclosure is reported in some studies (Riahi-belkaoui, 2001; Ho & Taylor, 2007; Ahmed & Courtis, 1999; Akano et al., 2013, Hassan, 2014). Consistent with legitimacy theory, the expected positive relationship between corporate size and CSRD is dependent on the view that larger companies receive greater attention from society , and will consequently be subjected to more social pressure. Therefore, the following hypotheses can be formulated: H1a: Larger firms provide greater quantity of CSRD than smaller firms H1b: Larger firms provide higher quality of CSRD than smaller firms 3.10.2 Industry type The industry type of type of activity is also one of the most examined internal determinants of CSRD in literature. Husted & Allen (2007) justified this trend and argued that the industry environment has a large effect on corporate adaptation for social strategy . They indicated that one of the key indicators of managerial engagement in social strategy is how managers perceive the terrain of the competitive industrial environment. Industries have different degrees of legitimacy, based on a variety of actions and consequences derived from the collective action of industry members. Many well-established industries have a high level of legitimacy, including sectors such as banking and medicine (Aerts & Cormier, 2006).In line with legitimacy theory, the expected association between industry type and CSRD is dependent on the view that companies activities influence public expectations about the role of the company in the society and the environment as a whole University of Ghana http://ugspace.ug.edu.gh 55 (Aerts & Cormier, 2006; Nasir, 2010). Therefore, the following hypotheses can be formulated: H2a: Manufacturing and mining companies provide greater quantity CSRD than non- manufacturing and mining. H2b: Manufacturing and mining companies provide a higher quality of CSRD than non- manufacturing and mining companies. 3.10.5 Degree of multinational activity Hassan (2014) argued that the degree of multinational activities appears to influence both the expectations concerning corporate social responsibility activities and the level of corporate disclosure. With regards to the impact of multinational activity on CSR, Newson & Deegan (2002) argued that, based on legitimacy theory, companies will respond to the expectations of relevant members of the public, and for multinational corporations, this response is not restricted to the home country, but their global orientation as well. Muller (2006) argues that in the multinational environment, corporations are faced with a potentially divergent home-country , host-country situation, and international pressures that affect their self-regulation strategies. Kolk (2005) also argues that international operations of a company have a substantial impact on the formulation and implementation of its business ethical principles such as codes of conduct. Previous research provides mixed results on the relationship between multinationality and disclosure. For instance, Webb et al. (2008) study showed a positive association between multinationality and social disclosure, while Gelb et al. (2008) study indicated a negative association between the two. Hassan (2014) study revealed that the degree of multinational activity is not related with CSRD. Also, previous literature concerning CSRD University of Ghana http://ugspace.ug.edu.gh 56 revealed little examination of the impact of multinationality on CSRD, compared with other corporate characteristics. Toms et al. (2007) and Toms (2008) found no relationship between the number of foreign countries and CSRD. Branco & Rodrigues (2008) found no relationship between the ratio of foreign sales and CSRD, while Stanny & Ely (2008) found an association between them. Based on previous literature, it can be argued that due to their geographical extension, multinational companies are more likely to face greater social pressure. This geographical extension creates more pressure from host societies on multinational companies, with regards to their social responsibilities. It can therefore be argued that the more foreign countries in which the company operates, the more pressure there is on the company . Consequently, the company could increase the level of CSRD as a tool to face this pressure, and legitimise its activities, so the following hypotheses will thus be examined: H3a: Firms with more multinational activities will provide greater quantity of CSRD H3b: Firms with more multinational activities will provide high quality of CSRD. 3.10.6 Board size According to Hassan (2010) there is a theoretical debate surrounding the size of a board of directors. Ho & Williams (2003) posits that while some researchers argue that a larger board promotes more effective decision-making and enhances information-processing capabilities, others argue that a larger board leads to less participation among members, and increases the opportunity for manipulation on the part of corporate management. While the board’s monitoring capacities increase as the board size (number of members on the board) increases, the incremental costs of poorer communications, that are often associated with large groups, may offset this benefit (Hassan, 2014). In addition, University of Ghana http://ugspace.ug.edu.gh 57 Cheng & Courtenay (2006) argues that there is no consistent evidence to suggest a relationship between corporate board size and voluntary disclosure. Their study found that there is no empirical association between board size and voluntary disclosure. With regards to the expected impact of board size on CSRD, Halme & Huse (1997) posits that with a large board, there is a higher probability of a broader range of stakeholders, which indicates that a higher level of environmental attention can be expected. In line with this view, it can be argued that increasing the number of directors on the board could provide better communication with the community, and consequently more probability that companies will react positively to social pressure. Given this theoretical and empirical debate, the following hypotheses will be examined: H4a: Firms with more directors on the board provide greater quantity of CSRD H4b: Firms with more directors on the board provide a higher quality of CSRD 3.10.7 Board Composition According to Hassan (2014) non-executive directors are considered a governance mechanism that enhances the board’s capacity to ameliorate agency-conflicts between owners and managers. These conflicts occur mostly in the decision to voluntarily disclosed information in annual reports (Barako et al., 2006). Chen (2006) indicates that in the US, non- executive directors are shown to play a more important role in monitoring managers, than executive board directors. In addition to the above, Ajinkya et al. ( 2005) posits that non-executive directors can play an important role in determining and monitoring voluntary corporate disclosure. Anderson & Reeb (2004) argued that independent directors can defend the minority shareholders by protecting their rights against large- shareholders’ opportunism. To this end, independent directors play an important role in University of Ghana http://ugspace.ug.edu.gh 58 balancing the interests of competing shareholders and act as an influential governance mechanism in protecting outside shareholders from large shareholders’ expropriation (Hassan, 2014). Previous research provides mixed evidence on the relationship between disclosure and the independence of the board of directors. For instance, Beasley (1996), Chen & Jaggi (2000) and Xiao et al. (2004) found a positive association between disclosure and the independence of the board, while Eng &Mak (2003) and Gul & Leung (2004) find a negative relationship. It can be argued that the presence of non-executive directors on the board is a tool to link the company with various stakeholders and the community as a whole, and therefore they represent one of the factors that drive the company to deal with the community ’s concerns regarding social responsibility (Hassan, 2010). Rose (2007) argued that new regulations, requiring more independent directors, are a major step in improving corporate ethics and social responsibility. Thus, the increasing percentage of non- executives directors on the board will encourage companies to react positively to social pressure, and consequently to increase the level of CSRD. Therefore, the following hypotheses can be formulated: H5a: The quantity of CSRD is positively related to the number of non-executive directors. H5b: The quality of CSRD is positively related to the number of non-executive directors. 3.10.8 Presence of Corporate social committee The existence of the corporate responsibility committee as a board committee is, in itself, a sign of the company’s interest in social responsibility (Hassan, 2014) . This committee reflects the company’s desire to perform its activities in line with social-responsibility guidelines and rules. It can be argued that the presence of the corporate responsibility committee, as one of the board committees, is one of the factors that may drive University of Ghana http://ugspace.ug.edu.gh 59 companies to react positively to social pressure concerning the company’s social responsibility and, consequently , to increase the level of CSRD. H6a: The quantity of CSRD is positively related to the presence of corporate responsibility committee. H6b: The quality of CSRD is positively related to the presence of corporate social responsibility committee 3.10.9 Corporate ownership diffusion Agency theory supports the argument that ownership-diffusion is positively related to corporate disclosure (Hassan, 2014). This means that the more diffuse the ownership, the more there is corporate disclosure because this helps owners to monitor the behaviour of management. Previous studies have established a negative relationship between block ownership and disclosure (Mitchell et al., 1995; Schadewitz & Blevins, 1998; Kelton & Yang, 2008; Hassan, 2014). This implies that a greater percentage of substantial shareholder-ownership, leads to less need for monitoring and transparent disclosure. Reverte (2009) supports this view when he argues that companies with widely -held shares are more likely to improve their financial reporting policy by using their social responsibility disclosure. On the other hand, companies with concentrated ownership are less motivated to disclose additional information on their CSR. It can be argued that more ownership- diffusion encourages management to react positively to social pressure, by increasing the level of CSRD to acquire owners’ satisfaction, and consequently the negative association between CSRD and block ownership. H7a: The quantity of CSRD is negatively related to block ownership H7b: The quality of CSRD is negatively related to block ownership University of Ghana http://ugspace.ug.edu.gh 60 3.10.10 Consequence of CSRD on market value Rahman (2002) posited that corporate voluntary disclosure is one of the major determinants of market value. This he argued was in line with both internal and external corporate governance factors. Hassan (2010) also argued that theoretically, an increased level of disclosure (either increased quantity of disclosure, or quality of disclosure, or both) reduces the information asymmetry between the company and its shareholders, or among potential buyers and sellers of company shares. This in effect should reduce the discount at which company shares are sold and hence lower the costs of issuing capital (Leuz & Wysocki, 2008). Plumlee, et al. (2007) argued that voluntary disclosure quality, influences firm value through direct effects on a company’s cost of equity capital, and/or indirect effects on a company’s cash flow. Fauzi, et al. (2007) argued that corporate voluntary disclosure is considered one of the determinants of market value, with internal and external corporate governance factors. In previous literature, the relationship between CSRD and the economic performance of a company was studied from two points of view. One examined economic performance as a determinant of CSRD, and the other examined economic performance as a result of CSRD. The empirical results of the studies provided inconclusive evidence about the relationship between CSRD and economic performance. Given the ambiguous theoretical opinions and empirical results, the null hypothesis appears to be most appropriate, and more consistent with legitimacy theory . H8a: The quantity of CSRD has no impact on the corporate market value. H8b: The quality of CSRD has no impact on the corporate market value. University of Ghana http://ugspace.ug.edu.gh 61 CHAPTER FOUR RESEARCH METHODOLOGY 4.0 Introduction Determining the research methodology, deciding on the relevant philosophical assumptions and methodological choices, and constructing the appropriate research design, portray the necessary steps for proceeding in an empirical study .This chapter presents three empirical models that were used to examine the framework of CSRD determinants and consequence. The first two models examined the determinants of social responsibility disclosures at the firm level whiles the last model examines the consequence of CSRD. Each model includes several variables which will be measured through secondary data. The focus of the framework is CSRD and as such its measurement is very critical. The measurement of CSRD is particular in annual reports is a controversial subject according to previous studies. To this end, this chapter focuses on clarifying ways of measuring CSRD. 4.1 Research philosophy and design Research process builds on development of knowledge of the subject matter under study. Saunders et al. (2011) presented three views on research philosophy, namely; interpretivism, realism and positivism. Interpretivism posits that to develop knowledge, interpreting and understanding the situation being studied is required in order to appreciate the motives, actions and intentions of the research participants. According to this view, knowledge is developed from the understanding that organizations experience frequent changes that makes what was relevant yesterday irrelevant to the same organization in the future. The second view is realism. Realism develops knowledge on the premise that external social factors and University of Ghana http://ugspace.ug.edu.gh 62 processes independent of human thoughts and beliefs exists, which make people interpret situations differently and arrive at different conclusions on the same subject matter, without their knowledge of existence of such factors. The last one is positivism. Positivism assumes that the researcher is independent of, and neither affects nor is affected by the subject of the research. Another assumption is that, the researcher is an objective analyst who interprets collected data in an unbiased manner emphasizing the use of highly structured method and quantifiable observations that uses statistical analysis. Under positivism approach, knowledge is built from using quantitative data which undergoes statistical processing, analysis and interpretation. Since the causal relationships between the dependent and independent variables of CSRD and its determinants as well as consequence are to be tested and analysed statistically, positivism approach is selected for the development of knowledge of this study. With the research philosophy for the study in place, the choice of research approach is the next logical step. Saunders et al. (2011) identifies two research approaches: deductive and inductive. Inductive approach first collects data and develops theory based on the results of data analysed. The deductive approach on the other hand explains causal relationships between variables. Research hypothesis are developed and research strategies designed to test them. Quantitative data is used in testing hypotheses even though qualitative data can also be used. In deductive approach, the researcher is independent of observed phenomenon and the research uses highly structured methodology to facilitate replication (Gill and Johnson, 2010). A further argument is that the concepts of the research are operationalized to measure quantitatively the relationship between relevant variables. This study establishes the causal relationships between the dependent and independent variables of CSRD through quantitative operationalization of the identified variables. The quantified observations are processed and University of Ghana http://ugspace.ug.edu.gh 63 the results measured statistically to obtain evidence on the relationships between the variables. Quantitative research is an approach for testing objective theories by examining the relationship among variables (Creswell, 2013). These variables can be measured using instruments so that numbers can be analysed using statistical procedures. In sum, this study adopts the positivism philosophy through a deductive approach using quantitative approach for testing the variables identified to establish the relationship that exist among them. 4.2 Data and source The study relied on secondary data from the annual reports of listed firms in Ghana. The data used for this study was collected through “content analysis “of annual reports of these companies. A study of prior literature on social responsibility disclosures revealed that majority of studies on corporate social responsibility disclosure used content analysis of annual report (Ernes & Ernest 1978; Gray et al 1995; Harkston and Milne 1996; Gurthrie and Parker 1990 Ratanajongkol et al., 2006, Hassan, 2014; Khlif et al. 2015). Santaon & Santon (2002) defined the annual reports as a formal public documents produced by companies in response to the mandatory corporate reporting requirements existing in most western economies. The annual report is sometimes referred to as a company business card which provides readers with a comprehensive picture of the publishing organization (Daub, 2007). Nasir (2010) argues that the annual reports are commonly divided into two; one for statutory required matters and the other for non-statutory matters. It has been argued that content changes in annual reports have been partly mandatory and partly voluntary. For instance, Nasir (2010) argues that in the UK, the mandatory content increased rapidly between 1970 and 1990 as a result of changing demands from several University of Ghana http://ugspace.ug.edu.gh 64 regulatory bodies. Santon & Santon (2002) stated that voluntary disclosure increased as the corporate report moved toward being a public relations documents. Research has also shown that compared with one or two decades ago, annual reports are more professional with colourful graphs and diagrams, fancy designs and exotic paper. This is because; there is a growing realization that annual reports have a disclosure function which can serve as a good public relations function for companies (Hassan, 2010). White & Hanson (2002) justifies the extensive use of annual reports by researchers when they stated that no other medium offers the same blend of consistency, accessibility and wide applicability like the annual report. They further argue that no other medium yields the same access to corporate communication with key audience. Majority of research in the area of CSRD use the annual report as the principal focus of disclosure and some of the justifications are stated below: According to Yousof & Lehman (2005), annual reports are the most accessible source of information for listed companies in hard copies and electronically. Tilt (1994) argues that the annual report provide a high degree of credibility to information disclosed within them. The annual report is a statutory document and produced regularly, and it presents what are probably the most important documents which companies used in construction of its own social imagery (Hassan, 2014). Gray et al. (1995) sums it all up when they stated that the financial image of a company is critical in terms of how companies are seen and judged, the social and environmental factors will produce conflict with financial ambitions of the company and its owners. This therefore means that the presentation of financial information within the same report becomes an important element in demonstrating how the company reconciles these matters. University of Ghana http://ugspace.ug.edu.gh 65 4.3 Sample and selection criteria The study sampled firms listed on the Ghana stock exchange. The study chose listed firms because they are the largest firms in Ghana and according to Brammer and Pavelin (2006), the larger the firm, the more likely they will disclose voluntary social information and also because of public availability of data. The sample was drawn from listed companies because complete data was relatively available for listed firms as compared to their non-listed counterparts. Copies of the annual reports of listed firms are deposited at the stock exchange and one could secure copies. Also, according to Dhaliwal et al. (2009), one great characteristic in corporate disclosure is that a company generally provides information to release specific obligations: to society, investor, supplier, creditors and legal authorities. These stakeholders are more demanding for listed firms than non-listed firms and as such the choice of the sample frame. Fu et al. (2012) also argue that financial statements filed with the Securities and Exchange Commission (SEC) are accorded a substantial degree of prominence and attention than other firm and hence the need to use listed companies. At the moment there are 35 companies listed on the Ghana stock exchange. The selection criteria were based on firms listed on the Ghana stock exchange from the year 2008. This is because research has established that the adoption of IFRS leads to an increase in voluntary disclosures (Ashbaugh, 2001; Ashbaugh and Pincus, 2001; Leuz and Verrecchia, 2000) and all companies on the stock Exchange fully complied with the adoption by the end of 2008. Since social disclosures are also an aspect of voluntary disclosure, it is important to examine companies’ social disclosures after the adoption of IFRS. In all, 33 firms made the sample selection criteria over a six year period from 2008 to 2013. This added up to 198 sample year observations. University of Ghana http://ugspace.ug.edu.gh 66 4.4 Research approach and execution strategy The research method that is commonly used in literature to assess the quantity of CSRD is content analysis. According to Krippendorf (1978), content analysis can be defined as “a research technique for making replicable and valid inferences from data according to their context”. The paper further argues that, the main characteristic of content analysis is that, data are coded and measured in a reliable and systematic manner. Abbott (1979) defines content analysis as a technique for gathering data that consist of codifying qualitative information in anecdotal and literacy form into categories in order to derive quantitative scales of varying levels of complexity. Studies in this area use a disclosure index as a guide to quantify social information. To achieve this objective, the study also adapted a comprehensive disclosure index from literature to quantify CSRD. Based on some previous studies, the study adapts the disclosure index by Hackston & Milne (1996) as adapted by Deegan et al. (2002) and Hassan (2014) in similar studies. These authors argue that, this index is the most comprehensive in literature and were applied in a similar study in New Zealand. To operationalise the index, a pilot study was conducted to examine the reliability of the disclosure index. To achieve this, the study randomly sampled companies from each industrial sector listed on the Ghana stock exchange. A modified social responsibility disclosure was constructed based on information from annual reports of listed firms on the Ghana stock exchange which was used to measure CSRD for the study. The modified disclosure index is presented in the appendix. After the index, the study determined the unit of analysis which served as the basis for measuring CSRD. Corporate social responsibility disclosure studies have used various units of analysis including number of character, words, sentences, pages and proportion of volume University of Ghana http://ugspace.ug.edu.gh 67 of CSRD to total disclosure (Hassan, 2014). Recent studies have indicated the weaknesses in using number of words as words in themselves do not convey any meaning (Hackstone & Milne, 1996; Nasir, 2010; Hassan, 2014). These studies advocate for the use of sentence, graphs and photos to quantify CSRD. For the purpose of this study, the measurement of the quantity of CSRD in annual reports was done using the number of sentences as well as graphs and photographs that represent CSRD disclosures. The objective of this choice is to reduce the weaknesses in relying on only written words to the neglect of photos which is argued to be an effective way of CSRD. Also, the study did not use the number of pages because CSRD reports in annual reports are not always presented in a structured and presented in a stand- alone manner. Sometime they are contained in the directors’ reports or other reports which cannot be isolated in terms of number of pages (Hassan, 2014). On the measure of the quality of CSRD, previous studies have relied on scoring method to measure the quality of CSRD (Al-Tuwajiri et al., 2004; Hammond & Miles, 2004; Cormier et al., 2005; Hassan, 2014). These studies have argued that social information that is quantified, graphed, or disclose a specific measure that has been taken by companies is of higher quality. From the discussion above and in line with previous literature and especially the work of Hassan (2014), the study used a 2-point scale system to assess the quality of social disclosure in annual reports which was presented as follows; 1, if disclosure is quantified, graphed or narrative disclosure which reports the policies and activities of the company with respect to social responsibility. 0, otherwise University of Ghana http://ugspace.ug.edu.gh 68 The quality score was measured by evaluating each sentence on social disclosures in accordance with previous rating by Hassan (2014) and calculated as the average score using (total score/ number of sentences). 4.5 Empirical model estimation The study employed the following econometric models to test the hypotheses formulated CSRDarsit= ß0 +ß1CSit +ß2PROit +ß3TAit +ß4DMAit + β5BSit+ß6NEDpit +ß7SRCit + ß8SSit + ß9OWNit+ εit ………………… (1) CSRDarqit= ß0 +ß1CSit +ß2PROit +ß3TAit +ß4DMAit + β5BSit+ß6NEDpit+ß7SRCit + ß8SSit+ ß9OWNit + εit ……………………… (2) TQit =ß0 +ß1CSRDarsit +ß2CSRDarqit + β3CSit +ß4AFMit +ß5LEVit +ß6PROit +ß7IGit +ß8IDit +ß9DMAit + ß10AUDit + εit ……………… (3) Where; Variable Meaning Measurement TQ Firm value Measured by Tobin’s q CSRDars Quantity of CSRD No. of sentences & photographs CSRDarq Quality of CSRD Total score/ No. of sentence CS Corporate size Log of total assets TA Type of Activity (Industry) Dummy, 1 if firm is manufacturing or mining, 0 otherwise. DMA Degree of multinational activity Number of countries a firm operates University of Ghana http://ugspace.ug.edu.gh 69 BS Board size Number of directors on the board NED Non-executive directors Proportion of non-executive directors SRC Social responsibility Committee Dummy, 1 if there is SRC committee on the board, 0 otherwise SS Substantial shareholders (ownership diffusion) Percentage of shares held by substantial shareholders (3% or more) OWN Foreign ownership Dummy, 1 if the firm is foreign owned 0 otherwise. AFM Access to financial market Dummy, 1 if firm paid dividend, 0 otherwise LEV Leverage Total debts/ total assets PRO Corporate profitability Return on assets IG Investment growth Ratio of capital expenditure to sales ID Industry diversification Dummy, 1 if firm operates in more than one industry, 0 otherwise. AUD Auditor type Dummy, 1 if firm is audited by big 4, 0 otherwise. Tobins q = Market value of assets Book value of assets 4.6 Methodological issues Regression models that examine the dependent variables CSRDars, CSRDarq, TQ, can be conducted using the linear regression model, which is considered to be the most common method in disclosure literature. Without verifying that the data have met the assumptions underlying OLS regression, the results may be misleading. A number of University of Ghana http://ugspace.ug.edu.gh 70 assumptions underlie OLS regression; normality, homogeneity of variance (homoscedasticity) and collinearity. 4.6.1 Normality of residuals This assumption refers to the fact that the residuals (errors) should be normally distributed. Normality of residuals is required for assurance that the P-values for t-tests and F-test are valid. The normality assumption is not required in order to obtain unbiased estimates of the regression coefficients (Hassan, 2014). This assumption can be examined using the Stata programme as follows: After conducting regression analysis, the predict command was used to generate residuals (predict r, resid). Then, the kdensity command was used to create a kernel density plot with the normal option(density r, normal). The plot indicated that the data is not normally distributed. The remedy for this is to include dummy variables and increase sample size. 4.6.2 Heteroskedasticity Heteroskedasticity simply means that the error variance should be constant, as one of the main assumptions for OLS regression is the homogeneity of the variance of residuals. If the variance of the residuals is non-constant, then the residual variance is said to be heteroskedastic. To examine the heteroskedasticity problem using the Stata programme, after conducting regression analysis, the command hettest was used to run the Breusch-Pagan test. The null hypothesis of this test is that the variance of residuals is homogenous, so if the P-value is small the null hypothesis will be rejected, and will accept the alternative hypothesis that the variance is heteroskedastic. The results from the test for the Breusch-Pagan test showed a significant chi2 for all the models which indicates the presence of the heteroskedasticity. To resolve this problem using University of Ghana http://ugspace.ug.edu.gh 71 stata, the command for the regression included robust which will eliminate the problem of heteroskedasticity. 4.6. 3 Multicollinearity Multicollinearity means that the independent variables are correlated, which can cause problems in estimating the regression coefficients. The possible existence of multicollinearity is tested based on the correlation matrix incorporating all the independent and control variables. Both Pearson and Spearman 's rank correlation matrices show that correlation coefficients are less than 0.8, the limit or cut off correlation percentage commonly suggested by prior studies after which multicollinearity is likely to exist (see Gujarati, 2003). To further assure the absence of multicollinearity, the study examined the multicollinearity problem using the Stata programme, after conducting regression analysis, the command VIF will be used. The following table provides the values of VIF which are less than 2 meaning no problem for multicollinearity (Hassan, 2014). These values of VIF indicate a limited problem of multicollinearity. Table 4.6.3: VIF of independent variables Variable VIF 1/VIF CS 1.90 0.5263 PRO 1.13 0.8850 TA 1.15 0.8696 DMA 1.10 0.9091 BS 1.50 0.6667 NED 1.15 0.8696 SRC 1.14 0.8772 University of Ghana http://ugspace.ug.edu.gh 72 SS 1.20 0.8333 OWN 1.36 0.7353 MEAN VIF 1.29 Finally, the study concluded on the methodological issues by deciding on using a fixed effect model or random effect model. In stata, after running each regression model for both fixed and random effect, you type the command estimate store fixed or random. Then you type Haussmann fixed, random. If the prob>Chi is significant at a 5% significance level, we use fixed effect and vice versa. After conducting the Haussmann test, all the three models had settled on the random effect model because it’s provided more consistent results using the R2 values and the outcome of the hausman test in line with previous studies in the area (Hassan, 2014, Khlif et al., 2015). The table below presents a summary of the outcome of the hausman test Table 4.6.4: Results of Hausman test Model 1 Chi2 (9)= 2.29 0.9860 Model 2 Chi2 (9)= 2.81 0.9455 Model 3 Chi2 (9)= 4.31 0.8901 University of Ghana http://ugspace.ug.edu.gh 73 CHAPTER FIVE ANALYSIS AND DISCUSSION OF FINDINGS 5.0 Introduction The analysis of the finding is divided into three sections. The first section provides a descriptive statistics of CSRD which is aimed at addressing the first objective of the study. The section will seek to highlight key issues relating to the nature of CSRD disclosure is Ghana, the quality of those disclosures as well as the industry and other key variables. The second section used the statistical model of correlation and panel regression to examine the theoretical model of determinants of CSRD at the firm level. The objective of the section is to present an analysis of the determinants of CSRD using various variables and how those determinants relate to quantity and quality of CSRD. The last section will present an analysis to explain the consequence of CSRD in terms of both quality and quantity on firm value. The overall objective is to use legitimacy theory to explain the determinants of CSRD and conclude that indeed, legitimacy theory provides an appropriate theoretical background for explaining CSRD in Ghana. 5.1 Descriptive statistics of CSRD Table 5.1 Descriptive statistics on CSRD Variable obs Mean std. Dev. Min Max CSRDars 198 42.36869 65.25227 0 406 CSRDarq 198 0.6407531 0.2700218 0 0.948718 Source: Computation from research data, 2015 University of Ghana http://ugspace.ug.edu.gh 74 Table 5.1 provides a descriptive statistics of CSRD for the period under study. With regards to the quantity of CSRD in annual reports (CSRDars), the results show that the minimum disclosure of CSRD in annual report is 0. This indicates that some companies did not provide any social information at all in their annual report for the study period. In all, 9% of the companies sampled did not provide any social information in their annual report reflecting that 91% provided some form of CSRD in their annual report. This percentage is very high especially in the context of developing economies as Hassan (2010) argues that few companies in developing economies provides CSRD. In relation to previous literature, a study by Guthrie & Parker (1990) found that 98% of UK firms provide CSRD as against 85% and 56% of United States and Australian firms respectively. However, in recent times, Hassan (2014) found that 99% of UK firms provide CSRD in their annual report. Andrew et al. (1989) found that of 119 publicly-listed companies in Malaysia and Singapore, only 26% of the firms provide CSRD. Lynn (1992) found that only 17 (6.4%) of 264 companies in Hong Kong provide any CSRD information in their 1989 annual reports. The quantity of CSRD was measured using the number of sentences and photographs that represent CSRD. The number of sentences and photograph of disclosure of CSRD in annual report was between 0 and 406 and an average of 42. Hackston & Milne (1996) study in New Zealand had an average of 23.4. However, Hassan (2014) study of UK firms had an average of 102 and a maximum of 691. The result confirms the assertion that CSRD in developing economies is low as compared to developed economies. With regards to the quality of social disclosures in annual reports (CSRDarq), the score of disclosure quality is between 0 and 0.95 and an average score of 0.640. This implies that on average, majority of social disclosures in annual reports in Ghana comprises specific activities and not general activities. The maximum score of 0.95 is below that of firms in UK University of Ghana http://ugspace.ug.edu.gh 75 with a score of 1 but the average score is above that of UK firms which have an average score of 0.43 (Hassan, 2014). The study also analysed the trend of CSRD disclosure over the six year period. The figure below is a graph showing the mean disclosures of social information from 2008 to 2013. Figure 5.1: CSRD over the study period Source: Research data, 2015 The result from figure 5.1 above shows that, social disclosures have trended upward over the study period. This suggests that companies in Ghana are getting more interested in social disclosures as the years go by. 5.2 Analysis of CSRD in annual reports To obtain a better understanding of CSRD in annual reports, table 5.2 provides descriptive statistics for the categories of CSRD in annual reports. It is expected that environmental information could be regarded as privileged information, due to global concerns regarding environmental issues (Hassan, 2014). Owen (2003) posits that the growing practice of environmental reporting among companies throughout the world is not surprising, due University of Ghana http://ugspace.ug.edu.gh 76 to the wide variety of influences that have been brought to bear on the process, from a myriad of sources. Table 5.2: Categories of CSRD in Annual reports Variable obs Mean std. Dev. Min Max ENV 198 4.909091 10.1262 0 60 COMM 198 8.060606 11.34222 0 68 EMPL 198 5.681818 10.28995 0 60 CUST 198 1.436548 2.92984 0 20 ETHICS 198 2.611111 7.928953 0 75 PROD 198 1.857868 3.060629 0 13 H&S 198 2.050761 4.956696 0 45 OTHERS 198 2.328283 4.421717 0 26 Source: Computation from research data, 2015 The results from table 5.2 show that community information (with an average mean of 8.06 was readily available contrary to expectation by Hassan (2014) that environmental information should dominate. The second category is employee information with an average mean of 5.68. In a study by Hassan (2014) using UK firms’ employee information was found to be more readily available. This is followed by environmental information with an average mean of 4.91 sentences and photographs. The category with the least disclosures is product with a mean of 1.86. The results are to some extent consistent with previous studies. For instance, Hackstone & Milne (1996) study in New Zealand found that Human resources or University of Ghana http://ugspace.ug.edu.gh 77 employee information and community information were the most disclosed category. Rizk et al. (2008) found that employee-related information was the most disclosed among Egyptian firms. Lynn (1992) also found employee information to be the most disclosed category in Hong Kong. Guthrie and Parker (1990) found that CSRD in US, UK and Australia were: 40% human resources, 31% community involvement, 13% environment, 7% energy and products and 2% other. Sobhani et al. (2009) found that human resource information is the most important information in Bangladesh companies. The results suggest that firms on the Ghana stock exchange are more concerned about giving back to society and satisfying their employees rather that reporting on the environmental impact of their operations. This also suggests that firms on the Ghana stock exchange care less about climate change concerns. Haron (2004) in his study concluded that between the periods 1979 to 1991, CSRD information had increased by 4 times and this he attributed to changes in disclosure regulations, which made disclosure of some themes mandatory. The study further argued that the importance of disclosure themes has changed. Employee-related disclosure dropped from approximately 90% to about 78% and community and environmental reporting rose from approximately 10% to 32%. Hasnah et al. (2006) also noted that the increase in health and safety disclosure is associated with the general rise in environmental concerns, and the increase in major widely-publicised accidents. A very low level of energy and customer information was found, regardless of the need for the economical and efficient use of energy (Haron, 2004). University of Ghana http://ugspace.ug.edu.gh 78 5.3. The incidence of CSRD categories Table 5.3 Incidence of CSRD categories Variables Frequency percent Environment 69 38.85 Community 127 64.14 Employee 108 54.45 Customers 56 28.28 Ethics 55 27.78 Product 68 34.34 Health and safety 62 31.31 Others 78 39.40 Source: Computation from research data, 2015 The results from table 5.3 above show that a higher number of companies are interested in community information than other categories. It also provides additional evidence that companies are more concerned about community, employees, environment, health, safety and other information, while they are less concerned about product, customer and ethical information. This result is inconsistent with similar study by Hackstone & Milne (1996) in New Zealand where most companies disclosed human resource information. In the study of Sobhani et al . (2009) in Bangladesh, 100% of companies were found to disclose at least one item of disclosure on human resources, with other categories as follows: community issues (47%), consumer issues (23%) and environmental issues University of Ghana http://ugspace.ug.edu.gh 79 (19%). Imam, (2000) found that all companies in Bangladesh, provided employee information, 25% of the sample companies provided community, and 22.5 %, environmental disclosure. Only 10 per cent of companies provided consumer-related disclosure. The incidence rates of CSRD categories in Malaysian companies show that the employee category is the most important category, as 42% of companies in 1996 and 43% of companies in 2002, disclosed one or more items concerning employees. The incidence for other categories shows: environment category 11% and 17% in 1996 and 2002 respectively, community category 9% and 21% in 1996 and 2002 respectively and product category 16% and 17% in 1996 and 2002 respectively (Haniffa & Cooke, 2005). Saleh (2009) found that human resources disclosure has the highest disclosure level and environmental disclosure was less disclosed in Malaysian companies. Abu-Baker & Naser, (2000) found that the majority of Jordanian companies have disclosed social responsibility information on human resources. The results therefore suggest that community information is the most important category from the perspective of quantity of disclosure, which is consistent with various studies in other countries. This is not consistent with various studies in other countries. The proportion of companies that make some form of environmental disclosure is 54%, and this percentage is lower than has been found in some earlier studies in the UK, e.g. The figure of 75% found by Gray et al . (1995) and 57% found by Brammer & Pawelin (2006), reflecting the increasing interest of UK companies in environmental issues. The figure below shows a graphical display of the above discussion University of Ghana http://ugspace.ug.edu.gh 80 Figure 5.3: Disclosing Companies CSRD for each category Source: Research data, 2015 5.4 Disclosure location in Annual reports According to Gray et al. (1995) the disclosure location of CSRD has received little attention in previous literature. Yusoff & Lehman (2005) argued that the study of disclosure location may contribute positively to the literature. To analyse the selection of locations in annual reports for disclosing social responsibility information, the annual reports are divided into six sections: chairman report (CR), chief executive statement (CEO), business strategy (BS), directors’ report (DR), corporate governance report (CG) and a specific section for corporate responsibility disclosure (CSR) as used by (Hassan, 2014). Figure 5.4 below presents the disclosure location in annual report over the sample period. University of Ghana http://ugspace.ug.edu.gh 81 Figure 5.4: Disclosure location in the annual report Source: Research data, 2015 The results from figure 5.4 above show that the most important location for social responsibility disclosures for firms listed on the Ghana stock exchange (GSE) is the separate section for social reporting, accounting for 42% of the disclosure location. This is followed by the Chairman report accounting for 24% of the disclosure location; Chief executive officer Report is next with 15%. The locations with the least disclosures are the corporate governance report with 5% and directors report with 4% of the disclosure location. The finding is inconsistent with the study by Guthrie & Parker (1990) and Hassan (2014) studies in the UK where directors’ report was the most location for social responsibility reporting in the annual report. The devotion is a separate section for CSRD shows the importance listed firms put on CSRD. Comparing this result with Harte & Owen (1991), who analysed the annual reports of 30 companies, where it was found that 10 reported on environmental matters in a separate section of their annual reports, indicates increasing attention to University of Ghana http://ugspace.ug.edu.gh 82 providing this separate section in UK companies which is consistent with the findings of this study. This result is also consistent, to some extent, with the result s of Holland & Foo (2003), which indicated that the separate section is the most noticeable means of disclosure in UK companies (58% of UK companies have done so, compared with only 28% of US companies), and with the study of Yusoff & Lehman, (2005) which indicated that director’s report and separate section tended to be the favoured locations for Australian companies and a review of operations and separate section for Malaysian companies. The use of the chairman section for social disclosure is because in most cases being a non- executive director, it is expected that they will be more interested in corporate responsibility issues. In addition, both directors statements and corporate governance reports are less interested in corporate responsibility information. The reason why the corporate governance section and directors reports were the least section for reporting CSRD is because, the content of these sections are to some extent regulated. For instance, the Companies Act of 1963, Act 179 specifies information to be included in the directors’ report. 5.5 CSRD in different sectors To analyse CSRD according to different sectors, the sample companies are divided into four categories based on the study by Hassan (2014). 1. Industry (manufacturing) sector – this sector comprises the following industries: automobiles & parts, beverages, chemicals, construction & building material, diversified industrials, electricity , electronic & electrical equipment, engineering & machinery, food producers & processors, forestry & paper, mining, oil & University of Ghana http://ugspace.ug.edu.gh 83 gas, personal care & household products, pharmaceuticals, real estate, steel & other metals. 2. Financial sector – this sector comprises the following activities: banks, insurance, investment companies, life assurance and speciality & other finance. 3. Services sector – this sector comprises the following activities: business support services, computer software services, gas distribution, health, information technology, leisure, entertainment & hotels, media & photography, packaging, restaurants, pubs & breweries, support services, telecommunication services, transport and travel & leisure. 4. General retailers sector – this sector comprises the following activities: food & drug retailers and general retailers. Figure 5.5 provides the average score of disclosure variables according to different sectors. Generally, it is expected that, based on previous literature, the industrial companies could produce a higher quantity of social disclosure. Contrary to this expectation, Chambers et al. (2003) expected the services sector to be more associated with CSR than the agricultural and industrial sector, because services companies tend to be more conscious of their customer image. University of Ghana http://ugspace.ug.edu.gh 84 Figure 5.5: Analysis of CSRD in different sectors Source: Research data, 2015 The results from figure 5.5 shows that manufacturing or industrial companies provides more CSRD information than their financial and service counterparts. It appears that due to the nature of their activities, industrial companies could face more public questions about the impact of their activities on the environment and community as a whole, so they provide more social disclosure in annual reports. This finding is consistent with the study by Hassan (2014) where he concluded that industrial companies provide greater quantity of CSRD than other sectors. The financial institutions also provide more social information than the general retailers and the service firm. The reason for this trend is because the financial institutions are among the most profitable institutions in Ghana and as such have higher societal expectation to give back to society. Akano et al.(2013) argued that financial institutions in Nigeria disclose CSR information because; first, the citizen perceived that banking sector is earning abnormal University of Ghana http://ugspace.ug.edu.gh 85 returns on investment, hence their believe that they should give back part of that abnormal earnings to the citizen through corporate socially responsible activities. Besides financial institutions enjoy high visibility and are very sensitive about the corporate image. 5.6 CSRD categories in different economic sectors This section presents further analysis of CSRD in different economic sectors. Table 5.6 below provides the incidence of disclosures in different economic sectors. Tagesson, et al, (2009) present some expectations concerning CSRD in different economic sectors as follows: Manufacturing companies negatively influence the environment and consequently disclose more environmental information than companies in other sectors. Finance and services companies disclose little information on social and environmental issues, while mining companies, oil companies and chemical companies maintain a leading position regarding such reporting. Mining, oil and chemical companies emphasise information regarding environmental, and health and safety issues, while the finance and services companies seem to report more on social issues and philanthropic deeds (Tagesson, et al ,2009). University of Ghana http://ugspace.ug.edu.gh 86 Table 5.6: The incidence of disclosure categories in different economic sectors Financial Manufacturing Gen. Retailers services ENV 19(27.54%) 45(65.22%) 5(7.25%) 0(0%) COMM 55(43.31%) 45 (38.58%) 19 (9.45%) 11(6.66%) EMPL 40 (37.04%) 52(48.15%) 10(9.26%) 6(5.56%) CUST 30(53.57%) 13(23.21%) 7(12.50%) 6(6.91%) PROD 29(42.65%) 29(42.65%) 10(14.71%) 0(0%) H&S 8(12.90%) 47(75.81%) 7(11.81%) 0(0%) OTHERS 17(21.79%) 38(48.72%) 9(11.54%) 14(17.95%) Source: Computation from research data, 2015 The result shows that most of the disclosures are done by either manufacturing or industrial companies or financial companies. Manufacturing companies disclose more environmental information that their other counterparts. Also, they are more interested in employee information which their disclosure constitute 48.15%, health and safety which constitute 47.81% and health and safety which constitute 75.81% of the disclosure in that category. The result shows the impact of manufacturing companies on the society and the need to legitimize their activities by disclosing more information on environmental, employees, health and safety etc. For the financial sector, they are more interested in community information with their disclosure making 43% of the total disclosure in that category. They are also interested University of Ghana http://ugspace.ug.edu.gh 87 customer information where their disclosure constitutes 53% of the total disclosure in that category. At the same time, the other companies are less interested in the product information category leaving the financial and manufacturing companies dominating that category. Financial companies also show noticeable attention to community information, indicating that community activities are the most important social responsibility activities for financial companies. These results are partly consistent with Hamid (2004) who found that financial services companies focus on customers and employees as primary stakeholders. With regards to the industrial or manufacturing sector, it appears that employee information is the most important information category for industrial companies, indicating that employees are considered the most important stakeholders for these companies. This result is consistent with the results of Andrew, et al (1989) which found that the most important category in Malaysian and Singaporean industry companies was human resources information. They argued the interest in employee information is on the basis that, to improve working conditions and living standards of workers is an important concern for the governments of these countries. The results show that environmental information is not the most important category for industrial companies, and is not consistent with the result of Tagesson, et al , (2009) which indicated that Swedish companies within the raw-material industry disclose more environmental information. For the services sector, employee information is the most important category for these companies, which is consistent with Saches,et al, (2006) who found that the Communication Company in Switzerland has devoted much attention to employees. University of Ghana http://ugspace.ug.edu.gh 88 These results can be viewed from the perspective of disclosure categories. Concerning environmental information, the industrial sector is more interested in this category, while the financial sector is less interested in this category . Community and employee information are similar among different sectors and financial companies are less interested in both categories. 5.7 Relationship between key disclosure variables The measure of CSRD was done using different measure that represented both quality and quantity. To better understand the phenomenon of CSRD, a correlation test was conducted between the quality of CSRD and the quantity of CSRD. The correlation results show a weak correlation between the quantity of CSRD (CSRDars) and the quality of CSRD (CSRDarq) with a correlation coefficient of (β0.1129). This results in consistent with the findings of Hassan (2010) on UK firms where the results showed a weak correlation with a coefficient of 0.298. This weak correlation between the quantity and quality of disclosure in annual reports reflected that the extension of social disclosure in annual reports did not match, in most cases, with the quality of disclosure. In summary, this first section addressed the first objective of the study which was to examine the nature of CSRD in Ghana. The results show that CSRD has increased over the six year study period from 2008 to 2013. Majority of companies provide social information in their annual report (about 90%). The quality of disclosure is generally low meaning majority of the disclosure are on general statement. Contrary to growing interest in environmental issues, community information and employee information are considered the most important information in annual report disclosure. A separate section for social reporting was the main University of Ghana http://ugspace.ug.edu.gh 89 location for CSRD in annual reports followed by the chairman report. Manufacturing or industrial companies provides the higher quantity of CSRD followed by financial institutions. Also, manufacturing companies are more interested in disclosing employee’s information, environmental, health and safety and community information with less interest in products, customers and ethics which is consistent to a large extent with previous literature. The second objective is aimed at examining the determinants of CSRD using corporate characteristics, corporate governance variables and degree of multinational activity 5.8 Descriptive statistics Table 5.8: Descriptive statistics of CSRD variables Variable obs Mean std. Dev Min max CS 198 7.995864 1.170009 5.31049 11.2458 PRO 198 -.0152971 .3537315 -4.21995 .324343 TA 198 .4494949 .4987036 0 1 DMA 198 3.588832 6.931772 1 39 BS 198 8.30303 2.600096 3 18 NED 198 .8622523 .8109746 .25 .9167 SRC 198 .1010101 .302106 0 1 SS 198 .7677525 .1412076 .286 .984 OWN 198 .4162437 .4941908 0 1 Source: Computation from research data, 2015 University of Ghana http://ugspace.ug.edu.gh 90 The average size (CS) of firms sampled for this measured by the logarithm of total assets is 7.99. With regards to the degree of multinational activity (DMA), the average number of foreign countries is 3.59. This is lower than the average Hassan (2014) had using UK firms (6.66 ) and also lower than the average number of foreign countries of 17.85 in Toms et al. (2007) and Toms (2008). The reason for the difference is the fact that Ghana is a developing country with few multinational companies as compared to the UK. Concerning corporate governance variables, the results show that the companies follow good corporate governance mechanisms, such as a high percentage of non-executive directors on the board among others. The average board size (BS) is 8 and rages between 3 and 18. Comparing this average with what has been found in other studies: 7.7 in Singapore (Cheng & Courtenay , 2006), 7.099 in South Africa (Mangena & Chamisa, 2008), 10.34 in the USA (Kanagaretnam et al., 2007) and 9.46 in India (Sarkar & Sarkar, 2008) may reflect that board size in the sample is within normal range. The average proportion of non-executive directors on the board (NED) is 0.86 indicating that more than 80% of the directors are non-executive directors. The average of NED is considered a high percentage compared with 0.65 in the UK (Hassan, 2014), 0.72 in the USA (Kelton & Yang, 2008), 0.06 in Hong Kong (Gul & Leung, 2004), 0.37 in Singapore (Cheng & Courtenay , 2004), 0.413 in South Africa (Mangena & Chamisa, 2008), 0.36 in Spain (Arcay & Vazquz, 2005), 0.74 in India (Sarkar & Sarkar, 2008) and 0.31 in a sample of some European countries (Krivogorsky, 2006). With regards to corporate social responsibility committee (SRC), majority of the companies (about 90%) did not have SRC on their board. This finding is similar that of Hassan (2014) study in the UK where on over 70% of the firms surveyed did not have SRC on the board. University of Ghana http://ugspace.ug.edu.gh 91 The average percentage of shares held by majority shareholders is 0.76. This rate is higher than the findings of Hassan (2014) in the UK. This finding suggests that ownership of most companies on the Ghana Stock exchange is closely held/owned firms. 5.8 Correlation results for CSRD variables Table 5.8.1 Correlation results for quantity of CSRD determinants CSDars CSDars CS PRO TA DMA BS NED SRC SS OWN CSDars 1.0000 CS 0.6333 1.0000 PRO 0.0698 0.2510 1.0000 TA 0.1110 - 0.1104 0.0300 1.0000 DMA 0.7169 0.4558 0.0450 0.0274 1.0000 BS 0.4446 0.6358 0.1784 - 0.1382 0.4638 1.0000 NED 0.0341 0.0034 0.0907 0.0953 - 0.0912 - 0.0537 1.0000 SRC 0.4309 0.3977 0.0465 0.3027 0.3457 0.1301 - 0.0352 1.0000 SS -0.4106 - 0.2414 0.1076 0.1628 - 0.2481 - 0.1635 0.1407 - 0.0734 1.0000 OWN 0.5029 0.4943 0.1962 0.3301 0.3540 0.2492 - 0.0332 0.3299 - 0.0170 1.0000 Source: Computation from research data, 2015. University of Ghana http://ugspace.ug.edu.gh 92 Table 5.8.2: Correlation results for quality of CSRD determinants CSDarq CSDarq CS PRO TA DMA BS NED SRC SS OWN CSDarq 1.0000 CS 0.4263 1.0000 PRO 0.2665 0.2510 1.0000 TA 0.0581 - 0.1104 0.0300 1.0000 DMA 0.0604 0.4558 0.0450 0.0274 1.0000 BS 0.2914 0.6358 0.1784 - 0.1382 0.4638 1.0000 NED -0.0220 0.0034 0.0907 0.0953 - 0.0912 - 0.0537 1.0000 SRC 0.1099 0.3977 0.0465 0.3027 0.3457 0.1301 - 0.0352 1.0000 SS 0.0344 - 0.2414 0.1076 0.1628 - 0.2481 - 0.1635 0.1407 - 0.0734 1.0000 OWN 0.2982 0.4943 0.1962 0.3301 0.3540 0.2492 - 0.0332 0.3299 - 0.0170 1.0000 AUD 0.2306 0.2035 0.2301 0.2023 0.1372 - 0.0118 0.1908 - 0.0724 0.3353 1.0000 Source: Computation from research data, 2015 The results from table 5.8.1 shows that profitability is not correlated with the quantity of CSRD as expected. However, other corporate characteristics such as corporate ownership (OWN), proportion of majority shareholding (SS), type of activity (TA) and degree of multinational activity (DMA) are correlated with the quantity of CSRD. Regarding corporate characteristics variables (PRO, CS TA and DMA, OWN), the results, in general, show that corporate characteristics variables are significantly correlated with social disclosure, with the exception of corporate profitability, which appears not to be University of Ghana http://ugspace.ug.edu.gh 93 correlated with disclosure. This result is largely consistent with previous literature which indicates, in most cases, a significant correlation between both corporate size and type of activity, and disclosure, while there is no correlation between corporate profitability and disclosure (Fifka, 2013; Hassan, 2014). The findings with respect to profitability is consistent with Hackston & Milne (1996), who used four different measures for corporate profitability and found that none of these four profitability measures is significantly correlated with the amount of disclosure. Corporate size (CS) appears to be highly correlated with quantity and quality of social disclosure as shown in table 5.81 and 5.8.2 with a correlation coefficient of 0.63 for quantity and 0.43 for quality. The result is consistent with the results of Hackston & Milne (1996), who used three different measures for corporate size, and found that all size measures were highly correlated with the amount of social disclosure, and consistent with Parsa & Kouhy (2008), who used three different measures of size and found a significant correlation between two measures of size and social disclosure and Hassan (2014) who used two measure a found a significant relationship between corporate size and social disclosures. Type of activity (TA) appears to be less correlated with social disclosure even though it has a weak correlation with the quantity of CSRD. It appears to be correlated with quantity of disclosure in annual reports (0.11) (CSDars) and not correlated with quality of disclosure in annual reports (0.056)(CSDsaq). This result is somehow inconsistent with the findings of Hassan (2014) who found the type of activity to be more correlated with quality than quantity. Concerning the degree of multinational activity (DMA), it is correlated with the quantity of CSRD with a correlation coefficient of 0.72 and not correlated with the quality of CSRD with University of Ghana http://ugspace.ug.edu.gh 94 a coefficient of 0.05. This results appears to be consistent with the work of Hassan 92014) where DMA was strongly correlated with the quantity of CSRD and weakly correlated with the quality of CSRD. On the corporate governance variables (BS, SRC, NED), the results show that board size and presence of social committee on the board are correlated with the quantity and quality of CSD. However, they are more correlated with the quantity of CSRD (O.44, 0.43) than the quality of CSRD (0.1099, 0.2924). There appear to be no correlation between the proportion of non-executive directors and CSRD in terms of both quality and quantity. This is consistent with the findings of Hassan (2014) where board size and social committee with correlated with CSRD but NED was not. In addition, the results concerning the corporate shareholding ownership variable (SS) appears to be negatively correlated with the quantity of CSRD (-0.41), as expected. The negative correlation between SS and disclosure indicates that the greater dispersion of corporate ownership is related to a higher quantity of disclosure. Finally, auditor type (AUD) is positively correlated with the quality of CSRD with a coefficient of 0.23 even though the correlation is weak. 5.9 Panel data analysis According to Hassan (2010) the advantage of panel data regression is that it takes the time effect into account. Cormier et al. (2005) supported this view when he stated that; if an OLS regression for one period provides a picture, panel data provide a sequence of pictures. University of Ghana http://ugspace.ug.edu.gh 95 5.9.1 The Quantity of CSRD disclosure Table 5.9.1 provides the results of the regression model that examined the quantity of CSRD. The Adjusted R2 of 0.7270 which suggests that the independent variables have the explanatory power. The F-statistic measure by the Wald Chi2 in the case of random effect also further confirms that the model is well fit. Table 5.9.1: Panel regression results for quantity of CSD. Variable Coefficient Std. Error t- Statistic CS 15.99463*** 2.686999 5.95 PRO -7.485298 7.738823 -0.97 TA 11.20651*** 2.816565 3.98 DMA 4.363872*** 0.5773599 7.56 BS -3.973256 1.12419 -0.35 NED 47.94207*** 7.281955 6.58 SRC 14.88024* 8.199111 1.81 SS -114.2394*** 18.41417 -6.2 OWN 20.91674*** 4.038027 5.18 Ad. R2 within 0.7285 Between 0.8758 Overall 0.727 Wald Chi2 (9) 498.05 Prob>Chi2 0.000 (*** significant at 0.01level, ** significant at 0.05 level, 0.1 significance level) Source: Computation from research data, 2015 University of Ghana http://ugspace.ug.edu.gh 96 The results from the panel regression above show that corporate size (CS) and type of activity (TA) are significantly associated with the quantity of CSRD. However, profitability is negatively associated with profitability as expected but statistically insignificant. The degree of multinational activity is also significantly associated with the quantity of CSRD. Also, the test of whether a firm is Ghanaian owned or foreign owned (OWN) is also significantly associated with the quantity of CSRD. With respect to the corporate governance variables, the results show that, in general corporate governance is associated with the quantity of CSRD with the exception of statically insignificant association between board size and the quantity of CSRD. In addition to the above, the substantial shareholders (SS) is significantly associated with the quantity of CSRD. It is however negatively associated with the quantity of CSD as predicted by literature. The results in general are to a large extent consistent with previous literature. The results also seem to support the suggested framework to a large extent as it shows that corporate characteristics, corporate ownership, and corporate governance practices are important factors in determining the quantity of CSRD. The table below presents a summary of relationship between the different dependent variables and the independent variable (CSDars University of Ghana http://ugspace.ug.edu.gh 97 Table 5.9.2: The impact of different variables on the quantity of CSRD Variables relationship with CSRD Corporate size Positive significant association Corporate profit Negative insignificant association Type of activities Positive significant association Multinationality Positive significant association Board size Negative insignificant association Ownership Positive significant association Non-executive directors’ Positive significant association Social responsibility committee Positive significant association Substantial shareholders Negative significant association 5.9.3 Discussion of findings on determinants of the quantity of CSRD The results concerning corporate characteristics (CS, PRO TA OWN) are to a large extent consistent with the dominant trend in previous literature. The majority of previous studies indicate that there is no significant association between profitability and CSRD, while there is significant association between both corporate size and type of activities and CSRD (Fifka, 2013). The measure of whether a firm is Ghanaian owned or foreign owned is a new variable introduced because of the context of the study (developing) where Fifka (2013) argues that corporate social disclosures in these countries are significantly influenced by developed countries directly through policies or indirectly through multinational University of Ghana http://ugspace.ug.edu.gh 98 companies operating in these countries. The results with regards to this variable also showed a significant positive relationship with CSRD confirming the argument by Fifka (2013). Corporate size and type of activities are positively associated with quantity of CSRD, indicating that large industrial companies provide more quantity of social responsibility information. This result is consistent with findings from previous countries. In the African context, previous studies (Villiers & Barnard, 2000; Rizk et al., 2008; Akano et al., 2013) indicated a positive association between both size and activity, and quantity of CSRD. In the UK context, previous studies (Brammer & Pavelin, 2004; Brammer & Pavelin, 2006; Toms et al., 2007; Toms, 2008; Parsa & Kouhy, 2008; Hassan, 2014) all indicated a positive association between both size and activity, and quantity of CSRD. The results are consistent with legitimacy theory in the sense that large industrial companies are more visible and consequently they face more social pressure; hence they are more likely to provide a greater quantity of social responsibility information. This result provides evidence that supports hypotheses 1a and 2a which indicated a positive association between both size and activity , and quantity of CSRD. The results also provides evidence to support hypotheses 8a which indicated that firms that are foreign owned provide greater quantity of CSRD. These firms provide more quantity of CSRD because of the degree of social pressure they face as a result of pressure from their home country and the need to attract capital from socially and ethical investors from these developed countries. Corporate profitability is negatively associated but statistically insignificant, as expected, with quantity of CSRD, suggesting that companies with high profitability do not care to increase their social and environmental disclosure. This result is consistent with previous mainstream literature (Cowen et al., 1987; Patten, 1991; Ahmad et al., 2003; University of Ghana http://ugspace.ug.edu.gh 99 Ghazali, 2007; Reverte, 2009; Stanny & Ely , 2008; Garcia-Sanchez, 2008; Monteiro & Aibar-Guzman, 2009; Jinfeng & Huifeng, 2009), while it is not consistent with a small number of previous studies that indicated a significant association between profitability and CSD (Hackston & Milne, 1996; Hossain et al., 2006). This result is however consistent with legitimacy theory, which considers CSRD as a legitimacy tool without direct financial benefits. The results with regards to the degree of multinational activity showed a positive and significant relationship with the quantity of CSRD. This result is consistent with the findings of Stanny & Ely (2008) which indicated a significant relationship with CSRD. The results is however inconsistent with the findings of Branco & Rodrigues (2008) and Hassan (2014), which indicated the absence of a relationship between multinationality and CSRD in annual reports as well as the findings of Toms et al. (2007) and Toms (2008) whose results do not provide evidence on a clear relationship between multinationality and quantity of social disclosure. Therefore, with regard to hypothesis 3a, which proposes an association between DMA and quantity of CSRD, the regression results provide evidence to support this hypothesis. The presence of a relationship between multinationality and CSRD, which is consistent with a suggested framework based on legitimacy. This mean that may be due to the probability that the geographical extensions more often exist in developing countries, it compel these companies to pay attention to the social responsibility disclosures. It can therefore be argued that multinational companies face less social pressure in home countries, than they face in foreign countries and consequently they focus their attention on the concerns of the community in the foreign country . This result appears so especially for developing economies as the findings in developed economies provide otherwise. University of Ghana http://ugspace.ug.edu.gh 100 In summary, previous results concerning corporate characteristics ( CS, TA and OWN) revealed that size, type of activities and foreign ownership determine the social pressure that faces companies. It can be argued therefore, that the size of companies, type of activities and foreign ownership are the most important factors in attracting the attention of the community to pursue corporations to be more socially responsible. These findings support, to a large extent, the general argument that corporate characteristics play an important role in determining the degree of pressure faced by companies, and consequently the level of CSRD. The degree of multinational activity (DMA) was also statistically significant with the quantity of CSRD On the issue of corporate governance variables, the regression results show that, in general, governance mechanisms are associated with quantity of CSRD. Board size (BS) appears not to be statistically associated with quantity of CSD. The insignificant association between BS and CSDars, indicates that increasing the number of directors on the board is not associated with a greater quantity of CSRD in annual reports. This result is consistent with the results of Halme & Huse (1997), whose study found the absence of a relationship between board size and environmental disclosure in annual reports. With regards to hypothesis 4a, which indicated a positive association between board size and quantity of CSRD, the overall regression results does not provide evidence to support this hypothesis. This result does not supports the theoretical view that board size has an impact on corporate disclosure policy . The proportion of non-executive directors (NED) appears to be associated with quantity of CSRD. This result indicates that companies with a high proportion of non-executive directors are more likely to provide more CSRD. This finding is consistent with the results University of Ghana http://ugspace.ug.edu.gh 101 of Hassan (2014). However, the positive association between NED and CSRDars is not consistent with the findings of Haniffa & Cooke, (2005), which indicated that non- executive directors are negatively associated with CSRD, Prado- Lorenzo et al. (2009) and, in the UK, Brammer & Pavelin (2006), which indicated no relationship between the number of non-executive directors and environmental disclosure. The overall results provide support for this hypothesis (H5), which states that there is an association between NED and quantity of CSRD. This result can be understood in the context of the suggested framework, as increasing the proportion of non-executive directors on the board, increases the likelihood that companies respond positively to social pressure and consequently provide more social and environmental information. This result supports the view that non-executive directors are seen as the balance mechanism in ensuring that companies act in the best interests of all stakeholders, and provide additional windows on the world (Hassan, 2014). Consequently , non-executive directors can exert pressure on companies engaging in CSRD, to achieve congruence between organisational actions and societal values (Haniffa & Cooke, 2005). The reason for having outside presence on the board, according to Adrian Cadbury, is that outside directors bring awareness of the external world and the changing nature of public expectations, to board discussions (Clarke, 1998). Clarke (1998) argued that non-executive directors are often considered to bring valuable external business experience, which can contribute to the strategic success of a company. On the last corporate governance variable which is the presence of a social responsibility committee on the board, Hassan (2010) argues that the presence of a corporate responsibility committee tends to be associated with quantity of CSRD, which is positively associated with quantity of social disclosure in annual reports. The results show that the presence of University of Ghana http://ugspace.ug.edu.gh 102 social committee on the board is positively and significantly associated with the quantity of CSRD. This results is consistent with the study by Hassan (2014) but not consistent, to a large extent, with the study of Cowen et al. (1987). With regard to hypothesis 6a, which indicates the relationship between the corporate responsibility committee and CSRD, the overall results support this hypothesis, as the results state that there is an association between the corporate responsibility committee and quantity of social disclosure. The findings in short suggest that the presence of social responsibility committee or its equivalent is in itself enough evidence of the company’s desire to be socially responsible. With regard to corporate ownership-diffusion (SS), there is a negative association between substantial shareholders and quantity of CSRD, reflecting that companies with fewer block shareholders are more likely to provide greater CSRD. The result reflects a positive association between ownership-diffusion and CSRD. Ownership-diffusion means that there is a diversity of shareholders’ needs for information, and from that, information about the social responsibility of companies. This result is not consistent with the findings of Halme & Huse (1997), Ghazoli (2007) and Reverte (2009). With regard to hypothesis 7a, the regression results provide evidence to support this hypothesis. In summary, the overall results concerning corporate governance and ownership, provide support for a general argument that corporate governance and ownership play an important role in determining how companies respond to social pressure and consequently, in determining the level of CSRD. It also support to a large extent the suggested framework based on legitimacy theory that companies use good corporate governance mechanism to respond to social pressure by legitimizing their actions. University of Ghana http://ugspace.ug.edu.gh 103 5.9.4 The quality of CSRD disclosure Table 5.9.4 presents a panel regression models that examine the determinants of quality of CSRD. The Adjusted R2 of 0.2979 suggests that the independent variables have a lower explanatory power. However, the F-statistic confirms that the model is well fit. Besides previous studies have had a lower Adjusted R2. For instance, Hassan (2014) had an adjusted R2 of 0.1218. Table 9.9.4 Panel regression results for quality of CSRD Variable Coefficient Std. Error t- Statistic CS 0.0522028 0.0374305 1.39 PRO -0.0041974 0.002788 -1.51 TA 0.0241883 0.0724951 0.33 DMA -0.0069421** 0.0030521 -2.27 BS 0.0162258 0.012358 1.31 NED 0.0039446 0.0111859 0.35 SRC 0.0458834*** 0.0044666 10.27 SS -0.0074219** 0.0022534 -3.29 OWN 0.0603405 0.0569888 1.06 AUD 0.0579851 0.1095347 0.53 Ad. R2 within 0.0711 Between 0.3753 Overall 0.2979 Wald Chi2 (10) 678.06 Prob>Chi2 0.0000 (*** significant at 0.01level, ** significant at 0.05 level, 0.1 significance level) Source: Computation from research data, 2015 The results show that corporate size is positively associated with the quality of CSRD but statistically insignificant. Profitability is negatively associated with the quality of CSrD but also statistically insignificant. However, the type of activity and foreign ownership are not University of Ghana http://ugspace.ug.edu.gh 104 statistically significant with the quality of CSRD even though both are positively related with the quality of CSRD. The degree of multinational activities is significantly associated with quality of CSRD. On the corporate governance variables, both board size and the proportion on non-executive directors are not significantly associated with the quality of CSRD. However, the presence of a social responsibility committee on the board is associated with the quality of CSRD. The presence of substantial shareholders (SS) is also statistically significant with the quality of CSRD. Table 5.9.5: The impact of different variables on the quality of CSD Variables relationship with CSD Corporate size Positive insignificant association Corporate profitability Positive insignificant association Type of activities Positive insignificant association Multinationality Negative significant association Board size Positive insignificant association Ownership Positive insignificant association Non-executive directors’ Positive insignificant association Social responsibility committee Positive significant association Substantial shareholders Negative significant association Auditor type Positive insignificant association 5.9.5 Discussion of findings on determinants of the quality of CSRD The results concerning quality of disclosure are similar, to some extent and consistent with the results concerning quantity of disclosure. However, the impact of a suggested University of Ghana http://ugspace.ug.edu.gh 105 framework is clearer with regard to the quantity of disclosure, than the quality thereof. With regards to corporate characteristics, the results show that corporate size and corporate profitability are statistically insignificant with the quality of CSRD. In addition, the type of activity and foreign ownership are not significantly associated with the quality of CSRD. The results show that manufacturing and mining companies provide greater quantity of CSRD but less quality on CSRD. However, larger firms provide greater quality of CSRD than smaller firms based on the findings even though statically the difference is insignificant. This results is partly inconsistent with the findings of Brammer & Pavelin (2006 and 2008), Hassan (2014) which indicated a positive association between both size and type of activities, and quality of environmental disclosure. With respect to hypotheses 1b and 2b, the regression results does not provide evidence that support both hypothesis 1b but not for 2b. Corporate profitability is not associated with quality of disclosure, indicating that more profitable companies are not concerned about providing high disclosure quality in annual reports, or using reporting guidelines or independent reviews. The degree of multinational activities appears to be associated with quality of disclosure, but the association is negative indicating that geographical extension has no influence on the quality of social disclosure. With regard to hypothesis 3b, which indicates a positive association between multinationality and quality of CSRD, the results does not provide evidence to accept this hypothesis. These results are consistent, to a large extent, with a general argument that corporate characteristics determine social pressure on companies and consequently the level of CSRD. However, it is low in terms of the quality of CSRD as compared to the quantity of CSRD. Therefore, it can be argued that companies will University of Ghana http://ugspace.ug.edu.gh 106 focus their attention on increasing the quantity of social information, without paying attention to the quality of that information. On the corporate governance variables, only the presence of a social committee on the board that is significantly associated with the quality of CSRD. Board size and the proportion of non-executive directors are not significantly associated with the quality of CSRD. The results show that corporate governance practices have less impact on quality , than on quantity . The board size seems not to be associated with quality of CSRD, indicating that increasing the number of directors on the board, does not influence the type of information in annual reports, and encourages the use of reporting guidelines and independent auditors in preparing corporate responsibility reports. The results does not provide evidence to support hypothesis 4b which states that firms with more directors on the board provide higher quality of CSRD. The proportion of non-executive directors is not significantly associated with the quality of CSRD. This result is consistent with Brammer & Pavelin (2008) and Hassan (2014), who indicated a non-association between the number of non-executive directors and the quality of environmental disclosure. The results provide evidence that rejects hypothesis 5b. Regarding the corporate responsibility committee, it appears to be associated with quality of CSRD, indicating that the presence of a corporate responsibility committee, as a board committee, plays an important role in preparing corporate responsibility reports. Comparing the results concerning quantity and quality of disclosure, it reveals that association between the corporate responsibility committee and quantity of disclosure, University of Ghana http://ugspace.ug.edu.gh 107 is stronger than association with quality of disclosure. It can thus be argued that corporate responsibility committees are more interested in the quantity of social and environmental information, than their quality . With regard to hypothesis 6b, which indicates a positive association between SRC and quality of CSRD, the regression results support this hypothesis. Concerning corporate ownership, it appears that a substantial number of shareholders is significantly and negatively associated with quality of disclosure in annual reports. This finding is consistent with the findings of Hassan (2014) who found a negative significant association between SS and quality of social reporting. It can be argued that widening the ownership base is an incentive for the intention of preparing corporate responsibility reports, according to reporting principles, and using an independent auditor to review these reports. This result is inconsistent with the findings of Brammer & Pavelin (2008). With regard to hypothesis 7b, which indicated a negative association between SS and quality of disclosure, the results provide evidence that supports this hypothesis. These findings of the study provide support, to some extent, to the general argument that governance practices play an important role in determining how companies respond to social pressure, and consequently determine the level of CSRD. At the same time, the results actually reflect a weak association between corporate governance and corporate characteristics on quality of CSD. Therefore, it can be argued that the influence of the suggested framework on quantity of CSRD is clearer than its influence on quality of CSRD. The weak association with the quality of CSRD could be attributed to the work regulatory and institutional framework for CSRD in Ghana. In summary, concerning the second objective and based on the correlation and regression analysis of both determinants of the quality and quantity of CSRD, it can be concluded that University of Ghana http://ugspace.ug.edu.gh 108 the level of quantity and, to a lesser extent, the level of quality of CSRD can be interpreted in the following context: Larger industrial companies that are foreign owned attract more attention from the community , and consequently they are subject to more scrutiny concerning their social responsibilities. These companies thus face a high level of social pressure. Again, companies that have operations in more countries faces more pressure and scrutiny and hence disclose more social information that companies that operates in only Ghana. In addition, high proportion on non-executive directors and wide ownership, that have a corporate responsibility committee, are more likely to respond positively to this pressure. Therefore, the level of CSRD will be determined according to corporate characteristics (size, type of activity and foreign ownership ), degree of multinational activities, governance practices (non-executive directors and corporate responsibility committee) and ownership-diffusion. Based on these results, the following figure explains the modified determinants of level of social disclosure in a given company according to empirical results: University of Ghana http://ugspace.ug.edu.gh 109 Figure 5.6: Modified determinants framework of CSRD in Ghana The degree of social pressure facing companies The degree of companies’ response to social pressure CSRD Ownership diffusion Corporate Governance (NED, SRC) Size Industry (Type of activity) Multinational Activity University of Ghana http://ugspace.ug.edu.gh 110 5.10 consequences of CSRD The last objective was aimed at evaluating the consequence of CSRD on firm performance using firm value as a proxy. To achieve objective, the study began by looking at the correlation results of the third model that examined the relationship between CSRD and firm value which is measure using the Tobin Q. Table 5.10.1: Correlation results between corporate social disclosure and market value TQ TQ CSDars CSDarq CS AFM LEV PRO IG DMA TQ 1 CSDars 0.0147 1 CSDarq -0.006 0.1129 1 CS -0.0378 0.6328 0.4263 1 AFM 0.0507 0.2098 0.25 0.4818 1 LEV -0.0157 -0.0766 -0.2015 -0.2195 -0.1903 1 PRO -0.0023 -0.0185 -0.0233 0.0652 0.1568 -0.1655 1 IG 0.012 -0.0274 0.0216 0.0197 0.0513 0.002 -0.0019 1 DMA -0.0232 0.7172 0.0591 0.4553 0.1021 -0.0404 -0.0128 -0.0245 1 Source: Computation from research data, 2015 The results from the correlation analysis indicate that there is no significant relationship between corporate social responsibility disclosures (CSRD) and firm value (TQ). University of Ghana http://ugspace.ug.edu.gh 111 Table 5.10.2 Panel regression results on the consequence of CSRD Variable Coefficient Std. Error t-Statistic CSDars 0.016205 0.0172274 0.94 CSDarq 0.3725974 0.9828823 0.38 CS -1.896288 1.541192 -1.23 AFM -0.1695959 0.3850303 -0.44 LEV -0.1500945*** 0.036084 -4.16 PRO 0.0897572 0.1268001 0.71 IG 0.0224464** 0.0105704 2.12 DMA 0.1918495* 0.098717 1.94 Ad. R2 within 0.262 Between 0.4687 Overall 0.2644 Wald Chi2 (8) 45.00 Prob>Chi2 0.000 (*** significant at 0.01level, ** significant at 0.05 level, 0.1 significance level) Source: Computation from research data, 2015 The Adjusted R2 of 0.26 suggests a lower explanatory power but is consistent with previous studies. For instance, Hassan 2010 had 0.08; Kliff et al. 2015 had18.70. The panel regression results which is consistent with the correlation results show that to a large extent, there is no significant association between CSRD and corporate market value. The results appears to provide evidence that there are no financial consequences for CSRD. Three control variables such as leverage (LEV), investment growth (IG) and degree of multinational activity (DMA) are statistically significant with firm value. The results suggest that investors of firms listed of GSE neglect issues like social and environmental University of Ghana http://ugspace.ug.edu.gh 112 disclosures and focus on financial problems such as leverage. This result is consistent with the findings of Hassan (2010) in the UK where he argued that there is no significant association between CSRD and firm value. It is also consistent with the finding of Kliff et al. (2015) study using South African companies and Moroccan companies where the result revealed an insignificant association between CSRD and firm value. However, the results are considered inconsistent with some studies which indicated an association between CSRD and some financial aspects. For instance, Gozali et al. (2002) reported a significant relationship between environmental disclosure and share price, but their study distinguished between good and bad environmental news. Blacconiere & Northcut (1997) also in their study found a relationship between stock prices and the extensive use of environmental disclosure in chemical companies. In addition, Murray et al. (2006) indicated that over a period of time, total social and environmental disclosure is significantly related to market returns, even after adjusting for the size effect. Finally, Richardson & Welker (2001) in their study indicated in a contrasting hypothesis that there is a statistically significant positive relationship between the level of social disclosure and the cost of equity capital. With regard to hypothesis 8a and 8b, which indicated no impact of CSRD on market value, the overall results provide evidence to support this hypothesis. The insignificant between CSRD and corporate market value provides an indicator that investors are not actually interested in social disclosure information when they make their investment decisions. It can be argued that despite the growing importance of corporate social responsibility practices, suggesting that the investors could draw attention to the social responsibility information in the investment decision-making process, the studies conducted to University of Ghana http://ugspace.ug.edu.gh 113 examine the extent of investor interest in social disclosure provided mixed results. This inconsistency could be attributed to the differences in regulatory regimes as well as the differences in the sophistications of investors. For instance, Anderson & Frankle (1980) and Epstein & Freedman (1994) indicated that investors are interested in social disclosure, while Chan & Milne (1999) indicated that there is no significant reaction to good environmental performance, and Milne & Chan (1999) show that the decision impact of CSRD is small. In a survey by Economist Intelligence Unit (2005) the results showed that 85% of executive and investors surveyed ranked corporate social responsibility as a central consideration in investment decisions. Furthermore, Murray et al. (2006) argued that despite investors exhibiting an increasing demand for social responsibility information, there is no established link between the price-sensitivity of social disclosure and the substantial changes in economic circumstances that this information could be signalling. Epstein & Freedman (1994) also argued that social information should theoretically be of use to various stakeholders, and their survey results regarding the importance of social and economic information for various stakeholders, show that most shareholders prefer economic information (83.55% prefer additional information on economic impacts and 69.83% prefer social information), while other stakeholders (employees, community groups, customers and environmental groups) prefer social disclosure, more than economic disclosure. In summary, the empirical results and previous literature supported the hypotheses that CSRD has no impact on firm value. The absence of a financial impact of CSRD can be interpreted in the light of the following reasons: first, the probability that investors are not interested in social and environmental information when the y make investment decisions, second social responsibility investment still represents a small proportion of University of Ghana http://ugspace.ug.edu.gh 114 the business environment, third, the financial market still does not pay appropriate attention to corporate social responsibility and its impact on the financial future of the company . The findings of this sections seem to be consistent with the argument of Bebbington et al., (2008) that one of the emerging explanations of CSRD is that it could be conceived as both an outcome of, and part of, the reputation risk-management process. This explanation can be understood in the context of legitimacy theory, and it is not considered as an alternative for legitimacy theory , in the view of Adams (2008). It can be argued that both legitimacy and reputation are related concepts, and corporate reputation is considered an outcome of the legitimisation process. University of Ghana http://ugspace.ug.edu.gh 115 CHAPTER SIX SUMMARY, CONCLUSIONS AND RECOMMENDATION 6.0 Introduction The increasing global concern for social and environment impact of corporations and the consequent academic interest in researching corporate social and environmental disclosure practices have given tremendous impetus for initializing the current research. The study presented a framework that explained the determinants of CSRD in terms of both quality and quantity and its consequence on firm value. The framework is based on legitimacy theory which argues that CSRD is a function of public pressure on companies concerning their social responsibilities. The study is based on the limitations of previous studies which according to Hassan (2014) do not provide the accurate picture regarding CSRD. Previous literature, in most cases, has concentrated on environmental disclosure which is considered one category of CSRD. Also, previous studies concentrated extensively on the quantity of CSRD with less attention paid to its quality. Apart from these, the study incorporated new variable that have received little attention in literature such as the degree of multinational activity and corporate governance variables as well as foreign ownership. The study used content analysis technique to extract social and responsibility information of listed firm in Ghana over a six year period. A variety of statistical tests and analyses, University of Ghana http://ugspace.ug.edu.gh 116 including descriptive statistics, correlation analysis and regression analysis, are undertaken in order to measure the determinants of CSRD in terms of both quality and quantity and its consequence on firm performance. The next section presents the summary of the key findings 6.1 Summary of findings The key findings on the study in relation to the nature of social responsibility disclosures can be summarised as follows; First, The level of CSRD is increasing over time, reflecting growing attention to this type of disclosure. Generally, this increase in the level of CSRD is focused on increasing the quantity of this disclosure without paying the same attention to the quality of disclosure. Second, Despite the growing interest in environmental issues across the world and the focusing of the majority of studies on environmental disclosure, it appears that community information as well as employee-related information is the dominant category of social disclosure in annual reports. Consequently, it appears that the community and employees are the most important stakeholders for companies listed on the Ghana stock exchange. Third, Industrial or manufacturing companies appear to be more interested in CSRD than other economic sectors. The manufacturing companies are more interested in disclosing employee’s information, environmental, health and safety and community information with less attention in products, customers and ethics. University of Ghana http://ugspace.ug.edu.gh 117 Fourth, a separate section for social and environmental disclosure was the main source or preferred location in the annual report for CSRD disclosure among the listed firms on the GSE. With respect to the determinants of CSRD of firms listed on the Ghana stock exchange, the following findings were obtained; Corporate characteristics and presence of corporate responsibility committee on the board are significantly associated with quantity of disclosure in annual reports and to some extent the quality of CSRD in annual reports. Concerning corporate characteristics, the study found that corporate characteristics (CS, TA and OWN) determine the social pressure that faces companies and hence statically significant with the quantity of CSRD. It can be argued therefore, that the size of companies, type of activities and foreign ownership are the most important factors in attracting the attention of the community to pursue corporations to be more socially responsible. These findings support, to a large extent, the general argument that corporate characteristics play an important role in determining the degree of pressure faced by companies, and consequently the level of CSRD. The degree of multinational activity has a significant relationship with both quality and quantity of CSRD which is in line with legitimacy theory. It can be deduced that corporate governance mechanisms play an important role in making companies closer to community needs and ensuring that companies are acting in the interest of all stakeholders not only shareholders. Corporate governance variable such as NED and SRC were associated to a large extent with both the quality and quantity of CSRD. University of Ghana http://ugspace.ug.edu.gh 118 Finally, ownership diffusion was also significantly associated with both the quality and quantity of CSRD among the listed firm on the GSE. On the last objective, which examined the consequence of CSRD on firm value, the results showed that, both the quality and quantity of CSRD are not significantly associated with firm value. The implication is that, investors of stocks on the Ghana stock exchange do not take into consideration the social reputation of firms before investing in those companies. This view is supported by the legitimacy theory and previous literature as well. 6.2 Conclusions It can be concluded from the discussion so far that, corporate characteristics, corporate governance as well as the degree of multinational activity influences the level of social responsibility disclosures of firms listed on the Ghana stock exchange. However, they do not necessarily influence the quality of CSRD especially in the same way as they influence the quantity. In addition to the above, it can be concluded that investors of stocks on the Ghana stock exchange are not socially and environmentally sensitive in their investment decisions. The study provides evidence to support the argument that legitimacy theory provides a better theoretical background for determinants and consequence of CSRD. The results of the study also show a weak association between the quantity of CSRD and the quality of CSRD. This result draws attention for the need to focus on the development of clear standards or guidelines for the preparation of corporate responsibility reports, and develop the process of reviewing these reports in terms of the existence of independent parties to carry out that process and develop appropriate criteria for such a review University of Ghana http://ugspace.ug.edu.gh 119 6.3 Recommendations The following recommendations are made for future research; First, In-depth study of the relationship between corporate governance and CSRD through analyses of the impact of various corporate governance practices on CSRD is needed. It is important to analyse this relationship in different environments, as corporate governance practices are varied among different environments especially in developing economies. A comparative study among different countries concerning the relationship between corporate governance and CSRD can be conducted. A study on the development of guidelines for the assurance process of corporate responsibility reports can be conducted. A comprehensive study on the consequence of CSRD is also important especially a study that incorporates social reputation of firms as well as other critical market variables and financial measurement. On policy and practice, the study reveals low level of environmental awareness of both customers and the firms themselves. Given the global campaign on climate change and adaption, policy makers in Ghana should adopt a clear–cut policy on environmental and other social disclosure to ensure that firms are held accountable for their actions. In addition, audit firms in Ghana should enhance their capacity to allow them audit social and environmental disclosures to help improve the quality of disclosure as the study revealed a low quality of CSRD among the firms. University of Ghana http://ugspace.ug.edu.gh 120 Finally, consumers as well as investors should take into consideration the social and environmental responsiveness of firm before taking their investment and purchasing decisions. This will compel firms to be more socially responsible. University of Ghana http://ugspace.ug.edu.gh 121 References Abbott, W. F., & Monsen, R. J. (1979). On the measurement of corporate social responsibility: Self-reported disclosures as a method of measuring corporate social involvement. Academy of management journal, 22(3), 501-515. Abu-Baker, N., & Naser, K. (2000). Empirical evidence on corporate social disclosure (CSD) practices in Jordan. International Journal of Commerce and Management, 10(3/4), 18-34. Abusbaiha, A. M. A. (2012). Corporate Social Responsibility (Csr) Information Disclosure Among Industry Classification By Annual Reports Of Public Companies Listed At Indonesia Stock Exchange (Idx). In Conference In Business, Accounting, And Management (CBAM) 2012 (Vol. 1, No. 1, pp. 77-88). Adams, C. A. (2002). Internal organisational factors influencing corporate social and ethical reporting: Beyond current theorising. Accounting, Auditing & Accountability Journal, 15(2), 223-250. Aerts, W., & Cormier, D. (2006). The association between media legitimacy and corporate environmental communication. Cahier de recherche, 07. Aerts, W., Cormier, D., & Magnan, M. (2008). Corporate environmental disclosure, financial markets and the media: An international perspective.Ecological Economics, 64(3), 643-659. Ahmad, N. N. N., & Mohamad, N. A. (2013). Environmental disclosures by the Malaysian construction sector: Exploring extent and quality. Corporate Social Responsibility and Environmental Management. Ahmad, Z., Hassan, S., & Mohammad, J. (2003). Determinants of environmental reporting in Malaysia. International Journal of Business Studies,11(1), 69-90. Ahmed, K., & Courtis, J. K. (1999). Associations between corporate characteristics and disclosure levels in annual reports: a meta-analysis. The British Accounting Review, 31(1), 35-61. Ajinkya, B., Bhojraj, S., & Sengupta, P. (2005). The association between outside directors, institutional investors and the properties of management earnings forecasts. Journal of accounting Research, 43(3), 343-376. Akinpelu, Y. A., & Ogunbi, O. J. (2013). Corporate Social Responsibility Activities Disclosure by Commercial Banks In Nigeria. European Journal of Business and Management, 5(7), 173-185. Anderson, J. C., & Frankle, A. W. (1980). Voluntary social reporting: An iso-beta portfolio analysis. Accounting Review, 467-479. University of Ghana http://ugspace.ug.edu.gh 122 Anderson, R. C., & Reeb, D. M. (2004). Board composition: Balancing family influence in S&P 500 firms. Administrative Science Quarterly, 49(2), 209-237. Balabanis, G., Phillips, H. C., & Lyall, J. (1998). Corporate social responsibility and economic performance in the top British companies: are they linked?.European Business Review, 98(1), 25-44. Baughn, C. C., & McIntosh, J. C. (2007). Corporate social and environmental responsibility in Asian countries and other geographical regions. Corporate Social Responsibility and Environmental Management, 14(4), 189-205. Beattie, V., McInnes, B., & Fearnley, S. (2004, September). A methodology for analysing and evaluating narratives in annual reports: a comprehensive descriptive profile and metrics for disclosure quality attributes. In Accounting forum (Vol. 28, No. 3, pp. 205-236). Elsevier. Belkaoui-Riahi, A. (2004). Accounting theory. Thomson Learning. Bhatia, A., & Mahajan, P. (2013). Determinants of Corporate Social Disclosure: An Empirical Study of BSE-SENSEX Companies in India. IIMS Journal of Management Science, 4(2), 191-204. Boesso, G., & Kumar, K. (2009). An investigation of stakeholder prioritization and engagement: who or what really counts. Journal of Accounting & Organizational Change, 5(1), 62-80. Brammer, S., & Pavelin, S. (2004). Voluntary social disclosures by large UK companies. Business Ethics: A European Review, 13(2‐3), 86-99. Brammer, S., & Pavelin, S. (2006). Voluntary environmental disclosures by large UK companies. Journal of Business Finance & Accounting, 33(7‐8), 1168-1188. Brammer, S., & Pavelin, S. (2008). Factors influencing the quality of corporate environmental disclosure. Business Strategy and the Environment, 17(2), 120-136. Branco, M. C., & Rodrigues, L. L. (2008). Factors influencing social responsibility disclosure by Portuguese companies. Journal of Business Ethics, 83(4), 685-701. Carroll, A. B., & Shabana, K. M. (2010). The business case for corporate social responsibility: a review of concepts, research and practice. International Journal of Management Reviews, 12(1), 85-105. Chambers, E., Chapple, W., Moon, J., & Sullivan, M. (2003). CSR in Asia: A seven country study of CSR website reporting. Cheng, E. C., & Courtenay, S. M. (2006). Board composition, regulatory regime and voluntary disclosure. The international journal of accounting, 41(3), 262-289. University of Ghana http://ugspace.ug.edu.gh 123 Cho, C. H., & Patten, D. M. (2007). The role of environmental disclosures as tools of legitimacy: A research note. Accounting, Organizations and Society,32(7), 639-647. Cormier, D., & Gordon, I. M. (2001). An examination of social and environmental reporting strategies. Accounting, Auditing & Accountability Journal, 14(5), 587-617. Cormier, D., & Magnan, M. (2007). The revisited contribution of environmental reporting to investors' valuation of a firm's earnings: An international perspective. Ecological economics, 62(3), 613-626. Cormier, D., Magnan, M., & Van Velthoven, B. (2005). Environmental disclosure quality in large German companies: economic incentives, public pressures or institutional conditions?. European accounting review, 14(1), 3-39. 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. Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods approaches. Sage publications. Crisóstomo, V. L., de Souza Freire, F., & de Vasconcellos, F. C. (2011). Corporate social responsibility, firm value and financial performance in Brazil. Social Responsibility Journal, 7(2), 295-309. da Silva Monteiro, S. M., & Aibar‐Guzmán, B. (2010). Determinants of environmental disclosure in the annual reports of large companies operating in Portugal. Corporate Social Responsibility and Environmental Management,17(4), 185-204. Darrell, W., & Schwartz, B. N. (1997). Environmental disclosures and public policy pressure. Journal of accounting and Public Policy, 16(2), 125-154. Daub, C. H. (2007). Assessing the quality of sustainability reporting: an alternative methodological approach. Journal of Cleaner Production, 15(1), 75-85. Daub, C. H. (2007). Assessing the quality of sustainability reporting: an alternative methodological approach. Journal of Cleaner Production, 15(1), 75-85. Deegan, C. (2002). Introduction: the legitimising effect of social and environmental disclosures-a theoretical foundation. Accounting, Auditing & Accountability Journal, 15(3), 282-311. Deegan, C. (2002). Introduction: the legitimising effect of social and environmental disclosures-a theoretical foundation. Accounting, Auditing & Accountability Journal, 15(3), 282-311. University of Ghana http://ugspace.ug.edu.gh 124 Deegan, C., & Gordon, B. (1996). A study of the environmental disclosure practices of Australian corporations. Accounting and business research, 26(3), 187-199. Depoers, F. (2000). A cost benefit study of voluntary disclosure: Some empirical evidence from French listed companies. European Accounting Review, 9(2), 245-263. Dhaliwal, D. S., Radhakrishnan, S., Tsang, A., & Yang, Y. G. (2012). Nonfinancial disclosure and analyst forecast accuracy: International evidence on corporate social responsibility disclosure. The Accounting Review, 87(3), 723-759. Dhaliwal, D., Li, O. Z., Tsang, A., & Yang, Y. G. (2009). Voluntary non-financial disclosure and the cost of equity capital: the case of corporate social responsibility reporting. Available at SSRN 1343453. Dobers, P., & Halme, M. (2009). Corporate social responsibility and developing countries. Corporate Social Responsibility and Environmental Management,16(5), 237-249. Du, S., Bhattacharya, C. B., & Sen, S. (2010). Maximizing business returns to corporate social responsibility (CSR): The role of CSR communication. International Journal of Management Reviews, 12(1), 8-19. El Mosaid, F., & Boutti, R. (2012). Relationship between Corporate Social Responsibility and Financial Performance in Islamic Banking. Research Journal of Finance and Accounting, 3(10), 93-103. Eng, L. L., & Mak, Y. T. (2003). Corporate governance and voluntary disclosure. Journal of accounting and public policy, 22(4), 325-345. Epstein, M. J., & Freedman, M. (1994). Social disclosure and the individual investor. Accounting, Auditing & Accountability Journal, 7(4), 94-109. Esa, E., & Anum Mohd Ghazali, N. (2012). Corporate social responsibility and corporate governance in Malaysian government-linked companies. Corporate Governance: The international journal of business in society, 12(3), 292-305. Fauzi, H., Mahoney, L. S., & Abdul Rahman, A. (2007). The link between corporate social performance and financial performance: evidence from Indonesian companies. Issues in Social and Environmental Accounting, 1(1), 149-159. Fifka, M. S. (2013). Corporate Responsibility Reporting and its Determinants in Comparative Perspective–a Review of the Empirical Literature and a Meta‐analysis. Business strategy and the environment, 22(1), 1-35. Fu, R., Kraft, A., & Zhang, H. (2012). Financial reporting frequency, information asymmetry, and the cost of equity. Journal of Accounting and Economics,54(2), 132-149. University of Ghana http://ugspace.ug.edu.gh 125 Gao, S. S., Heravi, S., & Xiao, J. Z. (2005, June). Determinants of corporate social and environmental reporting in Hong Kong: a research note. InAccounting Forum (Vol. 29, No. 2, pp. 233-242). Elsevier. Gao, S. S., Heravi, S., & Xiao, J. Z. (2005, June). Determinants of corporate social and environmental reporting in Hong Kong: a research note. In Accounting Forum (Vol. 29, No. 2, pp. 233-242). García‐Sánchez, I. M. (2008). Corporate social reporting: segmentation and characterization of Spanish companies. Corporate Social Responsibility and Environmental Management, 15(4), 187-198. Gelb, D., Holtzman, M. P., & Mest, D. (2008). International operations and voluntary disclosures by US-based multinational corporations. Advances in accounting, 24(2), 243-249. Gill, J., & Johnson, P. (2010). Research methods for managers. Sage. Golob, U., & Bartlett, J. L. (2007). Communicating about corporate social responsibility: A comparative study of CSR reporting in Australia and Slovenia.Public Relations Review, 33(1), 1-9. Gomez, L. M., & Chalmeta, R. (2011). Corporate responsibility in US corporate websites: A pilot study. Public relations review, 37(1), 93-95. Gray, R., & Milne, M. (2002). Sustainability reporting: who's kidding whom?.Chartered Accountants Journal of New Zealand, 81(6), 66-70. Gray, R., Javad, M., Power, D. M., & Sinclair, C. D. (2001). Social and environmental disclosure and corporate characteristics: a research note and extension. Journal of Business Finance & Accounting, 28(3‐4), 327-356. Gray, R., Kouhy, R., & Lavers, S. (1995). Corporate social and environmental reporting: a review of the literature and a longitudinal study of UK disclosure.Accounting, Auditing & Accountability Journal, 8(2), 47-77. Griffin, J. J., & Mahon, J. F. (1997). The corporate social performance and corporate financial performance debate twenty-five years of incomparable research. Business & Society, 36(1), 5-31. Hackston, D., & Milne, M. J. (1996). Some determinants of social and environmental disclosures in New Zealand companies. Accounting, Auditing & Accountability Journal, 9(1), 77-108. Hahn, R., & Kühnen, M. (2013). Determinants of sustainability reporting: a review of results, trends, theory, and opportunities in an expanding field of research. Journal of Cleaner Production, 59, 5-21. University of Ghana http://ugspace.ug.edu.gh 126 Halme, M., & Huse, M. (1997). The influence of corporate governance, industry and country factors on environmental reporting. Scandinavian Journal of Management, 13(2), 137-157 Halme, M., & Huse, M. (1997). The influence of corporate governance, industry and country factors on environmental reporting. Scandinavian Journal of Management, 13(2), 137-157. Hammond, K., & Miles, S. (2004, March). Assessing quality assessment of corporate social reporting: UK perspectives. In Accounting Forum (Vol. 28, No. 1, pp. 61-79). Elsevier. Haniffa, R. M., & Cooke, T. E. (2005). The impact of culture and governance on corporate social reporting. Journal of accounting and public policy, 24(5), 391-430. Hasnah, H., Sofri, Y., Sharon, M., & Ishak, I. (2006). Level of corporate social disclosure in Malaysia. Hassan, N. (2010). Corporate social responsibility disclosure: an examination of framework of determinants and consequences (Doctoral dissertation, Durham University). Hassan, N. T(2014). Corporate Governance, Degree of Multi-nationality, and Corporate Social Responsibility Disclosure: Evidence from the UK. Journal of Empirical Research in Accounting & Auditing,1(1). Hasseldine, J., Salama, A. I., & Toms, J. S. (2005). Quantity versus quality: the impact of environmental disclosures on the reputations of UK Plcs. The British Accounting Review, 37(2), 231-248. Hinson, R., Boateng, R., & Madichie, N. (2010). Corporate social responsibility activity reportage on bank websites in Ghana. International Journal of Bank Marketing, 28(7), 498- 518. Ho, C. A., & Williams, S. M. (2003). International comparative analysis of the association between board structure and the efficiency of value added by a firm from its physical capital and intellectual capital resources. The International Journal of Accounting, 38(4), 465-491. Hoeffler, S., & Keller, K. L. (2002). Building brand equity through corporate societal marketing. Journal of Public Policy & Marketing, 21(1), 78-89. Hooghiemstra, R. (2000). Corporate communication and impression management–new perspectives why companies engage in corporate social reporting. Journal of business ethics, 27(1-2), 55-68. Hope, O. K. (2003). Firm‐level disclosures and the relative roles of culture and legal origin. Journal of International Financial Management & Accounting, 14(3), 218-248. Hughes, S. B., Anderson, A., & Golden, S. (2001). Corporate environmental disclosures: are they useful in determining environmental performance?.Journal of accounting and public policy, 20(3), 217-240. University of Ghana http://ugspace.ug.edu.gh 127 Husted, B. W., & Allen, D. B. (2007). Corporate social strategy in multinational enterprises: Antecedents and value creation. Journal of Business Ethics, 74(4), 345-361. Ingram, R. W. (1978). An investigation of the information content of (certain) social responsibility disclosures. Journal of Accounting Research, 270-285. Jennifer Ho, L. C., & Taylor, M. E. (2007). An Empirical Analysis of Triple Bottom‐Line Reporting and its Determinants: Evidence from the United States and Japan. Journal of International Financial Management & Accounting, 18(2), 123-150. Jinfeng, Z., & Huifeng, X. (2009). Empirical Research on Factors Influencing Level of Environmental Protection Information Disclosure in Annual Reports by Listed Companies. Chinese Journal of Population Resources and Environment,7(1), 15-22. Kelton, A. S., & Yang, Y. W. (2008). The impact of corporate governance on Internet financial reporting. Journal of accounting and Public Policy, 27(1), 62-87. Kemp, M. (2001). Corporate social responsibility in Indonesia: quixotic dream or confident expectation?. Khan, M. H. U. Z., Halabi, A. K., & Samy, M. (2009). Corporate social responsibility (CSR) reporting: a study of selected banking companies in Bangladesh. Social Responsibility Journal, 5(3), 344-357. Kolk, A. (2005). Corporate Social Responsibility in the Coffee Sector:: The Dynamics of MNC Responses and Code Development. European Management Journal, 23(2), 228-236. Kolk, A., & Pinkse, J. (2010). The integration of corporate governance in corporate social responsibility disclosures. Corporate Social Responsibility and Environmental Management, 17(1), 15-26. Krippendorff, K. (2012). Content analysis: An introduction to its methodology. Sage. Lambert, F. H., & Webb, M. J. (2008). Dependency of global mean precipitation on surface temperature. Geophysical Research Letters, 35(16). Lee, T. M., & Hutchison, P. D. (2005). The decision to disclose environmental information: A research review and agenda. Advances in accounting, 21, 83-111. Leuz, C., & Wysocki, P. D. (2008). Economic consequences of financial reporting and disclosure regulation: A review and suggestions for future research. Available at SSRN 1105398. Lina, D. & Gokiene, R. (2011).Valuation of corporate social responsibility reports. Economics and Management, 16. University of Ghana http://ugspace.ug.edu.gh 128 Liu, X., & Anbumozhi, V. (2009). Determinant factors of corporate environmental information disclosure: an empirical study of Chinese listed companies. Journal of Cleaner Production, 17(6), 593-600. Milne, M. J., & Chan, C. C. (1999). Narrative corporate social disclosures: how much of a difference do they make to investment decision-making?. The British Accounting Review, 31(4), 439-457. Mohd Ghazali, N. A. (2007). Ownership structure and corporate social responsibility disclosure: some Malaysian evidence. Corporate Governance: The international journal of business in society, 7(3), 251-266. Moon, J. J. (2007). In Good Companies? A Critical Evaluation of the Corporate Social Performance-Corporate Financial Performance Link. Working paper, University of Pennsylvania. Morimoto, R., Ash, J., & Hope, C. (2005). Corporate social responsibility audit: From theory to practice. Journal of Business Ethics, 62(4), 315-325. Morimoto, R., Ash, J., & Hope, C. (2005). Corporate social responsibility audit: From theory to practice. Journal of Business Ethics, 62(4), 315-325. Morsing, M., Schultz, M., & Nielsen, K. U. (2008). The ‘Catch 22’of communicating CSR: Findings from a Danish study. Journal of Marketing Communications, 14(2), 97-111. Muller, A. (2006). Global versus local CSR strategies. European Management Journal, 24(2), 189-198. Murray, A., Sinclair, D., Power, D., & Gray, R. (2006). Do financial markets care about social and environmental disclosure? Further evidence and exploration from the UK. Accounting, Auditing & Accountability Journal, 19(2), 228-255. Nelling, E., & Webb, E. (2009). Corporate social responsibility and financial performance: the “virtuous circle” revisited. Review of Quantitative Finance and Accounting, 32(2), 197- 209. Newson, M., & Deegan, C. (2002). Global expectations and their association with corporate social disclosure practices in Australia, Singapore, and South Korea. The International Journal of Accounting, 37(2), 183-213. Ofori, D. F., & Hinson, R. E. (2007). Corporate social responsibility (CSR) perspectives of leading firms in Ghana. Corporate Governance, 7(2), 178-193. Parsa, S., & Kouhy, R. (2008). Social reporting by companies listed on the alternative investment market. Journal of Business Ethics, 79(3), 345-360. Patten, D. M. (1992). Exposure, legitimacy, and social disclosure. Journal of accounting and public policy, 10(4), 297-308. University of Ghana http://ugspace.ug.edu.gh 129 Patten, D. M. (2002). The relation between environmental performance and environmental disclosure: a research note. Accounting, Organizations and Society, 27(8), 763-773. Perrini, F., Russo, A., Tencati, A., & Vurro, C. (2011). Deconstructing the relationship between corporate social and financial performance. Journal of Business Ethics, 102(1), 59- 76. Plumlee, M., Brown, D., Hayes, R., & Marshall, S. (2010). Voluntary environmental disclosure quality and firm value: Further evidence. Unpublished paper. Prado‐Lorenzo, J. M., Gallego‐Alvarez, I., & Garcia‐Sanchez, I. M. (2009). Stakeholder engagement and corporate social responsibility reporting: the ownership structure effect. Corporate Social Responsibility and Environmental Management, 16(2), 94-107. Prado-Lorenzo, J. M., Rodríguez-Domínguez, L., Gallego-Álvarez, I., & García-Sánchez, I. M. (2009). Factors influencing the disclosure of greenhouse gas emissions in companies world-wide. Management Decision, 47(7), 1133-1157. Raar, J. (2002). Environmental initiatives: towards triple-bottom line reporting.Corporate Communications: An International Journal, 7(3), 169-183. Rahaman, A. S. (2000). Senior management perceptions of social and environmental reporting in Ghana. Social and Environmental Accountability Journal, 20(1), 7-10. Rahaman, A. S., Lawrence, S., & Roper, J. (2004). Social and environmental reporting at the VRA: institutionalised legitimacy or legitimation crisis?. Critical Perspectives on Accounting, 15(1), 35-56. Razek, M. A. (2014). The Association between Corporate Social Responsibility Disclosure and Corporate Governance–A Survey of Egypt. Research Journal of Finance and Accounting, 5(1), 93-98. Reverte, C. (2009). Determinants of corporate social responsibility disclosure ratings by Spanish listed firms. Journal of Business Ethics, 88(2), 351-366. Riahi-Belkaoui, A. (2001). The extent of environmental disclosures: effects of regulatory costs and level of exposure to environmental risk. International journal of environmental technology and management, 1(1), 75-86. Richardson, A. J., & Welker, M. (2001). Social disclosure, financial disclosure and the cost of equity capital. Accounting, Organizations and Society, 26(7), 597-616. Rizk, R., Dixon, R., & Woodhead, A. (2008). Corporate social and environmental reporting: a survey of disclosure practices in Egypt. Social Responsibility Journal, 4(3), 306-323. Roberts, R. W. (1992). Determinants of corporate social responsibility disclosure: an application of stakeholder theory. Accounting, Organizations and Society, 17(6), 595-612. University of Ghana http://ugspace.ug.edu.gh 130 Robertson, D. C., & Nicholson, N. (1996). Expressions of corporate social responsibility in UK firms. Journal of Business Ethics, 15(10), 1095-1106. Rose, J. M. (2007). Corporate directors and social responsibility: Ethics versus shareholder value. Journal of Business Ethics, 73(3), 319-331. Said, R., Hj Zainuddin, Y., & Haron, H. (2009). The relationship between corporate social responsibility disclosure and corporate governance characteristics in Malaysian public listed companies. Social Responsibility Journal, 5(2), 212-226. Salama, A. (2005). A note on the impact of environmental performance on financial performance. Structural Change and Economic Dynamics, 16(3), 413-421. Saunders, M. N., Saunders, M., Lewis, P., & Thornhill, A. (2011). Research methods for business students, 5/e. Pearson Education India. Shafer, W. E., Fukukawa, K., & Lee, G. M. (2007). Values and the perceived importance of ethics and social responsibility: The US versus China. Journal of Business Ethics, 70(3), 265- 284. Sharma, N. (2013). Theoretical Framework for Corporate Disclosure Research. Asian Journal of Finance & Accounting, 5(1), 183-196. Sharma, U., & Davey, H. (2013). Voluntary disclosure in the annual reports of Fijian companies. International Journal of Economics and Accounting, 4(2), 184-208. Stanton, P., & Stanton, J. (2002). Corporate annual reports: research perspectives used. Accounting, Auditing & Accountability Journal, 15(4), 478-500. Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of management review, 20(3), 571-610. Sweeney, L., & Coughlan, J. (2008). Do different industries report corporate social responsibility differently? An investigation through the lens of stakeholder theory. Journal of Marketing Communications, 14(2), 113-124. Tilling, M. V. (2004). Some thoughts on legitimacy theory in social and environmental accounting. Social and Environmental Accountability Journal,24(2), 3-7. Tilt, C. A. (1994). The influence of external pressure groups on corporate social disclosure: some empirical evidence. Accounting, Auditing & Accountability Journal, 7(4), 47-72. Toms, J. S. (2002). Firm resources, quality signals and the determinants of corporate environmental reputation: some UK evidence. The British Accounting Review, 34(3), 257- 282. Toms, S., Hasseldine, J., & Massoud, H. (2007). Political, social and economic determinants of corporate social disclosure by multi-national firms in environmentally sensitive industries. University of Ghana http://ugspace.ug.edu.gh 131 Tyler, T. R., & Blader, S. L. (2005). Can businesses effectively regulate employee conduct? The antecedents of rule following in work settings.Academy of Management Journal, 48(6), 1143-1158. Ullmann, A. A. (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. Unerman, J. (2000). Methodological issues-Reflections on quantification in corporate social reporting content analysis. Accounting, Auditing & Accountability Journal, 13(5), 667-681. Van der Laan Smith, J., Adhikari, A., & Tondkar, R. H. (2005). Exploring differences in social disclosures internationally: A stakeholder perspective. Journal of Accounting and Public Policy, 24(2), 123-151. Van der Laan, S. (2009). The role of theory in explaining motivation for corporate social disclosures: Voluntary disclosures vs ‘solicited’disclosures.Australasian Accounting, Business and Finance Journal, 3(4), 2. Vurro, C., & Perrini, F. (2011). Making the most of corporate social responsibility reporting: Disclosure structure and its impact on performance.Corporate Governance, 11(4), 459-474. Wahba, H. (2008). Does the market value corporate environmental responsibility? An empirical examination. Corporate Social Responsibility and Environmental Management, 15(2), 89-99. White, R., & Hanson, D. (2002). Economic man and disciplinary boundaries: A case-study in corporate annual reports. Accounting, Auditing & Accountability Journal, 15(4), 450-477. Yeganeh, Y. H., Amiri, M., Barzideh, F., & Barzegar, G. (2013). The effect of disclosure level of corporate social responsibility (CSR) On corporate financial performance (CFP) in Tehran Stock Exchange. Journal of Applied Sciences Research, 9(13). Ziek, P. (2009). Making sense of CSR communication. Corporate Social Responsibility and Environmental Management, 16(3), 137-145. Zimmerman, M. A., & Zeitz, G. J. (2002). Beyond survival: Achieving new venture growth by building legitimacy. Academy of Management Review, 27(3), 414-431. University of Ghana http://ugspace.ug.edu.gh 132 Appendix: CSRD disclosure index Corporate social responsibility disclosure index 1- Environment  Environmental pollution control  Compliance with pollution laws and regulations  Prevention or repair of environmental damage  Conservation of natural resources  Using recycled materials  Supporting anti-litter campaigns  Receiving awards on environmental issues  Preventing waste  Designing products and facilities harmonious with the environment  Contributions to beautify the beautify the environment  Supporting and undertaking environmental studies  Utilizing waste materials for energy production  Other environmental disclosures 2- Community development  Donations for community development  Internships or part-time employment for students  Sponsoring public health projects  Aiding medical research  Sponsoring educational facilities and conference  Funding scholarship programmes or activities  Supporting national pride/government sponsored campaigns  Other specific community related activities University of Ghana http://ugspace.ug.edu.gh 133 3- Employees information  Employees health and safety  Employment of minorities or women  Employee training  Employee assistance/ benefits  Employees remuneration  Employees profiles  Employee share purchase schemes  Industrial relations 4- Products  Product development (research and development)  Product safety  Product quality 5- Customers  Quality customer service  Customer satisfaction  Other customer focused initiatives 6- Ethical  Code of conduct 7- Health and safety  General health and safety information that are not specific to any other category  Corporate policies on health and safety standards  Report about the presence of corporate social responsibility or its equivalent on the board and its related members and activities  Other health and safety disclosures University of Ghana http://ugspace.ug.edu.gh 134 8- Others  Information about awards received by the company concerning corporate social responsibility  Other corporate policies on sustainability issues not in the above categories. University of Ghana http://ugspace.ug.edu.gh