CSR GOVERNANCE IN THE NETHERLANDS

CSR GOVERNANCE IN THE NETHERLANDS

Amsterdam, August 2008

Rijksuniversiteit Groningen

Faculty of Economics and Business

Master of Accountancy

Author: J.J. Seraus

Supervisor 1 RuG: R.B.H. Hooghiemstra

Supervisor 2 RuG: C.A. Huijgen

Supervisor 1 PwC: K.J.R. Aben

Supervisor 2 PwC: A.C. ten Klooster

PREFACE

This thesis has been written in the final phase of my study of Accountancy at the Rijksuniversiteit Groningen.

Early on I started to gain interest in the study of corporate governance. Topics as Board structure, remuneration of top executives and the role of the Supervisory Board have always intrigued me, and especially in the context of the financial frauds a few years ago. At the same time I noticed an increase in social and environmental concerns. Companies are increasingly trying to convince the society that they are contributing to a better world. See e.g. the many TV commercials aimed at this objective. I found myself wondering. Why are they doing this? What are the financial rewards? What is the future of CSR?

I am very glad that I could have combined the two very broad topics of corporate governance and CSR in this thesis. As both terms have been gaining prominence, I believe CSR governance will become increasingly important in the future. I have enjoyed the process of writing this thesis and would like to thank the various people who have helped me during this process.

First of all, I would like to thank all the employees of PricewaterhouseCoopers that have taken the time to discuss various issues. Kim Aben, thanks for the many useful tips and suggestions. It was great discussing my thesis with you! Arco ten Klooster, I would like to thank you for the wonderful insights and constructive comments. I really appreciate the time you took to give me your insights on the topic, and always in an enthusiastic way! Finally, I would like to express my gratitude to Reggy Hooghiemstra, my thesis supervisor of the Rijksuniversiteit Groningen, for the guidance and comments on this research. Thanks for the quick and valuable feedback which have truly increased the quality of this study!

SUMMARY

As organisations are being forced to increase their accountability and transparency towards society, they have been increasingly reporting on corporate governance and corporate social responsibility (CSR). Nevertheless, it seems there is still a long way to go when it comes to the disclosure of corporate governance practice with regard to CSR. This study gives insight into current CSR governance reporting in The Netherlands by analyzing 36 separate CSR reports.

I find moderate evidence for the first hypothesis: companies operating in an environmentally sensitive sector disclose more CSR governance information than companies operating in a less environmentally sensitive sector. This hypothesis is based on legitimacy theory, which predicts that companies that are experiencing threats to their legitimacy will disclose more CSR information. Therefore, it can be expected that certain sectors will disclose more CSR information (in this case: the environmentally sensitive sector, containing e.g. companies in the oil sector, transport sector and chemical sector). I furthermore find strong evidence for the second hypothesis: companies with much CSR experience disclose more information on CSR governance than companies with little CSR experience.

However, the fact that there is moderate evidence that companies in the sensitive sector disclose more CSR governance information does not necessarily mean that the legitimacy theory holds, as there can be other unknown factors that have caused this difference. On the other hand, it is a very interesting result. The relation between environmental sensitivity and the extent of environmental information is far more logical. This study shows that the environmental sensitivity of a sector is related to CSR governance information!

This research finds a large variation in CSR governance information between companies, which does no good to the comparability of the CSR reports. Common standards, in the form of benchmarks, can lead to more uniform reporting. These benchmarks already exist for general CSR reporting, but it could be useful to develop a more specific benchmark related to CSR governance reporting. Benchmarks will also make it easier for smaller companies to produce CSR reports.

TABLE OF CONTENTS PREFACE... 1 SUMMARY ... 2 1. INTRODUCTION ... 5

1.1 Background ...5

1.2 Research Questions...6

1.3 Relevance...7

1.4 Structure...9

2.1 Introduction...10

2.2 Corporate Governance ...10

2.2.1 Importance of corporate governance...10

2.2.2 Stakeholder vs. shareholders ...11

2.2.3 The link with corporate social responsibility...12

2.3 Corporate Social Responsibility ...13

2.3.1 Importance of corporate social responsibility...13

2.3.2 Reasons for voluntary reporting ...14

2.3.3 Legitimacy theory...15

2.3.4 Some evidence of the legitimacy theory ...16

2.4 Evidence on CSR Governance Disclosures ...17

2.5 Hypotheses...18

2.6 Summary...19

3.1 Introduction...21

3.2 Data ...21

3.3 Sample...21

3.4 Methods...25

3.5 Summary...26

4. IMPORTANT CSR GOVERNANCE ASPECTS... 27

4.1 Introduction...27

4.2 Determination CSR Governance Factors...27

4.3 Important CSR Governance Aspects ...28

4.3.1 Important themes in CSR governance...28

4.3.2 Structure...29

4.3.2. Strategy and engagement...30

4.3.3. Processes and controls ...30

4.3.4. Assurance and reporting...32

4.4 Interviews...32

4.5 Criteria ...33

4.6 Significance of CSR Governance Score ...36

4.8 Summary...37

5. RESULTS ... 39

5.1 Introduction...39

5.2 CSR Governance Disclosure: Structure...39

5.3 CSR Governance Disclosure: Strategy and Engagement ...42

5.4 CSR Governance Disclosure: Processes and Controls ...44

5.4 CSR Governance Disclosure: Assurance and Reporting ...47

5.3 Hypothesis Testing...48

5.3.1 Hypothesis 1...48

5.3.2 Hypothesis 2...49

5.4 Multiple factors...50

5.6 Conclusion ...52

6. DISCUSSION AND RECOMMENDATIONS... 53

6.1 Introduction...53

6.2 Comparison of Results with Previous Research ...53

6.2.1 Transparency Benchmark ...53

6.2.2 Other studies regarding CSR disclosure...55

6.2.3 Other studies regarding legitimacy theory ...56

6.3 Discussion of the Results ...57

6.3.1 Hypothesis 1...57

6.3.2 Hypothesis 2...59

6.4 Limitations and Recommendations for Further Research ...60

6.5 Recommendations for CSR Reporting...62

REFERENCES... 64

APPENDIX 1...69

APPENDIX 2...70

1. INTRODUCTION

1.1 Background

The research topic of this report is corporate social responsibility governance (CSR governance). In this chapter the reader will gain insight on the topic of this study. First, background information is given regarding CSR governance. Second, the research questions are discussed. Then, the relevance and constraints of this study are presented and the structure of this paper is made clear.

Over the past few years there have been many developments concerning corporate governance. These developments, triggered by the many financial scandals, have been the result of the decreased confidence of society in the integrity of corporate managers and the reliability of financial reports. New regulations have been imposed on corporate managers, auditors and organisations and various new codes of good governance have been developed all over the world (Aguilera & Cuervo-Cazurra, 2004). The academic world has been busy trying to answer questions as in whose interest companies should be run (see e.g. Friedman & Miles, 2002; Gamble & Kelly, 2001; Letza, Kirkbride, Sun & Smallman, 2008). At the same time, there was another development taking place, namely the increased attention to corporate social responsibility. The amount of companies releasing a CSR report has been steadily rising since 1993 (Kolk, 2005; KPMG, 2005). It is not hard to find a link between corporate governance and CSR. Corporate governance implies that attention should be given to ethical, social and environmental issues. In order for companies to claim that they are indeed acting in a socially responsibly way, there should be evidence of an embedment/organisation of that strategy in the organisation. Although the link between corporate governance and CSR is not hard to envision, a thorough understanding of the link and the implications for accountability and transparency of an organisation is still lacking.

The importance of good CSR governance has been emphasized in the past. According to Elkington (2006) the better the system of corporate governance, the greater the chance that we can build towards genuinely sustainable capitalism. The case for this argument is not difficult to understand: no matter how ambitious the views of the executive Board regarding social responsibility, if not enough attention is paid to governance issues, like the structure, responsibilities and control systems, we can not be assured of the degree of attention that is really devoted to CSR in organisations.

As organisations are forced to increase their transparency and accountability towards society, they have increasingly been reporting on corporate governance. Most annual reports in the world include a chapter in which the roles and responsibilities are specified, the various committees are described and in which the company explains why there is non-compliance to a particular principle of the best

practice code that has been followed. In the CSR reports however, this chapter is less common. Although most companies include a section on corporate governance, the information is often at general level and the approach towards corporate governance is not discussed (KPMG, 2005).

Which corporate governance aspects are crucial when it comes to social responsibility? How are companies currently reporting on corporate governance in their sustainability reports? What are the current issues that according to theory can be improved? In which direction should reporting go in the future? This research can not give answers to all the remaining questions regarding CSR governance, but may provide the reader with useful knowledge that can be used to improve the governance structures in the future, which can help companies maximize value for al their stakeholders.

1.2 Research Questions

Most people recognize the importance of governance in the CSR context, but there is still not much evidence of true embedment of CSR in organisations (see Elkington, 2006; Kolk, 2005). Previous research indicates that most sustainability reports lack specifics regarding how corporate governance is structured (Kolk, 2005). Furthermore, there is a lack of understanding concerning how companies fail to be accountable for some aspects of their sustainability performance (Adams, 2004). That corporate governance is important in the sustainability context is widely known, but how to put this in practice is still a question many companies are dealing with. In order to gain insight on CSR governance disclosure in The Netherlands the following research question and sub-questions have been formulated.

1. To what extent are Dutch companies transparent about their CSR governance and how can we explain the differences?

a. Which corporate governance aspects are of main importance for CSR? b. To what extent do Dutch companies report on these important aspects?

c. To what extent can the legitimacy theory explain differences in the disclosure of important CSR governance aspects?

d. To what extent is there a relation between the experience companies have with CSR and the amount of important CSR governance factors disclosed?

e. Which recommendation can we make in order to enhance the quality of CSR reports as to corporate governance?

With this study I try to gain insight on governance disclosure practice in The Netherlands, but I will also try to explain the results of this study by testing some hypotheses. One of the most referenced

theories when trying to explain why CSR disclosures are being made is legitimacy theory (see e.g. Deegan & Gordon, 1996; Guthrie & Parker, 1989; Patten, 1991). According to this theory an organisation’s survival will be threatened if society believes that the organisation is not operating in a legitimate manner. In this case managers will react by disclosing more information in order to change the perceptions of the society (Deegan, 2002). If this view holds, then companies that operate in sectors that receive a lot of negative attention would also disclose more CSR information (Deegan & Gordon, 2006).

The purpose of the study is to acquire more insight into the current CSR governance disclosure in The Netherlands. By firstly analyzing which corporate governance aspects are important, and then analyzing the transparency of organisations judged by these important aspects, knowledge is gained on how we can improve the current reporting by improving governance structures and auditing guidelines. Furthermore, by testing the hypotheses, insight is gained on the motivations for CSR governance disclosure.

1.3 Relevance

Research in the area of social and environmental accounting has been becoming increasingly popular over the past years. As companies started to globalize, the public has been giving more attention to the roles companies can and should play in the world. Many organisations and institutions, such as the Global Reporting Initiative and the World Business Council for Sustainable Development, have published guidance documents for CSR reporting. Fact is that there is still a lack of knowledge and consensus regarding CSR (see e.g. Adams, 2004; Deegan, 2002; Kolk, 2005 ;) and therefore research in this area can be very relevant.

In the past much research has been conducted regarding CSR and CSR reporting (see e.g. Adams & Larrinaga-Gónzalez, 2007; Aguilera et al, 2006; Deakin & Whittman, 2007; Gray, 2006). Much attention has been given to topics like the reasons for voluntary reporting (e.g. Deegan, 2000; Patten, 1992; Roberts; 1992) and the current trends in CSR reporting (e.g. Acca, 2004; Kolk, 2005). However, after reading many of these articles I can conclude that relatively little attention has been paid to corporate governance.

A notable exception to this is the study by Kolk (2005). She conducted a remarkable research, regarding the incorporation of corporate governance aspects in sustainability reports of the year 2004. She concludes that while most companies refer to corporate governance in sustainability reporting, the references are still very general and there is not much detail regarding the actual structures and

responsibilities. She calls for follow-up studies that extend to other aspects of corporate governance that have not been analyzed before in previous research. This research responds to this call by including some other factors and by extending the study from a descriptive study to an empirical study.

Dutch Sustainability Research (DSR) also conducted a study regarding the sustainability reports (DSR, 2006). In this case, various elements of the sustainability reports of the year 2006 of 24 Dutch companies, listed on the AEX-index, were analyzed. One of these elements was corporate governance. A conclusion of this study is that when it comes to corporate governance in general, Dutch companies score well above average when compared to other countries. However, when taking into account the incorporation of sustainability at Board level, things are very different. They find that only three companies mention performance targets related to sustainability for their executives. Furthermore, they conclude that only 13 companies make use of the reporting framework of the Global Reporting Initiative (GRI).

The studies of Kolk (2005) and DSR (2006) both illustrate that there is still a long way to go when it comes to the disclosure of corporate governance practice with regard to CSR. As CSR reporting is increasing at a very high rate, it would be interesting to study the current situation. Adams (2002) reminds us that increased reporting does not necessarily mean improved accountability; this is an issue to keep in the back of our minds. Increased reporting on governance is only a benefit if it is the result of increasing corporate governance mechanisms that truly make it easier for companies to reach all their CSR goals.

In The Netherlands there is a yearly evaluation of the transparency of companies regarding CSR, namely the Transparency Benchmark. By giving companies the opportunity to compare their reports on various criteria, companies gain insight into best practices. Thus, there is already knowledge concerning the transparency of Dutch companies regarding CSR. This research brings another point of view by focusing exclusively on the transparency of governance aspects. Why is this focus so important? Previous research has concluded that social and environmental activities are not focused at the Board level, nor is there understanding of how Boards and corporate governance systems work (Elkington, 2006) and that a significant cause of breaching responsibility standards lies in the misaligned performance management systems (Mackenzie, 2007). CSR governance is crucial for achieving CSR goals and this research provides insight on how Dutch companies are currently reporting on some important governance aspects.

1.4 Structure

The following figure illustrates the structure of this study. Notice that the process was an iterative process: the research design phase overlapped much with the literature. And after analyzing some CSR reports, I realized it was necessary to reformulate/change some of the criteria points.

Nevertheless, figure 1 gives a good representation of the different phases of this research.

Figure 1 I Research Design VI Analysis Results V Analysis CSR Reports III Development Criteria IV Validation Criteria II Literature Study VII Reccomendations

The structure of the paper is as following. In the next chapter I will discuss the important concepts and theories regarding corporate governance and CSR. In the third chapter the various methods and procedures used in this study will be presented. The fourth chapter addresses the development of the criteria by which the CSR reports will be judged. Chapter five contains the results of this study and chapter six ends with a conclusions and recommendations that can be made with regard to CSR reporting and further studies.

2. THEORETICAL BACKGROUND

2.1 Introduction

In this chapter the most important theoretical concepts regarding CSR governance are described. In the next section I discuss corporate governance concept, paying specific attention to the importance of governance, the stakeholder vs. shareholder debate and the link of corporate governance with CSR. The third section addresses CSR, namely the increasing importance of CSR, the reasons for voluntary reporting, the legitimacy theory and evidence on the legitimacy theory. Then, section four contains the results of previous research regarding the disclosure of CSR governance issues. Next, the hypotheses are presented and section six ends with a summary.

2.2 Corporate Governance

2.2.1 Importance of corporate governance

‘Corporate Governance is concerned with holding the balance between economic and social goals and between individual and communal goals. The governance framework is there to encourage the efficient use of resources and equally to require accountability for the stewardship of those resources. The aim is to align as nearly as possible the interest of individuals, corporations and society.’

Sir Adrian Cadbury

‘Corporate governance is concerned with the regulation, supervision, or performance and the conduct oversight of the corporation’ (Letza et al 2008, p. 18). Over the years the term corporate governance has been used quite a lot, as the confidence in the management Board and supervisory Board reached a low point. The scandals of the past few years and the increasing importance of social responsibility have led to many corporate governance reforms. Governments, professional associations and stock exchanges have been increasingly developing ‘codes of good governance’. According to Aguilera & Cuervo-Cazurra (2004), the exponential growth in the number of codes of good governance started after the publication of the Cadbury Report in 1992. This report, published by the Committee on the Financial Aspects of Corporate Governance, was issued to tackle the problems with corporate governance and financial reporting within companies. The trend of code development is not only visible in countries that traditionally publish a great amount on corporate governance, like the United States, Germany and Japan. Corporate governance is gaining worldwide attention!

Aguilera and Cuervo-Cazurra (2004) identify two processes that have led to the increased attention to corporate governance. According to them, the globalisation and transformations in ownership structures of firms increased the importance of good corporate governance systems. The importance of corporate governance has been emphasized in the past. Research from the Economist Intelligence Unit (2001) reveals that more than 80% of European and American investors are willing to pay more for a company with good governance. Bushman & Smith (2001) emphasize the importance of governance research, as it can give relevant insight on the impact of accounting information on the allocation and utilisation of resources. They also call attention to the fact that the effectiveness of control mechanisms can have a great influence on the decisions of managers. Research from Bollen et al (2005) also highlights the importance of corporate governance. They conclude that 37% of high business failures are related to fraudulent and/or unethical behaviour. They also found that at most companies, where fraud had taken place, ineffective corporate governance mechanisms where in place.

2.2.2 Stakeholder vs. shareholders

Most discussions on corporate governance concern the two theoretical views on corporate governance: the traditional view, in which the shareholders take a central place, and the stakeholder view, in which managers should take into account the rights of all stakeholders. It is remarkable to note that although most people agree that corporate governance should be about the society, there is much less agreement to how to put this in practice (AccountAbility, 2007). According to Letza et al (2008, p 17): ‘The corporate governance malpractice such as the scandals of Enron and WorldCom and the dramatic decline of stock markets at the beginning of the new century have fuelled an old-aged debate on the fundamental issues of corporate governance: for what purpose the corporation exists and whose interests it serves’. The traditional view encompasses the view that corporate governance should ensure that shareholders’ value are transformed into company performance (Carver, 2007) and is supported by many academics (see Sundaram and Inkpen, 2004). The alternative view is that the concern of corporate governance should add value to as many organisational stakeholders as is practicable (Bain and Band, 1996) and is also supported by various scholars (see Freeman et al, 2004).

Baums (2005) suggests two premises that proponents of the shareholder view hold. Firstly, the Board has no legitimacy to make the trade-offs between stakeholders required when adopting the stakeholder view. In addition, the only guide the Board has in making the trade-off is its personal preferences. Second, he acknowledges that the shareholders play a central role in the functioning of the firm. Keasey et al (1997) suggests two ways that the efficiency of the stakeholder view can be demonstrated. They claim that the trust relation with the various stakeholders will lead to profitable

investments and mutually beneficial exchanges. According to them, the efficiency is also demonstrated by the success of countries, like Japan and Germany, that are adopting this approach.

In the past each side assumed that their own theory is superior, but Letza et al (2008) have been calling for a pluralistic view of corporate governance, where corporate governance is always moving back and forward between the two ends, depending on various factors. The reason for this is that corporate governance should always be changing due to internal processual impetuses and external environmental dynamics. According to this view, there is no right way of corporate governance and corporate governance should be formed by more factors than just economic logic. An example of this pluralistic view is the enlightened stakeholder model and enlightened value maximisation by Jensen (2001), where there is acceptance of the stakeholder theory, while at the same time believing in value maximisation for the stakeholders.

The Dutch corporate governance code (code Tabaksblat) builds on the view that ‘the management Board and supervisory Board have overall responsibility for weighting up the interests of all the stakeholders, generally with the view to ensure the continuity’, attempting to create long-term shareholder value (Corporate Governance Committee, 2003). The confidence of shareholders in the management and supervisory Board is crucial for organisations. According to the Dutch corporate governance code this confidence can be increased by good entrepreneurship and proper supervision thereof.

2.2.3 The link with corporate social responsibility

Why did the corporate governance agenda start to include more societal concerns? A key reason identified by Elkington (2006) is the increased attention to ‘triple bottom line’ (TBL) reporting, which encompasses that organisations can create value in three dimensions: economic, social and environmental. Elkington claims that the TBL agenda is the responsibility of the corporate Board, but acknowledges that at the moment the TBL activities within organisations are not focused at the Board, nor is there understanding of how Boards and corporate governance systems work. He concludes that the TBL agenda is moving from ‘factory fence to the Boardroom’, while recognizing at the same time that we are a long way from achieving this.

An interesting contribution comes from Michael Power (1991), who adapted the original principal-agent relationship, to include accountability to society as a result of the environment. The principal is the society and the agent is the company. This shows the complexity for managers in dealing with accountability to all stakeholders. According to Kolk: (2005, p.4): ‘It is here that reporting on corporate governance and sustainability seem to merge, most notably when companies start to report

on both aspects in one report, when issues become related because of these combined accountability efforts or due to the fact that corporate governance starts to include ethical (or social/environmental) concerns and sustainability covers control and accountability structures.’

Rossouw (2005) claims that corporate governance has an ethical character, as typical issues that have to do with corporate governance, for instance Board structure, are there to ensure that the corporation is acting in a manner that is fair, accountable, responsible and transparent. Not only academics see the ethical character, but also companies recognize this ethical character of corporate governance. A large study by PricewaterhouseCoopers and Nyenrode University (2001) among 2500 Dutch companies shows that almost one third of all the companies indicated that social and ethical issues are important in the corporate governance debate. Rossouw does however make a distinction between the ‘ethics of governance’ and the ‘governance of ethics’. This notion of ‘governance of ethics’ comes very close to CSR governance. It is about the management making sure that the ethical standards are followed throughout the companies. As ethical standards are very much related to social and environmental issues, it becomes clear that the ‘governance of ethics’ is the focus of my study and not the ‘ethics of governance’.

2.3 Corporate Social Responsibility

2.3.1 Importance of corporate social responsibility

According to the European Commission (2000): ‘Corporate Social Responsibility (CSR) is a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis’. The goal of sustainable development is to ‘meet the needs of the present without compromising the ability of future generations to meet their own needs’ (World Commission on Environment and Development, 1987, p. 43). At present there are many different terminologies to refer to topics of social, environmental and economic performance (KPMG, 2005). Not only are there different terminologies, but different meanings are given to terminologies.

The importance of CSR is illustrated by the large amount of research concerning this topic (see e.g. Adams & Larrinaga-Gónzalez, 2007; Aguilera et al, 2006; Deakin & Whittman, 2007; Gray, 2006). CSR is receiving increasingly more attention and companies react to this by publishing even more non-financial information (see e.g. Cormier & Magnan, 2003; Harte & Owen, 1991; Niskala & Pretes, 1995). Nevertheless, there are many uncertainties and there is lack of consensus regarding CSR. According to Deegan (2002) the reporting objectives are not clear and there is no consensus on the

qualities the information should possess. According to him, it is also not clear who the audience of the reports is, and what the best presentation format is.

It is not necessary to witness the increase in academic publications to notice the increased importance of CSR. By taking a look at the topics in the media and paying attention to commercials on TV, I notice an increased focus on social and environmental issues over the past years. Clearly, there must be an advantage to being socially responsible, because much money is spent on showing the society that companies are acting in a responsible way. Companies are not required by law to publish much of this information and this is probably why the reasons behind the disclosure of non-financial information have been analyzed so much in the past. The following section summarizes the most common identified reasons for voluntary reporting.

2.3.2 Reasons for voluntary reporting

The previous section illustrates that companies seem to believe that there is an advantage in reporting on social and environmental issues and therefore many academics have been keeping themselves busy with researching the motives behind CSR disclosure. Deegan (2002) identifies some motivations behind the decisions of managers to voluntarily undertake activities.

 Compliance with legal requirements (Deegan, 2000), borrowing requirements and industry requirements.

 Business advantages in doing the right thing: ‘economic rationality’ (Friedman, 1962)

 A belief in accountability or responsibility to report (Hasnas,1998; Donaldson and Preston, 1995)

 To attract investment funds  To avoid any more regulations  To win reporting awards

 To manage stakeholder groups (Ullman 1985, Roberts 1992)  To comply with community expectations

 As a result to threats to the organisation’s legitimacy (Deegan et al 2000, 2002, Patten 1992)

Grossman (1981) and Milgrom (1981) claim that when investors believe that managers are withholding information, investors believe that the undisclosed information is negative. The result is that investors will lower the value of the firm. Therefore they claim that corporations should reveal all the information that they possess.

When it comes to the thoughts of managers regarding the decision of voluntarily disclosing information, much has been suggested by academics. Berthelot, Cormier and Magnan (2003) suggest that managers are unlikely to release information if they expect that the information will lead to

proprietary costs. Li, Richardson and Thornton (1997) conclude that less information will be disclosed by firms facing serious environmental problems. Dye (1985) pays attention to the role of the media. He shows that when the media exposes information on environmental performance of a particular company, this company is more likely to reveal information about environmental incidents.

2.3.3 Legitimacy theory

When trying to explain why companies make disclosures regarding CSR, many academics refer to the legitimacy theory (see e.g. Deegan and Gordon 1996; Guthrie and Parker, 1989; Patten 1991), which suggests that managers disclose social and environmental information to legitimize various aspects of their organisations. Central in the legitimacy theory is the notion of a ‘social contract’, which is an agreement between the organisation and society regarding what is deemed to be acceptable and legitimizing behaviour (see Mathews, 1993). A basic assumption of this theory is that organisations do not have any inherent right to resources and exist only if the society deems them legitimate. So when managers think that they are breaching the social contract, this theory predicts that managers will disclose information in order to have effect on society. Lindblom (1994) suggests four courses of action that an organisation can take to obtain or maintain legitimacy:

 Education and information to the public on the changes of its performance and activities  Changing the perception of the ‘relevant publics’

 Manipulation of the perception by deflecting the attention away from the issue of concern  Changing of the external expectations of its performance

The legitimacy theory is the most used theory to explain why managers disclose information on social and environmental aspects. One very important paper regarding this theory is the paper of Guthrie and Parker (1989) in which they analyzed the relation between disclosure policies and significant events. Although they did not have enough evidence to prove that the legitimacy theory can be used to explain the disclosures, they motivated many other academics to do research on this topic (Deegan, 2002) and there is currently much evidence that managers adopt legitimizing strategies (see for example Patten, 1992; Grey, Kouhy & Lavers, 1995; Walden & Shwarts, 1997). Patten stated (p475): ‘…it appears that at least for environmental disclosures, threats to a firm’s legitimacy do entice firms to include more societal responsibility information in its annual report’.

This does not mean however, that this theory is well developed. According to Deegan (2002) the theory has many gaps: there is a general lack of knowledge regarding the difference between different groups in the society and there is a lack of knowledge concerning the terms of the ‘social contract’. Also, many other motivations have been used to explain social disclosures. One of these is that managers should only disclose positive information, as negative information would lead to a decrease of the cash flows (Verrechia, 1983; Dye, 1985). Research of O’Dwyer (2002) indicates that there

could be cultural differences in legitimization strategies. While he accepted that disclosures are reactive to social pressures, he also concluded that in Ireland legitimizing disclosures are unlikely to succeed as people did not tend to emphasize on the positive achievements. Recent research of Brammer and Pavelin (2008) finds that media exposure of companies plays no role in stimulating voluntary disclosures.

An interesting contribution comes from Parsons (2005) who asked organisations to identify why they disclose non-financial information. He concludes that the two most important reasons given, namely reputation enhancement (71%) and confidence of investors, insurance providers and financial institutions (61%), constitutes part of ‘impression management strategy’ (see Hooghiemstra, 2000) and ‘public relations effort’ (see Elkington, 1997). Other important reasons for the disclosure of non-financial information identified by Parsons are operational and management improvements (59%) and improvements in risk management (51%).

2.3.4 Some evidence of the legitimacy theory

As mentioned before, one very influential paper was the paper of Guthrie and Parker in 1989, where the disclosure of BHP Ltd (large Australian company) over 100 years was related to public concern. Although they could not provide evidence on the legitimacy theory, they have inspired many others to conduct further research. A remarkable contribution to the study of Guthrie and Parker is the study of Deegan et al (2002), where they updated the work of Guthrie and Parker by using a different measure of environmental concern. The corporate social and environmental disclosures over a period from 1983-1997 was correlated among a measure of public concern, namely media attention. They find support for the legitimacy theory by finding a positive correlation between media attention for particular issues and the disclosure on these particular issues. Their study highlights the power the media can have in influencing reporting.

Adams et al (1998) conducted research among a sample of 150 reports of six European countries and concluded that the legitimacy theory is important in explaining differences in reporting, even among countries. They find that large companies disclose more ethical, employee and environmental information. The starting point is the realisation of Bewley and Li (2000) that larger firms are more likely to be targeted by environmental pressure groups. Furthermore, the industry sector has an influence on the amount of employee and environmental information. (Adams et al also conclude that the reasons for the differences cannot be explained by the legitimacy theory and that there are certainly country-specific effects).

An interesting contribution comes from Deegan & Gordon (2006), where they investigated the disclosure practices of a large amount of Australian corporations. One of their objectives was to determine whether environmental sensitivity is related to the extent of corporate environmental disclosure. In order to classify the organisations into groups, they did not use previous research, but determined the sensitivity directly by asking environmental lobby groups to rate the industry according to the extent that is has been targeted by the lobby group over the past five years. They conclude that there is a positive correlation between environmental sensitivity and the level of corporate environmental disclosure.

O’Donovan (2002) also brings relevant and valuable insight on the legitimacy theory, by acknowledging that legitimization strategies will depend upon whether managers are trying to gain, maintain or repair the legitimacy of the corporation. Another important insight he brings is that organisations that rely more on legitimacy for commercial purposes, will have to react more to legitimacy threats. He finds evidence for the legitimacy theory, as the results show that managers are more likely to disclose information when there are legitimacy threats to the organisation.

2.4 Evidence on CSR Governance Disclosures

Although many academics recognize the importance of CSR governance (see Rossouw, 2005; Elkington, 2006) there has not been much research on the disclosure practices of companies concerning CSR governance. A notable exception is the research of Kolk (2005). According to Kolk, there are two angles that lead to the increased call for transparency. First of all, she mentions the expansion of accountability requirements regarding corporate governance, leading to the inclusion of ethical aspects. Secondly, the increased attention to sustainability has led to the broadening of the scope of sustainability. While sustainability used to be mostly about environmental performance, it now also includes issues as ethics, social issues and organisational structure.

Kolk (2005) examined to what extent the largest 250 companies in the world incorporate governance aspects in their non-financial reports of the year 2004. Table 1 summarizes the results. Her results show that organisations do recognize the link between governance and CSR, but she acknowledges that the information given is very general. She sees much opportunity for more research on this topic and calls for further research on aspects of corporate governance that have not been covered in her study. However, no concrete examples are given for further aspects of corporate governance that can be studied.

TABLE 1

Governance Factors % of Companies

1. Structure of sustainability within organisation 38.0%

2. Existence of separate sustainability department/unit 31.0% 3. Indication of person/body finally responsible for sustainability 34.1% 4. Separate section on corporate governance in the report 55.0% 5. Corporate governance specially linked to sustainability issues 51.2%

6. The existence of code of conducts/ethics 65.9%

7. Existence of complaint mechanisms linked to sustainability 30.2%

8. External verification of sustainability reports 34.1%

*Adapted from Kolk (2005), the percentages refer to sustainability reports and not to the integrated reports. Of the 250 companies, 129 companies had a separate report.

The Dutch Sustainability Research (2006) concludes that the incorporation of CSR governance still has a long way to go. Although general corporate governance seems to be well integrated into the Dutch companies, the specific link of CSR and governance is still something that can be improved in the future (DSR, 2006)

Various studies highlight the difficulties regarding the investigation of the link between corporate governance and sustainability. Aguilera et al (2006) underline the troubles that scholars find when trying to analyze the link between CSR issues and corporate governance. They complain about the lack of consistently presented, comparable, internal data. Kolk (2005) accentuates that it remains unclear how the linkage between corporate governance and sustainability can be best disclosed.

2.5 Hypotheses

The legitimacy theory predicts that companies that are experiencing threats to their legitimacy would disclose more CSR information (see e.g. Adams et al 1998; Deegan and Gordon, 2006). In this case, it can be expected that certain sectors would be more prone to threats to their legitimacy and therefore these sectors can be expected to reveal more CSR information. In this study I will call this sector the ‘environmentally sensitive’ sector. This sector can be expected to receive much attention from the media regarding corporate responsibility. Companies, that receive fewer threats to their legitimacy, belong in this study to the ‘less environmentally sensitive’ sector. The next chapter gives further information regarding the classification of corporations into the two sectors. If we want to study whether the legitimacy theory holds when trying to explain motivations for voluntary CSR governance disclosure, the following null hypothesis can be stated:

H0: Companies operating in environmentally sensitive sectors disclose as much CSR governance information as companies operating in less environmentally sensitive sectors

The alternative hypothesis is:

H1: Companies operating in environmentally sensitive sectors disclose more CSR governance information than companies operating in less environmentally sensitive sectors

Although the legitimacy theory has been supported many times (see e.g. Adams et al, 1998; Deegan et al 2002; Patten, 1992), research from O’Dwyer (2002) suggests that legitimization strategies might vary between countries, as in some countries, reporting to fulfill the legitimacy gap might be counterproductive. For these countries corporate disclosures must be explained by other theories than the legitimizing theory. I believe that there could be a very simple reason explaining the amount of CSR disclosure, namely the amount of experience a company has regarding CSR experience. The more experience the company has with CSR, the more internal information the company will possess that can eventually be disclosed. This thought has been proven right by ACCA (2004), where they find that Western Europe, having the greatest amount of experience regarding CSR reporting, has the greatest productivity regarding qualitative and quantitative non-financial reporting. Western Europe is also found to have the most external assurance given on non-financial reports. It makes sense that companies with much CSR experience will have CSR better incorporated in the organisation, by being much further with the governance of CSR.

The following null hypothesis can be stated.

H0: Companies with much CSR experience disclose as much information on CSR governance as companies with little CSR experience

The alternative hypothesis:

H2: Companies with much CSR experience disclose more information on CSR governance than companies with little CSR experience

2.6 Summary

It is clear that there is a growing interest in the research of corporate governance, social and environmental accounting. As corporate governance has been increasingly expanding to include societal concern, CSR has increasingly been associated with governance issues. Research on CSR

governance is therefore a logical consequence of the former and latter expansion. As the legitimacy theory is a very popular theory when it comes to explaining managerial motivations behind voluntary disclosure, this theory will be tested by a hypothesis. Experience with CSR can also have an influence on the amount of information that companies will disclose, and therefore this also will be tested.

3. RESEARCH DESIGN

3.1 Introduction

This chapter contains the research design of the study. The second section deals with the selection of data for this research. An important factor for the success of this research is the sample, and this is addressed in the third section. The methods and procedures to answer the various sub-questions are discussed in the fourth section. Finally, section five ends with a conclusion.

3.2 Data

The data selection is an important part of the research design, as it will have a great effect on the results of this research. Corporate social and environmental information can be found in a variety of sources: the straightforward annual reports and CSR reports, but also in interim reports, newspaper advertisements, interim reports, specific reports on issues and press releases. In order to get a complete view of the practice of one company all of these sources would have to be checked. As the former is not the objective of my study, I only examined the CSR reports of the year 2006. As companies are increasingly publishing a separate CSR report, it would be logical to focus on these CSR reports. Prior research by Unerman (1999) furthermore suggests that an exclusive focus on annual reports is likely to result in an incomplete picture of reporting practices, as companies provided more CSR information in other reports than annual reports. Logically, it would be more interesting to limit my study to the CSR reports, than to annual reports. Websites currently also contain much CSR information and many CSR reports contain references to the website of the company. However, I choose to only analyze the content of the CSR reports due to time constraints. I analyzed the content of the CSR by focusing on categories of data rather than the more common analysis of single words or phrases. This way, insight is acquired on whether particular information is disclosed rather than the amount of information disclosed.

3.3 Sample

The intention of this study was to focus on one country in order to exclude cultural effects. Previous research has indicated that cultural factors play an important role in the disclosure of companies (see for example Adams et al, 1998; Gray, 1998; Hofstede, 1980). To exclude these cultural differences, I analyzed the CSR reports of Dutch companies. Cultural differences are surely not the only reason behind the choice of Dutch companies. As separate CSR reporting is something of the last few years, there is currently not much knowledge on what is being reported by various Dutch companies. This is logical, as the many rules, regulations and guidelines that are common for financial reporting are not

developed for non- financial information; at least not to the extent of financial reporting. Therefore, this research could provide us with new information regarding current CSR governance disclosure in The Netherlands. The Netherlands ranked first in the 2006 research by Dutch Sustainability Research (DSR, 2006), where the sustainability performance of several countries was compared to each other. We know that Dutch companies are very far regarding CSR reporting, and this enables us to get more in-depth insight into CSR governance reporting. Also, the fact that this same study finds that the Netherlands are doing very well regarding general corporate governance reporting, but still have a long way regarding CSR governance information, makes this study very interesting. I expect that by studying the reports in the Netherlands, which is far in CSR reporting, interesting conclusions can be made in the end (rather than having many companies revealing almost no information on CSR governance).

The sample is also limited to the availability of a separate CSR report. Findings of Unerman (1999) suggest that more information on CSR is found in other reports than the annual report, but many Dutch companies do not have a separate report and publish the CSR information in other sources than the CSR report (or do not publish any CSR information at all). However, I still decided to only analyze the CSR reports for many reasons. First, and this is the most important reason, the current trend is that companies are increasingly publishing a separate report. The KPMG International Survey of Corporate Responsibility Reporting 2005 finds that while in 2002 45% of the top 250 companies in the Fortune 500 (G250) and 23% of the largest 100 companies in 16 countries (N100) companies published a separate report, this percentage has risen to 52% respectively 33% in 2005 (KPMG, 2005). The separate CSR report is thus gaining importance and an exclusive focus on these reports could give relevant insight in current CSR reporting. Second, the sample containing companies with a separate CSR reports is surely large enough, taking in consideration the large amount of criteria points that are being analyzed. Third, for the companies publishing a separate CSR report the findings of Unerman still hold. It would be logical that when companies have a separate CSR report, it would contain more CSR information that the annual report.

With an internal database, I searched for all the companies that have published a CSR report over the year 2006. This database contained information on 174 large Dutch companies. Further information on this internal database:

 98 listed companies

 76 non-listed companies (largest companies in the Reach database on July 1st_{, 2005)}

To gain reasonable assurance on the completeness of the database I checked whether all the companies on the AEX index and AMX index were included in the database (this was the case). After searching through the database I found 40 companies with a separate CSR report.

In order to test the legitimacy theory it was necessary to further divide the sample into two groups. This was done with the use of previous research of Adams et al (1998) and Deegan and Gordon (2006). The classification of the companies into industry groups is crucial in this study, as the results are very much influenced by this classification. Therefore I paid careful attention to this process of classification. It is very important that the companies that have been classified in the environmentally sensitive group truly receive more attention from the media and lobby groups. For the classification in different industry groups my starting point was the method used by Adams et al (1998), as it is supported by various previous findings (Dierkes & Preston, 1977; Roberts, 1992). I started by categorizing all the companies that have published a CSR report over the year 2006 by this method. Adams et al (1998) classified the companies in four groups:

 Oil, chemicals, metals and power  Manufacturing and autos

 Engineering and construction  Service, food and retail

Hereby they used the findings of Cowen, Ferreri & Parker (1987) to divide the groups into two groups. The findings of Cowen et al (p. 113) are that ‘consumer oriented industries can be expected to exhibit greater concern with demonstrating their interest in social responsibility since corporate image among the mass market consumers is likely to have an influence over the amount of sales generated’. According to Adams, this classification is supported by the results of prior research in this area. The first two groups can be regarded as functioning in ‘sensitive’ areas and the last two groups as functioning in ‘less sensitive’ areas.

I started by adding Dutch companies that operate in the oil, chemicals, metals and power industry, to the environmentally sensitive group. I did not take consideration of the group ‘manufacturing and autos’ as there were no Dutch organisations operating in the auto sector. The classification in the sector manufacturing I found to be very dubious, so I did not take this in consideration. I found five companies that belonged in the environmentally sensitive group, based on this initial categorization.

Another classification is given by Deegan & Gordon (2006). By asking lobby groups directly to rate the industry groups according to the extent that they have targeted them over the past few years, similar results are obtained. The most sensitive industries are uranium mining, chemicals, coal, transport, oil/gas explores, plastics manufacturing, oil/gas producers, paper merchants and timber products. Most of these industries fall under the most sensitive group ‘oil, chemicals, metals and power’. I used the insight of Deegan & Gordon, to add some more companies to the environmentally sensitive group. In my opinion these companies did not clearly fit in one of the categories by Adams

et al (1998). These were three companies in the transport sector and five in the print media/paper sector.

However, the combination of insights of Adams et al (1998) and Deegan and Gordon (2006) still did not leave me satisfied that I had good criteria to categorize the groups. My main concern was the two dredging companies in my sample, which were the only two Dutch companies in the sample that fell under the initial categorization ‘engineering and construction’. In my view, these companies clearly operate in environmentally sensitive sectors, as their decisions and operations can have a big impact on the environment (see e.g. Subba Rao, Srinivasa Rao, Iyer & Chittibabu, 2008). Dredging activities cause much pollution to land, air and water and creates much waste. Therefore I added these two companies to the environmentally sensitive sector. In the end, I classified fifteen companies in the environmentally sensitive sector and the other 25 companies were placed in the less environmentally sensitive sector.

The final categorization is as following:

 Environmentally sensitive sectors: oil, chemicals, power, transport and print media/paper (15 companies)

 Less environmentally sensitive sectors: service (banking and assurance providers), food, retail, telecommunication, personal hygiene, IT hardware and electronic production (25 companies)

See appendix 1 for more specific information

However, after further research, I discovered that four companies published a social report. Further analysis of their reports made me exclude three of these reports out of the sample, as these reports focused only on social issues, and excluded environmental and economic issues. These reports focused mostly on employee issues within the company and the broader view of corporate responsibility was not encompassed. Lastly, I excluded one company that published a separate environmental report and a separate social report. This was also due to the reason that this company did not have a broad view of corporate responsibility. None of the excluded companies reported on the structure of CSR within the company and the governance mechanisms. In the end, the sample consisted of 12 companies operating in the environmentally sensitive sectors, and 24 companies operating in the less environmentally sensitive sector.

This final sample contains 24 listed companies and 12 non-listed companies. Most listed companies included are on the AEX index (16 companies) and one company is on the Euronext-100. Furthermore, the sample contains five midcap companies and two smaller listed companies. Various industries are represented in the sample. The largest representation comes from the

food/beverages/daily products (7), the banks (5) and medical insurance (5). For more information on the sample see appendix 1.The different industries make it possible to get insight in industry practice regarding CSR governance disclosure and could bring other relevant insights. The same counts for the classification in listed and non-listed companies. To ensure that the two samples (environmentally sensitive vs. less environmentally sensitive) are similar regarding size, I tested whether there were differences between the samples regarding revenues and number of employees. No differences were found when testing with a parametric technique and a non-parametric technique.

3.4 Methods

In order to test hypothesis 1 I used a Mann-Whitney test to see whether there are differences in the probability distributions of the two samples. The Mann-Whitney test is a non-parametric statistical test, which means it does not rely on assumptions about the distribution of the data. Because I do not have much knowledge of the distribution of the data, I decided to choose for this method.

Hypothesis 2 concerns the amount of CSR experience that companies have. I decided to determine the amount of CSR experience of each company by checking how many separate CSR reports they have published. There is one disadvantage to this method: companies could have been disclosing social and environmental information long before publishing a separate annual report. But to analyze all the annual reports to determine when the companies started to report on CSR issues would be very time consuming. It would also require a subjective view on when companies really started to report on CSR, because even very old reports would contain at least some social and environmental information. The amount of years of experience that companies have in CSR reporting was compared to the amount of information they provide by a regression analysis. To be more certain about the completeness of the number of years, I only take in consideration separate CSR reports from 1990-2006. This means that the maximum years of experience that a company can have would be 17. To gain information about the number of years a corporation has been publishing a separate CSR report, I first checked the CSR report of the year 2006. When the information was not found, the website was visited and the earliest report was analyzed to verify whether this is the first report. So, I did not just take the year of the earliest report, but checked whether in that report is explicitly stated that it is the first report. However, for six companies I did not find the information in the reports or on the website. I contacted these six companies and received an answer from two companies. Therefore, hypothesis two was tested taking in consideration 32 companies. The following companies are excluded from the analysis: Unive, CZ groep, Menzis and Reed Elsevier.

3.5 Summary

This chapter addressed the design of the study. The hypotheses will be tested by analyzing the 2006 CSR reports of 36 Dutch companies. For the purpose of testing whether the legitimacy theory holds regarding CSR governance disclosure, the sample was divided in two sectors based on previous research (environmentally sensitive sector vs. less environmentally sensitive sector). A regression analysis will be conducted to study the influence of the amount of years a corporation has been publishing a CSR report on CSR governance information. The following chapter deals with the important governance factors and the development of criteria by which the 36 companies were judged.

4. IMPORTANT CSR GOVERNANCE ASPECTS 4.1 Introduction

This chapter addresses the important CSR governance aspects and gives an answer to the first sub-question of this study. This sub-question is an important starting point for this research, as before any conclusions can be made about CSR governance disclosure we should determine which CSR governance aspects are important. The next section identifies the difficulties in trying to determine relevant CSR governance factors and describes the method by which these factors were determined. Section three contains four important themes when it comes to CSR governance and includes the link of these themes to academic literature and business practice. The fourth section explains how the interviews with experts were conducted. In the next section, the 60 criteria points are presented and in the following section the significance of the scores of the companies is discussed. Lastly, this chapter ends with a conclusion.

4.2 Determination CSR Governance Factors

The first sub-question of this study concerns which governance aspects are important for CSR. As I will analyze the CSR reports, this study does not specifically focus on governance practice within companies, but on the transparency of governance in the CSR reports. Since previous research (e.g. Deegan, 2002) indicates that there is no consensus on the qualities that information in CSR reports should possess, the important governance factors are not straightforward. There are no theories that describe which CSR governance factors companies should disclose and this makes it difficult to give an opinion regarding CSR governance within companies.

Fortunately, there are many theories and guidelines regarding both corporate governance and CSR. These provide relevant insight on which governance aspects might be important. Currently the GRI guidelines are the dominant guidelines in CSR reporting. The mission of GRI is to fulfil the current need for more transparency on economic, environmental and social impacts and the need for more open and clear communication on sustainability by providing a credible framework for sustainability reporting (GRI, 2006). In this framework, there are ten principles regarding CSR governance. (See appendix 2) However, these principles apply to CSR reporting in general (in annual report and/or CSR report) and are not specifically for the CSR report.

In The Netherlands there is currently a Transparency Benchmark, with the objective of giving insight on the transparency of social and environmental information of the largest Dutch corporations. The main motive for the development of this benchmark was to give an impulse to the dialogue between

corporations and their stakeholders. In this benchmark companies are judged in ten categories, relating to environmental, social and economic aspects. In this benchmark, there are some factors present that are related to CSR governance.

By combining the GRI and the Transparency Benchmark, using insight from research (for example Kolk, 2005) and getting input from experts, relevant governance topics have been identified. In order to be able to accept/reject the hypotheses, points will be allocated to the various sub-components of these important topics (each criteria point is worth one point). These sub-components of the important governance topic form the criteria, by which the CSR reports will be judged. When constructing the criteria the following was kept in mind:

 Even though the total score of a company does not necessarily mean that the company has better governance mechanisms, I tried to incorporate in the criteria the fact that good governance action is better than just governance disclosure.

 The importance of gaining insight into current CSR reporting was deemed more important than developing criteria that is the best description of good governance. This point can be illustrated with the following example: Although reporting on the amount of violations of a code of conduct, might not be crucial for good governance, or might not be crucial for certain sectors, reporting on the amount of violations is still included in the criteria. This is so we can gain insight in current reporting. Therefore, the main goal of the criteria is to be able to measure the degree of reporting, and not mainly to suggest the best way of CSR reporting. Based on the results of the study, best practice might be able to be suggested though.

4.3 Important CSR Governance Aspects

4.3.1 Important themes in CSR governance

‘Corporate governance is concerned with the regulation, supervision, or performance and the conduct of oversight of the corporation’ (Letza et al 2008, p. 18). In order to answer the first sub question, I identified four topics that I believe are important for CSR governance reporting:

1. Structure

2. Strategy and Engagement 3. Processes and Controls 4. Reporting and Assurance

An important insight into corporate governance comes from Hart (1995) who states that corporate governance issues only arise when there is agency problem and when transaction costs are so high, that the agency problem cannot be solved through a contract. Keeping this in consideration, the

following part explains the choice of these topics and links these themes to academic literature on corporate governance and CSR, but also to current business practice.

4.3.2 Structure

In this study the theme ‘structure’ refers to the internal organisation of CSR (= functions and responsibilities, including the role of the Management Board and the Supervisory Board). It is then not hard to envision why structure is important in the CSR context: in order to be able to supervise and conduct oversight over (CSR) performance, it is necessary that people can be held accountable for their actions.

The many accounting scandals in the beginning of this century have taught us that there was a lack of proper supervision of corporations. These scandals have started the debate on corporate governance, where much attention has been paid to the corporate governance structure. Since the nineties a great deal of interest has been paid to the ownership structure (see e.g. La Porta et al, 1998), but also to the internal structure of corporations, and especially to Board composition (see e.g. Brickley, Coles & Terry, 1994; Daily and Dalton, 1994). An interesting result comes from the study of Campbell (2002), where he concludes that some of the variability in the volume of social disclosure can be explained by changes in the chairperson of the corporation. The academic link between structure and CSR can also be made. An interesting study, which links governance structure to CSR performance, comes from de Graaf and Herkströter (2007). They show that there is a link between corporate governance structure (in this specific case: the role of stakeholders) and corporate social performance. In the context of structure the presence of committees immediately comes to mind. I believe committees can give valuable input to the Board regarding relevant CSR issues and could lead to better CSR performance within a company. This has been recognized as early as in 1987, when Cowen et al. found a positive relation between the existence of a CSR committee and the number of social disclosures.

Business practice reveals the importance of the corporate governance structure. The Dutch corporate governance code for example, pays specific attention to the importance of transparency of decision-making and proper supervision within a company. Governance codes over the world all include principles that relate to the roles and responsibilities of the Management Board, Supervisory Board and shareholders (see e.g. the Combined Code, 2006; German Corporate Governance Code; 2007, the Dutch Corporate Governance Code, 2003) As in the financial reports, we would expect that the readers would like to gain information on the structure of the corporations. Questions as who is ultimately responsible for CSR within the company, or the role of the supervisory Board, are relevant to the public. Both the GRI and the Transparency Benchmark include criteria on governance structure.

4.3.2. Strategy and engagement

A logical step after the decision of a particular corporation to pay specific attention to CSR is the integration of this desire into the strategy of the corporation. After deciding what the mission of the organisation is, this should be translated into a strategy that clearly sets the direction for the organisation. A report of PricewaterhouseCoopers (2006) shows that this is easier said than done. The old financial models used to set strategy in the past, might not be useful in a time when CSR is becoming increasingly important. According to PricewaterhouseCoopers, the models will have to include new scenarios and risk factors. Katsoulakos and Katsoulacos (2007) have tried to do just this. They developed a strategic management framework that incorporates corporate responsibility principles into business strategy. Stakeholders and readers of CSR reports want to be informed about the CSR mission and CSR strategy. Financial reports as well as CSR reports at many times start with laying out the mission and strategy of the corporation; it is an important starting point to assess the performance of the company. However, having a great mission translated into a strategy is not enough. What is the point of behaving in a socially responsible way when all your suppliers are not behaving the same way? Therefore, society expects to have insight into the policy regarding the responsibilities of the suppliers. This is demonstrated by the large amount of companies publishing a separate chapter on their suppliers.

According to the Dutch business practice (incorporated in the Corporate Governance Code 2003) the Board and supervisory Board have the overall responsibility for weighting up the interests of all the stakeholders. Taking this in consideration, the role of stakeholders is very important for good corporate governance. Consequently, stakeholders should be included in the whole CSR process, to ensure that companies pay attention to issues that are relevant to stakeholders. Research conducted by AccountAbility (2007) also illustrates that stakeholder engagement is very important in the CSR context. They claim that companies and stakeholders reported that ‘participating in the panel had accelerated their own learning, and strengthened their ability to understand and influence the links between business strategy and societal issues’ (p 2). This research illustrates the importance of processes to ensure that important CSR issues are discussed at the top before they become problems. One thing that was found to very important was the composition, ground, rules and mandate of the panel. Therefore the developed criteria points pay much attention to the process behind stakeholder engagements.

4.3.3. Processes and controls

After the strategy is set is important to have processes and controls in place to ensure that the CSR goals are reached. The agency theory states that managers and executives cannot be expected to act in the best interest of shareholders at all times and therefore predicts that the compensation plan will be