Environmental and Social Reporting After Incidents
The effect of the Deepwater Horizon oil spill /
Andrew Maarschalkerweerd
01-03-2012
Student nr: 315863am
Supervisor: Dr. K.E.H. Maas
Erasmus University Rotterdam
Erasmus School of Economics
Master Thesis Accounting, Auditing and Control

Acknowledgements

This Master thesis is the conclusion of the Master degree Accounting, Auditing and Control at Erasmus University Rotterdam. By means of thisacknowledgements page I would like to thank those who assisted me either personally or academically.

I would like to thank both supervisors that assisted me in the writing process: From the Erasmus School of Economics, Dr. Karen Maas; From PwC, Maarten Hoornweg RA. Both supervisors gave me good feedback and ideas that I could use in my master thesis.

I also would like to thank PwC Rotterdam for offering me a place to write my thesis and a chance to get an insight into what they do.

I also would like to thankDr. C.D. Knoops, who supervised me in the first part of the writing process during the Seminar Advanced Financial Accounting. I also would like to thank my fellow students for giving feedback on my paper during this seminar.

I would like to thank my family and friends for supporting me during my studies.

A special thanks to Professor Colbert, one of my lecturers during my exchange at Trinity College Dublin, who inspired me to continue my studies in the field of accountancy.

Rotterdam, 01-03-2012

Andrew Maarschalkerweerd

Executive Summary

In the last couple of decades there hasbeen a large increase in the voluntary disclosure of environmental information by companies. In this thesis I specifically look at these disclosures surrounding incidents and how companies change their disclosures.

The main question that is posed in this thesis is “What are the reasons for voluntary disclosure and how do companies respond to certain incidents?”. This main question is then split up into sub-questions and these are discussed in several chapters.

Firstly the theoretical fundament of this thesis is set out. Several frameworks used for voluntary disclosure are discussed. These help companies to disclose information in a structured manner. This helps several stakeholders by making the information more comparable. Also theories are discussed that can explain voluntary disclosures by companies, of which legitimacy theory and media agenda setting theory are the most important for the study conducted later.

Legitimacy theory predicts that companies, when they are threatened to breach their ‘social contract’ with society, will disclose more information. Patten (1992) takes this theory even further and tests whether companies in the same industry as a threatened company, will increase their disclosures. This isspecifically done for companies in the oil sector after the Exxon Valdez oil spill relating to their environmental disclosures. He finds that companies, although unrelated in a business sense with Exxon, increase their disclosures.

Also companies that are larger (measured in log revenues) show a larger increase in their scores than smaller companies. Companies that arepartly responsible for the slow clean-up response also showed a larger increase in their scores.

Media agenda setting theory predicts that media convey the concerns of the public but also influence the public opinion on certain topics, especially if they are far away from the public. Therefore managers are also influenced by topics that are discussed in the media.

I replicate Patten’s study by examining the disclosures of companies in the oil sector after the BP oil spill in the Gulf of Mexico, to see whether the relation still holds. I quantify the disclosures by means of an environmental disclosure index model as developed in Aerts and Cormier (2009) for the years 2009 and 2010 (before and after the incident). Firstly, I test the hypothesis that the disclosures significantly increase and secondly, whether there is a difference when companies are larger and whether there is a difference for US companies opposed to non-US companies.

I find that companies still disclose significantly more after an incident,even if they are not responsible for the incident (the only relation being that they operate in the same environmental sensitive industry). When this information is split-up into economic (reactive, related to litigation) disclosures and social (soft claims) I find that both are significant.

I cannot find any evidence that larger companies also show a larger increase in the scores of both years, as Patten found in his study. This might be explained by how companies deal with environmental disclosures today compared with the end of the 1980’s or beginning of the 1990’s.

I cannot find any evidence that US companies increase their disclosure to a larger extent than non-US companies. A reason for this could be that all companies are large listed companies in the US and that their historic location has become irrelevant.

Table of Contents

Acknowledgements

Executive Summary

Table of Contents

1. Introduction

1.1 Topic introduction

1.2 Topic and research question

1.3 Structure

2. Theories and methods behind this research

2.1 Introduction

2.2 Social and environmental disclosures

2.3 The Global Reporting Initiative

2.3.1 Reporting principles for defining content

2.3.2 Reporting principles for defining quality

2.3.3 Reporting guidance for boundary setting

2.3.4 Standard disclosure

2.3.5 Oil & gas sector supplement

2.3.6 The future

2.4 Other standards

2.4.1 AccountAbility framework

2.4.2 ISO 14001

2.4.3 ISO 26000

2.5 Theories

2.5.1 Political economy theories

2.5.2 Media agenda setting theory

2.5.3 Voluntary disclosure theory

2.6 Models

2.6.1 Measuring the quality of voluntary disclosures

2.6.2 Limitations

2.7 Research in relation to the theories

2.8 Conclusion

3. Literature review

3.1 Introduction

3.2 General literature

3.2.1 Capital market reactions

3.2.2 Characteristics of firms and countries

3.2.3 Mandatory reporting

3.2.4 Different disclosures

3.3 Specific incidents

3.4 Environmental disclosure index models

3.4.1 Financial disclosures

3.4.2 Environmental disclosures

3.4.3 Additional factors: litigation and monetary and non-monetary information

3.4.4 Model for this study

3.5 Conclusion

4. The BP oil spill

4.1 Introduction

4.2 An overview

4.3 Responsibility

4.4 Conclusion

5. Hypotheses development and research design

5.1 Introduction

5.2 Hypotheses

5.3 Research and sample

5.4 Methodology

5.5 Conclusion

6. Results and analysis

6.1 Descriptive results

6.2 Statistical results

7. Conclusion

7.1 Summary and findings

7.2 Limitations and suggestions for future research

7.3 Generalisability

8. Bibliography

Appendix 1

Appendix 2

Appendix 3

Appendix 4

Appendix 5

1. Introduction

1.1 Topic introduction

Companies have been presentfor a very long time. A main facet is reporting on the company performance. Companies engage in this reporting to inform parties that have provided them the resourcesor chance to come to a certain performance. The focus of this reporting process used to be relatively limited: Most of the attention of the reporting process was on showing how well the company performed financially, while other (non-financial) factors were often omitted, although these factors are often very important to those affected.

In the last decades a change has occurred in this reporting process. A change that has affected how companies and other entities report their performance to the people affected (also known as stakeholders). More and more, the focus of the reporting process has shifted away from only financial reporting to a broader form of reporting. This form of reporting would generally include how the company interacts with society as whole and other stakeholders instead of only internal stakeholders such as employees, lenders and shareholders. Issues such as the environment and the usage of labour in developing countries are topics that increasingly can be found in disclosures of companies.

The main article of reference within this thesis will be Patten (1992). Patten discusses the change in disclosures after an incident in the oil sector. It was found that companies increased their disclosures significantly after the Exxon Valdez oil spill. This did not only hold for Exxon, but for companies throughout the industry.

Approximately twenty years later we will try to assess whether this relation still holds in a world where additional disclosure is becoming common. This will be done by replicating Patten’s study but with an updated model and a new incident: The BP oil spill in the Gulf of Mexico in 2010.

1.2 Topic and research question

This thesis will lookinto the disclosuresby companies and their place with respect to a broader set of stakeholders. Several theories that can explain why companies voluntarily disclose information are discussed. Companies are more and more making additional disclosures, although scholars still dispute why a company would do so.

Next to that, I will investigate the current literature on how companies deal with their disclosures around incidents that might affect them. This thesis will be structured into several chapters and in these chapters, each (sub)-question will be answered.

The main questionis: What are the reasons for voluntary disclosure and how do companies respondto certain incidents?

The sub questions to be answered in theseparate sections are:

-What are the theories behind voluntarydisclosure?

-What does the literature say on voluntary disclosure and more specific in response to incidents?

-What went wrong at the time of the BP oil spill?

-How is this research on environmental disclosure at the time of the BP oil spill conducted?

-Are there any changes in the environmental disclosures of oil companies since the BP oil spill?

This thesisgives a clear insight to anyone who wants to investigate voluntary disclosure practice. Next to that, the field of accountancy can see it as an addition to the increasing literature on voluntary disclosure around incidents. In general voluntary disclosure has increased; therefore it might be relevant to see whether incidents still influence a company’s reporting behaviour.

1.3 Structure

The structure of this thesis is as follows: in the secondchapter, theories and methods of assessing the voluntary disclosure process will be discussed.

First a general assessment will be made of voluntary social and environmental disclosure. When did it start? This is further explained by means of a four step framework.

Also the Global Reporting Initiative will be discussed. The GRI is one of the first real frameworks that givesguidance to entities on how to report on their non-financial performance. In this section the guidelines of the framework will be discussed. Next to the GRI, other frameworks are also reviewed

The first theories that will be discussed are: legitimacy theory, stakeholder theory andinstitutional theory: Every theory explains a different way in which companies respond to the stakeholders around them.These three theories are embedded into the so called ‘Political economy theory’ (Gray, Owen and Adams, 1996) that states that politics, economics and society cannot be seen separate from each other but have to be seen as overlapping concepts that will influence each other when one of the three factors is changed. Therefore they should not be seen as competing but as complementary explanations.

Next to these theories (that all relate to the Political economy theory), the media agenda setting theory will also be discussedbriefly. This theory credits the media for bringing up many issues that are subsequently taken over by companies as important issues.

In the last part the attention shifts to how one can examine environmental disclosures. This subject is still at this moment open to debate since most of the social and environmental disclosures are all qualitative instead of quantitative. This brings an extra difficulty to the research of the subject, as some text cannot be compared with others.

Over the last thirty to forty years, academics came up with measures to solve this problem and by that, have madevoluntary disclosures more comparable to each other: Some methods just count the number of pages or sentences in which a company discloses their policies, while others are more sophisticated.

In the thirdchapter of this thesis, the current literature in the field of voluntary disclosures is discussed. Why do companies disclose voluntarily? Do they benefit from this and are there contradicting explanations for this behaviour?

The first part provides a general overview of the literature that relates to voluntary disclosures, environmental reporting and CSR (Corporate Social Responsibility) reporting. This is done by means of several sub-chapters.

Next to the general literature in the field of voluntary disclosure, an overview is presented of a more specific field. The literature on voluntary disclosure around the occurrence of incidents is looked at. One of the first scholars to study this field was Patten (1992). He examined the effect that the Exxon Valdez oil disaster had on the voluntary disclosure that the oil sector provided. He did not take Exxon into account but still he found a significant increase in the disclosure of companies in the oil sector after the incident happened. An explanation is sought for why this happens and why companies have to do this. Even companies that are unrelated in a business sense to Exxon, were found to increase their disclosure. This was a major finding as a narrow economic view predicts that every company is responsible for its own mistakes and that others will not have to alter their behaviour.

It is found that many previous studies used models that were derived from one of the first environmental disclosure index models from Wiseman (1982).

In the fourthchapter an overview is presented on the Deepwater Horizon oil spill in the Gulf of Mexico. Next to this, a separate paragraph looks at who could be held responsible for the oil spill.

In the fifth section of this thesisthe research that will be conducted is introduced.

The hypotheses are presented on the change in voluntary disclosure after incidents. The research design and methods will be presented.

In the sixth chapter the results and analysis are presented. Firstly, the descriptive statistics from the sample are given and discussed. After this, the results and outcomes of this study are presented and discussed.

In the seventh chapter this thesis is concluded. This is done by giving an overview of the chapters, the outcomes of the study and listing the limitations of the study. Some suggestions for future research will be given and the contribution of this study will be discussed.

2. Theories and methods behind this research

2.1 Introduction

In this section of thisthesis, the underlying theories of social and environmental described. Over the last decades some theories have been defined, and they have in common that they try to explain why companies provide voluntary disclosures on these topics.

In the secondparagraph, social and environmental reporting is reviewed.The history of the process will be discussed and then this is examined by the why-who-for what-how framework set out by Deegan and Unerman (2006, pp.309-365).

In the thirdparagraph the Global Reporting Initiative will be discussed. This is a way of reporting for companies: Not only on financial aspects, but also on environmental and social aspects.

In the fourth paragraph a brief overview is given of other frameworks used in environmental reporting.

In the fifth section, the theories that relate to voluntary disclosures are discussed. The theories described are: legitimacy theory, stakeholder theory, institutional theory and media agenda setting theory.

The methods that are used when studying disclosures will also be discussed. Several options exist, varying from subjective to semi-objective.

2.2 Social and environmental disclosures

Although social and environmental reporting, next to financial reporting, is common for public listed companies nowadays (KPMG, 2005; KPMG, 2008; KPMG, 2011;See Table 1), it only started to become common in the last decade of the 20th century. In the last two decades (since approximately the 1990’s) more and more companies have focussed on explaining to their stakeholders what they do with respect to the non-financial outcomes of their business (ICAEW, 2004).

Year / 1993 / 1996 / 1999 / 2002 / 2005 / 2008 / 2011
N100: Percent of Companies with CR report / 12% / 18% / 24% / 28% / 41% / 53% / 64%
G250: Percent of companies with CR report / - / - / 35% / 45% / 64% / 83% / 95%

Table 1, The rise of corporate responsibility reporting

Source: (KPMG, 2005; KPMG, 2008; KPMG, 2011)

In the research of Patten (1992) one can see that many companies used to provideadditional information on their non-financial performance in the annual reports but many companies now also publish separate CSR reports next to the additional disclosures in their annual reports.

Although this sudden increase in additional voluntary disclosure has really taken offin the last twenty years approximately, some companies already provided information on their environmental performance. In studies it was found that for example US Steel already had social reporting practices from the year 1905 (Hogner, 1982). Also some others companies reported on these matters.

In the ‘why’ section in the framework of explaining environmental reportingby Deegan and Unerman (2006, pp.315- 335),they consider why companies voluntary report environmental and social information. Does the company have a responsibility? Are they accountable for their actions?

This can be viewed from two sides: a narrow view and a broad view.

The narrow view would relate to the traditional capitalistic economic view: The only responsibility the company has, is theresponsibility to make a profit and outperform their competitors. This view is in line with the thoughts of the free-market economists. This would increase the general wealth of each and the profits could for example be used by individuals to spend on the environment. One can reject this view by saying that the main focus would be on the owners of the company instead of the society and stakeholders as a whole.

The second view is the broad view. This view does not focus on the financial side only but also considers social and environmental issues. If these are systematically ignored, it might result in the company being excluded in some parts of society. This might increase the costs of capital for the companies and subsequently result in lower return on investment. Companies that endorse the broad view report and communicate on the three factors (environmental, social and financial disclosures) by for example employing techniques such as ‘triple bottom line’ and the ‘global reporting initiative’. These will be examined later.