The influence of Corporate Social Responsibility reporting ...



Erasmus University RotterdamErasmus School of EconomicsThe influence of Corporate Social Responsibility reporting on reputationThesis: Accounting, Auditing and ControlAuthorStephanie TermeulenStudent number304834SupervisorK.E.H. MaasDate July 2011PrefaceDeze scriptie heb ik geschreven ter afronding van mijn Master opleiding, Accountancy, Auditing and Control aan de Erasmus Universiteit. Graag wil ik mijn begeleidster, Karen Maas bedanken voor haar kritische feedback op mijn scriptie. De afgelopen jaren heb ik intens genoten van mijn studententijd aan de Erasmus Universiteit. Vooral mijn studietijd aan de Haskayne School of Business in Calagry was onvergetelijk. Deze mooie tijd had ik niet kunnen beleven zonder steun van mijn ouders, zussen, vriend en vrienden. Hen wil ik hiervoor graag bedanken.Stephanie TermeulenOude Wetering, 15 juli 2011AbstractIn this paper the impact of Corporate Social Responsibility (CSR) reporting on reputation is investigated. Disclosed CSR information is observed by stakeholders. By publishing a CSR report a company provides its stakeholders with more information about the company. When a CSR report is of higher quality, stakeholders are provided with more transparent information. The agency theory, information asymmetry, legitimacy theory and stakeholder theory suggest that due to a CSR report an impact on the reputation is expectable. In this paper it is to be expected that higher CSR reporting quality increases the reputation of a company. CSR reporting quality is measured with the application level that is provided by the Global Reporting Initiative (GRI) Report List. The reputation index from Fortune 500 is used as proxy for the reputation. The CSR reporting of current year affects the reputation of the next year. The timeframe of this paper contains CSR reporting from 2005 till 2009. From this selected time period 85, 106, 138, 167 and 175 respectively data points are observed.Unlike prior studies, this paper concludes that CSR reporting has no influence on reputation. The outcome of this paper indicates that other variables than CSR reporting has an influence on the reputation of a company. Table of Contents TOC \o "1-3" \h \z \u Chapter 1 Introduction PAGEREF _Toc299353356 \h 111.1Research question and research method PAGEREF _Toc299353357 \h 121.2Relevance PAGEREF _Toc299353358 \h 141.3Structure PAGEREF _Toc299353359 \h 15Chapter 2 Theoretical Background PAGEREF _Toc299353360 \h 162.1 Disclosures PAGEREF _Toc299353361 \h 162.1.1 Financial disclosure PAGEREF _Toc299353362 \h 162.1.2 Voluntary disclosure PAGEREF _Toc299353363 \h 172.1.4 Summary disclosures PAGEREF _Toc299353364 \h 172.2 Corporate Social Responsibility PAGEREF _Toc299353365 \h 182.2.1 Definition of Corporate Social Responsibility PAGEREF _Toc299353366 \h 182.2.2 Developments in Corporate Social Responsibility PAGEREF _Toc299353367 \h 202.2.3 Corporate Social Responsibility reporting PAGEREF _Toc299353368 \h 212.2.4 Corporate Social Responsibility reporting guidelines PAGEREF _Toc299353369 \h 212.2.5 Summary Corporate Social Responsibility PAGEREF _Toc299353370 \h 232.3 Positive Accounting Theory PAGEREF _Toc299353371 \h 232.3.1 Agency theory PAGEREF _Toc299353372 \h 242.3.2 Information asymmetry PAGEREF _Toc299353373 \h 262.3.3 Legitimacy theory PAGEREF _Toc299353374 \h 282.3.4 Stakeholder theory PAGEREF _Toc299353375 \h 302.3.5 Summary Voluntary Disclosures and Positive Accounting Theory PAGEREF _Toc299353376 \h 31Chapter 3 Literature Review PAGEREF _Toc299353377 \h 323.1 Incentives of Corporate Social Responsibility strategy and reporting PAGEREF _Toc299353378 \h 323.1.1 Corporate Social Responsibility strategy and reporting incentives PAGEREF _Toc299353379 \h 323.1.2 Implementing Corporate Social Responsibility strategy and report PAGEREF _Toc299353380 \h 373.1.5 Summary incentives Corporate Social Responsibility strategy and report PAGEREF _Toc299353381 \h 393.2 Impacts of Corporate Social Responsibility Report PAGEREF _Toc299353382 \h 393.2.1Forecast accuracy PAGEREF _Toc299353383 \h 393.2.2 Public sentiment PAGEREF _Toc299353384 \h 423.2.3 Financial impacts PAGEREF _Toc299353385 \h 443.2.4 Summary impacts PAGEREF _Toc299353386 \h 48Chapter 4 Hypotheses development PAGEREF _Toc299353387 \h 504.1 Development of the main hypothesis PAGEREF _Toc299353388 \h 504.2 Difference compared to previous studies PAGEREF _Toc299353389 \h 524.3 Development of sub hypotheses PAGEREF _Toc299353390 \h 544.4 Summary hypothesis development PAGEREF _Toc299353391 \h 57Chapter 5 Methodology PAGEREF _Toc299353392 \h 585.1 Sample PAGEREF _Toc299353393 \h 585.1.1 Used databases PAGEREF _Toc299353394 \h 585.1.2 Development sample and time period PAGEREF _Toc299353395 \h 595.2 Methodology of statistical research PAGEREF _Toc299353396 \h 605.3 Data PAGEREF _Toc299353397 \h 61Chapter 6 Analytical Results PAGEREF _Toc299353398 \h 646.1 Descriptive analysis PAGEREF _Toc299353399 \h 646.2 Statistical Results PAGEREF _Toc299353400 \h 676.2.1 Correlation PAGEREF _Toc299353401 \h 676.2.2 Regression PAGEREF _Toc299353402 \h 72Chapter 7 Conclusion PAGEREF _Toc299353403 \h 807.1 Discussion PAGEREF _Toc299353404 \h 807.2 Implications PAGEREF _Toc299353405 \h 827.3 Limitations and further research PAGEREF _Toc299353406 \h 83References PAGEREF _Toc299353407 \h 84Appendix A: Companies included in sample PAGEREF _Toc299353408 \h 88Appendix B: Distribution CSR report levels over all variables PAGEREF _Toc299353409 \h 98Chapter 1 IntroductionThe last decades environmental and social related impact of companies is under rising attention of the public. The ‘Deepwater Horizon Oil Split’ of April 2010 resulted in negative public attention of the oil company, British Petroleum (BP). During three months oil was continuously spilling into the Gulf of Mexico. The enormous environmental damage due to BP’s negligence was widely described in the worldwide media. Besides companies’ environmental issues, social consequences and social responsibility also face attention of the public. The public attention to social issues is reflected in the scandal of the British newspaper ‘News of the World’. The newspaper was suspected of illegally monitoring private phone calls. This scandal was widely described in the media. The associated negative attention about ‘News of the World’ made the owner of the newspaper decide to immediately abolish the newspaper. The rise of attention for environmental and social related impact is reflected in increasing attention for Corporate Social Responsibility (CSR). Besides the society, companies also became more aware of their environmental and social side. The increased awareness of environmental and social impact results in the implementation of CSR activities within a company. CSR activities are implemented to contribute to the economic performances of the company and to benefit society. In a CSR report the CSR performances are summarized, when the CSR report is disclosed stakeholders are provided with more transparency. Since the early nineties a lot of research has been done on the topic of CSR. The impact of CSR is mainly examined on the financial performances of the company. The study of Margolis and Walsh (2003) analyzed the outcomes of 127 prior studies and found a positive relation between CSR performances and financial performances. Wood (2010) argued that the relation between CSR performances and financial performances is already well established and that CSR research should be more focused on society and stakeholders. Where the studies of Margolis et al. (2003) and Wood (2010) investigated the CSR performances, this paper investigates the CSR reporting. In this paper the impact of CSR reporting is examined, the quality of CSR performance is not taken into consideration, but the quality of the CSR report is.The quality of a CSR report benefits the company’s stakeholders. When a CSR report is disclosed, stakeholders acquire CSR information of the company. If the CSR report is of high quality, stakeholders are provided with reliable CSR information. This reliable information increases the transparency. The stakeholders’ impression of the company changes due to the provided information. In this paper the impact of CSR reporting quality on the reputation is examined. This reputation is the impression of a company of all stakeholders together. The impression of stakeholders of a company is influenced by CSR information. Existing studies used the term public sentiment, to investigate stakeholders’ impression of a company. In this paper reputation is used, to investigate stakeholders’ impression. Reputation is a more direct term to examine stakeholders’ impression about a company. In this paper no distinction is being made between public sentiment of existing studies and reputation of this paper.Research question and research methodThe research question of this paper is:What is the impact of Corporate Social Responsibility reporting on reputation?To answer the research question, different CSR related topics are investigated in advance. First the definition of CSR should be defined. Also the characteristics of CSR reporting should be investigated. The positive accounting theory provides theories which predicts and explains CSR. Description of the agency theory, information asymmetry, legitimacy theory and stakeholder theory are provided to investigate incentives of CSR and CSR reporting. To implement CSR into the company, the company should go through different stages, these different stages are described. Finally, the investigated influence of CSR reporting on reputation from prior research is described.The research question is answered by empirical research. The impact of CSR reporting is measured with CSR reporting quality. The CSR report quality is observed from the CSR report level provided by the Global Reporting Initiative (GRI) Report List. This CSR report level indicates the degree of conformity with GRI reporting guidelines of the CSR report. The application level is determined for CSR reports which are voluntary submitted in the GRI report list. The used proxy for reputation is the reputation index. This index is based on nine components and is yearly compiled. The CSR report is disclosed after the reported year and influences the reputation of the year after the reported year. In the reputation index database, reputation indices from 2006 are available, therefore this paper uses the CSR reporting time period 2005 up till and including 2009. The research question is answered by the main hypothesis and sub hypotheses. These hypotheses are tested with a regression and a correlation. The hypotheses are:Main hypothesis: CSR reporting quality increases the reputation of a company.Sub hypotheses: The relationship between CSR reporting quality and the reputation is positively influenced by larger companies.The relationship between CSR reporting quality and the reputation is positively influenced by better financial performances.The relationship between CSR reporting quality and the reputation is positively influenced by higher general selling and administrative expenses to sales.The relationship between CSR reporting quality and the reputation is influenced by regions of companies.The relationship between CSR reporting quality and the reputation is influenced by industries of panies for which all relevant data were available are included in the sample. For the year 2005 up till and including 2009; 85, 106, 138, 167 and 175 respectively data points were available. The years are individually tested, otherwise the results will be biased. The influence of CSR reporting quality on reputation is tested by the correlation and the regression of the reputation index and the CSR report level.RelevanceThe investigated relation between CSR reporting quality and reputation provides benefits for the scientific world, companies and society. Below the relevance for scientific world, companies and society is explained. In the study of Clarkson, Hua Fang, Li and Richardson (2010) a positive relation is found between CSR reporting and public sentiment. But in this study, public sentiment is measured with the Janis Fadner coefficient. This coefficient is based on the amount of negative and positive environmentally and socially related articles. The Janis Fadner coefficient is not the most reliable measurement to investigate the public sentiment. This coefficient suggests that companies with equal amounts of negative and positive articles have the same public sentiment. The study of Clarkson et al. (2010) takes the influence of CSR reporting into consideration but the studies of Brown and Dacin (1997) and Ittner et al. (1998) take the influence of CSR on public sentiment into consideration. These studies investigate the relation between CSR and public sentiment with a field study. The field study evaluates products based on the customer satisfaction. Product evaluations by (potential) customers’ satisfaction levels are no exact measurements to investigate the public sentiment. It is hard to accurately evaluate services of companies. Some companies producing more than one product, when one product is evaluated not a right representation is provided of the company.The scientific world also benefits from the investigated relation between CSR quality reporting and reputation because of the lack of CSR research which is focused on stakeholder and society. From this paper, society benefits afford, there is investigated if CSR reporting influence the reputation. An investigation is performed, where the reputation is dependent on. When no relationship is found, society does not observe the CSR reporting quality. This paper is also of relevance for companies because it researches if CSR reporting quality leads to an improvement of a company’s reputation. If evidence is found for a positive relation, companies benefit from providing high quality CSR reports. If no evidence is found, companies can change their way of CSR reporting to achieve better results with their CSR disclosure. Investigating the relation between CSR reporting quality and reputation is therefore beneficial for society. StructureThis paper is outlined as follows, in the second chapter the theoretical background of CSR reporting is discussed. Disclosures in general, the definition of CSR and the relation of CSR with the positive accounting theory are described. In the third chapter the literature review is provided. Also the incentives and impacts of CSR and CSR reporting are presented. The hypotheses are introduced in chapter four. The methodology, which is used to test the hypotheses, is described in chapter five. Methodology, statistical research and data are part of chapter five. In chapter six the analytical results are presented, these results are composed of the descriptive and statistical results. Finally in chapter seven, the conclusion is stated. Chapter 2 Theoretical BackgroundTo provide a complete understanding of CSR reporting, in this chapter a theoretical background is given about communication information. This chapter starts with describing, different types of communicating information, financial disclosures and voluntary disclosures are introduced. Thereafter CSR and CSR reporting are discussed. The definition of CSR is explained, the CSR report and the guidelines for providing a CSR report are discussed. The final part of this chapter explains the positive accounting theory and the relation with CSR reporting.2.1 DisclosuresA disclosure communicates company’s information, the disclosure is provided by the company. Rules and regulation exist to provide all obligated information. Companies should make required information publically available. The financial disclosure is a well known mandatory disclosure. Also voluntary disclosures exist, in voluntary disclosures information is provided which is, next to mandatory disclosure, not obligated to communicate. First, a short description is given of the financial disclosures, followed with a description of voluntary disclosures. This description gives a background for understanding the CSR report disclosures.2.1.1 Financial disclosureCompanies communicate information in different ways, these announcements are called disclosures. The most common known disclosure is the financial annual report. In these disclosures, information is provided about past financial performances and the current financial position of the company. In the financial report the profit and loss account, balance sheet and cash flow statement are included. Also other important information for users should be incorporated in a financial report. Financial disclosures provide useful information to investors and other stakeholders, based on the disclosed information investors and stakeholders make decisions. Financial reports are valuable for investors, this associated in the studies of Kothari (2001), Louhichi (2008) and Su (2003). In these studies evidence is found that financial reports affect stock price reactions. The reaction of investors can be observed due to the changes in stock prices. Specific requirements exist for the financial reporting of listed companies. All listed companies are required to make the financial report publically available. These listed companies have a big influence on the society, so the disclosures should be free from errors and fraud. Therefore the financial reports should be audited by a Certified Public Account (CPA) firm. Providing a financial report is the responsibility of company’s management. Management is primarily responsible for the information provided in the report. The financial report should be in conformity with standards and regulations. The General Accepted Accounting Principles (GAAP) and the International Financial Reporting Standards (IFRS) are common used standards. For GAAP different country versions exist. The IFRS services a broader public, this reporting standards is applied in an international setting. 2.1.2 Voluntary disclosureManagement is able to choose, to disclose more information than is required. The non mandatory disclosed information is called voluntary disclosures. Mandatory disclosed information is imposed by laws and regulations. But no laws and regulations exist for providing voluntary disclosures. When on a voluntary base information becomes available, investors are provided with additional information, this will lead to improved decision making by investors. Also companies can distinguish themselves from other companies, when providing voluntary information. Different forms of voluntary disclosures exist, examples are: conference calls, press releases, interviews and sustainability or CSR reports.2.1.4 Summary disclosuresCompanies communicate information to stakeholders by disclosures. In a financial disclosure required information is provided. For this required information rules and regulations exist. Voluntary disclosure is a disclosure, that communicates non mandatory information and no rules and regulation exist for these provisions2.2 Corporate Social Responsibility The subject of this thesis is to investigate the impact of CSR reporting. Before investigating this impact, first the definition of CSR and CSR reporting should be clear. In a CSR report the performance of the CSR strategy is communicated. First the definition of CSR is provided, this is followed with an introduction of CSR reporting. Besides that CSR reporting is a voluntary disclosure, guidelines exist for providing a report. The guidelines of CSR reporting are provided in the third part of this paragraph. 2.2.1 Definition of Corporate Social ResponsibilityCompanies which implement CSR strategy into the organization, put attention to environmental and social activities. When a company implements CSR strategy into the organization, a switch is made in the focus of the company. Not only the financial performances are important, but also the environmental and social performances face attention. These companies are concerned about their negative influences.Many CSR definitions exist, one of the definitions is coming from the World Business Council for Sustainable Development (WBCSD). This is a platform developed by 200 companies, out of 30 countries. The WBCSD deals with business and sustainable development, they help companies exploring a new CSR strategy in the company. Also sharing knowledge and experiences are important purposes of the WBCSD. Their definition of CSR is:“The continuing commitment by business to behave ethically and contribute to economic development while improving the quality of life of the workforce and their families as well as of the local community and society at large.”A stricter explanation of CSR is provided by Naylor (1999) and used by Douglas, Doris and Johnson (2004). This definition is more in line with WBCSD, where CSR is defined as a continuously commitment. The stricter definition of Naylor (1999) is:“The obligation of managers to choose and act in ways that benefit both the interest of the organization and those of society as a whole.”CSR is makes use of activities, different activities exist for different companies. For example, manufacturing companies that use for production natural materials can implement CSR activities. One of the goals of their CSR strategy can be, lowering the amount of used natural materials by setting maximum levels of use. Minimizing the pollution rate is a well know CSR activity too. The social side activities of CSR are, for example, related to providing sponsorships to foundations. The Dutch bank Rabobank provides much sponsorships to sport and cultural organizations. Companies have also the possibility to implement CSR activities which focus more on the intern side of the company. Examples of intern CSR activities are, stimulating of employees to go by bike to office and serving healthy food in the canteen. The health of employees is an important factor for companies, healthy employees are better performers (Biotech Week, 2004). Above some examples are given of CSR activities, these CSR activities leads to the CSR performances or outcomes of the CSR strategy. A CSR strategy includes many CSR activities, the CSR activities all together have the purpose of reaching a final goal. The outcomes or performances of CSR activities are corporate social performance (CSP). Wood (2010) provides a framework which made the distinction clear between CSP and CSR. Following the study of Wood (2010) there is claimed that CSP is more focused on the outcomes and impacts of performances. The CSP is defined by the principles of CSR, the CSR strategy sets out the expected performances. Wood (2010) implies that, CSP result from CSR and that CSP proves from social responsiveness. She identified the social responsiveness as a process where environmental scanning, stakeholder management and public affairs management are incorporated. This all implies that, CSP is the outcome of the CSR activities.In the study of Wood (2010) there is made use of CSR, when measuring the corporate social performances. Before, already was mentioned that Wood (2010) sees CSP as the outcome of CSR. In this framework CSR is defined on three principles: legitimacy, public responsibility and managerial discretion. In this study a definition about the legitimacy theory is given: “Businesses that abuse the power society grants them will lose that power.” A company obtained power, achieved from society. The company uses this power in their daily production and decision making. If a company harms the environment of society due to the production or decision making, they misuse their power. The society will judged the company in a negative way, the company will obtain less power finally. The definition of the public responsibility is:”Businesses are responsible for outcomes related to their primary and secondary areas of involvement with society.” This definition implies that, the company is full responsible for the environmental and social performances. Not only the positive performances are under company’s responsibility but also the negative performances are under company’s responsibility. The managerial discretion is defined by Wood (2010) as: “Managers and other employees are moral actors and have a duty to exercise discretion toward socially responsible, ethical outcomes.” This implies managers and other employees should act in the way the CSR strategy prescribes. This extended framework with the three included principles, designed by Wood (2010), suggests that CSR is driven by managers and employees, towards the company’s performances which benefit society. When a company abuses power from society, the company will be harmed in losing power. Companies are free to choose whether or not to implement a CSR strategy into the organization. No mandatory requirements exist for implementing a CSR strategy. CSR is a strategy that provides benefits not only for the company, but for the society too. When implementing CSR, the company acts in their own preferences (for example, high financial performances) and also takes the preference of society into account (for example, reducing the company’s pollution). 2.2.2 Developments in Corporate Social Responsibility In the previous decades the use of CSR was rising. This is mainly due to the rise in awareness about environmental and social issues (Quaak, Aalbers and Goedee , 2006 and Kakabadse, Kakabadse and Lee-Davies, 2006). The focus of companies on environmental and social issues has changed over time. In the first decades of the twentieth century a little start of CSR was already been made. Companies became more aware of their social side, attention to housing of employees and health security was growing. Later on, the government takes the responsibility for this social guidance. The consequences of the economic growth became visible in the seventies of the last century, big economic growth was harming the environment too much. There was coming more pressure on the social and environment involvement of the company. The last decades continues globalization leads to a growing change in the social and environmental problems, from local perspective to global perspective. Companies now, have to solve problems on three different dimensions; economic, social and environmental. Nowadays, companies can solve these problems by the triple P-strategy. (Quaak et al., 2006)2.2.3 Corporate Social Responsibility reportingPerformances related to CSR can be communicated by several CSR disclosures. One of these CSR disclosures is the CSR report. In a CSR report the CSR performances of underlying period are summarized. The CSP are communicated in this CSR report. In the study of Dawkins and Stewart (2003) is prescribed that companies should disclose CSR information. In this study the demand of CSR reporting was investigated. The demand to provide CSR reporting was represented by 86 percent of the British public. But what contains the concept of CSR reporting? Different descriptions exist for this concept, different definitions of CSR reporting are discussed.In the study of Sutantoputra (2009) companies are assessed on the level of CSR reporting, when providing a rating of the social performances of the selected companies. The assessment of the CSR reports is based on the GRI framework. In this study there is made use of the following definition of CSR reporting:“The process of providing information designed to discharge social accountability”In another study of Douglas et al. (2004) the CSR reports of financial institutions from Ireland are analyzed. In this study is argued that CSR reporting is a way to provide stakeholders information about social performances of the company. In this study of Doris et al. (2004) corporate social reporting is seen as a tool for developing a positive image among the stakeholders. Sutantoputra (2009) notice that CSR performances are also communicated in annual reports instead of a standalone CSR report. But a rising trend is observed in providing CSR information, a growing rate of CSR disclosures is disclosed in a standalone CSR report. The CSR reporting behavior is different among countries. Douglas et al. (2004) argued that these differences are based on the government policies, cultural differences and stage of the economical development. They claim too, the quality of CSR report is not related to the volume of disclosed information.2.2.4 Corporate Social Responsibility reporting guidelinesSince a CSR report is a voluntary disclosure no laws and regulations exist when developing the report. But when no rules and regulations exist to provide a CSR report, too much freedom occurs when formatting a report. This can follow in a diversity of high and low quality levels among CSR reports. Low quality level reports provide less transparent and less reliable information then high quality reports. To provide high quality CSR reports, a framework for developing CSR reports should be constructed. A framework with guidelines to establish high quality CSR reports provides CSR reports which achieve a satisfied quality level. Below, descriptions and tools are given about guidelines, provided by the Global Reporting Initiative (GRI).The GRI is an important institute for the development of CSR reporting. They provide a framework for sustainability reporting. This framework is based on stakeholders’ perceptions. Principles and objectives underline the GRI framework. When managers follow the framework benefits arise for the organization it selves and the users of the report. The GRI framework is global, the most used framework for social reports. On their website they refer to their mission:“To create conditions for the transparent and reliable exchange of sustainability information through the development and continuous improvement of the GRI Sustainability Reporting Framework.” The provided framework is developed for every company independent of size and sector characteristics of companies. The principles guarantee qualified provided information. The principles include standardizations of the information disclosures. Different reporting levels are specified by the GRI framework. The application levels are A, B and C, these levels measures the extent of coverage of the GRI framework. This level framework provides benefits for report users and makers. This level framework provides incentives for report makers, to increase their reporting skills.Following the objectives of GRI, the CSR report should measure and publish the company’s responsibilities to stakeholders about sustainable growing performance. These reports can be used for different purposes:Comparing and assessing the sustainable performance of legislation, official norms, codes, performance standardization and voluntary initiatives.Assessing how the company is influenced by and influencing the expectations of sustainable paring different performance of a company and the difference between companies on a long turn.2.2.5 Summary Corporate Social Responsibility Several definitions exist CSR, the CSR definitions describe that CSR covers responsibility of environmental and social side, which benefits the company and society. When a CSR strategy is implemented in the company, CSR activities support the goal of CSR strategy. The outcome of CSR is CSP. When the CSR strategy is implemented in the organization, managers have the opportunity to communicate the CSR information in a CSR report. When the CSR information is disclosed, stakeholders are provided with more information. No laws and regulation exist to prove a CSR report. But the GRI provides guidelines for developing a report. This framework provides benefits for users and makers of CSR report. 2.3 Positive Accounting TheoryThe positive accounting theory is important in disclosing voluntary information. The positive accounting theory is useful to specify the reasons of disclosing a CSR report. The positive accounting theory includes theories which explains and predicts circumstances. As argued by Deegan and Unerman (2006) the positive accounting theory focuses on the relationship between stakeholders and the company. Deegan et al. (2006) explains that positive accounting theory is used to assists this relationship. The agency problem, information asymmetry, legitimacy theory and stakeholder theory are explained in this paragraph. First, the information asymmetry and the agency problem are introduced. The information asymmetry is deducted from the agency problem. The legitimacy theory and stakeholder theory are explained as well. The incentives of CSR reporting which are based on the positive accounting theories are more discussed in next chapter. 2.3.1 Agency theoryThe agency problem arises among stakeholders and management. This problem is designed by Jensen and Meckling (1976). The agency problem is driven by self interest. Stakeholders provide resources to the company, for example, shareholders or investors provide financial resources and the environment provide environmental resources to the company (fresh air and natural production materials). The resources provide by environment are represented by environmental foundations and government Management (insiders) make decisions based on their self interest. But in the decisions making, concerns of stakeholders (outsiders) are involved. Because stakeholders provide resources to the company. Stakeholders are defined by Freeman and Reed (1983) as:“Any identifiable group or individual who can affect the achievement of an organization’s objectives, or is affected by the achievement of an organization’s objectives.” Different stakeholder groups have different concerns of the company. Environmental foundations have environmental concerns. For example, Environmental foundations do not appreciate management decisions to open a new polluting factory near a natural area. The environmental foundation is affected by the decision making, because it looks after the environment. But the environmental foundation has no opportunity to decide whether or not building the new factory. Management benefits from the new factory, if the new factory results in higher profits, this higher profits drives management to decide to build the new factory. Managers make decision based on their self interested incentives. Management and stakeholders differs in their incentives. In the decision of new projects, resources of stakeholders are involved. But this stakeholders does not always play an active role in decision making, stakeholders are not able to make decisions on their provided resources (employees are not able to make decision which insure their work). Jensen et al. (1976) argue that when managers and stakeholders do not own the same incentives and when stakeholders are not able to monitor the management’s behavior, the agency problem occurs. (Healy and Palepu, 2000)The agency problem arises due to different incentives among management and stakeholders. A lot of different stakeholders exist, with different concerns. Investors benefit from high company returns, although environmental foundations are more concerned about the negative damage of the company and labor unions wants to retain the amount of work as before. All this different concerns leads to different incentives. But outcomes exist to solve issues, about conflicting incentives. Healy and Palepu (2000) offer ways to decrease the agency problem. The manners that Healy et al. (2000) are described are: contracting, disclosing, corporate governance, information intermediaries and corporate control contest. Requirements can be made about issues. Healy et al. (2000) argue that compensation requirements of managements and rights and obligations of different stakeholders decrease the agency problem. Contracts should be made, to describe the requirements. For proving if management honors the requirements in the contract, disclosures are important. Healy et al. (2000) argue if information is disclosed, stakeholders have the opportunity to monitor if management meets the contractual agreements. Also the board of directors can influence the agency problem, which is structured in the corporate governance of the company. Following Healy et al. (2000), this board has the aim to control management, they make sure if management acts properly. Information intermediaries as rating agencies and financial analysts will probably detect misuse of resources and funds which are provided by investors. When detection by rating agency or financial analyst occurs, the reputation of management will be harmed and the market values the company lower. Healy et al. (2000) investigate that the threat of market undervaluation reduces the incentive of management to make decisions in favor of their self. Finally the threat of taking over company’s stocks by competitors reduces the agency problem between investors and managers. If investors are not satisfied, they have the opportunity to sell stocks to competitors. This threat is called the corporate control contest. (Healy et al., 2000)Healy et al. (2000) argue that whether contracting, disclosure, corporate governance, information intermediaries and corporate control contest eliminate agency problem is questionable. Economical and institutional factors are the reason for the questionable effectiveness. These economical and institutional factors are, for example, the ability to develop requirements in contracts. Also Healy et al. (2000) argue that it is possible that the board of directors can own the same incentives as management. The writers also state that when the corporate control market is social and friendly, competitors are less motivated to take over the stocks. These economical and institutional factors have influences on the effectiveness of the solutions to solve the agency problem. Economical and institutional factors are continuously changing, so solutions for agency problems should be continuously be adjusted.As Healy et al. (2000) argue that the provided solutions of the agency problem are questionable, this implies not, that the provided solutions of the agency problem are not useful. At least the solutions separate will face a too less impact on the agency problem. But when taken the solutions together, the effect will be larger. Voluntary disclosures and also CSR reporting are one of the solutions and can be seen as a tool for reducing the agency problem when also the other solutions are taken into considerations. The problem exists due to the fact that managers and investors have different incentives with the company. Also both parties act in their own interest. When disclosing information, management behavior can be monitored, for example CSR report disclosure monitored management on their CSR performances. Disclosures provide more openness in the behavior of management. This implies a way to monitor acts and decisions made by management. Now it’s possible for stakeholders to control and assess management behavior. When stakeholders disapprove management behavior, it will probably result in an undervaluation of the company. The undervaluation is harming the company, therefore management tries to make a good reputation, voluntary disclosures can be used as a tool for achieving this reputation. This result of voluntary disclosure, the possibility that investors judge management has the result of more openness.2.3.2 Information asymmetryThe explanation of information asymmetry is based on the difference in information hold by stakeholders (outsiders) and management (insiders) of the company. Management makes decisions in favor of continuing the company. For making accurate decisions, management is completely informed. Although stakeholders do not own all information. The company is dependent on stakeholders. Stakeholders provide resources to the company. Before, the agency problem was explained. The information asymmetry can be seen as a result of the agency problem. The agency problem is driven by self interest. Management and stakeholders make decisions based on this self interest. Providing information in a disclosure is one of these decisions. Provided Information is used in decision making, when management influences the amount of provide information they have the possibility to influence stakeholders decisions. Management sometimes benefits from disclosing less information. When management provides less information, the information asymmetry is high. The information asymmetry shows the openness of the company. When a company is full open, management and stakeholders owns the same amount of information. But in practice a distinction in information hold by management and stakeholders exist. This difference in holding information is called the information asymmetry. Healy et al. (2000) argue that in the information asymmetry risk is involved. Stakeholders make decisions based on the provided information. When relevant information is not provided to stakeholders, the decision of stakeholders can be different. Stakeholders want receive as much information as possible, to make optimal decisions. When management provides all information the competition status of the company can be affected. (Healy et al., 2000)Argued by Akerlof (1970) managers bias the provided information in a positive way. When providing positive information, the company will attract more investors. When management judge less well and bad information also in a positive way, more investors are attracted. Akerlof (1970) described that this result in a situation where investors are not able to make a distinction between good and bad information. Information announcements about good investments are undervalued and the announcements of bad investments are overvalued. This problem is called the Lemons problem. Akerlof (1970) argue the lemon problem will finally leads to a breakdown of the capital market. The lemon problem arises from the information asymmetry. If no information asymmetry exists, outsiders own all detailed information and are able to value the information content. If a situation of no information asymmetry occurs, investments are valued on unbiased information.The information asymmetry problem and lemon problem can be reduced in different manners. Healy et al. (2000) explains three different tools for reducing the information asymmetry and lemon problem. Information intermediaries, optimal contracts and regulation result in a lower information asymmetry. Ratings agencies and financial analysts, which are information intermediaries, are searching for information privately owned by management. If information intermediaries make their findings publically available, the distinction between information hold by managers and outsiders is reduced. Optimal contracts are also solutions for reducing the information asymmetry. Optimal contracts incorporate management requirements to disclose optimally all private information. When managers should announce all private information, more information becomes available for investors. This leads to a reduction in the information asymmetry. Besides contracts, regulations that require management to disclose all private information, results in a reduction of the information asymmetry. When stakeholders own more information, they are more able to measure the difference between good and bad information. The increase in information hold by stakeholders, results in a reduction of the lemon problem. (Healy et al., 2000)Voluntary disclosures and CSR reporting can use as tool to reduce the information asymmetry. Management can decide to provide voluntary disclosures, the voluntary disclosures will decrease the distinction in information hold between insiders and outsiders. Voluntary disclosures are announced by management it selves. No rules and regulation exist that requires management to disclose the provided information. This negative relation between information asymmetry and voluntary disclosures is observed in the study of Akhtaruddin and Haron (2010). The information asymmetry was measured with the agency costs and found evidence that high agency costs are associated with low level of voluntary disclosures. This result implies that voluntary disclosures reduce the information asymmetry.2.3.3 Legitimacy theoryThe legitimacy theory belongs to the positive accounting theory, the legitimacy theory explains why management decides to take action in favor of society. For understanding the legitimacy theory it is important to know, that a company operates in the society. And that the company is judged by the society on their performances. A detailed explanation of the legitimacy theory is provided by Deegan and Unerman (2006):“Legitimacy theory relies upon the notion that there is a ‘social contract’ between the organization in question and the society in which it operates.”Deegan and Unerman (2006) explained the social contract concept by: “… a concept is used to represent the multitude of implicit and explicit expectations that society has about how the organization should conducts its operations. “This explanation implies that, the legitimacy theory is driven by the public expectation of companies about their operations. This expectation is not only based on the production process, but also on the performances of corporate governance and other processes which assist the operations of the company. Companies obtain legitimacy from the public, if they are performing in accordance of public. In the study of Lightstone and Driscoll (2008), where companies behavior is tested to obtain legitimacy. In this study Canadian companies were investigated on ethics and legitimacy. They found that companies are using ethical language to inform stakeholders. When bad news was communicated also positive information was provided. The authors suggest that this positive information misleads the stakeholders about the bad news. When stakeholders are provided with bad news and positive information, stakeholders make more positively expectations about the company than when only bad news is communicated. This is in conformity that companies will achieve legitimacy from the public. When only the single bad information was announced, the public will value the company at a lower level. But when also positive news is announced, the valuation of the public improves. From the disclosure of this positive news, the company is judged legitimate by the public. Lui and Taylor (2008) found evidence that management achieves legitimacy from the society. In their research they investigate the voluntary disclosures of executives’ remuneration in Australia between 2003 and 2004. Starting in 2002, scandals about corporate governance were detected. This scandals result in a rising demand of stakeholders to provide more information about executives’ remuneration. In the study of Lui et al. (2008) was investigated that management legitimated the company, by providing more detailed information about executives’ remuneration. These details were voluntary provided in the annual report. This study implies that companies are driven by the expectations of the public to legitimate their operations. After disclosing detailed information about executives’ remuneration, the company achieves from the society the ‘license’ to operate.Cho and Patten (2007) found evidence that environmental disclosures are used as a tool for legitimacy. This study is based on companies which are separated out in different groups, the environmental sensitive and non-environmental sensitive companies and good and bad environmental performers. In this study is concludes that companies which are bad environmental performers are disclosing more environmental information. The findings that environmental disclosures is a tool for legitimacy and that bad environmental performers provide a higher level of environmental information is consistent with Dowling and Pfeffer (1975). In the study of Dowling et al. (1975) three tools are meant to obtain legitimacy, these tools are:“The organization can adapt its output, goals and methods of operation to conform to prevailing definitions of legitimacy.”“The organization can attempt, through communication, to alter the definition of social legitimacy so that it conforms to the organization’s present practices, output, and values.”“The organization can attempt, again through communication, to become identified with symbols, values, or institutions which have a strong base of social legitimacy.”This implies that the findings of Cho et al. (2007) are in consistent with the second tool of Dowling et al. (1975). Poor environmental performers are using a higher amount of environmental disclosures. The poor environmental performance companies need more communication to obtain a legitimacy level and therefore these companies communicate more environmental related information. This described studies of Lightstone et al. (2008), Lui et al. (2008), Cho and Patten (2007) and Dowling et al. (1975) provide the incentives to disclose voluntary information. Because public provide legitimacy to the company. The company should act in a way that they deserve the legitimacy. Voluntary disclosures are a way to become legitimate. Argued by Dowling et al. (1975), through communicating, the company can define the taken legitimacy operations. The public observes the information from the voluntary disclosure and provides legitimacy. 2.3.4 Stakeholder theoryThe stakeholder theory has some similarities with the legitimacy theory. For understanding both theories there should be taken into account that the company operates in the entire society. The company has to deal with different stakeholder powers concerns. Argued by Deegan and Unerman (2006), the stakeholder theory can be distinguished into the ethical branch and the managerial branch. From an ethical point of view, all stakeholders should be treated equal. All stakeholders have the same rights and no difference in stakeholder power should exist. The rights of the stakeholders should not be violated. This result that all stakeholders should be informed about the impact of the company on their individual concerns. (Deegan et al., 2006)The managerial branch of stakeholder theory emphasizes different stakeholder groups. The different groups should be managed in different ways, this suggest that not all stakeholders should be treated equally. Demands of powerful stakeholders are rather meet than less powerful stakeholders. This is conformity with the study of Ullman (1985) which argues that stakeholders’ power to influence the company is dependent on the stakeholders’ possibility to control over required resources of the company. This implies that stakeholders which have much power over the company also have more influences on the company. In a recent study of Huang et al. (2010) the drivers for environmental disclosures are investigated in Taiwan. There is made a distinction into internal and external stakeholders. There is found evidence that external stakeholders have a larger power in environmental disclosures. Internal stakeholders provide additional pressure to the environmental disclosure. Huang and Kung (2010) argue that larger companies and companies that in the past harmed the environment observe more pressure from government to disclose environmental information. They suggest companies that harmed the environment in the past, should improve their legitimacy. In this study is also claimed that larger companies facing more impact of stakeholders, on environmental disclosures. The final conclusion of this study is that, stakeholders put an important pressure on management to disclose environmental related information.2.3.5 Summary Voluntary Disclosures and Positive Accounting TheoryIn this paragraph the positive accounting theory is explained. The positive accounting theory explains and predicts company’s behavior. The agency problem explains that management and stakeholders behave out of their self interest incentives. This problem predicts that managers and stakeholders make decisions in favor of their self interest. A voluntary disclosure is a tool to reduce the agency problem. In the disclosed information, the behavior of management is visible. The information asymmetry is also reduced by voluntary disclosures. The level of information asymmetry reflects the gap between information of the company hold by managers and stakeholders. After a disclosure, more information becomes available to stakeholders, this reduces the information asymmetry.The legitimacy theory explains that company made decisions to favor society. Companies obtain power from society, when they abuse this they will lose power. So this implies the fact that companies made decision to legitimate each self. CSR disclosure is used to achieve legitimacy, a CSR report provides more openness in the CSR activities of the company. The society will judge the company in a positive way, when CSR information is provided. At least the stakeholder theory explains that company is influenced by some stakeholders. Two types of branch of the stakeholder theory exist. The ethical branch explains that all stakeholders should be treated equal independent of their power. But the managerial branch explains that differences exist in the power of stakeholders to influence the company. But the main message of the stakeholder theory is that different stakeholder groups have different demands. The company pays attention to all different demands of stakeholders. Every company has stakeholders which are interested in the CSR performances. The demand of these stakeholders is met when, a CSR report is disclosed with information about CSR performances. Chapter 3 Literature ReviewBefore a CSR report is provided, managers first choose to implement a CSR strategy in the company. When the strategy is operative, the CSR performances and other relevant CSR information can be summarized in a CSR report. By the provided CSR report, stakeholders get an inside view of the related CSR activities and results. But why do managers choose to implement a CSR strategy and to disclose a CSR report? The reasons of providing information about CSR activities and results are important to know. Also the impact of CSR reporting is interesting. Using prior research, the incentives and impacts of CSR reporting is discussed in this chapter. In the first paragraph the incentives of management to implement CSR in the organization are given. The remainder of the chapter provides an overview of existing literature, related to the impact of CSR disclosures. First the impact on the forecast error is discussed. Afterwards, literature related to the impact on the public sentiment is described. Finally, the influences on the financial performances are discussed.3.1 Incentives of Corporate Social Responsibility strategy and reportingA CSR strategy leads to another focus of the company, the triple P-strategy has an important role in here. CSR can be a way for implementing the triple P-strategy into the organization. Stakeholders are important parties for the implementation of the CSR strategy, this is discussed in the part about the incentives of the CSR strategy implementation. The CSR reporting incentives are also discussed in this part. The final part of this paragraph provides a description of the implementation of the CSR strategy and the CSR report. 3.1.1 Corporate Social Responsibility strategy and reporting incentivesThe triple P-strategy is a strategy performed by commercial companies, to have more focus on the social and environmental sides of the firm. This triple-P strategy is an aspect of CSR. The triple-P is coming from, People, Planet, Profit. Quaak et al. (2006) argued that this triple-P strategy is value creating by the company, not only value creating in terms of profit but value creating for society and environment too. This value creating comes from the possibility that, a cost reduction can occur when focusing on the society and the environment, but this cost reduction is not necessary the case. Conceivable cost reduction is decreases in potential litigation and pollution costs. More practical cost reduction is savings due to efficient paper use. The view of the company how finally the triple P-strategy should be implemented in the organization is dependent on stakeholders. All stakeholders have different concerns with the company (Quaak et al., 2006).For providing more transparency, companies should answer the demand of CSR related information from stakeholders. Dawkins et al. (2003) suggest that stakeholders are asking for more openness regarding CSR related performances. The company is able to inform stakeholders on different ways. The CSR report is one way, to make CSR information available for public. The CSR report improves the transparency of the company (Quaak et al., 2006). The CSR report is part of the CSR strategy. Determined on the CSR strategy, management decides whether or not to provide CSR information in a CSR report.As before already was mentioned, stakeholders put pressure on the company. The CSR strategy (and thus the CSR reporting) is influenced by stakeholders. But not only stakeholders have an impact on the CSR strategy. In the study of Quaak et al. (2006) different influences on CSR strategy are summarized. There is made a distinction between internal and external influences. Internal influences are coming from inside the company. Internal influences are further divided into actors and factors. Actors are person or people which are organized in groups within the company. This persons or groups have the power to determine and influence the CSR strategy. The internal factors are not person or organized person, but are elements within the company and have an effect on the CSR strategy. Outside influences are only coming from people or organized people, no factors from outside influences the CSR strategy. These people can make use of their power to influence managements’ decisions about CSR strategy. Since CSR reporting is part of the CSR strategy, internal and external actors and internal factors are influencing CSR reporting. Below in table 1 the internal and external influences are stated. Table 1: Internal and external influences on CSR strategy?Internal ?ExternalActorsInvestorsShareholdersConsumers / CitizensCustomerManagementSuppliersEmployeesNon- government OrganizationsNeighborhoodMediaGovernmentCompetition?FactorsCostsComplexityKnowledgeThe level of influences of actors and factors can be different for each company, the level of power of the actors and factors are different. Companies which operate in rural areas are facing a lower level of neighborhood power, compared to companies in urban areas. Expected from the stakeholder theory, companies benefit if they satisfy CSR reporting demands of stakeholders. Stakeholders have the opportunity to misuse their power, stakeholders are able to put pressure on their demands. Companies are part of the society, the society and companies are affecting each other. Decision made by the society has an influence on the company and also decision made by companies influence the society. This implies society and companies are dependent of each other. The implemented CSR strategy influences the society as well. Porter and Kramer (2006) called that society and companies should make decisions based on the shared value. If they taken the shared value into account, the decision adds value for both parties. Porter et al. (2006) separate out the two types of linkages: inside-out linkages and the outside-in linkage. The inside-out linkage represent the influence of the company on the society, an example is the pollution effects of the producing. The outside-in linkage represents the effect of the society on the company. An example of the outside-in linkage is incentives provided by government to invest in social projects. For developing a CSR strategy into the company, the company should implement the inside-out and outside-in linkages in the CSR strategy. Porter et al. (2006) claim that the implementation of inside-out and outside-in linkages supports the shared value. When allow for the dependences of company and society, the company is able to design a CSR strategy with benefits for both parties. When the CSR strategy is implemented in the company, the company can decide to provide a CSR report. In the CSR report the CSR performances are showed, also future CSR related information can be incorporated in the CSR report. Since companies are free to choose for communicating CSR information, the CSR report is a voluntary disclosure. The incentives based on agency theory and information asymmetry, legitimacy theory and stakeholder theory are capable to explain the incentives of a CSR report. In paragraph 2.3 the agency theory, information asymmetry, legitimacy theory and stakeholder theory are explained, in this explanation is already described that these theories are in conformity with voluntary disclosures.Healy and Palepu (2001) described incentives to provide voluntary disclosures, based on the positive accounting theory. These incentives are also applicable for CSR reporting. The capital market incentive can be linked to a CSR report. The capital market incentive explains that due to the disclosure the information asymmetry is reduced and this results in a reduction of costs. Managers which decide to disclose a CSR report, can base this decision on the capital market incentive. For example, in the provided CSR performances, information is given about the investments in sustainable production. This reduces the information gap between stakeholders and management and reduces the agency costs. The cost of equity decreases, investors are more able to expect future pollution and litigation costs. If a threat undervaluation of the company occurs, managers put more focus on the CSR performances. The focus on CSR performances reduces the risk of undervaluation. It is imaginable that a company has bad financial performances due to an economical downturn, but has invested in social education projects. Social projects are beneficial for the society as a whole, stakeholders value these projects positive. When company puts attention on the CSR report and if good projects are performed, the magnitude of the undervaluation is reduced. This is in conformity with the corporate control incentive of voluntary disclosures of Healy et al. (2000). The legitimacy theory is the driver of this corporate control incentive. The company wants to receive legitimacy from the society when communicating CSR performances. The corporate control contest is observed in the findings of Lightstone et al. (2008) they found that companies communicate bad information together with positive information. Managers are affected by the performances of the stock prices. CSR performances are quite well judged by (potential) investors, there is less risk involved when investing in CSR reporting companies, then in comparison with not CSR reporting companies. The litigation costs are reduced, the company faces a more open view to investors, due to the reduction in information asymmetry. There is a better view of the possibility if a third party will litigate the company, when the company harms the society or environment. This decrease of litigation costs implies an increase in the future cashflows, which finally results in an increase of the stock prices. This described stock compensation incentive of Healy et al. (2001) is also applicable for the CSR reporting incentives. Similar to the control contest incentive, the stock compensation is based on the legitimacy theory. The stock compensation incentive is not based on the stakeholder theory because, the demand of providing CSR information comes not from the demand of stakeholders. But the stock compensation incentive for CSR reporting is also a result of the agency problem and the information asymmetry. Also the standalone threat that stakeholders litigate managers is an incentive for the decision of providing a CSR report. When companies announce more information about the influences on the environment, stakeholders are more informed about the impact of the company. These CSR performances can be positive or negative judged by the stakeholder (also dependent on the stakeholder interest). So managers should be very careful in the decision of providing certain information in a CSR report, or even providing a CSR report. This is in conformity with the litigation cost incentive of Healy et al. (2000). This incentive is based on the legitimacy and stakeholder theory. And also the agency problem is taken into account when CSR reporting decisions are based on the litigation cost incentive. The management talent signaling incentive of Healy et al. (2000) argues that managers are better judged if they provided more information. When voluntary CSR performances are announced, it is expectable that managers are better judged due to the more provided information. This incentive is based on the incentive that the manager wants to achieve a license to operate. Also when managers disclose a CSR report based on signaling. They take the agency theory into consideration, the management makes the CSR reporting decision based on their own interest. The sixth incentive of Healy et al. (2000), implies a reduction in provided information, this is also applicable to CSR reporting. For example, a company implements a specific project to improve the health of employees. The designed health project is unique and no other companies have a comparable project. This project influences in the first place the quality of work provided by the employees. But also talented employees are attracted to the company due to the good second working conditions. For the company its beneficial to do not communicate specific project information about the content, to stakeholders, it can decrease the competitive position of the company. This proprietary cost incentive should also be taken into account when managers decide to provide CSR information. The stakeholder theory is relevant for this proprietary cost incentive, because the decision whether or not to disclose certain information, is based on the power of stakeholder groups. And also the agency problem is important in this incentive, the self interest of managers to keep the information inside the company should be noticed.CSR reporting is a voluntary disclosure, companies are not obligated to disclose CSR information. The above described incentives argue a disclosure of a CSR report. But when incentives exist to provide a CSR report, the question rises, why CSR reporting is not mandatory. In the study of Rodriquez and LeMaster (2007) this question is answered. Rodriguez et al. (2007) claimed that CSR reporting should stay voluntary disclosures. The Security and Exchange Commission (SEC) should not require companies to disclose CSR reports, CSR reporting increases creditability and transparency (Rodriguez et al., 2007). When CSR disclosures become required, few gains are reached but this outweighed not the cost of the requirement. 3.1.2 Implementing Corporate Social Responsibility strategy and reportThe development of a CSR strategy is very important, when the strategy is well developed and implemented, the company achieves benefits. Porter et al. (2006) argue that a chosen CSR strategy incorporates environmental, social and economical aspects. The environmental, social and economical aspects are comparable with triple-P strategy. With the chosen strategy the company is able to differentiate from competitors. An example where the strategy differentiation results in many benefits is provided by Porter et al. (2006) this is the strategy of the Prius produced by Toyota. Toyota designed a strategy for the hybrid Prius and these results in a competitive advantage and in well environmental performance.To implement a CSR strategy, the company should go through different phases. The whole company is effected by the CSR strategy, therefore the company should take different phases into account when implementation a CSR strategy. These different phases are developed by Kabadse et al. (2006). The first phase is the decision phase, in this phase the knowledge about CSR is important and different stakeholders concerns. The outcomes of the decision phase are the designed CSR strategy goals. When the decision for implementing a CSR strategy is completed, the adoption phase will follow. CSR is a structure that should be implemented in the whole organization. This implies the aim of the adoption phase, all employees should have the same incentives for the CSR strategy. Kakabadse et al. (2006) argues that, to make the adoption phase successful there should be a persuasive sub stage included. The possibility exist, that two or more contrasting incentives are conflicting with each other. To handle these conflicts, paradoxes and conflicting priorities should be prepared. Also it is important that through the company’s activities the CSR message is consistent implemented. Monitoring has the aim to justify if the CSR message is consistent implemented. This monitoring will give accountability to the CSR performance. The third phase of implementing CSR strategy in the company is the commitment phase. Following Kakabadse et al. (2006), all employees should have enough discipline and passion for results to follow the CSR from initiative to application. This phase suggest there is a will to act. This last phase has the aim to continuing the CSR strategy in the company. Below in table 2, the different stages are presented. Table 2: Different phases that lead to the CSR strategy (Kakabadse et al. 2006)3.1.5 Summary incentives Corporate Social Responsibility strategy and reportOutside and inside factors and actors have an influence on the implementation of CSR into the organization. The power of stakeholder influences the impact. The capital market, corporate control, stock compensation, litigation cost, managers talent signaling and proprietary cost incentive which are based on the general voluntary disclosures, are useful to explain CSR report incentives. The incentives to provide a CSR report are based on the agency theory, legitimacy theory and the stakeholder theory. The decision, adoption and commitment phase are important phases for implementing CSR. The decision and adoption phase influence the implementation of CSR. The commitment phase influences the going concern of CSR. 3.2 Impacts of Corporate Social Responsibility Report Companies implement CSR because benefits are expectable from the CSR strategy. This benefits or incentives of CSR are described in previous paragraph. When CSR performances are disclosed, a reaction from stakeholders is expectable. Stakeholders’ value providing of additional information. In this paragraph the impacts of CSR reporting are discussed, based on existing literature. The impacts on the forecast accuracy are first discussed. The public sentiment or the reputation will be influenced too, this is described in the second subparagraph. At the end of this paragraph the financial impacts are discussed. In the summary a table is provided, which summarize the results of impacts from the used literature. 3.2.1Forecast accuracyWhen CSR information becomes available, stakeholders are provided with more information (Dhaliwal, Radhakrishnan, Tsang and Yang 2010, A). This additional information produces different effects. The forecast accuracy is one of the variables that are affected by CSR disclosures. First a describing will follow about expectations of the impact on the forecast accuracy. Thereafter studies are described which provide evidence on the impact of the forecast accuracy. CSR disclosures which are provided by management include relevant information for stakeholders. This additional information adds value to the stakeholders based on the information asymmetry and the agency problem (Healy et al., 2010). The information asymmetry will be reduced after disclosing CSR performances. The distinction in information hold by management and stakeholders decreases. The increase in level of information hold by stakeholders results in a reduction of information asymmetry. CSR reporting is provided to all stakeholders, also analysts achieve this CSR information. In CSR reports, relevant information is stored, which influences future financial performances. Analysts make forecast based on observed information. When more information is disclosed, analysts achieve more information, this result in more accurate forecasts. Due to the more provided information, analysts are able to make fewer errors in the predictions. Therefore there can be expect the forecast error decreases. (Dhaliwal et al. 2010, A)CSR information is not directly related to financial performances of the company, but although it is valuable information for investors and other stakeholders. CSR information is environmentally and socially related and does not provide direct information about financial performances. The provided information gives a view in the contribution of the company activities on the society and environment. The provided information is useful to make expectations about future financial performances. The CSR report provides information about future costs (pollution and litigation). Analysts are able after the disclosure of the CSR report, to analyze the company in a more certain way. When less risk is involved in an analyze, more accurate forecasts results. (Dhaliwal et al, 2010, A)The more accurate forecasts will influence the society. Almost all society behavior faces partly influences of forecasts. Management and citizens are making decisions based on expectations. Forecasts are expectations about future financial performances of a company. Many issues are dependent on expected financial performances. A practical examples is, the extension of company (social) activities, this affects working employees and potential employees. In the study of Dhaliwal (2010, A) the relation between CSR disclosures and analyst forecast accuracy is studied. Evidence is found that disclosing a CSR report leads to a significant lower level of analyst forecast error. This means, forecasts are of better quality (more accurate) when a CSR report is disclosed. When a company provides CSR performances, more company’s information becomes available. Analyst inserts this information in the forecasts, which results in more accuracy of the forecasts.This positive relation observed in the study of Dhaliwal et al. (2010, A) between CSR reporting and forecast accuracy is stronger for countries with a higher level of stakeholder orientation. In countries with a high level of stakeholder orientation, companies’ social performances are important. In countries with a high level of stakeholder orientation, fewer errors are made in the forecasts. The evidence in the study of Dhaliwal et al. (2010, A) is found by analyzing CSR reports and forecast accuracy data, out of 24 countries. The result that circumstances of countries influencing the impact of a CSR report is also observed in other studies. Williams and Aguilera (2008) argue that differences exist in importance of social performances of companies. This importance of social issues is related to forcing norms. Countries which put more attention to stakeholders’ issues, are more dealing with stakeholder influences on decision making. Companies which operate in countries with lower levels of natural resources are more concerned with the impacts of using these resources. This scarcity of natural resources will result in more stakeholder orientation.In the study of Dhaliwal et al. (2010, A) evidence is found that the relationship between CSR disclosing and forecast accuracy is stronger for companies with less transparency. This implies if companies with a low level of transparency, which provide CSR information face a bigger impact on the forecast accuracy. These lower level companies have a bigger positive change in forecast accuracy compared with companies which are more transparent. Obviously companies which are already open (more transparency), will achieve less additional transparency compared to less open companies. Dhaliwal et al. (2010, A) argued that due to providing CSR information, analysts made fewer errors in their forecast, this is the result of a more transparent view. Investors and other stakeholders benefits from this transparent view.The increased transparency due to providing a CSR report, is observed in the study of Dhaliwal et al. (2010, A) but also in the study of Lang and Lundholm (1996). The study of Lang et al. (1996) is based on older data out of 1980. The older study also includes financial disclosures. They argue that ratings made by analysts, based on financial and non financial disclosures, are positively related to forecast accuracy. They found evidence that companies which have more informative disclosure policies (so disclosing more information), are followed by more analysts, this analysts providing more accurate earnings forecast. The accurate earnings forecast are observed from less distribution among individual forecasts and less volatility in the forecasts. The level of information in disclosures is determined by Financial Analyst Federation (FAF) ratings. These ratings are based on annual and quarterly published information, other published information and investors’ relations. The level of followed analysts is measured by the actual amount of followed analysts. The error in forecasts is measured by the difference in actual earnings per share and the median of forecasts divided by the price per share. Hope (2003) studied the influence of disclosures on forecast accuracy. Positive evidence on this relationship was found, based on a sample out of 22 countries. To measure the information contained in an annual report, the CIFAR disclosure score is used. This CIFAR score incorporates 85 annual reports variables. The forecast error is based on the difference between actual earnings per share and forecasted earnings per share, divided by the stock price at the beginning of the fiscal year. The described studies of Lang et al. (1996) and Hope (2003) takes the financial disclosures also into consideration when measuring the impact on the forecast error. But the study of Dhaliwal et al. (2010, A) focus on the voluntary CSR disclosures and the forecast accuracy effect. 3.2.2 Public sentimentStakeholders put pressure to implement more social activities into the organization. This implies the support of stakeholder for a CSR strategy (Quaak et al., 2006). If a company implements the stakeholders demand in the organization, stakeholders are satisfied. Stakeholders’ satisfaction has an impact on company’s sales, which is dependent on stakeholder power. (Chen, 2009) In the study of Chen multi stakeholder satisfaction data is used for proving the positive relation. Stakeholder satisfaction influences the future revenues. The influences of stakeholders on the future revenues are dependent on stakeholder power. This pressure of stakeholders to provide a CSR report is in conformity with the legitimacy theory and the stakeholder theory. In this subparagraph the effect of CSR on the satisfied stakeholders or on the public sentiment is discussed. The effect on the satisfied stakeholders has also an impact on financials, this will be discussed in next part. Not only the fact that stakeholders are asking for a CSR strategy (stakeholder theory), have influences on the public sentiment. But also the reduction in information asymmetry and agency problem is a result too and influences the public sentiment. As already been mentioned in describing the impact of forecast accuracy, CSR disclosure results in a reduction of information asymmetry. Stakeholders are satisfied with this reduction, they hold more company information, after the disclosure. This results in more certainty about future performances of the company. Also the agency problem is reduced after a CSR disclosure. Due to the provided information stakeholders are able to monitor management. Now stakeholders have the possibility to control management on the performed activities. This more transparent view and way to monitor management improve public sentiment. This improved sentiment is also reflected in the legitimacy theory, companies wants to achieve a ‘license’ to operate. With a CSR report this ‘license’ can be obtained.In the study of Clarkson et al. (2010) different relationship were tested. One of tested relationship was between stakeholder perception and environmental voluntary disclosures. In this study the economical impact of environmental disclosures was also investigated. The CSR report was a proxy for environmental voluntary disclosures. The Janis-Fadner coefficient was used as proxy to measure the stakeholder perception. This coefficient is build on the relationship between positive and negative environmental related articles. This study was based on the years 2003 and 2006. The sample was based on companies from the five most polluting industries of the US. For measuring the voluntary environmental disclosures, they made use of CSR reports and the quality of these reports was measured with a self constructed index. In this study an important control variable was included the Toxic Release Inventory (TRI) emissions. Emissions are mainly negative for the environment. There can be expected, the public judges this emissions in a negatively, when making related information available. The overall conclusion of this study was, CSR reports have a positive influences on the public sentiment CSR reports increased the Janis-Fadner coefficient. In the study of Brown and Dacin (1997) is argued that knowledge about a company, hold by consumers, and affects the attitude and behavior towards the company. This affects was measured with a field study, potential consumers were asked to evaluate products. CSR disclosures give consumers insight knowledge about the social activities of the company. This knowledge will affect the consumer’s attitude and behavior towards the company. A positive relation was observed, CSR disclosures influence reputation. Especially an improvement of reputation takes place in countries where consumers have a high level of awareness on social issues. Following the study of Lev, Petrovits and Radhakrishnan (2010) this effect is extended to higher sales. In this study is argued that companies with better reputation face an improvement in sales. This relation is stronger for companies who operate in the consumer market, these companies have to deal with high levels of consumer’s sensitivity.Ittner and Larcker (1998) found a positive association between consumers’ satisfaction and company’s financial performances. Ittner et al. (1998) studied if financial performances are dependent on non-financial measurements. Customer satisfaction is a proxy for measuring the non-financial performance. The used customer satisfaction is highly dependent on CSR. In this study of Ittner et al. (1998) a positive relation between CSR and customer satisfaction is concluded.3.2.3 Financial impactsCSR information provides information about future company’s performances. Expectations about future performances have a financial impact. Financial performances are affected by a CSR report. The reduction in information asymmetry and agency problem is influencing financial performances. This has an influence on the public sentiment. Before already is described that the impact on the public sentiment is affected by the agency theory, information asymmetry, legitimacy theory and stakeholder theory. Due to improvements in public sentiment, total sales improve. Total costs can be reduced by a decrease of pollution and litigation risks. Information holds by stakeholders increase, this will lower the information asymmetry. Due to, the provided information, stakeholders are able to control management on environmental and social behavior. There will be expected that this shall face a reduction in cost of equity and cost of debt. Investors receiving more certainty about the company’s performances. In the study of Lev et al. (2010) is argued that the improvement in public sentiment is reflected in the increase of sales. The increase in sales has an influence on the financial performances of the company. In the study of Dhaliwal, Li, Tsang and Yang (2010, B) the relation between voluntary nonfinancial disclosure and the cost of equity capital is researched. In this study there is made use of CSR reporting as an indicator for voluntary nonfinancial disclosures. They only incorporate standalone CSR reports, coming from U.S. companies. The level of the reports are ranked, this is based on the KLD social performance ranking. If no standalone report is disclosed, the indicator variable will have a value of zero. The used cost of equity is calculated by three different models, the mean of these models is used to indicate the cost of equity. The founded variables are incorporated in two different regression models. The cost of equity plays in important role in financing the company, this financing finally faces results in the operational activities. The first regression model of Dhaliwal et al. (2010, B) test if companies with high cost of equity are more likely to disclose CSR information. This is researched by including all first-time reporting company variables. The second model tests the negative relation between cost of equity and CSR reporting. In the study of Dhaliwal et al. (2010, B) a negative relation between cost of equity and CSR reporting was observed. Cost of equity is reduced after the implementation of the first CSR report. For well CSR performing companies, this association is stronger. Companies which disclose better financial CSR performances are more attractive for investors. (Dhaliwal et al., 2010, B) The founding’s of Dhaliwal et al. (2010, B) are in conformity with the findings of Frankel, McNichols and Wilson (1995). In their study they investigated the reason of providing voluntary disclosures. They argue that companies which disclose voluntary information, wants to raise capital in the future. If companies are providing more information than required, they will be followed by more investors. When providing more information the cost of future raised capital will decrease. In this study also evidences is found about the negative relationship between voluntary disclosures and cost of equity. These findings are based on 1880 companies, from the New York Stock Exchange. Since CSR reports are voluntary disclosures, this results are applicable to CSR reporting.Richardson and Welker (2001) investigated in their study the relation between cost of equity and financial and social disclosures. In this sample Canadian companies are included, in the period 1990 till 1992. For measuring the financial and social disclosures there is made use of a score, which ranks the quality and quantity of the reports. A negative relation between cost of equity and financial disclosures has been found. But surprisingly a positive relation between cost of equity and the quality and quantity of social disclosures. This positive relation faces a reduction, when company has good financial performances. There is argued that this surprising result is coming from bias in the social disclosures. They claim, social disclosures add value to the company, not only the cost reduction is of importance but also other benefits exist. Financial investors are not the only party which is affected by the social disclosure, but also other stakeholders, for example consumers, are affected. The improvements in customers’ satisfaction also play a role in CSR reporting. The relation between firm value and voluntary environmental disclosures is examined in the study of Plumlee, Brown and Marhshall (2008). The firm value is composed out of the cost of equity and the future expected cashflows. The sample of this study includes 5 industry types (oil & gas, chemical, food & beverage, pharmaceutical and electric utilities) from the United States. To measure the quality of the environmental disclosures, they use a self constructed index, based on the GRI Report List. In this study environmental announcement provided in the annual report are incorporated as well. The findings were consistent with a positive relation between environmental disclosures and firm value. This implies a negative relation between cost of equity and the quantity of environmental disclosures. More information provided for the investors implies a reduction in the cost of equity. But the association between future expected cashflows and environmental disclosures is positive, found in the study of Plumlee et al. (2008). The present quality of the environmental performances, represent high financial performances in the future. The recent study of Clarkson et al. (2010) takes the economical and the social impact of voluntary environmental disclosures into considerations. This study is based on the previous discussed studies of Dhaliwal et al. (2010, B), Richardson et al. (2001) and Plumlee et al. (2008). Again in the study of Clarkson et al. (2010), the CSR disclosures are taken as measurement for voluntary environmental disclosure. The used sample in this study covers companies from the most polluting industries in the U.S. The impacts of CSR disclosure on cost of equity, firm value and public perception are researched. The researchers choose to investigate the effects of pollution industries, because these companies have relatively high environmental expenditures. These high expenditures possibly affect firm value. In this study also the environmental performances are taken in consideration. For measuring this, there is made use of the Toxics Release Inventory (TRI) emissions. For measuring the impact of CSR quality, the CSR report quality is based on a self constructed index of Clarkson, Li, Richardson and Vasvari (2008). The firm value is measured by the stock prices at the end of the fiscal year. The cost of equity is based on the average of two different models. In the study of Clarkson et al. (2010) there is found evidence about the positive relationship between CSR disclosures and firm value. The provided information adds incremental information for investors, these results from the changes in stock prices. The control variable toxic emissions have a negative influence on the stock price. This is obviously, more toxic emissions have a negative influence on the environmental performances of the company and this is reflected in the stock price. The toxic emission data provides insight to the risk and future environmental liabilities of the company. In contrast with the study of Dhaliwal et al. (2010, B) the study of Clarkson et al. (2010) found no evidence for a reduction in cost of equity. This difference can be explained by the differences in studies, the study of Clarkson et al. (2010) studied the most polluting companies. Polluting companies are more involved with risk related due to the pollution. This risk will increase the cost of equity of polluting companies. Carbon emissions are an important factor in the environmental performances of companies. Providing information about this variable can be done with a CSR report. The influences of carbon emissions are studied in the research of Matsumura, Prakash and Vera-Munoz (2010). In this study there is investigated whether carbon emissions is associated with firm value and cost of equity and cost of debt. The researched studied if disclosures of carbon emission have an impact on the firm value, cost of equity and debt. Carbon emission is an important environmental performance, carbon emissions is a measurement for the pollution of a company. This pollution amount is furthermore related to the amount of future claims. When stakeholders are provided with carbon emission information, an effect on financial variables is expectable. For investigating the relationship between carbon emission on firm value, cost of equity and cost of debt, Matsumura et al. (2010) made use of S&P 500 companies for the period 2006-2008. The data about carbon emissions are observed from the Carbon Disclosure Project (CDP) database. CDP is an organization which holds the climate change information of companies. In this study also a separation is made between high and low carbon emitters. The highly carbon emitters are mainly the chemical, utilities and computer and electronic product manufacturers. For testing the influences there is made use of a regression, in this regression several control variables are implemented. The results show a negative relation between firm value and carbon emissions. Companies with a higher level of carbon pollution will face a lower level of firm value. This relationship is showed for the high carbon emitters, but no negative relationship was found for the low carbon emitters. This conforms the expectation, that carbon emission is more a concern of high carbon intensive industries. The negative effect of carbon emission is reflected in lower level of firm value. Surprisingly a negative relation is observed between carbon emission and the cost of equity. This relation is significant for the highly carbon emitters and not significant for the low carbon emitters. At beforehand a positive relation was expected, because the negative effect of carbon emissions should compensate investors. Therefore a higher level of cost of equity due to an increase in carbon emissions was at beforehand expected. The surprisingly negative relationship is showed for high and low carbon emission companies.Some studies also investigate the financial impact of CSP. CSP is more focused on the performances and the outcomes of the corporate social activities. Wood (2010) uses CSP to examine the impact of the business-society relationships with financial results. The study of Wood (2010) is defined as a Meta analyses, based on the findings of other studies the financial impact of business-society relationships is investigated. The financial impact of CSP contains the impact of the content of a CSR report. The studies of Orlitzky, Schmidt and Rynes (2003) and Margolis, Elfenbein and Walsh (2007) studied the impact of CSP on financial performances. They both used a Meta analysis to examine the relationship. Orlizkey et al. (2003) takes 52 studies with 33.878 results into account and found a positive relationship. Margolis et al. (2007) also found a positive relationship, based on 167 studies. But the writers of this study claim that the founded relationship was a ‘mildly positive relationship’. Wood (2010) judged prior research on the relationship between CSP and financial performances. She concluded that the research on this relationship is already well established. Wood (2010) concludes that the research on CSP should be more relevant for stakeholders and society. She observed a limitation in present studies, that they are now more focused on the company’s itself. This limitation can be found in the study of Orlitzky et al. (2003) in this study the CSP are measured by 4 company’s components, the CSR disclosure, the reputation ratings, the social audits, CSP processes and observable outcomes and the managerial CSP principles and values. These components don’t include measurements of stakeholders and society. At least Wood (2010) concludes that the research in CSP should be more linked to other research fields, for example, human resources, sociology and business ethics. 3.2.4 Summary impactsIn previous parts the results of previous literature are discussed. The tested relationship is between CSR reporting and forecast accuracy, CSR reporting and public sentiment and between CSR reporting and financial performances. For providing an overview the results of these studies are summarized in the tables below.Table: Summarization of prior research of impact on forecast accuracyStudyTested relationshipUsed proxiesTime periodFounded relationshipDhaliwal, Radhakrishnan, Tsang and Yang(2010) CSR disclosures related to the forecast accuracy (among countries)Standalone CSR report and actual forecast error1990-2007PositiveLang and Lundholm (1996)Disclosures related to forecast accuracyFinancial and non financial disclosures, rated to FAF rating and actual forecast error1985-1989PositiveHope (2003)Disclosures related to forecast accuracyFinancial and non financial disclosures, rated to CIFAR score and actual forecast error1991 and 1993PositiveTable 4: Summarization of prior research of impact on public sentimentStudyTested relationshipUsed proxiesTime periodFounded relationshipClarkson, Hua Fang, Li and Richardson (2010)Environmental voluntary disclosures related to social (public perception) and financial impactCSR report, quality is based on index and Janis Fadner coefficient2003 and 2006PositiveBrown and Dacin (1997)Knowledge of company related to consumers behavior and attitudeField study, evaluation of products1996PositiveIttner and Larcker (1998) Non-financial performances and financial performancesCustomer satisfaction (costumer, business unit and firm level data) 1994 and 1995PositiveTable 5: Summarization of prior research of impact on financial performancesStudyTested relationshipUsed proxiesTime periodFounded relationshipMargolis and Walsh (2003) Corporate social performances related to cost of equityResults of prior studies1972-2003PositiveDhaliwal, Li Tsang and Yang (2010Environmental voluntary disclosures related to cost of equityStandalone CSR report, rated on KLD ratings and cost of equity1993-2007NegativeRichardson and Welker (2001) Social and financial disclosures, related to cost of equitySocial and financial reports ranked with a quantity and quality score and cost of equity1990-1992Negative (financial disclosures) Positive (social disclosures)Plumlee, Brown and Marhshall (2008Environmental voluntary disclosures related to firm valueCSR report, ranked on GRI score and cost of equity and future expected cashflowsnegativeClarkson, Hua Fang, Li and Richardson (2010Environmental voluntary disclosures related to social and financial impactCSR report, quantity is based on index, cost of equity and firm value and TRI emissions2003 and 2006No relationship (cost of equity) and Positive (firm value)Matsumura, Prakash and Vera-Munoz (2010). Carbon emissions related to environmental performancesDisclosed carbon emission data, firm value and cost of debt and equity2006-2008Negative (firm value), Negative (cost of equity) and Positive (cost of debt)Chapter 4 Hypotheses developmentIn chapter 3, the incentives and implementation of a CSR strategy are described. The CSR report is seen as the last step in the implementation of a CSR strategy. In a CSR report the CSR performances are disclosed. The third chapter also described the impacts of CSR reporting. Financial impacts can be observed as a result of a CSR disclosure, but impacts on the forecast and the public sentiment are also observable. This paper focuses on the impact on the reputation. The terms reputation and public sentiment are used interchangeably in this paper. In previous research the term public sentiment is often used, but stakeholders’ impression is more directly reflected with reputation. In this paper the impact of CSR reporting on reputation is examined.This chapter describes the development of the hypothesis. The hypothesis is based on the impact of the reputation, the effect on the reputation was already investigated in prior research. But additional research should be done to test the impact of CSR reporting on reputation. To test the relation between CSR reporting and reputation, five sub hypotheses are formulated. The first paragraph of this chapter describes the development of the main hypothesis. In paragraph two, the differences between the developed hypothesis and the existing literature are described. In the third paragraph a description is provided about the sub hypotheses.4.1 Development of the main hypothesis Changes in financial performances can be observable after disclosing a CSR report. Information which is disclosed in a CSR report provides an overview of the past CSR performances of the company, a reaction on the financial side is expectable. The financial impacts of a CSR report are already well investigated in existing literature. In this thesis, the focus is not on the financial impacts but more on the social side of the company, the reputation.In a CSR report, the performances related to the CSR strategy are described. When making CSR performances publically available, all stakeholders will be influenced. If a CSR strategy is implemented in the company, the company is aware of the social and environmental consequences of production. The performances related to the CSR strategy influence the investors’ perception of the company, but also the perception by customers and environmental foundations are influenced (Clarkson et al., 2010). CSR strategy is a social concept, which incorporates different stakeholders. The whole society is indirectly affected by the CSR performances of the company. The impact contains the whole society, this makes it interesting to investigate the impact of CSR reporting.Prior studies examined the environmental and social performances of companies, for example the study of Clarkson, Li and Richardson (2004). This study distinguished two types of environmental performers: good environmental performers and poor environmental performers. Companies that can be considered as good environmental performers are exceeding current environmental regulations. On the other hand, poor environmental performers are companies that are meeting the minimum level of environmental regulations. Clarkson et al. (2004) argue that companies that are good performers benefit from the exceeding. But poor performers do not receive benefits related to environmental regulations, but instead face obligations to incur future expenditures. When a company reaches a minimum level of environmental regulations, it can be expected that future pollution and litigation costs increases. Clarkson et al. (2004) described, that poor environmental performance companies have lower environmental benefits. But due to the lower level of regulations they also have lower costs. Clarkson et al. (2004) described only the financial effects of good and poor environmental performances, but social effects are also expectable. The disclosure of a CSR report, which incorporates environmental and social performances, provides insight in the distinction between good and poor performances. The good and poor CSR reporting performers are observable by the quality of the CSR report. When a company provides a high quality CSR report, CSR reporting guidelines are followed in a proper way. CSR reports that meet lower levels of CSR quality are to a lesser degree in conformity with CSR reporting guidelines. In the remainder of this chapter the above described expectation of CSR reporting quality on reputation is discussed.The agency problem will be reduced after the CSR report disclosure. When a CSR report is of higher quality, the report takes more guidelines into consideration. Guidelines of reporting increase the transparency, so it can be expected that higher quality CSR reports result in a lower agency problem. Stakeholders observe more transparent information, to monitor management’s CSR behavior. The information asymmetry reduces. (Healy et al., 2000) The company makes decisions in favor of society, to obtain power from society (Rodriguez and LeMaster, 2007). Communicating CSR information, is in conformity with the legitimacy theory (Lightstone et al. (2008) and Lui et al. (2008), Cho et al. (2007) and Dowling et al. (1975). The public is more satisfied when it is provided with more reliable information. The quality of the CSR information is represented by the quality of the CSR report. Based on the legitimacy theory it can be expected that managers wants to develop a high quality CSR report, to obtain more legitimacy from the public. The stakeholder theory implies that companies disclose a CSR report based on the pressure of stakeholders (Rodriguez and LeMaster (2007) and Huang et al. (2010). The public will benefit from this, when management acts more in conformity with the CSR desires of the public. The public will value this in a positive way, which leads to an increase of the reputation (Clarkson et al., 2010). The public values CSR positively (Formbrund and Shanley, 1990). This implies that the quality of the information provided in a CSR report will also increase the reputation. The agency problem, information asymmetry, the legitimacy theory and the stakeholder theory expects that from CSR reporting the reputation increases. This expectation results in the following hypothesis: CSR reporting quality increases the reputation of a company. Reputation is the impression the public has about the company. CSR reporting has an impact on the reputation, this impact can be positive or negative, dependent on the published information. Since CSR reports provide information about past CSR performances, an influence on reputation is expected. Argued by Clarkson et al. (2004) and Patten (2002) public sentiment can be influenced in a positive way due to environmental performances. 4.2 Difference compared to previous studiesThe public sentiment in relation to CSR has already been investigated in the studies of Brown et al. (1997), Ittner et al. (1998) and Clarkson et al. (2010). They all found a positive relation between CSR disclosure and the public sentiment. But there are several differences between the design of these prior studies and the design of this paper. One of the differences is located in the used proxies of measuring reputation. Clarkson (2010) used the Janis-Fadner coefficient. This coefficient uses positive and negative environmentally related articles that are published about the company. This is not the most accurate tool for measuring public sentiment. These environmentally related articles are published by different media and read by a different public. Clarkson et al. (2004) suggest that, to formalize a judgment the public will use the positively and negatively related articles. The judgment of the whole public together is reflected in the public sentiment. Clarkson et al. (2004) measure this public sentiment by a formula that incorporates the number of negatively and positively related articles. So they suggest that companies, which have exactly the same number of positive and negative environmental articles also, have the same public sentiment. With this formula it is suggested that all articles have the same impact on public sentiment. This is not an accurate representation, since articles face different levels of impact on the public sentiment. For instance, if a negative environmental article is published in countries most popular newspaper, a bigger influence on public sentiment is expected, in comparison with the article being published in a small local newspaper. Therefore it is not accurate to suggest that all negative (or positive) articles have the same influence on public sentiment.In the study of Brown et al. (1997) a field study is used for analyzing the public sentiment. In this field study, potential consumers are asked to evaluate products on the CSR performances. The products are separate analyzed. It is very difficult to evaluate all existing single products. When companies provide a service, not a real tangible product is delivered, analyzing an intangible product is more difficult. Also it can be expected that companies that deliver products or services to specific industries and not to consumers, are not well evaluated by public. The public does not own enough detailed information to evaluate the products or services in an accurate way. When the public is evaluating these unknown companies, the evaluation is based more on the image of the company. For example, when the public only knows the logo of the company, the evaluation is mainly based on the associations with this logo. The study of Brown et al. (1997) focuses more on the marketing side. The evaluating of products by the public will be based on all information held by the public of the particular product. This information can be influenced by marketing. Marketing is an important tool, it has the purpose to influence the information known by the public about products in a positive way. Therefore evaluating products to observe the impact of CSR is not a right method. Consumers are not able to evaluate correctly the companies CSR performances based on product evaluation.Ittner et al. (1998) used the proxy customer satisfaction as a tool for measuring the non financial performances of the company. In their study the relation between financial performance and non financial performances are investigated. When customer satisfaction is taken as a proxy for non financial performances, they assume a perfect relation between customer satisfaction and non financial performances. In this study the level of customers’ satisfaction is build upon three different variables, the customer, business unit and firm level data. It cannot be assumed that customers’ satisfaction has a perfect relationship with non financial performances. Because customers satisfaction is dependent on many criteria. Therefore, the study of Ittner et al. does not make the right assumption about the perfect relation between customer satisfaction and non financial performances. The used proxies in previous studies are not the only reason for developing this paper for examining the relationship between reputation and CSR reporting. Also the focus of the studies differs. In the study of Clarkson et al. (2010) it is investigated whether a positive relation between public sentiment and environmental disclosures exists. For this study they researched companies that are operating in the five most polluting industries in the US. The outcome was a positive relation between the Janis-Fadner coefficient and the CSR disclosures. The polluting industries do not represent all companies, so the outcomes are only useful for the polluting industries. In the developed hypothesis no restriction of companies or industries has been made, so the outcomes will be representative for all companies and industries. 4.3 Development of sub hypothesesThe relation between the reputation and CSR reporting is influenced by many factors. A selection of the expected influencing criteria has been made. In this paragraph the criteria are described and the expected influences are provided.The size of a company is a factor that has an impact on the reputation and the CSR report. Larger companies face more public attention, through their larger size they are well known. More stakeholders are involved in the processes of the company, the company reaches a larger public. When more stakeholders are involved in the company, the company is facing more pressure to pursue a good reputation. If the company implements a CSR strategy, the company is aware of negative environmental and social influences of their production. A CSR report provides more transparency in the CSR performances of a company. From the agency theory and information asymmetry it can be expected that public values the CSR reporting positively. This results in the expectation that CSR reporting improves reputation. The public is more aware of the impact of larger companies (Huang et al., 2010) these larger companies have more incentives to provide a complete and high quality CSR report. Besides the role that size influences company’s public attention, size also plays an impact on the financial resources. In the study of Dhaliwal et al. (2010, A) it is argued that larger companies have more financial resources available to develop a CSR report. This implies that larger companies are more willing to invest in CSR activities and disclose these performances in a CSR report. So larger companies will achieve a reduction in information asymmetry and the agency problem which results in an increase of the reputation. The following sub hypothesis can be inferred. The relationship between CSR reporting quality and the reputation is positively influenced by larger companies.Financial performances of companies are also an important element in the relation between CSR reporting and the reputation. Argued by Dhaliwal et al. (2010, A) companies with better financial performances have more financial resources available to implement a CSR strategy and disclose a CSR report. In the study of Dhaliwal et al. (2010, A) the impact of the cost of equity on CSR reporting was studied. In this study evidence is found that cost of equity decreased after the first CSR disclosure. So a financial advantage due to the CSR disclosure was observable, investors valued the CSR disclosure positively. The CSR disclosing companies face a higher amount of new equity, then in comparison with not disclosing CSR companies. Also it can be expected when companies have a good financial performances, they have an improved influences on the society. Good financial performers provide benefits for the society. For example, the companies pay more taxes and also have more employees. This leads to an increase in reputation. The effect of financial performances on CSR reporting and reputation is incorporated in the following sub hypothesis:The relationship between CSR reporting quality and the reputation is positively influenced by better financial performances.The reputation of companies is dependent on the level of customer satisfaction. Companies are able to influence the satisfaction levels. Companies have the opportunity to provide additional services when selling products or services. When these products or services are delivered with additional services, customers appreciate this, resulting in a higher customer satisfaction. An improved reputation follows due to the provided additional services. A company can differentiate from other companies when building a good reputation. If a furniture shop provides a well structured delivery service, the reputation of the company improves. Due to this delivery service the company can differentiate from other furniture shops. The expenses that represent the services which are provided to the sold products and services are the selling general and administrative (SG&A) expenses. Argued by Palepu, Healy and Peek, (2010) the amount of the SG&A expenses will represent the level of services which lead to customer satisfaction. When a bigger amount of SG&A expenses are provided per amount of sales, reputation faces an improvement. This expectation is expressed in the following sub hypothesis:The relationship between CSR reporting quality and the reputation is positively influenced by higher general selling and administrative expenses to sales.Awareness of social and environmental issues is dependent on many circumstances. For example, the economical situation of countries is important for CSR related activities. When in countries a good economical environment exist, more financial resources are available to invest in CSR strategy. This will influence the reputation. Also many other country specific circumstances are influencing the decision making process, from choosing a CSR strategy to disclosing a CSR report. When the government stimulates companies to implement CSR strategy and to provide information about the CSR performances in a CSR report, more companies have incentives for the implementation of CSR strategy and the disclosure of a CSR report. This country specific circumstance is observed in the study of Dhaliwal et al. (2010, A). In this study evidence is found that the power of the relation between CSR disclosures and forecast accuracy is different among countries. Countries with more stakeholder orientation are facing a stronger relation. Countries can have a hidden information culture, in these countries significantly less information is provided to stakeholders. These country specific factors will influence the CSR report disclosure behavior, which finally have an influence on the public reputation. The following sub hypothesis will show this expectation:The relationship between CSR reporting quality and the reputation is influenced by regions of companies.Not only country specific circumstances have influences on the CSR reporting behavior, but also industry circumstances have influences on the CSR report behavior. In the study of Richardson et al. (2010) the control variable Toxic Release Inventory (TRI) emissions is included. The study investigates companies from the five most polluting industries. This TRI variable is seen as a measurement for relative environmental performance and is used to rank the companies within an industry. Evidence is found that the TRI emission has a negative impact on firm value and a positive impact on the cost of equity. These findings are explained by the possible future liability claims, which lower the firm value. It can be assumed that TRI emissions differ among industries. For example, financial services companies face a lower level of toxic emissions than industrial companies. Companies that operate in an industry with high levels of TRI emissions are harming the environment. This is valued in a negative way by the public and results in a lower reputation. From the legitimacy theory it is expected, that companies operating in high emissions industries have a high awareness of social and environmental issues. The expected effect on reputation and the awareness of CSR reporting among industries implies that the investigated relationship will differ. This expectation is implied in the following sub hypothesis:The relationship between CSR reporting quality and reputation is influenced by industries of companies.4.4 Summary hypothesis developmentThis chapter introduced the main hypothesis of this paper. The expectation is that a CSR report positively influences reputation. This reputation is dependent on the available information of the company. Due to a CSR report more information becomes available and this improves the reputation. Some previous studies already observed a positive relation between these two variables, but used a less accurate proxy for the reputation or measured the impact of only particular industries.Chapter 5 MethodologyIn the previous chapter the hypotheses are developed, these developed hypotheses are tested in the empirical research. In this chapter the design of the empirical research is described. In the first paragraph the development of the sample is provided. In the second paragraph the methodology for investigating the relationship between reputation and CSR report disclosure follows. The third paragraph provides an explanation of the used data.5.1 Sample In this paper the impact of a CSR report on the reputation of companies is tested, related data about the reputation and CSR reporting is used. Two important databases are used to obtain the relevant data of the reputation and the CSR report. In this paragraph, first the used databases are discussed and thereafter the relation with the observed data is explained. The final part of this paragraph explains the sample size and the time period. 5.1.1 Used databasesFor observing the reputation, the reputation index is used. The reputation index ranks companies based on their reputation. The reputation index formulates public attitude to companies. For obtaining this reputation index, the Fortune 500 database is used. The reputation index and components of the reputation index of the Fortune 500 are used in the studies of Griffin and Mahon (1997), Brown and Perry (1995) Conine and Madden (1987), Formbrund and Shanley (1990) and Mc Guire, Schneeweis and Branch (1998). The Fortune 500 database is a service provided by CNN Fortune & Money. Each year, the Fortune database ranks companies on reputation. CNN Fortune & Money formulate a ranking list which is called Most Admired Companies and is based on the reputation index. The reputation index is composed on nine key components which influence the reputation index. Directors, security analysts and executives define the reputation of companies on the nine different components. Results of respondents are taken together, which leads to the final reputation index of the Fortune 500.The second important database is the Global Reporting Initiative (GRI) Report List. This database asses the level of a CSR report. The GRI Report List is provided by the GRI. In this list the level of CSR reports are provided, this level is based on the degree of applying the guidelines of the GRI. The GRI calls this level the application level. In the application level, no judgments about the CSR performances are included. Companies are free to decide if their CSR report is included in the GRI Report List. This implies that it is not certain that companies which are not included in the GRI Report List are not disclosing a CSR report. 5.1.2 Development sample and time period In the Fortune database, reputation indexes of more than 700 companies over the time period 2006 up till and including 2011 are incorporated. The GRI Report List includes more than 2700 disclosed CSR reports, in the timeframe from 1999 up till and including 2010. Of companies which are included in the sample, the two core variables reputation index and level of CSR report should be known. Therefore the companies, which occurred in both databases, are included in the sample. But also the data should be matched with the expected time lag of the reaction of the CSR reporting on the reputation index. This implies that the reputation of next year is affected by the CSR report of prior year. As a consequence the reputation index of 2006 should be matched with the CSR report information of 2005. The Fortune database starts in 2006, which matches with the CSR reporting quality of year 2005. Because a very limited amount of companies have registered their CSR report of 2010 on the GRI Report List, year 2010 is not incorporated in the time period. Finally this leads to studying the relation between CSR reporting quality and reputation between 2005 and 2009.After finding common companies of the Fortune database and the GRI Report List, the sample included almost 300 companies. In the sample only companies should be included which match the CSR report level and next year reputation index, there for only 250 companies were finally incorporated. This sample is further declined by companies of which the control variables are not observable, for example, due to the acquisition of Air France-KLM the financial control variables of this company are not observable in the selected time period. For a limited amount of companies (32) data about level of CSR report and reputation index are known for the whole selected time period, but all companies are incorporated in the sample. For the years 2005 up till and including 2009, 85, 106, 138, 167 and 175 respectively data points are included, which have a total sample size of 671 data points. These data points are companies of which all variables are known, including the control variables. In appendix A the total sample is showed. 5.2 Methodology of statistical researchTo investigate the relation between CSR reporting quality and the reputation linear regression analysis is used. When making use of a regression formula, the existence of a relationship is observable. In the hypothesis is predicted, that CSR reporting quality influences the reputation. Therefore in the developed regression model, the reputation is predicted by the CSR reporting level. This is called the basic model. This basic model is shown below: REPt = α + β1CSRt-1 + εFor providing more evidence on the regression, control variables should be included. The control variables represent the sub hypotheses. The use of control variables is important, because it is not plausible that CSR reporting is the single variable that has influence on the relation between CSR reporting and reputation. When a single control variable is included in the basic model the influence of the control variable is tested. Finally all control variables are included in the regression model. This leads to the following regression model: REPt = α + β1CSRt-1 + β2SIZEt-1 + β3FPERt-1 + β4SGASt-1 + β5REG + β6IND + εREP Reputation of companies (observed from the reputation index provided by the Fortune 500)CSRCSR report level (observed from the application level provided by the GRI Report List)SIZESize of the company, based on the total sales (observed from Thomson One Banker)FPERFinancial performances, based on the return of assets (observed from Thomson One Banker)SGASSelling general and administrative expenses related to sales (observed from Thomson One Banker)REGRegion where the company is from (observed from the GRI Report List)INDIndustry wherein the company operates (observed from Thomson One Banker)A linear regression is used to test the expected relationships. The dependent variable REP, is predicted on basis of the independent variables. In the regression model qualitative variables are included. The influence of the qualitative variables in general are measured by making use of the above described method, this method is called Method 1. The variable CSR reporting level is a qualitative variable. In method 1 is measured if CSR reporting level has an influence on the reputation. Besides the general influence of the qualitative variables, it is also interesting to investigate if a specific CSR report level has an influence on the reputation. These influences of the specific categories in a qualitative variable are measured with method 2. In method 2 dummy variables are used. The amount of dummies which represent the qualitative variable is always one less the amount of categories in the qualitative variable. The coefficient of the dummy variables reflects the difference in impact with the reference variable. This reference variable is the variable which is excluded in a dummy variable. The influence of the category wherefore no dummy is developed, contains the constant variable. Besides a regression a correlation is used to investigate the relation between two variables. The correlation represents the relation between two variables in both ways. These both ways implies that the influence of the dependent variable is also investigated on the independent variables. The outcome of the correlation, examines if two variables can predict each other. A correlation with a categorical variable is impossible, therefore the One Way Anova test will be used to investigate the relation. 5.3 DataIn the previous paragraph the regression formula was introduced. In this paragraph the developed variables are introduced. Also the expectations about the influences of these variables in the regression formula are described. The REP is stated for the reputation, this is the dependent variable of the model. The reputation index is used as a proxy for reputation. In the first paragraph of this chapter already a description has been given about database and the development of this reputation index. The taken reputation which is corresponding with the CSR disclosure is one year later than the CSR disclosure. The variable CSR is stated for the quality level of a CSR report. As has already been discussed, this level is observable in the GRI Report List. Nine different application levels exist in the GRI Report List, the A+, A, B+, B, C+, C, U, IA and CI. The levels A+, A, B+, B, C+, C and U are application levels for reports that conform the GRI G3 Guidelines, which were launched in 2006. The A+ rating is the highest rating and the C is the lowest rating. CSR reports obtain the U rating when the quality of a CSR report is not defined. The quality levels IA and CI are based on the GRI G2 Guidelines from 2002. The IA level is for CSR reports which are ‘In Accordance’. The CI level is applicable for CSR reports which are ‘Content Index Only’ checked. Since the level of a CSR report is a qualitative variable, the CSR variable in the regression formula is tested with a dummy variable. The control variable SIZE is stated for the size of the company. The size of a company is measured with the logarithm of the total sales. Total sales of a particular year have an influence on the reputation of the following year. In the study of Dhaliwal et al. (2010, A) the total sales are also used to measure the size of a company. The logarithm has the result that the total sales are normal distributed, which is an important condition of regression. The total sales are observed from the financial database Thomson One Banker. In the hypotheses development is already stated that larger companies are facing more public attention. This will result in more pressure to disclose a well qualified CSR report. Therefore a positive sign of the size coefficient is expected. Financial performances are also included in the regression, this is represented by the FPER variable. To measure the financial performances return on assets (ROA) is used as a proxy. This variable reflects the financial performances compared with an item to reflect the company size. The financial performances should be controlled for size, because larger companies have better absolute financial performances. The ROA is a relative financial performance measurement, it measures how beneficial the assets of the company are. The financial performances of a particular year influence the reputation of the next following year. The ROA is observed from Thomson One Banker. It is to be Expected that good financial performances have a positive influence on the reputation, other variables can be used to measure the financial performances. A positive sign of the financial performance coefficient is expected.The SGAS variable is stated for the percentage of selling general and administrative expenses to sales. In the hypotheses development was predicted that selling general and administrative expenses expectantly have a positive influence on the reputation. The SGAS control variable is stated for the reputation expenses. This variable is controlled for the sales, because, there can be assumed that the percentage of selling general and administrative expenses relative to sales makes a difference in the level of reputation. Also the variable has influence on the reputation of next year. The selling generals and administrative expenses to sales are observed from Thomson One Banker. For this variable a positive sign of the coefficient is expected. The hypothesis about the influence of countries on the relation between reputation and CSR reports is tested with the variable REG. The companies in the sample come from six different regions and 26 countries. The regions are Europe, Northern America, Latin America, Africa, Asia and Oceania. The countries are clustered in this six regions, the sample is too small to investigate the influence of each individual country. This variable is also a categorical variable, in the regression formula dummies are used to investigate the influence of a specific region relative to another region. For this categorical variable no prediction can be made on the direction of the influence on the reputation and CSR reporting. At least, expectations can be made that there are differences observable among the different regions.The final variable IND contains the hypothesis of the differences of influence between industries. A separation of the types of industry is made based on the Industry Classification Benchmark (ICB). This benchmark divides sectors in ten industries. These industries are: Oil & Gas, Basic Materials, Industrials, Consumer Goods, Health Care, Consumer Services, Telecommunications, Utilities, Financials and Technology. This categorical variable is in the second method represented with dummies. No expectations can be made about the influence of specific industries on the reputation. None the less differences in the influence on the reputation among industries are expectable.Chapter 6 Analytical ResultsIn the previous chapter the data and the method for the statistical analysis is explained. In this chapter the results of the analysis are described. First, a description is provided about the data. The changes in data are summarized during the time period. In the second paragraph, the results of the statistical analysis are described. The statistical analysis contains the correlation and the regression between CSR reporting and reputation.6.1 Descriptive analysisThe whole sample over the entire time period includes 669 data points. In this paper there is no focus on the results of the entire period, because the separate years and their data are more important. The individual years have different amounts of data points, year 2005 has 85 matching data points, where the matching data points in 2009 increased to 175. When all years are taken together, the results are biased due to years which provide more data. Also some companies have matching data for the whole time period (for example, Henkel and Volvo). But other companies have matching data for a few selected years in the time period (for example, Reckitt Benckiser Group and Bridgestone). When there is focus on the entire time period, the results are biased because companies which have all data of the entire time period are more represented in the sample. Therefore the results of the individual five years are explained. This paragraph has the purpose to provide a description of the data in the five different years. In table 6 the distribution of the different CSR report levels over the years are summarized. This table shows IA and CI level do not appear in the years after 2006. This is in conformity with the IA and CI level assessed according to the G2 guidelines. The remainder levels, A+ up till U are assessed according to the G3 guidelines, which are initiated in 2006. After 2006 the CSR reports are judged on the level of the G3 guidelines, this explains the lack of IA and CI level reports after 2006. Before 2006, some CSR reports (18,8% for 2005 and 56,6% for 2006) already took the new G3 guidelines into consideration.Table 6: Distribution of the CSR reporting levels from 2005 till 2009. When the years are taken into consideration which includes reports that are judged on the G3 guidelines, no increase in CSR report level is observed. The percentages of reports that meet the A+ level indeed increases over time (from 13,0% in 2007 till 19,4% in 2009), also the B level increases (from 12,3% in 2007 till 18,9% in 2009). From these single level increases it is not possible to conclude a general increase in CSR report level. In this table it is not possible to assess, of which CSR quality level the increases in A+ and B levels come from. None the less it is possible to conclude, that the percentage of reports of which no level is defined (the U level) decrease over time. When the reports which are conform the G3 guidelines are taken into consideration over the years 2005 and 2006, the percentage of undefined report level is 53,3% for 2005 and 37,5% for 2006. After 2006 the amount of the U level reports increase once in 2007. A presumption can be made that the undefined level reports in 2007 also include CSR reports which were still in conformity with the G2 guidelines. Since no additional information is given from the GRI Report List on which characteristics the U quality level is judged, no statements can be made to assess the increase in 2007. Table 6 provides the participation of the different CSR report levels in the five selected years. It is important to show the distribution of the different variables over the CSR reporting levels. This is shown in Appendix B. The distribution of CSR reporting levels over the control variables, (size, financial performances, selling general & administrative expenses to sales, region and industry) is shown. The qualitative variables region and industry are divided in the predetermined regions and industries, where the continuously variables size, financial performances and selling general & administrative expenses to sales are divided into categories. In the distribution the average, medium and maximum amounts are considered. In the appendix is shown how much of the CSR reports, which meet a certain quality level, meet a specific category of the remainder variables. This implies 1,2% of the companies included in the samples have a CSR report in 2005 which meet the A+ level and have sales less than 50 billion Dollar. Appendix B indicates that companies which disclose a CSR report conform the A+ level, mainly come from Europe. With this assumption a note should be made, most included companies come from Europe, Northern America follows directly on Europe, Asia is in third place, thereafter Latin America, Africa and Oceania follow. When taking this distribution into account for all years, Europe scores better than Northern America on the A+ level. From this appendix can be obtained that Northern America provides more CSR reports which meets the levels B, C+, C and U than Europe does.Remarkable is that a relative high percentage of the A+ reports meet the lowest category of the selling general & administrative expenses to sales. In 2009 36,1 % of the companies in the sample have expenses to selling general & administrative to sales of less than 15%. Also in 2009 19,4% of the CSR reports meet the requirements of the A+ level. Therefore it can be expected that approximately 7% (36,1*19,4%) of the CSR reports in 2009 meet the A+ level and the lowest selling general & administrative expenses to sales category. In reality this is 9,7%, in comparison with the expectation of 7% this is a difference of almost 40% ((9,7-7)/7*100). This implies that the CSR reports which have a low selling general & administrative expenses to sales and the A+ level is relative high. The above described example of the relative high existence of A+ level CSR reports in the lowest level of the selling general & administrative expenses to sales in 2009, is also applicable for the years 2005 till 2008. In the distribution of all CSR reporting companies over the different industries no big differences occur. Also the distribution of the quality level of CSR reports over the different industries, are no big changes in time. This also applies for the variables size and financial performances.6.2 Statistical ResultsTo answer the main hypothesis about the impact of the CSR reporting on the reputation, the methodology which is developed in chapter 5 shall be followed. This is done by using the statistical program SPSS. In this paragraph the results of the SPSS output are presented. All these outputs investigate whether the expected relationship occurs. First the results of the correlation are explained, thereafter the results of the regression.6.2.1 CorrelationThe correlation is a measurement to observe the relation between two variables. For measuring the relation between continuously variables, the correlation should be used. The correlation cannot be used if one or two of the variables are qualitative variables, another measurement should be used to investigate this relation. First the correlation between the continuously variables, (reputation, size, financial performances and general selling & administrative expenses to sales) is shown. Thereafter the relation between the qualitative variables (CSR report level, region and industry) on the reputation is discussed. To measure the relation between a continuously variable and qualitative variable One Way Anova test is used.The correlation can be positive or negative, dependent of the sign of the coefficient. A positive coefficient implies a positive relationship and a negative coefficient implies a negative relationship. The degree of connection is dependent of the value of the coefficient. When the coefficient is zero, no connection is observable. A value of minus one or plus one, implies a perfect relationship.In this paper a significant level of 10% is used, this implies a correlation coefficient with p-values above 10% no connection is demonstrated. In table7 below, the correlation coefficients are provided between the reputation, size, financial performances and the general selling & administrative expenses to sales. In this table the significant relations are divided into the degree of significantly. When a coefficient is displayed with 3 stars, the coefficients is also significant at a lower level (p < 0,01).In this paper the effects of different variables on the reputation are examined, therefore the significant results of the first column are explained. Table 7: Correlation between the continuously variables?Rep.SizeFin. PerfSG&A to SalesRep.200510,236**0,237**-0,141200610,146**0,263***0,082200710,1640,129*-0,057200810,0810,432***-0,024200910,0670,365***-0,011Size20050,236**1-0,292**-0,15220060,146**1-0,066-0,11120070,1641-0,057-0,248***20080,08110,017-0,263***20090,0671-0,078-0,039Fin. Perf.20050,237**-0,292**1-0,05420060,263***-0,06610,15520070,129*-0,05710,04820080,432***0,01710,148**20090,365***-0,07810,177**SG&A to Sales2005-0,141-0,152-0,054120060,082-0,1110,15512007-0,057-0,248***0,04812008-0,024-0,263***0,148**12009-0,011-0,0390,177**1*** Significant at p < 0,01** Significant at p < 0,05* Significant at p < 0,1This table demonstrates the years 2005 and 2006 size and reputation have a positive relation. Larger companies have higher reputation indices, which are conform the legitimacy and stakeholder theory. Larger companies face more public attention and more pressure to reach a high reputation. The connection is not very strong, but still some connection between the two variables exists in the years 2005 and 2006. Between financial performances and reputation a relation is observable for all years. This relation is positive. This relation implies financial performances of a company have a positive influence on reputation. When a company has better financial results the reputation index of this company improves. The value of the relation improves over time from 0,236 till 0,365. This implies an increase of the relation between financial performances and reputation over time. Financial performances become stronger criteria for reputation. The above discussed correlation coefficients measure the connection between two continuous variables. In the designed methodology qualitative variables are also included. To investigate the connection between reputation (continuously) and CSR report level, region and industry (all qualitative variables) the One Way Anova test should be used. From the results of One Way Anova no statements are observable of the direction of the relation.When testing the relation between reputation, CSR report level, region and industry, the relation between the qualitative variables in common on the reputation, is taken into consideration. For the region variable first the relation between region in common and reputation is tested. The outcomes show whether region has an influence on reputation. Thereafter is tested if the different categories of the qualitative variables have a connection with the reputation. This implies for the region variable, whether a specific region has a relation with the reputation. In table 8 the outcomes are shown. The provided data of the test are the p-values. From the p-value a relation can be observed between the reputation and the variable. An actual relation is observed when the p-value reaches the significant value of 10%. The p-values marked with a * shows a connection between the two variables. Also in this table a distinction is made in the levels of significance.Table 8: p-value of the outcomes of the One Way Anova test. ?20052006200720082009CSR Report LevelCommon0,2110,6800,065*0,9690,714A+0,5680,084*0,063*0,5620,252A-0,7600,1740,9790,285B+0,1330,8370,9480,6590,897B0,7540,3080,3930,8920,983C+0,066*0,5040,6870,6160,872C-0,6310,086*0,9810,917U0,1850,1590,019**0,3070,129IA0,2890,716---CI0,9430,912---RegionCommon0,005**0,000***0,6300,001***0,000***Africa0,6390,8070,6740,875-Asia0,2030,002**0,9230,000***0,000***Europe0,013**0,013**0,092*0,3010,333Latin America0,6330,9040,7470,5980,303Northern America0,000***0,000***0,1670,001***0,000***Oceania-----IndustryCommon0,1580,3590,013**0,3200,017**Oil & Gas0,4360,4260,1860,013**0,009***Basic Materials0,1860,1300,008***0,4880,861Industrials0,6200,080*0,4100,4920,805Consumer Goods0,1070,3160,5690,3810,127Health Care0,8090,7280,9690,4630,307Consumer Services0,8160,8870,7520,8720,088*Telecommunications0,3360,8580,017**0,9630,851Utilities0,5080,9300,5050,6460,190Financials0,047**0,3600,8580,1690,064*Technology0,037**0,4260,010**0,3290,057**** Significant at p < 0,01** Significant at p < 0,05* Significant at p < 0,1Observed from the outcomes presented in table 8, evidence is found for a relation between CSR report level and the reputation index of 2007. This is in conformity with the agency theory, legitimacy theory and the stakeholder theory. The level of the CSR report has an influence on the reputation. When the CSR report is conform specific guidelines the stakeholders are provided to more reliable information. This decreases the agency problem and the information asymmetry, which is valued by the stakeholders in an increase of the reputation. The relation between CSR report level and reputation is also caused by the legitimacy theory. Due to a disclosure of a CSR report, the company achieves legitimacy of the public, this results in an improvement of reputation. The relation between CSR report level and reputation is in conformity with the stakeholder theory. The environmental and social related stakeholders of a company prefer a high CSR quality report. Management will react on this demand and an increase of the company’s reputation follows. From this table a connection between some of the specific CSR reporting levels and the reputation is observable. The A+ level has in 2006 and 2007 a significant relation with the reputation. In 2005 the C+ level also has a relation with the reputation. Besides the A+ level in 2007, the C and U level also have a significant relation on the reputation. With the One Way Anova test is investigated that a relation between region and reputation is observable in the years 2005, 2006, 2008 and 2009. This confirms different circumstances are influencing the reputation. Circumstances are different among certain regions, and have an influence on the reputation, when the relation of the different regions on the reputation is taken into account. The outcomes show Asia has a significant connection in 2006, 2008 and 2009. For Europe a relation is investigated in the years from 2005 till 2007. The relation between Northern America and the reputation is significant for all years except 2007. For Oceania too few data was available to examine the relation with reputation.A relation between industry and reputation is observable in 2007 and 2009. This implies the difference in industry has a relation with the reputation. The oil & gas industry has a relation with the reputation in 2008 and 2009. The basic materials and the telecommunications industries both have just in 2007 a relation with reputation. Industrials only have a connection with reputation in 2006. Consumer services have a relation with reputation in the year 2009. A relation between the financial industry and reputation is observable in the years 2005 and 2009. For technology the relation with the reputation is investigated in the years 2005, 2007 and 2009. The industries consumer goods, health care and utilities have no relation with reputation. 6.2.2 RegressionThe regression results are based on the developed methodology of chapter 5. The regression analysis is provided by testing the basic and extensive model. In the extensive model all control variables are included. In the basic model only the CSR report level is used as independent variable to predict reputation. To test the sub hypotheses, the represented control variables should be implemented in the basic model. By including the control variables in the basic model the influence of the control variables on the relationship between CSR report level and reputation is tested. In the extensive model all control variables are included, in this model is tested if all variables have an impact. If in the basic model an influence of the control variable on the relationship is proved, this relation is also proved in de extensive model.First, the outcomes of method 1 are described. This method takes the general influence of the qualitative variables into consideration. The second method takes the influence of the different categories inside the qualitative variables into considerations. The second method is discussed after the first method. Method 1The first regression results are shown in table 9. In this regression the influences of variables on the reputation is shown. For the qualitative variables: CSR report level, region and industry in this regression, no separation has been made on the different categories inside the qualitative variables. This implies that only the influence of CSR report level, region and industry are tested. In this developed regression no statements can be made on the influence of one specific CSR report level, region or industry. Table 9: Regression results (dependent variable: reputation)?20052006200720082009Constant3,114**4,332***4,020***3,392***3,628***CSR Report Level0,0180,0370,0580,040-0,029Size0,556**0,2150,397**0,357**0,389**Fin. Performances0,051**0,0240,0150,047***0,037***SG&A to Sales-0,181-0,331-0,356-0,160-0,176Region0,1670,324***0,160*0,298***0,235***Industry0,0220,0210,0700,0000,021?Anova0,0120,0010,0120,0000,000R squared0,2450,2560,1370,3590,226VIF1,2111,2541,1691,1331,172*** Significant at p < 0,01** Significant at p < 0,05* Significant at p < 0,1To examine if a variable has influence on the reputation, the p-value should be less than the significant value of 10%. When the beta (coefficient) has a negative value, the variable has a negative value on the reputation and a positive value of beta investigates a positive influence. The anova value of each year is below the significant level. This proves the regression model measures the dependent variable in a good way. The anova value is a score for the entire model, not for the prediction value of one variable. The R squared value represents the variance in the reputation index which can be explained by the selected variables. In this model the value varies from 0,137 (2007) till 0,359 (2008). The R squared values are relatively high, but there is space for improve improvement, there might be other variables (which are not included in the model) which also have influence on the reputation index. In the regression results of all years no collinearity in the data exists. The VIF score is between 0,2 and 10 this implies the variables which are included have no correlation with each other.Resulting from this model, CSR reporting level has no influences on reputation. The p-values of the years 2005 up till and including 2009 are higher than the significant level of 10%. It is proved that the beta of the CSR reporting level has no predictive value on the reputation. When the basic model is tested for 2009, a significant positive influence of the CSR quality level on the reputation is observed. The R squared is very low in this model, therefore can be assumed that other variables should be included to provide a better prediction of the reputation. The positive influence of CSR reporting level on the reputation for 2009 is in conformity with the agency theory, information asymmetry, legitimacy theory and stakeholder theory. For all years, except 2006, the size coefficient has a significant influence on the reputation. And when the size variable is included in the basic model an influence on the reputation is proven in the years 2005 and 2006. In the model, where size is included in the basic model, the R squared increases, which implies a better prediction of the reputation index. These results prove size has an influence on the relationship between the CSR report level and the reputation. Larger companies are dealing with a more public attention. A CSR report can be useful for satisfying the public of the large company. The CSR report provides more transparency, and therefore the information asymmetry and the agency problem reduce. The public appreciate this reduction, which results in an improvement of the reputation. This is conforming the legitimacy and stakeholder theory. In this model the degree of the size coefficient varies for 0,215 up till 0,556.In the years 2005, 2008 and 2009 a significant influence of the financial performances is observable. This influence is positive and very small. When the financial performance variable is included in the basic model, a positive influence is observable too. The R squared increases as well, which indicates financial performance has a positive influence on the relation between CSR report level and reputation. This confirms the expectation that better financial performing companies have more financial resources available to disclose a well qualified CSR report. Also better financial performers having more influences on the society. The bigger influence on the society of better financial performers results in an improvement of reputation.From this regression is proven that general selling & administrative expenses to sales have no influences on reputation. When this control variable is included in the basic model, no influence is obtained. The general selling & administrative expenses to sales have no impact on the relationship between CSR reporting level and reputation.From the outcomes of method 1, is proven that regions have an influence on reputation, except for 2005 this influence is observable. When regions are included in the basic model an influence on the relationship between CSR report level and reputation is observable. For the year 2007 this influence on the relationship is not obtained. The influences of the region is positive and of middle degree. Overall the findings proved an influence of region on the relation between CSR report level and reputation. In table 9, no influence of industry on the reputation is observable, for all years. When the basic model is extended with the industry control variable no influences is proven. There is no influence of the industry on the relation between CSR report level and the reputation.Method 2Above method 1 is described, in this method the general influences of the qualitative variables are tested. But with method 1 the influences of the specific categories (which are inside the qualitative variable) are not observable, therefore method 2 should be used. In this model dummy variables are included. The dummy variables measure the difference in influence with the reference variable. The differences in influences of the dummy variables are observed from the p-value. The degree of different influences is observable from the coefficient. The IA level is the reference variable for the CSR report levels, for the regions Europe is the reference variable and for the industry the consumer goods is the reference variable.On the next page table 10 is displayed, in this table the regression results over 2005 up till and including 2009 are shown. In this method the single variables are included in the basic model. When including one variable in the basic model the influence on the relationship of this variable is directly tested. The outcomes of the influence of the variable, observed from the basic model, does not differ from the influence of the variable observed from the model where all variables are included. Table 10: Regression results (Categorical variables are included)?20052006200720082009Constant2,1394,448***4,875***5,039***3,130***CSR Report LevelA+0,396-0,6090,375-0,171-0,287A--0,1580,8640,177-0,049B+-0,574-0,5210,479-0,236-0,080B-1,335-1,085*0,240-0,107-0,019C+1,793*-0,189-0,010-0,179-0,510C--1,361*-0,172-0,0150,042U0,447-0,3530,6750,122-0,210IAEXEX---CI0,032-0,611---Size0,635**0,3000,2200,1210,475**Financial Performances0,090***0,0410,0130,049***0,040**SG&A to sales0,2660,023-0,157-0,0270,204RegionAfrica1,0150,734---Asia-0,018-0,169-0,0480,085-0,465*EuropeEXEXEXEXEXLatin America0,6580,4551,101*0,7650,708*Northern America0,602*0,895***0,1090,1580,434**Oceania--1,028-0,168--1,298IndustryOil & Gas-0,3840,2980,3130,720*0,735**Basic Materials0,8430,116-0,4680,2140,481*Industrials0,6841,000**0,3400,1600,320Consumer GoodsEXEXEXEXEXHealth Care-0,782-0,0640,1750,007-0,091Consumer Services0,2550,2070,3740,164-0,215Telecommunications1,3821,058-0,938*0,4040,037Utilities0,4310,0640,7340,2021,133*Financials-0,0440,6060,3830,126-0,003Technology0,0850,2940,728**0,2290,689***?Anova0,1210,0120,0240,0020,000R squared0,4560,4910,3080,2890,381VIF1,5811,6034,0121,3221,401*** Significant at p < 0,01** Significant at p < 0,05* Significant at p < 0,1When focusing on the anova, all years except 2005, provide a reliable model. The value of anova for 2005 is above the used significant level of 10%, which implies the designed model for 2005 is not an accurate prediction of reputation. The R squared is relatively high for all years, this implies that when there is a change in the dependent variables, the change is observable in the reputation. The R squares of this method are higher compared to the first method. The included categorical variables in the second method have an improved prediction on the reputation index. Still there is space for other variables to increase the predictability of reputation index. It is remarkable that the VIF score is higher in the second model. The collinearity of the second model is higher, this implies the results are less reliable compared to the first model. In 2007 there is much collinearity between variables, this is caused by the categorical variables of CSR report level. Following the results of table 10, the different CSR report levels have less different influences on reputation. Only in the year 2005 and 2006 difference in influences are observable. There is proven that C+ level in 2005 and the B and C level in 2006 have different influences on the reputation. The C+ level has, in 2005, a significant positive influences and the B and C level have a significant negative influence, compared with the IA level. The degree of different influences is pretty high, the C+ level in 2005 improved the reputation index with 1,793 in comparison with level IA. Besides the different influences of C+ in 2005, B and C level in 2006, there is no proof that the different CSR report levels have a different influence on reputation. In the results evidence was found that size has an influence on reputation. In the results of 2005 and 2009 a p-value less than 0,01 is observable. The size of a company has a positive influence on the reputation. When the size control variable is included in the basic model, the R squared improves. But when the size variable is included only in the years 2005 and 2009 a significant influence is observable. Also only in the years 2005 and 2009 the size has a significant influence. In these years the CSR report levels B+, U and C+ have a significant different influence on the reputation index, compared with the IA level.In the years 2005, 2008 and 2009 there is a significant influence of the financial performances on the reputation index observable. For these years is proven that a higher return on assets (with an amount of one) increased the reputation index with 0,09;0,049 or 0,04. When the financial performances variable is included in the basic model, a significant influence for the years 2005, 2008 and 2009 on the reputation index is proven. The R squared increases in the basic model when the variable of the financial performance is included. Including the financial performances variable in the basic model has also results in a different influence of the B+ and U level in comparison with IA level.For the general selling & administrative expenses to sales, no evidence for a significant influence on the reputation was found. When including this variable in the basic model, no influence was found. This implies that the general selling & administrative expenses to sales, have no influence on reputation and no influence on the relationship between CSR reporting and the reputation.Following the results of table 10, it is proven that regions have a significant influence on the reputation. For 2005, 2006 and 2009 it is proven that companies from Northern America have a positive different reputation than companies from Europe. In 2007 and 2009 Latin America has a significant different influence on reputation compared to Europe. The reputation index is higher in 2007 and 2009 for companies from Latin America compared to companies from Europe. When Asia is taken as reference category (not showed in table) it is proven that Europe has a significant different positive influence on reputation in 2009. When the different regions are included in the basic model, evidence is found that regions have an influence on the relationship between CSR report level and reputation. These results prove differences exist in the influence of regions. Northern America, Latin America, Asia and Europe have different influences on the reputation. When looking at specific regions, evidence is found that the countries USA, Brazil, Finland, Germany, Switzerland and the United Kingdom have a significant different influence.When focusing on the results of the industry categories evidence is found differences exist in the influence of industries on the reputation index. The differences in influence are not observable in 2005. From 2006 up till and including 2009 the different influences among industries are observable, but few industries differ significantly from the consumer goods. In 2006 only the industry type industrials, has a significant different influence on the reputation index. In 2007 telecommunications and technology has a significant different influence on the reputation index, compared to consumer goods. In comparison with consumer goods telecommunications has a negative influence on the reputation index and technology industry improves the reputation index. Oil & Gas industry has, in 2008, a significant different influence on the reputation index compared to consumer goods. Companies in the oil & gas industry in 2008 have higher reputation index than companies in the consumer goods industry. In 2009 companies operating in the oil & gas, basic materials, utilities and technology industry are facing a significant different influence on reputation index than companies operating in the consumer goods. These companies operating in the oil & gas, basic material, utilities and technology have a higher reputation index than the consumer goods. When only the industry variable is included in the basic model, also less evidence is found that different regions have a different influence on the reputation index. Also the R squared improves little, when regions are included in the basic model. This implies that the industry has less influence on the relationship between CSR quality and reputation. Chapter 7 ConclusionIn this chapter the outcomes of the research are summarized. The first paragraph discusses the main findings of the research. Thereafter the implications for the scientific world, companies and society are discussed. The limitations of this paper and opportunities of further research are discussed in the third paragraph.7.1 DiscussionIn this paper the relationship between CSR reporting and reputation is investigated. In prior research different impacts of CSR already have been researched. The financial impacts are already well established, but there is still room available to investigate the impact of CSR on society and stakeholders. Reputation is a measurement to investigate the effect of CSR on society. The society is affected after a CSR report. With this report society receives more CSR information about the company. As a result of the CSR report more transparency is provided. When the CSR report is of high quality, stakeholders are provided with more reliable information. It is to be expected that society will value a high quality of CSR reporting. The agency theory explains that the information asymmetry is reduced after the CSR report. Stakeholders will value this, because stakeholders are provided with more transparency. The legitimacy theory and stakeholder theory explain that companies act in favor of stakeholders’ interest to obtain legitimacy. Stakeholders are demanding high quality CSR reports, which increase transparency, thereby providing stakeholders with more reliable information. The willingness of management to provide a high quality CSR report is dependent on the power among different stakeholders. This implies the expectation that CSR reporting quality has a positive influence on reputation.Prior research of Clarkson et al. (2010) found evidence about the positive influence of CSR reporting on public sentiment. In this study the five most polluting industries from the United States are included. In the study of Clarkson et al. (2010) the GRI framework is used as proxy to investigate CSR reporting. The Janis Fadner coefficient is used as proxy for the reputation. This is not the most accurate measurement for reputation, the coefficient assumes that reputation is equal for all companies when the same number of positive and negative environmental articles is published. Also it is interesting to investigate the impact of CSR reporting quality on reputation for all types of industries. In this paper CSR reporting quality is measured with the CSR report level provided by the GRI Report List. From this list the application levels of CSR reports are observable, the report levels are A+, A, B+, B, C+, C, U, IA and CI. These levels contain the degree of conformity with CSR reporting guidelines. The reputation is measured with the reputation index, this index contains nine different components. The index is composed by directors, security analysts and executives of international companies. Combining the two databases resulted for the period 2005 till 2009 in 85, 106, 138, 167 and 175 data points respectively. The tested relation of CSR reporting quality and reputation is also dependent on other variables. These variables are size, financial performances, expenses to selling general and administration, region of company and the industry where the company operates.The impact of CSR report quality on reputation is tested with the correlation and the regression Correlation and regression makes use of different variables, the dependent variable is reputation. The independent variables are CSR reporting level, financial performance, size, percentage of selling general and administrative expenses to sales, region of company’s headquarter and industry where the company operates. Correlation measures the relation between reputation and an independent variable. With the regression the influence of more independent variables are measured on the reputation. The results show that there is no impact of CSR report quality on reputation. In this paper no evidence is found that a relation exists between CSR reporting quality and reputation. Public does not observe the quality of a CSR report. The CSR reporting quality is not incorporated in the reputation index. Evidence is found about the influence of size on the relation between CSR reporting quality and reputation. The found evidence of the size influence is small, it is not observed for all years. Financial performances do have an influence on the impact of CSR reporting quality on reputation. The reputation index is dependent on financial performance. When a relation between CSR reporting quality and reputation exist, financial performances increase the relationship. The percentage of general selling and administrative expenses to sales is a measurement to observe the degree of service expenses of sales. There is no evidence found, that general selling and administrative expenses to sales have an influence on the relation between CSR reporting quality and reputation. However, regions in which the headquarter of the company is seated do have an influence on the reputation index. A difference exists in the reputation index of companies from different regions. Evidence is found that Northern America has a significant different influence than Europe on the reputation index. Asia and Latin America also have a significant different influence than Europe on the reputation index. Asia has a negative influence, Northern America and Latin America have positive influences. The results of regions implies that difference between region exist on the relation between CSR reporting quality and reputation. No significant different influences among industries are observable on the reputation index. Industries have no impact on the relation between CSR reporting quality and reputation. 7.2 ImplicationsThe findings of this paper contribute to the scientific world, as well as to companies and society. No relationship between CSR reporting quality and reputation is observed in this paper, but this outcome sheds new lights on the study of Clarkson et al. (2010).In the study of Clarkson et al. (2010) a relation is found between CSR reporting and public sentiment. But from this paper contradictory evidence is found, because no relationship is found between CSR reporting quality and reputation. Former studies that investigate the relation between CSR reporting and reputation, did not incorporate the impact of financial performances, percentage of general selling and administrative expenses to sales, regions or industries. The found evidence that suggests an impact of financial performances and regions on the relationship contributes to the scientific world. In this paper it is proved that CSR reporting has no impact on reputation, but when stakeholders are putting more attention on CSR reporting, perhaps other results will be observed. Until now stakeholders do not incorporate CSR reporting in their impression of companies. When more attention is given to CSR, stakeholders will become more familiar with CSR reporting, which may result in an impact on the reputation. CSR reporting provides important company information, when a CSR report is of high quality more transparency is provided. This paper provides contribution to the society because there is proved that stakeholders should put more attention to CSR reporting. CSR reports provides transparency into the company’s information, society can benefit from this transparency.The findings of this paper provide benefits for companies, because it is investigated if certain variables have an influence on the reputation index. It is proven that size, financial performances and regions have an impact on the reputation. Awareness of companies is provided on the influence of size, financial performance and region on reputation. 7.3 Limitations and further researchThe data of the reputation is based on the reputation index that is provided by the Fortune 500. This reputation index is compiled by directors, security analysts and executives of international companies, which is not an accurate representation of total society and all stakeholder sentiment. It is difficult to find a database that represents the reputation index of companies that is representative of the impression of total society. But maybe in the future ways are found, to set a proxy for reputation which contains total society’s sentiment. For future research it is defiantly to find a proxy for reputation, which finds another public than directors, security analysts and executives. Interesting is to compile the reputation on judgments of environmental and social foundations. A foundation as Greenpeace has much knowledge about environmental related performances of companies. An idea is to investigate if the judgment of, for instance Greenpeace is affected by CSR reporting. It is also interesting to base the reputation on costumers. In the researches of Brown et al. (1997) and Ittner et al. (1998) the reputation is already based on customer satisfaction, but customer satisfaction is compiled of the evaluation of single products. The company is more than its products and also companies exist that produce more than one product. An opportunity for future research is to compile reputation on customers’ satisfaction on company level and not on product level.The application levels of 2005 and 2006 are based on the G2 and G3 guidelines. The applications levels of 2007, 2008 and 2009 are based on the G3 guidelines. This means that two different measurements of CSR report level are used in this paper. The comparability of results of all years decrease, when in 2005 and 2006 different guidelines are the basis of application level. But when the application level that are conform the G2 guidelines are excluded, the sample becomes very small.Another limitation is that in the total sample contains only a few companies that operate in Africa and Oceania. So it is not possible to observe the influence of these continents on the relation between CSR reporting quality and the reputation. This implies that for future research the sample should be increased with companies from Africa and Oceania.This study focuses on CSR reporting. It is also interesting to focus on the CSR performances and their influence on the content of a CSR report. In this paper evidence is found that directors, security analysts and executives do not observe the CSR report level. But maybe they do observe CSR performances. ReferencesKarloff, G., 1970, The market for ‘lemons’ quality uncertainty and the market mechanism, Quarterly Journal of Economics, 90, p. 629-650Akhtaruddin, M., Haron, H., 2010, Board ownership, audit committees effectiveness and corporate voluntary disclosures, Asian Review of Accounting, 18, p. 245-259Brown, T.J., Dacin, P.A., 1997. The company and the product: Corporate associations andconsumer product responses. Journal of Marketing 61: 68-84Chen, C.X., 2009. Who really matters? Revenue implications of stakeholder satisfaction in ahealth insurance company. The Accounting Review 84(6): 1781-1804Cho, C.H., Patten, D.M., 2007, The role of environmental disclosures as tools of legitimacy: A research note, Accounting, Organizations and Society, 32, p. 639-647Clarkson, P.M., Richardson, G.D., Li, Y., 2004, The market valuation of environmental capital expenditures by pulp and paper companies, The Accounting Review, 04/2004, p. 3289-3353Clarkson, P.M., Fang, X.H., Li, Y., Richardson, G., 2010, The relevance of environmental disclosures for investors and other stakeholder groups: are such disclosures incrementally informative?, Working Paper, University of QueenslandCrane, A., Mcwilliams, A., Matten, D., Moon, J., Siegel, D., 2008, oxford handbook of corporate social responsibility, Oxford University Press, OxfordDawkins. J., Stewart L., 2003, CSR in stakeholder expectations: and their implication for company strategy, Journal of Business Ethics, 44, p. 185Dhaliwal, D., Radhakrishnan. S., Tsang, A., Yang, Y.G., May 2010. Nonfinancial disclosure and analyst forecast accuracy: International evidence on corporate social responsibility (CSR) disclosure, working paper, University of Arizona and The Chinese University of Hong KongDhaliwal, D., Li, O.Z., Tsang, A., Yang, G.Y., 2010. Voluntary nonfinancial disclosure andthe cost of equity capital: The case of corporate social responsibility reporting. Workingpaper, Chinese University of Hong Kong.Douglas, A., Doris, J., Johnson, B., 2004, Corporate Social Reporting in Irish FinancialInstitutions, Total Quality Management Magazine 16(6), 387-395.Dowling, J., Pfeffer, J., 1975, Organizational Legitimacy: Social Values and Organizational Behavior, The Pacific Sociological Review, 18, P. 122-136Du. D., 2003, Stock price reactions to earnings announcements: evidence from Chinese markets, Review of Financial Economics, 12, p. 271-286Elder, R.J., Beasley, M.S., Arens, A.A., 2010, Auditing and Assurance Services, an integrated approach, 13th edition, Pearson, New JerseyFormbrund, C., Shanley, M., 1990, What’s in a name? Reputation building and corporate strategy, academy of Management Journal, 33, p. 233-258Freeman, R.E., Reed, D.L., 1983, Stockholders and stakeholders: a new perspective on corporate governance, Californian Management Review, 25, p. 88-106Healy, P.M., Hutton, A., Palepu K.G., 1999, Stock performance and intermediation changes surrounding sustained increases in disclosure, Contemporary Accounting Research, 16, p. 485-520Healy, P.M., Palepu, K.G., 2000, Information asymmetry, corporate disclosure, and the capital markets: A review of the empirical disclosure literature, Journal of Accounting and Economics, 31, p. 405-440Health Partners, 02/2004, Fitness; Active Physically fit employees have better work performance, Biotech Week ISSN, 1535-2757, p. 186Hope, O. 2003. Disclosure practices, enforcement of accounting standards, and analysts’ forecast accuracy: An international study. Journal of Accounting Research 41(May): 235-272.Huang, F.H., Kung, C.Li., 2010, Drivers of environmental disclosure and stakeholder expectation; evidence from Taiwan, Journal of Business Ethics, 96, p. 435-451Ittner, C., Larcker, D., 1998, Are non-financial measures leading indicators of financial performance? An analysis of customer satisfaction, Journal of Accounting Research, 36 (supplement), p. 1-35Jensen, M.C., Meckling, M.H., 1976, Theory of the firm: managerial behavior, agency costs and ownership structure, journal of financial economics, 3, p. 305-360 Kakabadse, N. K., Kakabadse, A. P., Lee-Davies, L., 2006, CSR leaders road-map, Corporate Governance, 9, p. 52Kavitha, W., Anita, P., 2011, Disclosures About CSR Practices: A Literature Review, The IUP Journal of Corporate Governance X(1): 45-55Kothari, S.P., 2001, Capital markets research in accounting, Journal of Accounting and Economics, 31, p. 105-231Lang, M., Lundholm, R., 1993, Cross-sectional determinants of analysts ratings of corporate disclosures, Journal of Accounting Research, 31, p. 246-271Lang, M. Lundholm, R., 1996. Corporate disclosure policy and analyst behavior. TheAccounting Review 71, p. 467-493.Lang, M., Lundholm, R., 2000, Voluntary disclosure and equity offerings: Reducing information asymmetry or hyping the stock?, Contemporary accounting research, 17, p. 623Lev, B., Petrovits, C., Radhakrishnan, S., 2010. Is doing good good for you? Yes, charitablecontributions enhance revenue growth. Strategic Management Journal 31(2): 182-200.Lightstone, K., Driscoll, C., 2008, Disclosing elements of disclosure: A test of legitimacy theory and company ethics, Canadian Journal of Administrative Sciences, 25, p. 7-21Liu, J., Taylor, D., 2008, Legitimacy and corporate governance determinants of executives’ remuneration disclosures, Corporate Governance, 8, p. 59-72Louhichi, W., 2008, Adjustment of stock prices to earnings announcements: evidence from Euronext Paris, Review of Accounting and Finance, 7, p. 102-115Margolis, J.D., Walsh, J.P., 2003, Misery loves companies: Rethinking social initiatives by business, Administrative Science Quarterly, 48, p. 268-305Matsumura, E.M., Prakash, R., and Vera-Munoz, S.C., 2010 Carbon Emissions and Firm Value, Working paper, University of Wisconsin-MadisonOrlitzky, M., Schmidt, F.L., Rynes, S.L., 2003, Corporate social and financial performance: a meta-analysis, Organization Studies, 24 p. 403-441Patten, D., 2002, The relation between environmental performance and environmental disclosure: Research note, Accounting, Organization, and Society, 27, p. 763-773Plumlee, M., Brown, D., Marshall, S., 2009, Voluntary environmental disclosure quality and firm value: the role of venue and industry type. Working paper, University of UtahPorter, M.E., Kramer, M.R., 2006, Strategy and society; the link between competitive advantage and corporate social responsibility, Harvard Business Review, 85Quaak. L., Aalbers, T., Goedee. J., 2006, Transparency of corporate social responsibility in Dutch breweries, Journal of Business Ethics, 76, p. 293-308Richardson, A., Welker, M., 2001. Social disclosure, financial disclosure and the cost of equity capital, Accounting, Organizations and Society. 26 (7/8), 597-615Sutantoputra, A.W., 2009, Social Disclosures Rating System for assessing firms’, CSR Reports, Corporate Communications: An International Journal 14(1): 34-48Ullman, 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, p. 540-557Watts, R., Zimmerman, J., 1983, Agency problems, auditing and the theory of the firm: some evidence, Journal of Law and Economics, 26, p. 613-634Wood, D.J., 2010, Measuring Corporate Social Performance: A Review, International journal of Management Reviews, 12, p. 50-84 Internet Resources accessed on 15 March 2011 accessed on 15 March 2011 accessed on 1 April 2011 accessed on 1 April 2011 accessed on 14 March 2011Richtlijnen voor duurzaamheidsverslaggeving, 2000-2006 GRI, version 3.0 accessed on 15 March 2011Appendix A: Companies included in sampleR.L. = CSR Report LevelR.I. = Reputation Index?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. 3MOceaniaIndustrialsC+7,43C+6,96C+6,87ABBEuropeIndustrialsIA5,50U5,65U6,62B+5,85B6,18AbbottNorthern AmericaHealth CareCI6,66CI6,75U6,55U6,78U6,68ABN AMRO HoldingEuropeFinancialsIA5,56A+4,97AccentureNorthern AmericaIndustrialsC7,35Advanced Micro DevicesEuropeTechnologyB4,43AegonCI5,36B5,31B5,26U4,69AeonAsiaConsumer servicesU5,34U4,78Aisin SeikiAsiaConsumer goodsU5,19Akzo NobelEuropeBasic MaterialsA+4,57C+5,54AlcanNorthern AmericaBasic MaterialsU7,41U7,94U6,29AlcatelEuropeTechnologyCI5,60AlcoaNorthern AmericaBasic MaterialsU6,78U6,69U7,05All Nipon Airways (ANA)EuropeConsumer servicesB+4,85AllianzEuropeFinancialsIA5,84B5,85B6,04B6,20B5,70AllstateNorthern AmericaFinancialsU5,91American Electric PowerNorthern AmericaUtilitiesB5,85B6,03U7,85AMRNorthern AmericaConsumer servicesC5,20C4,57C4,06Anglo AmericanEuropeBasic MaterialsCI5,10A+5,04A+5,95Anheuser-BuschNorthern AmericaConsumer goodsU8,00U8,17B7,97?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. AppleNorthern AmericaTechnologyU7,40U7,07Applied MaterialsNorthern AmericaTechnologyU6,77C6,68Arcelor MittalEuropeBasic MaterialsU7,74U6,89U6,78Asahi BreweriesAsiaConsumer goodsU4,18U4,20Asahi KaseiAsiaBasic MaterialsU3,65U5,10AshlandNorthern AmericaBasic MaterialsA+4,47Assicurazioni GeneraliEuropeFinancialsB4,49B4,89AstraZenecuEuropeHealth CareB+6,18AT&T inc. Northern AmericaTelecommunicationsC7,05C6,63AutodeskNorthern AmericaTechnologyC6,63Avon ProductsNorthern AmericaConsumer GoodsC5,82Baker Hughes CompanyNorthern AmericaOil & GasIA6,62A+6,61A+6,01Bank of AmericaNorthern AmericaFinancialsU7,59B6,69BarclaysEuropeFinancialsCI5,87BASFEuropeBasic MaterialsIA7,42A+6,93A+6,89A+7,06A+7,04Baxter InternationalNorthern AmericaHealth CareB6,74B+6,71B+6,78BayerEuropeBasic MaterialsB+6,53A+5,70A+6,29A+5,53A+6,87Best BuyNorthern AmericaConsumer servicesU7,20U6,63U6,63BG GroupEuropeOil & GasU6,05BHP BillitonOceaniaBasic MaterialsA+6,07A+6,39A+5,90A+6,92BMWEuropeConsumer GoodsB+7,38B+7,88A6,50BNP ParibasEuropeFinancialsCI5,77U5,96U4,89U5,03BombardierNorthern AmericaIndustrialsB5,47U5,91BPEuropeOil & GasC+8,17C+7,32A+6,60A+6,84A+6,63BallNorthern AmericaIndustrialsB6,50Becton DickinsonNorthern AmericaHealth CareU6,90Black & DeckerNorthern AmericaConsumer GoodsU5,91BoiseNorthern AmericaBasic MaterialsIA5,29U3,86?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. BridgestoneAsiaConsumer GoodsC5,34British American TobaccoEuropeConsumer GoodsA+7,71A+7,18Bristol-Myers SquibbNorthern AmericaHealth CareB+5,23B+4,98B+5,44B+5,40Brown-Forman CorporationNorthern AmericaConsumer GoodsC4,14BT GroupEuropeTelecommunicationsIA5,90IA6,59A+6,47A+6,46A+5,76BungeLatin AmericaConsumer GoodsCI7,14A+7,14A+7,08A+7,00A+6,55Cadbury SchweppesEuropeConsumer GoodsCI5,65CanonAsiaTechnologyCI6,88U6,56U6,05U6,12CarrefourEuropeConsumer ServicesIA6,91B+6,82U7,09B+6,35B+5,65ChevronNorthern AmericaOil & GasCI7,92U7,80U7,62U7,02China Mobile CommunicationsAsiaTelecommunicationsU5,06B+5,86B+5,41B+5,25ChubbNorthern AmericaFinancialsU7,09Cisco SystemsNorthern AmericaFinancialsU7,29U7,73U7,82U7,36U7,83Citi GroupNorthern AmericaFinancialsCI7,40U6,87B5,54B4,54B3,41Coca-ColaNorthern AmericaConsumer GoodsB6,09B6,09C6,90B6,84B6,98Coca - Cola EnterprisesNorthern AmericaConsumer GoodsB5,88B5,02B5,98Colgate-PalmoliveNorthern AmericaConsumer GoodsU6,81U6,76ConAgra FoodsNorthern AmericaConsumer GoodsB5,36Conoco PhilipsNorthern AmericaOil & GasU6,80Constellation EnergyNorthern AmericaUtilitiesU4,46ContinentalNorthern AmericaConsumer GoodsU7,80Credit SuisseEuropeFinancialsCI5,53U5,85U6,54A6,53A6,63CumminsNorthern AmericaIndustrialsU6,94U6,49C6,25Daimler ChryslerEuropeConsumer GoodsCI5,94Darden RestaurantsNorthern AmericaConsumer servicesU6,21Dean FoodsNorthern AmericaConsumer GoodsU4,71U4,71?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. Delhaize GroupEuropeConsumer ServicesB+6,13DellNorthern AmericaTechnologyCI6,12C6,03B5,58B5,62B6,14Delta AirlinesNorthern AmericaConsumer ServicesC6,31DensoAsiaConsumer GoodsIA6,38U6,26U5,62Deutsche BankEuropeFinancialsIA4,69A5,10A6,03A5,59Deutsche Post (DHL)EuropeIndustrialsB+5,50B+5,41B+5,17Deutsche TelekomEuropeTelecommunicationsCI5,97A+5,13A+5,60A+5,28Dexia GroupEuropeFinancialsU4,75DiageoEuropeConsumer GoodsA+4,66A+5,82A5,40Dow ChemicalNorthern AmericaBasic MaterialsA+6,43A+5,64A+6,11DuPontNorthern AmericaIndustrialsU6,91B7,03Duke EnergyNorthern AmericaUtilitiesB7,16B6,95B6,87B5,88B6,11E.ONEuropeUtilitiesIA6,79A+5,82B+6,20EADSEuropeIndustrialsB+7,06Eastman Chemical CompanyNorthern AmericaBasic MaterialsA+5,01Eastman Kodak CompanyNorthern AmericaConsumer GoodsA+5,54A+5,67EdisonEuropeUtilitiesC+6,46B+7,92A+6,96El Paso CorporationNorthern AmericaOil & GasC6,59C5,94ElectroluxEuropeConsumer GoodsCI6,03EMCEuropeTechnologyB6,86EnCanaNorthern AmericaOil & GasCI5,67B+6,32B+5,80B+6,46ENI S.p.A.EuropeOil & GasU5,70B+5,88B+5,92ExelonNorthern AmericaUtilitiesU7,11U6,35Exxon MobileNorthern AmericaOil & GasCI8,24U8,17U7,95U7,79U7,36FiatEuropeConsumer GoodsCI3,52CI4,38B+3,56FinmeccanicaEuropeIndustrialsU4,93FPL GroupNorthern AmericaUtilitiesU6,80FluorNorthern AmericaUtilitiesB7,15B6,24?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. Ford MotorNorthern AmericaConsumer GoodsIA5,09A+5,21A5,06A3,89A3,78FortisEuropeFinancialsB+5,01B+5,40France Telecom (-Orange)EuropeTelecommunicationsU5,41B+5,66A+5,16Freeport-McMoRan Copper & GoldNorthern AmericaBasic MaterialsIA5,21U5,25A+5,81FujitsuAsiaTechnologyU4,92B+5,06GapNorthern AmericaConsumer ServicesU5,92U6,28GazpromEuropeOil & GasB5,62General ElectricNorthern AmericaIndustrialsIA8,29A+8,24A8,40A+7,44A7,07Glaxo SmithKline (GSK)Northern AmericaHealth CareCI6,58CI6,68U6,33U6,48U5,94H.J. HeinzNorthern AmericaConsumer GoodsB5,29B4,95HalliburtonNorthern AmericaOil & GasC6,03A7,40Hanwha ChemicalsAsiaBasic MaterialsA+3,35HeinekenEuropeConsumer GoodsCI5,48U4,54U5,59U4,99HenkelEuropeConsumer GoodsCI5,03U6,03B5,25B7,12B5,79Herman MillerNorthern AmericaConsumer GoodsB7,83B6,39Hess corporationNorthern AmericaOil & GasB+5,74Hewlett - PackardNorthern AmericaTechnologyCI7,08B5,56B7,38B7,04B7,74HitachiAsiaIndustrialsU6,14C5,86B+5,38HochtiefEuropeIndustrialsB+6,48B+6,62HSBC HoldingsEuropeFinancialsCI6,92U6,50Hyundai MotorAsiaConsumer GoodsIA5,39A+4,66A+7,28U4,41U4,49ITT IndustriesNorthern AmericaIndustrialsC7,09C5,92Imperial Chemical IndustriesNorthern AmericaOil & GasCI4,64Imperial TobaccoEuropeConsumer GoodsB+6,68B+6,74B+5,81ING GroupEuropeFinancialsU6,16U6,15A+5,63A+4,39Ingersoll-RandNorthern AmericaIndustrialsCI6,96U6,51B6,37B6,09B+5,84IntelNorthern AmericaTechnologyIA7,88IA7,12B+6,98B7,57A7,96IBMNorthern AmericaTechnologyA7,57A7,50A7,55A7,60?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. Jabil CircuitEuropeIndustrialsC4,31C4,82Jacobs Engineering Grp. Northern AmericaIndustrialsU6,32Johnson & JohnsonNorthern AmericaHealth CareCI7,64U7,53U6,95U7,31U6,67Johnson ControlsNorthern AmericaConsumer GoodsCI7,03CI6,87A6,50A6,39A6,26Jones Lang LaSalleNorthern AmericaFinancialsU7,16U6,27Juniper NetworksNorthern AmericaTechnologyA7,11A+6,65A+5,76KAO CorporationAsiaConsumer GoodsCI5,26CI5,36U5,58KB HomeNorthern AmericaConsumer GoodsU7,30U6,58KelloggNorthern AmericaConsumer GoodsB6,39B6,43Kimberly - ClarkNorthern AmericaConsumer GoodsU6,34U6,09KingfisherEuropeConsumer ServicesB+5,17B+6,48Kirin BreweryAsiaConsumer GoodsU4,31U4,58Korea Gas CorporationAsiaUtilitiesB+6,24LafargeEuropeIndustrialsA+5,22Lenovo GroupAsiaTechnologyU4,77U5,21Lexmark InternationalNorthern AmericaTechnologyB5,68LG ElectronicsAsiaConsumer ServicesCI6,39CI6,14U6,41B+5,77U5,79Liberty Mutual Insurance GroupAfricaFinancialsCI5,96B+6,42B+6,07B+5,93LindeEuropeBasic MaterialsC4,53B4,74U4,75B+5,97L'Oreal EuropeConsumer GoodsCI7,26CI7,16U7,25U6,93B6,86Louisiana-Pacific CorporationNorthern AmericaIndustrialsA+4,20L.M. EricssonEuropeTechnologyU6,93B+6,93B+5,91A+6,37ManpowerLatin AmericaIndustrialsB6,06Marathon OilNorthern AmericaOil & GasU6,60Marks & SpencerEuropeConsumer ServicesU6,90MascoNorthern AmericaIndustrialsB5,24?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. Mazda MotorAsiaConsumer GoodsIA5,51IA6,03U4,34U3,76McDonald's Northern AmericaConsumer ServicesU7,72McKessonNorthern AmericaConsumer ServicesC6,63MedtronicNorthern AmericaHealth CareC6,66B6,52Merck GermanyEuropeHealth CareB5,88B6,67MetLifeLatin AmericaFinancialsU5,94U5,37B+6,26B+5,99B+6,47Metro GroupEuropeConsumer ServicesCI6,66U6,76MicrosoftNorthern AmericaTechnologyU6,37U6,54Mitsubishi ElectricAsiaIndustrialsU5,71U5,48Mitsubishi Heavy IndustriesAsiaIndustrialsU5,82MotorolaNorthern AmericaTechnologyCI7,41CI7,60U7,16U5,09U4,58NEC CorporationAsiaTechnologyCI5,84U5,35U4,86B4,93NestleEuropeConsumer GoodsC8,08B+7,63Newmont Mining Northern AmericaBasic MaterialsA+5,70NikeNorthern AmericaConsumer GoodsU7,82U8,02B8,15Nissan MotorAsiaConsumer GoodsU6,12U5,17U3,80NokiaEuropeTechnologyU7,80U6,96A+6,94Norsk HydroEuropeBasic MaterialsCI6,81B+6,68B+6,33Occidental Petroleum (Oxy)Northern AmericaOil & GasU7,89U8,04U7,27Office DepotNorthern AmericaConsumer ServicesCI5,64C+5,70C+6,13C+4,52U4,75Oji Paper GroupAsiaBasic MaterialsU5,71PPG IndustriesNorthern AmericaBasic MaterialsC5,46C6,96Pepsi CoNorthern AmericaConsumer GoodsU6,50U7,47U6,88PetrobrasLatin AmericaOil & GasCI5,37A+6,14A+6,63PfizerNorthern AmericaHealth CareB6,00B5,25POSCOAsiaBasic MaterialsIA6,41U6,04U6,48U6,72U6,37Protcer & GambleNorthern AmericaConsumer GoodsU8,39U7,69U7,94?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. ProLogisNorthern AmericaFinancialsC+7,62B+8,09B+6,09PSA Peugeot CitroenEuropeConsumer GoodsB+4,28A+4,39QualcommNorthern AmericaFinancialsC7,04C+7,16Reckitt Benckiser GroupEuropeConsumer GoodsB7,13RenaultEuropeConsumer GoodsU4,57U3,58Repsol YPFEuropeOil & GasIA4,99A+4,75A+5,89A+5,76Reynolds AmericanNorthern AmericaConsumer GoodsU7,26U7,01RicohAsiaTechnologyB5,63B5,23C5,22Rio TintoEuropeBasic MaterialsA+5,94A+5,97A+5,37A+6,30Royal AholdEuropeConsumer ServicesC5,74B5,42B+5,44Royal Bank of ScotlandEuropeFinancialsIA5,73B+6,46B+6,12Royal Dutch Shell GroupEuropeOil & GasA+7,22A+7,17A+7,28A+7,55A+7,05Royal Philips ElectronicsEuropeConsumer GoodsCI6,79IA6,63B+6,77B+5,98A+5,62RWEEuropeUtilitiesA+6,38A5,87U5,89A+6,54SABMiller UKEuropeConsumer GoodsCI3,78B+4,59B+5,49B+6,26Samsung ElectronicsNorthern AmericaTechnologyA+5,88A+6,74SAPEuropeTechnologyC6,49B+6,19A+6,41Sanofi - AventisEuropeHealth CareU6,07U5,56U5,54Sara LeeNorthern AmericaConsumer GoodsU4,67B4,73B4,85SCA GroupEuropeConsumer GoodsC5,61Sempra EnergyNorthern AmericaUtilitiesU6,40U6,45SiemensEuropeIndustrialsIA7,32U7,19U6,98U6,40A+6,12Smithfield FoodsNorthern AmericaConsumer GoodsB6,34B5,13B5,45Societe GeneraleEuropeFinancialsCI5,66U4,64SolvayEuropeBasic MaterialsU4,42Sony EuropeConsumer GoodsU6,53U6,49U7,01U6,30U6,29?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. SouthernNorthern AmericaUtilitiesU6,55Southwest AirlinesNorthern AmericaConsumer ServicesC+6,26Sprint NextelNorthern AmericaTelecommunicationsU4,11StarbucksNorthern AmericaConsumer ServicesCI8,11CI8,09B+8,12B+6,88StatoilEuropeOil & GasCI6,31U6,50U6,72A+6,42A+6,59State Street Corp.Northern AmericaFinancialsCI6,35B+6,63B+6,86B+6,23B+6,29SteelcaseNorthern AmericaConsumer GoodsU5,76Stora EnsoEuropeBasic MaterialsCI6,65CI6,44B+4,05Suez EnvironmentEuropeIndustrialsB+5,92Sumitomo ChemicalAsiaBasic MaterialsU5,40Sumitomo Electric IndustriesAsiaIndustrialsU5,42U5,18Swiss Re / reinsuranceEuropeFinancialsU6,18U6,21U6,18TargetNorthern AmericaConsumer ServicesCI6,66U6,90U6,92TechnipEuropeOil & GasB+6,41TelefonicaEuropeTelecommunicationsIA6,40A+6,40A+6,04A+6,47Teradata/TeredataNorthern AmericaTechnologyC+6,42Texas InstrumentsNorthern AmericaTechnologyCI7,95U8,21U7,26U7,16B7,30Thyssen Krupp GroupEuropeIndustrialsU6,69Time WarnerNorthern AmericaConsumer ServicesCI6,50TNTEuropeIndustrialsIA5,97C+5,39A+5,62A+5,87A+5,96ToshibaAsiaIndustrialsIA6,22B6,11U6,30U5,94B+5,67TotalEuropeOil & GasCI6,89U6,70U6,72Toyota MotorAsiaConsumer GoodsCI7,51U7,86A+6,25A+5,20Tyco InternationalNorthern AmericaIndustrialsU5,32Tyson FoodsNorthern AmericaConsumer GoodsU6,07B5,39U.S. Postal ServiceNorthern AmericaConsumer ServicesUBSEuropeFinancialsCI5,73U7,06U5,99A+4,79UnileverEuropeConsumer GoodsCI6,63CI6,46B+6,10B+6,55B+6,52?20052006200720082009?RegionIndustryR. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. R. L. R. I. United Parcel Servive (UPS)Northern AmericaIndustrialsCI8,17B8,42B7,39ValeLatin AmericaBasic MaterialsA+6,31Vivendi UniversalEuropeConsumer ServicesCI5,25U5,74U6,65U5,70Vodafone GroupEuropeTelecommunicationsIA6,47B+6,42B+6,38B+6,48B+6,63VolkswagenEuropeConsumer GoodsA+5,09A+6,75A+5,22VOLVOEuropeIndustrialsIA5,85U6,27U6,24B4,50B3,80Wachovia Corp. Northern AmericaFinancialsC6,01Wal-Mart Stores Northern AmericaConsumer ServicesC6,54B7,29B7,14WeyerhaeuserNorthern AmericaBasic MaterialsIA6,90A6,75A7,20A5,76A5,10WoolworthsOceaniaConsumer ServicesCI4,84A+4,82A+4,57WyethNorthern AmericaHealth CareCI6,43U5,89Xerox Northern AmericaTechnologyB6,92B7,28B6,51Appendix B: Distribution CSR report levels over all variables?2005 (N = 85)2006 (N = 106)CSR levelA+B+BC+UIACIA+AB+BC+CUIACIFrequency23218244516412114330521Percentages2,43,52,41,29,428,252,915,13,811,310,43,82,828,34,719,8Size (sales in million $)??< 50.0001,23,52,4-7,112,936,56,60,98,57,52,81,917,94,714,250.000 - 100.000----2,411,89,44,71,91,91,9-0,94,7-4,7> 100.0001,2--1,2-3,57,13,80,90,90,90,9-5,7-0,9Financial Performances (ROA)?? < 0%-----1,2-0,9-0,9------0% - 10%1,22,41,21,24,722,434,19,43,89,47,51,90,917,02,812,3> 10% 1,21,21,2-4,74,730,64,7-0,92,81,91,911,31,97,5SG&A expenses to sales??< 15%1,2--1,22,412,917,66,60,90,90,91,90,99,4-4,715% - 30% 1,21,2--4,78,214,12,80,93,84,70,91,913,22,83,8> 30%-1,21,2---8,21,90,90,92,8--0,9-6,6Unknown--1,2-2,47,112,93,80,95,71,90,9-4,71,94,7Regions??Africa------1,2--0,9------Asia----1,24,75,91,9-0,91,9--3,80,91,9Europe2,42,4-1,22,417,623,59,41,97,52,81,91,913,21,98,5Latin America----1,2-2,40,9-----0,9--Northern America-1,22,4-4,75,920,01,91,91,95,71,91,910,41,98,5Oceania-------0,9-------0,9Industry??Oil & Gas1,2--1,2-2,47,11,9---0,9-2,8-0,9Basic Materials-1,2--1,23,52,44,70,90,9--0,94,70,91,9Industrials-----7,11,20,9-0,90,90,90,93,8-0,9Consumer Goods--1,2-2,43,515,34,7-0,91,9--4,71,95,7Health Care-1,2----3,5--0,91,9--1,9-2,8Consumer Services-----1,27,1--1,9-0,9-1,9-3,8Telecommunications----1,23,5-0,9-1,9----0,90,9Utilities1,2-1,2--1,2-0,90,9-1,9-----Financials-1,2--2,44,79,40,90,93,81,90,9-5,7-1,9Technology----2,41,27,1-0,9-1,9-0,92,80,90,9?2007 (N = 136)2008 (N = 167)CSR levelA+AB+BC+CUA+AB+BC+CUFrequency186201731260297292821062Percentages13,04,314,512,32,28,743,517,44,217,416,81,26,037,1Size (sales in million $)??< 50.0008,02,210,18,72,28,028,37,81,810,212,01,23,624,050.000 - 100.0002,90,73,61,4--8,76,61,211,41,8--6,6> 100.0002,21,40,72,2-0,76,53,01,21,83,0-2,45,4Financial Performances (ROA)?? < 0%0,7-0,7---3,62,41,82,42,40,61,83,60% - 10%7,24,38,08,71,43,621,710,81,810,29,00,62,422,2> 10% 5,1-5,83,60,75,118,14,20,64,85,4-1,810,8SG&A expenses to sales??< 15%7,22,95,13,6-1,413,87,82,46,64,2-2,412,615% - 30% 2,21,45,85,80,72,97,24,20,64,88,40,61,815,6> 30%0,7-2,21,4-2,916,70,6-3,03,0-0,66,0Unknown2,9-1,41,40,75,15,84,81,23,01,80,60,63,0Regions??Africa--0,7------0,6----Asia0,7----0,75,10,6-1,20,6-0,69,0Europe7,2-8,02,90,71,413,810,81,812,03,6-1,29,6Latin America0,7-0,7----1,2-0,6----Northern America3,64,35,19,40,76,524,63,62,43,012,60,64,218,6Oceania0,7---0,7--1,2---0,6--Industry??Oil & Gas1,4-1,4--1,44,33,0-1,2--0,63,0Basic Materials6,50,7-0,7--3,63,60,60,6--0,63,0Industrials0,70,70,72,20,7-5,82,4-2,43,00,61,23,6Consumer Goods2,91,43,63,6-2,210,93,61,83,64,2-1,29,0Health Care--0,70,7-0,72,2--1,21,2--3,0Consumer Services--2,2-0,72,24,30,6-1,81,80,61,23,0Telecommunications1,4-0,7---1,41,8-1,8--0,6-Utilities---0,70,7-0,7--1,20,6--3,0Financials--3,62,2-0,75,81,21,22,42,4-0,63,0Technology-1,41,42,2-1,44,31,20,61,23,6--6,6?2009 (N = 175)CSR levelA+AB+BC+CUFrequency3410263361155Percentages19,45,714,918,93,46,331,4Size (sales in million $)?< 50.0009,72,98,614,92,95,722,950.000 - 100.0006,31,74,61,10,6-5,7> 100.0002,91,11,72,9-0,62,9Financial Performances (ROA)? < 0%1,71,14,02,9-2,36,30% - 10%14,33,47,49,12,93,419,4> 10% 3,41,13,46,90,60,65,7SG&A expenses to sales?< 15%9,72,92,36,30,63,410,915% - 30% 3,41,78,66,31,11,713,1> 30%1,1-2,34,60,60,62,9Unknown4,61,11,71,71,10,64,6Regions?Africa-------Asia0,6-2,90,6-0,610,3Europe13,11,78,05,71,11,16,9Latin America1,7-0,60,6---Northern America2,94,03,412,01,74,614,3Oceania1,1---0,6--Industry?Oil & Gas3,40,61,70,6--2,3Basic Materials4,60,61,7-0,60,62,9Industrials1,10,62,93,40,62,33,4Consumer Goods2,91,72,36,9-0,69,7Health Care--1,10,6--2,9Consumer Services0,6-1,71,10,61,14,0Telecommunications2,3-1,1--0,6-Utilities1,1-0,60,6--1,1Financials0,61,11,11,71,1-2,3Technology2,91,10,64,00,61,12,9 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download