The relation between corporate social responsebility and earnings management

Hele tekst

(1)The relation between corporate social responsibility and earnings management Master Thesis Accountancy and Control. Student: Student number: First supervisor: Date: Word count:. Lucie Sleper 6167314 Dr. G. Georgakopoulos June 12th, 2015 10.971. Master of Science Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam.

(2) Statement of originality This document is written by student Lucie Josephine Sleper, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economic and Business is responsible solely for the supervision of completion of the work, not for the contents.. 2.

(3) Abstract The research question of this thesis is: does the growth of CSR reporting influence earnings management? CSR reporting has grown fast the last decennia. But still has not found the ideal form and standards. Earnings management is a subject of discussion. The government tries to prevent earnings management with the introduction of SOX. On basis of the agency legitimacy and stakeholder theories is the following hypothesis developed and tested: Earnings management is negative related to CSR. Earnings management is tested on basis of discretionary accruals and measured with the modified Jones model with the data of the Compustat database. For the measurement of CSR are the KLD ratings and database used. The dataset consist of 810 non-financial U.S. firms. There is made a comparison between the time period 2004 till 2008 and 2009 till 2013. There is no significant negative correlation for both of the time periods found between earnings management and CSR. Therefore the answer to the research question is that CSR does not influence earnings management.. Key words: Earnings management, corporate social responsibility, discretionary accruals. 3.

(4) Dutch Summery De onderzoeksvraag van deze scriptie is: heeft de groei van de CSR-reporting invloed op earnings-management? CSR-reporting is snel gegroeid in de laatste decennia, maar heeft nog niet de ideale vorm aangenomen. De overheid heeft maatregelen genomen tegen earnings-management door middel van de invoering van SOX. Op basis van de agency legitimacy and stakeholder theorieën is de volgende hypothese ontwikkeld en getest: Earrings-management is negatief gerelateerd aan CSR. Earnings-management wordt getest op basis van de discretionary accruals en gemeten door middel van het modified Jones model met de gegevens van de Compustat database. Voor het meten van CSR zijn de KLD ratings en database gebruikt. De dataset bestaat uit 810 niet-financiële Amerikaanse bedrijven. Er wordt een vergelijking gemaakt tussen de periode 2004 tot 2008 en 2009 tot 2013. Er is geen significant negatieve correlatie voor beide perioden gevonden tussen earnings-management en CSR. Het antwoord op de onderzoeksvraag is daarom dat CSR geen invloed heeft op earnings-management.. 4.

(5) Table of contents 1. Introduction. Page 6. 1.1 Background. 6. 1.2 Research question. 8. 1.3 Motivation. 9. 2. Literature review. 10. 2.1 CSR reporting. 10. 2.2 Earnings management. 13. 2.3 Agency theory. 15. 2.4 Legitimacy theory. 16. 2.5 Stakeholder theory. 17. 2.6 Sarbanes-Oxley act. 18. 3. Hypothesis. 19. 4. Research methodology. 20. 4.1 Data and Sample. 21. 4.2 Measurement of variables. 22. 4.3 Methodology. 26. 5. Results. 26. 6. Conclusion. 35. 7. Discussion. 36. References. 37. 5.

(6) 1. Introduction. 1.1 Background Corporate social responsibility (CSR) reporting became important since 1989. A few companies began with publishing of environmental reports. Over the years the importance of this sustainability reports increased substantially. In 2002 35% of the Dutch companies were publishing a sustainability report compared to 5% in 1993 (Kolk, 2004). The CSR report has gone through a lot of changes. In 2002 the latest version of the governmental reporting initiative was released (Flower, 2015). There are many standards of CSR reporting but the three most important and widely accepted standards in the field of CSR reporting are Global reporting initiative (GRI) G4 standards, AccountAbility’s AA1000 series, and the united nations compact communication on progress (Tschopp and Huefner, 2014; Flower, 2015). The need for companies to become more transparent and more accountable increased. This request is coming from the shareholders and stakeholders of the companies. For this reason it is important for companies to respond to this. The CSR report offers the possibility to be more transparent and accountable and show this to the shareholders and stakeholders (Kolk, 2004). The pressure to meet shareholders expectations and to meet the set targets is high. When the targets are and/or the expectations of the shareholders are too high the managers of the companies have incentives to engage in earnings management. This means that they are intentionally influencing the process of financial reporting. This comes at the expense of earnings quality. There are two different types of earnings management accrual based earnings management and real earnings management. At accrual based earnings management the economic reality of the company is not affected. Real earnings management does affect the economic reality of the company and influences decisions that affect the economic position of the company (Dechow, 2012; Hong and Andersen, 2011). 6.

(7) There is done a lot of research after earnings management and the motivation to engage in earnings management. CSR reporting is also a subject that is done a lot of research after (Chan et al., 2014; Holder-Web et al., 2009; Murphy and Schlegelmilch, 2013). The companies that are engaging in CSR reporting increased in recent years (Kolk, 2004). The consequences of companies that engage in CSR reporting are related with less earnings management and have a higher earnings quality (Kim et al., 2012). Following Murphy and Schlegelmilch (2013) CSR reporting is subjective and the values related to CSR reporting are depending on the time period. In the fifty’s the focus of CSR was on the decision making of managers and the obligations to the community. Comparing this to the CSR nowadays where the focus is on the larger corporate society (Murphy and Schlegelmilch, 2013).. Following Cohen et al. (2008) there is evidence that after the introduction of the Sarbanes-Oxley (SOX) act the accrual based earnings management is reduced. Therefore it is interesting to look if this decrease is the result of the introduction of SOX or if there are other factors involved in this reduction. The factor that is looked after in this research is CSR. Are companies that are engaging in CSR engaging less in earnings management. In this thesis there is researched if these factors have a correlation and significant negative influence to each other. In this research there is going to be researched if companies that engage in CSR reporting are less likely to engage in earnings management. This is done for nonfinancial firms in the U.S. for two different time periods namely 2004 till 2008 and the second time period will be 2009 till 2013. In this time period the CSR reporting increased substantially and because of that it is possible to look at the increase or decrease of earnings management as a result of the CSR reporting (Kolk, 2004). This is done because CSR reporting is changing a lot in this period and went through a big growth. It is interesting if earnings management is decreasing in this period. Also it is important to detect the reason behind this decrease of earnings management. Are there other factors that influenced the reduction of earnings management besides SOX. (Cohen et al., 2008). 7.

(8) 1.2 Research Question CSR reporting has gone through a lot of changes and the companies that engage in CSR reporting increased substantially. Nowadays it is important for companies to be transparent and accountable, CSR reporting is an important factor of this process (Kim et al,; 2012; Lindblom, 1994; Murphy and Schlegelmilch, 2013). Earnings management has a negative influence on the quality of earnings and the transparency and accountability of the company. Accrual based earnings management. 1. is not influencing the economic reality of the company but does influence the financial reporting. Real earnings management2 influences the economic reality of the company. This ensures that the financial reports of the companies that engage in earnings management are not giving a true and fair view of the company (Scholtens and Kang, 2013). Because of the increase of the CSR reporting and the need for transparency and accountability there may be a relation between CSR reporting and earnings management. In this research the discretionary accruals are used to measure earnings management. Therefore only the accrual based earnings management is measured. This is because this is easier to influence for managers. Based on the literature it is interesting to test if there is a relation between this type of earnings management and CSR. (Hong and Andersen 2011; Kim et al., 2012) This leads to the following research question: Does the growth of CSR reporting influence earnings management?. 1. Accrual based earnings management is measured by the discretionary accruals. For example firms can. influence their warranty liabilities by changing the estimates. This has influence on the accruals but not on the actual future cash flow (Hong and Andersen 2011) 2. Real earnings management does influence the economic reality of the firm. An example is that managers. manipulate their sales revenue by modifying their credit firms. This influence the future cash flow and therefore the economic reality of the firm (Hong and Andersen, 2011). 8.

(9) 1.3 Motivation CSR reporting nowadays is far from ideal. It is going through a lot of developments and changes fast. Following the research of Carnegie and Napier (2002) there are seven relevant factors that are important for CSR and accounting. These factors are: period, place, people, practice, propagation, product and profession. These factors indicate that CSR is changing in periods of time and also in different places and professions. Period has an influence on CSR because the culture changes and the society values CSR different in different periods of time. Place is an influencing factor because there are different cultures and different standards and values for CSR in different cultures. This is dependent of the geographic area. By people is meant the accountants. To who have they report and what are their values. Practices means the accounts. How are these prepared and what do they include and what is excluded. This is important to understand for comparison of the accounts. Propagation is the how the accounting ideas are formed and spread. This has an influence on the image of accounting. Propagation can be formal and informal. Product is produced by the accountant. Is this the accountant report or is there additional information like a CSR report. The last factor is profession. What kind of image and activity’s does the accounting and CSR have in the community. How is the professional status of both (Carnegie and Napier, 2002). Following the paper of Tschopp and Huefner (2014) CSR reporting is far from perfect. There are a lot of different standards. This makes the reports difficult to compare and it is hard to see the structure. CSR reporting has to develop further, to make the reports easier to read for the users and regulate the standards (Tschopp and Huefner, 2014). For this research the growth and development of CSR reporting are important to compare for the companies that are engaging in CSR and the companies that do not engage in CSR. This research is an complement to the existing research by making the distinguish between two time periods namely the first time period is 2004 till 2008 and the second time period will be 2009 till 2013.. 9.

(10) The remainder of this thesis proceeds as follows. Section 2 will outline the theory and existing literature. Section 3 describes the hypothesis, section 4 explains the different variables and describes the research methodology. Section 5 will present the results of the statistics and test the hypothesis. In Section 6 the research question is answered and in the final section the discussion with the limitations and suggestions for further research is presented.. 2. Literature review In this chapter the most relevant concepts, literature and theories are discussed to answer the research question. First CSR reporting is explained followed by earnings management, agency theory, legitimacy theory, stakeholder theory and at last the SOX act is discussed.. 2.1 CSR reporting A dated but influential definition of CSR came from Carrol (1979): the social responsibility of business encompasses the economic, legal ethical and discretionary expectations that society has of organizations. CSR became a lot more popular recently (Kolk, 2004). Hong and Andersen (2011) found that CSR and financial performance are positively related. This is because of the stakeholder theory (Hong and Andersen, 2011). By engaging in CSR the company tries to satisfy the stakeholder expectations. By satisfying the stakeholder expectations the investors may be positively influenced and because of that financial performance will increase (Hong and Andersen, 2011). Companies with a good management can be socially responsible and turn that in a competitive advantage for the company (Cohen, 2009). CSR has grown in the U.S. between 1995 and 2002 and CSR expenditure tripled (Holder-Web et al., 2009). The demand for non-financial reporting is increasing and companies respond to this. Especially the stakeholders are asking for non-financial reports. There can be different stakeholders of the audit firms. The current issues in stakeholder dialogue and engagement are that it is difficult to identify the range of the stakeholders, for some stakeholders it is impossible to engage in the discussion. For. 10.

(11) these stakeholders there are NGO’s (Non-governmental organizations). NGO’s are representing the interest of these stakeholders. Another problem is addressing the homogeneous stakeholder groups and negotiating the consensus in these groups. These are problems by providing a CSR report (Rinaldi et al., 2014). By providing a CSR report an audit firm has two motives. An ethical motive and an economic motive. The ethical motive holds that all stakeholders have the same rights and that the CSR report has to provide a report of all these stakeholders by treating them equally. The economic motive holds that the CSR report is addressed to the stakeholders with the greatest economic influence. At the economic motive not all the stakeholders are treated equally (Rinaldi et al., 2014). Porter and Kramer (2006) found four motives to engage in CSR. These motives are strong related with the ethical and economic motive of Rinaldi et al. (2014). The first is moral obligation, this means that companies have to behave correctly to be good citizens. The second is sustainability, companies have to behave sustainable so that the society in the future will not be worked against by the operations nowadays. The third motive is license to operate. This holds that the companies have to get the permission from the government and other legal instances to get the license to operate. The last one is reputation. By improving the company’s image the value of the stock of the company may increase. The CSR reports are in most cases not mandatory. The only case where a CSR report is mandatory is South Africa (Dawkins and Ngunjiri, 2008). This means that the CSR reports do not have to comply with certain rules. It is difficult for the stakeholders to read these reports because of the lack of standards. For this reason the GRI provided performance regulators. To make these CSR reports more standard it would be easier for the stakeholders to see the difference between the CSR reports. In this way it is easier for the stakeholders to classify these reports and recognize the strong and weaknesses (Kolk, 2004). Because CSR reporting is not mandatory there is not a single standard to produce a CSR report. Although the three most widely recognized standards are the Global. Reporting. Initiative. (GRI),. the. United. Nations. Global. Compact’s. 11.

(12) Communication on Progress and the AccountAbility’s AA1000 series (Tschopp and Huefner, 2014; Flower, 2015) The latest version of the GRI is published in 2013. The guidelines are consisting of reporting principles, guidance and disclosures. The guidelines relate to five content areas: strategy and analysis, organizational profile, report parameters, governance commitment and engagement and finally management approach and performance indicators (Flower, 2015). The United Nations Global Compact’s Communication on Progress comes from 2000. It is based on ten principles of human rights, labor standards, environmental standards and anti-corruption principles (Tschopp and Huefner, 2014). This standard has no certain rules but through a statement a company can show their commitment to this standard. The last standard is the AccountAbility’s AA1000 series. This is a principle based standard. By following these principles a company can enhance their sustainability. The principles of transparency, responsiveness and compliance are improved when a company stays consistent with the AA1000 series (Tschopp and Huefner, 2014). CSR is dependent of a lot of different factors. The factors that are used in this research are: community, corporate governance, diversity, employee relations, environment, human rights and product. These factors are also used to measure if a company is engaging in CSR. The most important factor is that the corporate governance is perceived as a distinct construct from CSR. Corporate governance has a big influence on the financial reporting of a firm (Kim et al., 2012). The relation with the community where the firm is operating is also a factor that influences the CSR of a firm. Communities can ask for CSR engagement or they do not care about it. To keep the community and the stakeholders satisfied it is important that the firm will listen to the needs of the community. Diversity is important for CSR because if a company is diverse there are a lot of different interests, it is important for a firm to listen to a lot of different interests. Another factor is relationship with the employees. Has a firm good or bad relationship with the employees. Is the employee turnover small or large.. 12.

(13) Employees have a big influence on the firm. If the employees are strongly committed to the company can it influence the engagement from the firm in CSR. The environment is important because CSR and sustainability are inextricably linked to each other. The environment is mostly a key stakeholder. Because it hard to determine the wishes of the environment it is hard to know what is the best to do. There are a lot of action groups who represents the interests of the environment. By carrying about the environment the goodwill of the stakeholders will grow. Human right is also a factor. This means if the firm respects or violates the human rights. The last factor which is important is the product which the firm is selling. This has influence on the CSR of the company because if it is a sustainable product the company will care more about the CSR than when it is a polluting product (Kim et al., 2012).. 2.2 Earnings management As previously discussed there are two different types of earnings management, real earnings management and accrual based earnings management (Hong and Andersen, 2011). Managers have different reasons to engage in earnings management. One of the main reasons that they want to influence the earnings is that if their bonus is dependent of their performance (Hong and Andersen, 2011). In most of the cases this performance is measured by the earnings. By engaging in earnings management the management can influence the earnings and therefore also the bonus they will receive. For that reason it is important to look at which system the managers will be rewarded to see their motives. Accrual based earnings management does not influence the future cash flows. An example of accrual based earnings management is to change the estimates of the warranty (Hong and Andersen, 2011). By estimating the warranty the accrual part of the earnings will change but the actual future cash flows will not be influenced. Following the paper of Healy (1985) accruals and managers’ income-reporting incentives under their bonus contract are strong related. Managers are more likely to choose incomedecreasing accruals when their bonus plan upper or lower bounds are binding, and income-increasing accruals when these bounds are not binding (Healy, 1985). To find material risks it is important to look at the reward system of the managers. Another. 13.

(14) reason to engage in earnings management is to influence the outcome of contracts. Although accrual based earnings management does not influence future cash flows and future economic decisions it decreases earnings quality (Kim et al., 2012). Real earnings management means that companies modify their actions which results in that the economic reality of the company will be affected (Hong and Andersen, 2011). An example of real earnings management is that a company reduces the R&D expenses and therefore increases the earnings. Mostly the real earnings management actions increase the earnings only on short-term. For example reducing R&D costs may increase earnings now but looking at the long term the company will not come with innovations which are necessary to keep the economic growth on the long term. A reason that firms are engaging in real earnings management to meet forecasts of financial analyst. By meeting these expectations firms are not presenting the economic reality (Roychowdhury, 2006). Earnings management can have disastrous consequences. Examples of this are the scandals of Worldcom and Enron (Jiraporn et al., 2008). By introducing the SOX act the government wanted to reduce earnings management and accordingly reducing the scandals. Legislation makes is difficult to engage in earnings management. Before the introduction of the SOX act firms where more engaging in accrual based earnings management after the introduction of the SOX act firms are more engaging in real earnings management. This shift is because real earnings management is harder to detect (Cohen et al., 2008). Earnings management is hard to detect. On basis of accruals it is possible to measure if a firm is engaging in earnings management. There are different measures for accrual based earnings management and real earnings measurement. Accrual based earnings management can be measured on basis of discretionary accruals (Hong and Andersen, 2011). These are expenses that are non-obligatory that are yet to be realized but is recorded in the financial statements. For real earnings management there are different measures needed. The measures that affect the real economic situation of a firm are for example abnormal levels of operating cash flows and abnormal production costs. In this research is focused on the discretionary accruals because based on the. 14.

(15) agency, legitimacy and stakeholder theory CSR engagement may have an effect on accrual based earnings management.. 2.3 Agency theory The agency theory holds that one party, the principal delegates work to another party, the agent. In this theory there are two problems. The first problem is that the agent has different goals compared to the principal. The goal of the principal is to create a firm value as high as possible, the agent wants to create his own value as high as possible. These goals can be conflicting. The second problem of the principal agent theory is that the principal and the agent both have different risk appetite (Jones, 1995). There are different costs in the agency theory. Monitoring costs and moral hazard costs. Monitoring costs are the costs the principal makes to monitor the actions of the agent. The principal has to make a consideration between the costs of monitoring and the costs of shirking of the agent. In the case of a high monitoring level the principal reduces the possibility the agents shirks but the monitoring costs will increase. If the principal decreases the monitoring level and therefore the monitoring costs the possibility that the agents will shirk increases, this will also bring costs. It is not possible to monitor all the actions of the agent for the principal. Other costs that occur at the agency theory are moral hazard costs. The costs that occur if the agent has a lack of effort. Because there is an information advantage for the agent the principal cannot verify the effort of the agent exactly. This will bring costs (Jones, 1995). Efficient contracting is important to reduce these costs. There are two parts of an efficient contract. The explicit part and the implicit part. Under the explicit part falls the fixed salary of the agent. The explicit part of the contract is provable in court. If one of the parties shirks, for example the agents refuses to work or the principal refuses to pay it is provable for both parties in court (Pearce and Stacchetti, 1988). The implicit part of the contract is more complex. This contains the bonuses. This part of the contract is dependent of a variable mostly related to a target. It is very difficult to set a target. The target cannot be too easy, because when the agent reaches the target he will shirk. In contrast by setting a target that is too high and the agents. 15.

(16) realizes that the agents will not put effort in reaching the target (Pearce and Stacchetti, 1988). For efficient contracting both, explicit and implicit contracting are important. It is difficult to find the right composition. The most efficient contract will also be dependent on the risk appetite of the agent and principal and will therefore be different for every job (Pearce and Stacchetti, 1988). Following the paper of Jiraporn et al. (2008) the level of earnings management is positively related to the agency conflicts. This means that if there are relatively many agency problems in a firm the managers will engage more in earnings management. This is important because in some cases it is hard to detect earnings management. It will make it easier if the motives to engage in earnings management will be clear.. 2.4 Legitimacy theory Firms can choose for different legitimation strategies. This is explained in the paper of Suchman (1995) and Lindblom (1994). In the paper of Lindblom are four different broad strategies of legitimation. The first one is the one that the firm is actually changing its behavior. By changing the actual behavior the company wants to meet the expectations of the key stakeholders. At the second theory the company is not changing the actual behavior but the company tries to convince the stakeholders through social constructed knowledge that it has changed its behavior. The third theory explains that the company not changes its actual behavior but distract the attention of the stakeholders by focusing on positive projects. The last legitimation theory following Lindblom (1994) holds that the company tries to change the expectations of the stakeholders and do not change the actual behavior by changing the expectations of the key stakeholders the company does not have to change their own behavior (Lindblom, 1994). In the research of Suchman (1995) are the different kinds of legitimation theories divided in three primary forms: Pragmatic, moral and cognitive. For the term legitimation Suchman (1995) uses the following definition “Legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or. 16.

(17) appropriate within some socially constructed system of norms, values, beliefs, and definitions” Suchman (1995) makes the difference between these three forms of legitimation because these types are based on the difference in behavior dynamics in an organization. The pragmatic legitimacy is based on the audience and self-interest. The moral legitimacy is based on the normative approval and cognitive legitimacy is based on the taken-for-grandness. In the research of Milne et al. (2009) there is found that the communication of organizations are suggesting the companies are engaging in CSR and are therefore sustainable. In the research of Milne et al. (2009) there is taken a close look at these reports and the actual actions in CSR and sustainability. Therefore it is important to not take these CSR reports as taken-for-granted but actually take a closer look at the companies if the report is presenting a true and fair view. The lack of interest of investors is a risk in this case because the investors do not put that much effort in the research for in which organization they want to invest. In this case the reports can have a misleading role (Christensen et al., 2014). Legitimacy can be referred to the social contract. This is a contract that exists between organizations and individual members of the society. When the organization does not hold to the social contract society will punish the organization by for example not buying the product. For this social contract it is important that the organization acts legitimate and to report this to outsiders like investors or consumers it can use CSR reporting (Deegan, 2002).. 2.5 Stakeholder theory According to Salancik and Pfeffer (1978) organizations have to rely on stakeholders. These stakeholders will provide support to the organizations but may ask some actions in return. There are internal stakeholders and external stakeholders. Salancik and Pfeffer (1978) are discussing the external stakeholders above. The external stakeholders of an organization are the suppliers, society, government creditor’s shareholders and customers. An organization has also to do with internal stakeholders these are the employees, managers and the owner. It is difficult and almost impossible to satisfy all. 17.

(18) these stakeholders because they may have different conflicting interests. Therefore important for the stakeholder theory is to know who the key-stakeholders are. According to De Villiers et al. (2011) the stakeholder theory includes taking into account the interest of stakeholders. This can be done by carrying about the environment where the stakeholders are living or produce environmentally friendly goods and services. Taking into account the interests of shareholders and consider the environment can a firm also ward off disasters. For example the oil spill of BP, if BP considered more about the environment and their stakeholders the disaster would not be this big (de Villiers et al., 2011). Stakeholder theory means like CSR reporting that a management of the company does more than making profit and strategic decisions in favor of the company by taking into account the interest of the stakeholders and care about the environment (Van der Laan Smith et al., 2005).. 2.6 Sarbanes-Oxley act In 2002 the SOX act is in U.S. introduced. This act has as goal reducing earnings management. The result of SOX is that firms where less engaging in accrual based earnings management but more in real based earnings management because this is harder to detect (Cohen et al., 2008). The SOX act is based on penalties this in contrast to CSR which is based on the social contract, behaving sustainable and act in favor of the stakeholders and environment and works based of the legitimacy of the community. Following Cohen et al. (2008) there is significant evidence that after the introduction of SOX the earnings management decreased.. Concluding from the theories that are explained in this chapter, the relationship between the environment and the stakeholders of a firm is important but also the relationship between the principal and the agent is important. These are factors that have influence on the welfare of the firm and therefore on the earnings quality and CSR. To meet the expectations and take into account the interest of the stakeholders is important but. 18.

(19) unrealistic. Therefore it is important to identify the key stakeholders of the firm. Also it is important to look at the factors that may reduce earnings management besides SOX.. 3. Hypothesis In this chapter the development of the hypothesis is discussed. With explaining the relationship between earnings management as a result of CSR. After that the hypothesis is given by which the relationship between earnings management and CSR tests in this research. The relation between CSR and earnings management can be clarified in two different ways. The first is that a highly CSR concerned firm engages in earnings management to meet the expectations of the stakeholders and shareholders. These can be reputation concerns and/or financial performance (Kim et al., 2012). Mostly CSR is seen as a positive image a firm can create. But is also a costly strategy for firms. By being more sustainable there are matching activities and products that costs a lot of effort and money. This can be a motive for firms to engage in earnings management. Because the firms do not want to perform the financial statement with less earnings (Kim et al., 2012). Investors are having a lack of interest and do not take in general a close look to the changes in the financial statements (Christensen et al., 2014). Therefore they will not pay attention to the higher costs that are a result of the CSR but they will only see the decrease in earnings. Therefore firms will engage in earnings management as a result of CSR because they not want to show their decreasing earnings to the investors. In the research of Sloan (1996) there is made a distinguish between three types of earnings manipulation: expenses manipulation, revenue manipulation,. margin. manipulation. These types are important at CSR because by these three types of earnings manipulation will dissemble the extra costs of CSR and therefore these costs will not have an effect on the stock price. Why are firms engaging in CSR? It costs a lot of effort and can therefore can result in decreasing revenues (Chih et al., 2008). This can be seen in the stakeholder theory. It is important to satisfy the stakeholders of the firms. This can bring a lot of. 19.

(20) advantages (Chih et al., 2008). Advantages like litigation, because firms are more sustainable they will have less litigation problems. Another advantage can be that because of the firms meet the stakeholder expectations the stakeholder are more generous and will buy the product and stick to the firm. On the other hand earnings management can be seen as agency costs. Managers maximize their own wealth by managing the earnings and so reaching their targets and therefore get the bonus (Scholtens and Kang, 2013). Each firm is different in driving the agency costs as low as possible. A way of reducing these costs is minimize the difference in interest between the management and stakeholders. A way of minimize these interests may be improve their CSR. Seen from the point of view of the legitimacy theory it is important to look at the relationship between CSR and earnings management. The four different legitimacy strategies that a firm can choose (Lindblom, 1994) can be seen in the earnings management. If a firm decides to improves their CSR but still engages in earnings management it is likely that the firm is not changing their actual behavior but is convincing the stakeholders that it does. Theory points out that CSR gives an incentive for firms and management to be honest and act ethically right in their business processes. Managers want to do the right thing in a firm where CSR is important (Kim et al., 2012). In this case there is expected a negative relationship between CSR and earnings management. By this is meant that if a firm is involved in CSR the level in earnings management will decrease. This will be tested based upon the literature review. To research the relationship between CSR and earnings management the following hypothesis will be tested: H1 Earnings management is negative related to CSR 4. Research methodology To test the hypothesis the following control variables are used: size, return on assets, leverage, growth and big4. These variables will be explained in this chapter. The descriptives that are tested are CSR and discretionary accruals. These variables will also. 20.

(21) be explained. The chapter will be ended with the regression models to test the hypothesis. 4.1 Data and Sample The data that is collected is from non-financial firms in the U.S. There is chosen for non-financial firms because the nature of the earnings of financial firms is different than from non-financial firms. It is not possible to compare these firms with the nonfinancial firms. (Lucey and Zhang, 2011) Also the outcome of the research will be more interesting for non-financial firms because earnings management is relatively easy to detect and the engagement in CSR is increased in recent years. The choice for firms in the U.S. was made because there are many different companies active and a lot of differences between these companies. This ensures that it is a wide research. The time periods that are used are two different periods to compare the development of CSR reporting. This is because CSR is going through big changes in the last decennium therefore it is interesting if these changes are influencing earnings management. Also Cohen et al. (2008) shows that after the introduction of SOX earnings management is decreased. Is SOX the only reason for this decrease and what is the role of CSR in this decrease. For the data to compute the discretionary accruals there is made use of the Compustat database. The characteristic for the Compustat database is that it is standardized to ensure comparability. This is necessary for this research because there will be compared two time periods. The first time period will be 2004 till 2008 and the second time period is from 2009 till 2013. To compute the discretionary accruals for this period we need the financial data from the U.S. Non-financial firms from 2003 until 2008. This also applies to the second time period. This period is from 2009 till 2013. For this period is data needed from 2008 until 2013 to compute the discretionary accruals for this period. This is necessary because the changes in the data have to be computed therefore the data is needed from one year before the actual period that is going to be used to test the hypothesis. To measure CSR there is made use of the KLD database. This database is known for its social research for institutional investors. KLD researches benchmarks,. 21.

(22) compliance and consulting services. It is the largest database for CSR worldwide. The KLD provide high quality consistent research. This database is perfect to measure if a company is corporate social responsible. The data base contains data from 3000 large U.S. companies. To compare the data from the Compustat database and the data from the KLD database the data has to be filtered to connect properly to each other. Otherwise it is not possible to compare the datasets and to see if there is a relation between earnings management and CSR. In the Compustat database is financial information from 10.000 companies for this period. In the KLD database is data available from 5451 different companies. After deleting the companies with missing values, matching the two different datasets from KLD and Compustat and 5% of the outliers consists the dataset of 810 companies.. 4.2 Measurement of variables The variables that are used to measure the relation between earnings management and CSR are consisting of descriptive variables and control variables. The descriptive variables are CSR and discretionary accruals. The discretionary accruals are used to measure earnings management. The control variables are used to avoid the problem of correlated omitted variables (Kim et al., 2012). These control variables are chosen on basis of existing literature. (Chih et al., 2008; Kim et al., 2012; Roychowdhury, 2006). The first descriptive variable is discretionary accruals. The discretionary accruals are computed by the modified Jones model. The Jones model found in 1991 assumes that the discretionary accruals stay constant. An assumption indirect to the model is that revenues are non-discretionary. This model removes part of the earnings that are managed through the discretionary accruals. The modified Jones model is designed to measure the non-discretionary accruals with errors when the discretion is exercised over revenues (Dechow et al., 2012). Dechow et al. (2012) provides a test to detect earnings management between several models. In this research is found that the modified Jones model provides the most powerful test to detect earnings management. For this reason this test is used in this research to detect earnings management. To use the modified. 22.

(23) Jones model the total accruals have to be computed first. The total accruals are computed following Chih et al. (2008) with the following formula: = (∆. ) − (∆. − ∆. −. )−. −∆. After the total accruals are computed the coefficients from the following formula are determined. ∆. =. )+. (1/. (∆. )+. (. )+. =. Total accruals scaled by lagged total assets. =. Total assets at τ-1. =. Revenues in year t less revenues in year t-1 scaled by total assets at τ-. =. Gross property plant and equipment in year t scaled by total assets at. =. Firm-specific coefficients estimates. =. Error term. 1. τ-1 ,. ,. The coefficients that are determined in the equation stated above are used in the modified Jones model to compute the non-discretionary accruals. Non-discretionary accruals are measured by. =. +. 1. +. (∆. −∆. )+. (. )+. =. Estimated non-discretionary accruals at τ. =. A year subscript indication a year in the event period. =. Total assets at τ-1. ∆. =. Revenues in year τ less revenues in year τ-1 scaled by total assets at τ-. ∆. =. Net receivables in year τ less net receivables in year τ-1 scaled by. τ. 1. total. assets at τ-1. 23.

(24) =. Gross property plant and equipment in year τ scaled by total assets at. =. Firm-specific coefficients. τ-1 ,. ,. With the modified Jones model the non-discretionary accruals are computed. In this research the accrual based earnings management is tested by the hypothesis. For this type of earnings management the discretionary accruals are needed. To compute the discretionary accruals the following formula is used: DA =. −. The second descriptive variable in this research is CSR. To determine the engagement in CSR there is going to be made use of the KLD database ratings. In the KLD database is designated as the best source for measuring social responsibility and has many advantages (Hong and Andersen, 2011). These advantages are independency and the large number of companies that are developed in the database. This is important for this research to determine the CSR engagement of the non-financial U.S. firms. In this database there are seven issue areas. In these areas there is data for strengths and concerns. In this research only the strengths are used because the engagement in CSR is going to be researched and the influence on earnings management. For this reason there is not going to made use of the weakness data in the seven areas. These seven areas are: community, corporate governance, diversity, employee relations, environment, human rights and product. Besides the descriptive variables there are control variables needed to determine the relation between earnings management and CSR and finally to test the hypothesis. These control variables are needed to avoid the problem of correlated omitted variables (Kim et al., 2012). There is made use of four control variables and one dummy variable.. 24.

(25) The control variables are: Size The first control variable that is used is size. Size relates to the size of the firm and is measured by the log of the firm’s total assets. This is Item 6 in the Compustat database. Following Roychowdhury (2006) size is an important control variable because the size of the firm can explain the engagement in earnings management. In this research there is clarified the relation between size of the firm and earnings management. There is determined that the influence is significant. There is also a correlation between size and the engagement of CSR of a firm. Return On Assets (ROA) is the second control variable. ROA indicates the financial performance of the firm by the return on assets. This variable is computed by dividing net income by total assets. Net income is item 18 in the Compustat database and total assets is item 4 in the Compustat database. The control variable ROA is used to exclude the ethical aspect of CSR on earnings management after controlling for the potential effect of financial performance (Kim et al., 2012) The third control variable is Leverage. This variable relates to the total debt scaled by total assets. Total debt is item 9 in the Compustat database and total assets is item 4. Following Kim et al. (2012) the variable leverage is important because this controls for the equity related incentives to engage in earnings management. The last control variable is Growth and this variable indicates the growth of the company. This is measured by the percentage change in revenues from year t-1 to year t. this variable is included because fast growing companies have to legitimize their actions and financial statements more specifically so for them there is less opportunity to engage in earnings management. But to keep the growth they these firms have also incentives to engage in earnings management. Therefore this control variable (Chih et al., 2008) The dummy variable is Big 4. This indicates whether a firm is audited by a big 4 audit firm or not. If a firm is audited by a big four audit firm the outcome is 1 otherwise the variable will take the value of zero. This dummy variable is taken into account because earnings management might differ for firms that are audited by a big 4 (Kim et al., 2012).. 25.

(26) 4.3 methodology The hypothesis is tested by the following regression model: =. +. +. +. 4 +. +. +. +. In this model is DA the dependent variable which measures the earnings management. CSR is the total score of the seven areas. If. is negative and significant than. hypothesis 1 is supported. The other variables SIZE, ROA, LEV and GROWTH are control variables and big4 is a dummy variable which can take the value 0 or 1.. 5. Results In this chapter the results of the data are shown and discussed. First the descriptive statistics of the discretionary accruals and the CSR are calculated, then the descriptive statistics of all the variables are calculated for the two different time periods. After that the correlation matrix is given for the variables and the two different time periods. The chapter is closed by the regression model of CSR on discretionary accruals.. 26.

(27) Table A CSR index Year. Community. Diversity. 2004. Corporate governance 13,8%. 10.8%. 2005. 14,6%. 2006. Environment. 38,9%. Employee relations 22,5%. Product. 9,6%. Human Rights 0,4%. 5,2%. CSR total 60,1%. 11,7%. 42,5%. 22,8%. 14%. 0,6%. 5,3%. 62,3%. 16%. 12,5%. 42,9%. 27,4%. 17,1%. 0,9%. 6,1%. 64,4%. 2007. 16,5%. 13%. 44,3%. 31,4%. 20%. 1%. 6,5%. 67,3%. 2008. 16,1%. 12,4%. 45,3%. 36%. 22,9%. 1,4%. 6,4%. 70,4%. 2009. 7,7%. 12,4%. 45,9%. 36,1%. 22,9%. 1,4%. 6,3%. 70,9%. 2010. 18,2%. 25,3%. 37,6%. 25,4%. 44%. 3,2%. 5.9%. 60,4%. 2011. 18,2%. 25,3%. 37,6%. 25,4%. 44,1%. 3,3%. 6,4%. 61,2%. 2012. 16,6%. 24%. 26,3%. 36,9%. 24,9%. 5,8%. 7,2%. 63,9%. 2013. 3,4%. 21,9%. 28%. 51,6%. 36,7%. 5,3%. 6,4%. 67,8%. Table B Discretionary accruals Data Year - Fiscal. N. Minimum. Maximum. Mean. Std. Deviation. 2004. DA. 810. -26,24. 49,53. ,0277. 2,32006. 2005. DA. 810. -92205,43. 47544,28. -56,7590. 3697,54954. 2006. DA. 810. -10,13. 157,66. ,3132. 6,30296. 2007. DA. 810. -62,48. 9,64. -,1861. 2,57943. 2008. DA. 810. -10,72. 8395,21. 11,6253. 300,91813. 2009. DA. 810. -18768,57. 36,39. -24,1792. 673,31986. 2010. DA. 810. -30,54. 30,82. -,0408. 1,95783. 2011. DA. 810. -17,08. 92,11. ,0611. 3,47767. 2012. DA. 810. -7,52. 27,24. -,0096. 1,21674. 2013. DA. 810. -14,92. 35,72. ,0107. 1,78683. All tables are computed in SPSS. 27.

(28) In table A the CSR index for the 810 firms is calculated. In this table there is calculated the percentage of the CSR firms. The engagement for the companies is also calculated for each different category. From the table can be seen that the overall score of CSR engagement of the firms has increased from 60,1% to 67,8% looking at the percentages more closely at the different years. There is an increase until 2009 and after that the CSR engagement of the firms dropped down. But in the last years of the research CSR made a big comeback. This increase is namely in the categories employee relations and environment. A fall of CSR is also shown in employee relations. This is the most fluctuating factor together with corporate governance. This can be clarified on basis of the financial crisis. A lot of people got fired and the attention of the company’s where not on the corporate governance and employees but on surviving the crisis. In 2013 there is shown a big increase. Companies have the opportunity to strengthen the employee relations and to look after the environment. This is a trend in the community which is also a lot more important looking at the table. Can be clarified from the perspective of the social contract (Deegan, 2002) There are also fluctuations in the measure of the earnings management the discretionary accruals. The descriptive statistics of the discretionary accruals are calculated in table B. In the years with the big fluctuations in the mean of the discretionary accruals the standard deviations are very large comparing to the other years. This indicates that in years of economic uncertainty the value of the accruals fluctuate a lot across the different companies. This can be clarified through that some companies be harder hit by the economic uncertainty and the financial crisis than others. In this research the discretionary accruals are a measurement for earnings management. On basis of table B there is shown that earnings management is lying far apart between the different companies in the years 2008 and 2009. These are the years of the financial crisis. In the years starting from 2010 the discretionary accruals are more stable and lower. It is difficult to conclude if this is the result of the introduction of the SOX act. But the situation is improved these years.. 28.

(29) Table C.1 2004 -2008 Descriptive statistics of all the variables N. Minimum. Maximum. Mean. Std. Deviation. CSR. 4050. 0. 7,00. 1,8640. 2,50895. DA. 4050. -92205,43. 47544,28. -9,0193. 1659,20790. Size. 4050. 6,44. 17,69. 12,7486. 2,08314. ROA. 4050. -60,24. 318,62. ,4992. 6,80549. LEV. 4050. 0. 214,34. ,4156. 3,87625. Growth. 4050. 0. 3702,47. 4,1508. 65,30483. Auditor. 4050. 0. 1. ,96. ,204. Table C.2 2009 -2013 Descriptive statistics of all the variables N. Minimum. Maximum. Mean. Std. Deviation. CSR. 4050. 0. 7,00. 2,6620. 3,47948. DA. 4050. -18768,57. 92,11. -4,8018. 300,19924. Size. 4050. 5,48. 18,02. 12,8243. 2,17499. ROA. 4050. -71,40. 175,93. ,2913. 3,88852. LEV. 4050. 0. 66,71. ,3570. 1,87232. Growth. 4040. 0. 1131,04. 3,8500. 42,02837. Auditor. 4050. 0. 1. ,96. ,199. 29.

(30) In table C.1 and table C.2 are the descriptive variables calculated from all the variables. In table C.1 for the first time period 2004 till 2008 and table C.2 for the second time period 2009-2013. The number of observations is for each time period 4050. The CSR score is measured together for the seven different areas. Therefore the total score is CSR is taken. Comparing the two tables shows that the CSR is increased and the discretionary accruals decreased and less fluctuating resulting in a strongly decreased standard deviation. The other variables are shown because they are the control variables in the regression model. The last variable auditor is a dummy variable. It can have the outcome 0 or 1. The outcome 0 means that the firm is not audited by a big 4 auditor. If the outcome is 1 the firm is audited by a big for auditor. The firms in this research are almost all audited by a big 4 auditor. The standard deviation decreased a little in the second period. This can be clarified because there is needed a lot of financial and CSR information from the companies. Often this is available for the big companies that are automatically audited by a big 4 audit firm. Looking at the control variables to compare the two periods the size of the firms did almost not change. The return on assets decreased a little. The leverage decreased for the second period. This is also the case for the growth of the firms. This can be the result of the financial crisis which is affecting the second time period.. 30.

(31) Table D.1 2004 -2008 Correlation matrix Pearson Correlation DA. DA. CSR. Size. ROA. LEV. Growth. Auditor. 1. CSR. 0,013. 1. Size. 0,012. -0,248. ROA. 0,004. -0,015. LEV. 0,069. -0,026. 0,017. -0,003. Growth. -0,227. -0,020. 0,026. -0,002. 0. 1. Auditor. -0,001. 0,055. 0,000. 0,004. 0,016. 0,005. 1 0,035. 1 1. 1. Correlation is significant at the 0.05 level (2-tailed). Table D.2 2009 -2013 Correlation matrix Pearson Correlation DA. DA. CSR. Size. ROA. LEV. Growth. Auditor. 1. CSR. 0,012. 1. Size. -0,010. -0,292. 1. ROA. 0,001. 0,001. 0,020. 1. LEV. -0,608. -0,018. 0,033. -0,009. 1. Growth. 0,001. -0,002. 0,034. -0,001. -0,005. 1. Auditor. -0,003. 0,118. -0,024. 0,023. 0,029. -0,017. 1. Correlation is significant at the 0.05 level (2-tailed). 31.

(32) In table D.1 and D.2 are the correlations calculated of the variables of the regression model. Therefore is used the Pearson correlation. This gives an indication of the relationship between the variables. If there is a positive outcome the relationship holds that is one variables increase the other also increases. If the outcome is negative it holds that if one variable increases the other decreases and vice versa. The correlation between discretionary accruals and CSR is positive. This does not support the hypothesis of this research which tests the negative relationship between CSR and earnings management. There is almost no difference in the correlation between the two different time periods. The correlation ROA and CSR went from negative to positive comparing the two different time periods. This indicates that if the performance of firm is good than the CSR engagement of that company is positively affected. This does not apply for the first period but it does for the second period. If this correlation is compared with the relation between discretionary accruals and ROA which decreased between the time periods. This indicates that firms that are performing financially well are engaging more in earnings management. This correlation decreased which is a positive situation. The negative relationship between firm size and CSR is noteworthy. This is not according to the previous literature which indicates that the bigger the firm the more it is corporate social responsible (Kim et al., 2012). This is because they have to be accountable and have to legitimate their business. In this case the relationship is negative. Which indicates that the bigger a firm the less it participates in CSR Leverage has a big influence on earnings management in the second period. Leverage has less influence on CSR but the influence is in both cases negative. The growth of a firm is in the first period strong negative related to earnings management this is turned in a not so strong positive relation in the second period. The dummy variable auditor has a positive influence on CSR. This relation increased in the second period. There is a negative small relation between auditor and earnings management.. 32.

(33) Table E.1 2004-2008 Regression estimates of CSR on DA N R squire Adj. R squire. 4050 0,057 0,055 β. t. (Constant). Sig -1,027. ,304. Lower bound 95% -622,320. Upper bound 95% 194,462. Correlations partial. CSR. ,015. ,954. ,340. -10,606. 30,725. ,015. Size. ,021. 1,303. ,193. -8,440. 41,855. ,021. ROA. ,003. ,198. ,843. -6,669. 8,169. ,003. LEV. ,069. 4,475. ,000. 16,462. 42,133. ,071. Growth. -,227. -14,645. ,000. -6,619. -5,056. -,228. Auditor. -,002. -,134. ,893. -266,980. 232,845. -,002. Lower bound 95%. Upper bound 95%. Correlations partial. Table E.2 2009-2013 Regression estimates of CSR on DA N R squire Adj. R squire. 4050 0,370 0,369 β. t. (Constant) CSR Size ROA LEV Growth Auditor. Sig 30,914. ,724. -71,537. 49,681. ,003. 1,152. ,831. -2,012. 2,503. ,003. ,012. 1,848. ,385. -2,019. 5,228. ,014. -,005. 1,016. ,676. -2,417. 1,569. -,007. -,609. 2,192. ,000. -109,073. -100,476. -,608. -,002. ,089. ,872. -,190. ,161. -,003. ,014. 19,541. ,259. -16,266. 60,357. ,018. 33.

(34) In table E.1 and table E.2 is the regression model to test the hypothesis shown. The hypothesis of this research is: CSR is negative related to earnings management. This is tested on basis of the regression model with the dependent variable discretionary accruals which is the measure of earnings management in this research. The other variables CSR, size, return on assets, leverage, growth and auditor are the independent variables. To compare the two time periods there is a different regression model for each period. For the second period the R squired value has increased. This means that in the second period the model is explaining 37% of the variance and that the relation is positive. For time period one is this only 5,7% and the relations is positive. Looking at the significance there is no evidence for the hypothesis because it is insignificant for both of the periods 0,304; 0,724 respectively and the relation for both periods is positive. This in contrast to the hypothesis which indicates a negative relation between CSR and earnings management. Looking at the other independent variables the leverage and growth are significant for the first period. This means that these variables are correlated with the independent variable discretionary accruals. Leverage has a positive influence on earnings management and the growth of a firm has a negative influence on earnings management. This is because large firms has to clarify their earnings more specifically and are audited more closely so they have less opportunity to manage their earnings. For the second period only the variable leverage is significant and strong negative. This means that firms with high leverage are engaging less in earnings management. Concluding the hypothesis is rejected. There is no evidence that high CSR incorporated firms are engaging less in earnings management. There is no significant evidence that there is a relation between CSR and earnings management for the North American companies that are tested with the data from Compustat and KLD. This results in that the hypothesis is rejected.. 34.

(35) 6. Conclusion CSR reporting became a lot more important in the last decade. Even though that there is not one central regulation for CSR reporting much firms are providing CSR reports. Because there is not a regulated standard but many different standards it is difficult to compare the reports. Therefore CSR still have to go through a lot of developments. But has the growing importance of CSR influence on earnings management? This has led to the following research question: Does the growth of CSR reporting influence earnings management? By testing the hypothesis: Earnings management is negative related to CSR. This is tested for North American companies by using the KLD and Compustat database. To test the hypotheses the information from 810 firms is used. There is looked at the total CSR score by taking the score of the seven areas together. To measure earnings management the modified Jones model is used. By using this model the discretionary accruals are computed and on basis of these accruals the participation in earnings management is tested. This is tested for two different time periods. The first time period is 2004 until 2008 and the second time period is from 2009 until 2013. This period is taken because it is two years after the introduction of the SOX act to see the influence of this act on earnings management. The second period is chosen because it is interesting to look at the most recent developments in CSR. On basis of this method and data the hypothesis is rejected. There is no significant evidence that there is a negative relation between earnings management and CSR. There is evidence that the participation in CSR is growing in the compared periods and the discretionary accruals are fluctuating among the years but there is no significant evidence that there is a relation between these two.. 35.

(36) 7. Discussion There are several measures against earnings management. The introduction of the SOX act was one of the mature measures. It is important to reduce earnings management because the result of management can be big scandals which results in the big financial crisis in 2008. Nowadays firms are more sustainable and looking after their stakeholders and the shareholders. This can be reported in CSR reports. There are several standards for these reports which make it difficult to compare these reports amongst different firms. But CSR is still going through a big development. That is why it is interesting to look at the influence from CSR on earnings management. In this research this is done on basis data from KLD and Compustat. The limitations of this research are that the data is from the non-financial firms of the U.S. therefore the results cannot be generalized for other continents like Europe. Another limitation is that there was a lot of missing data and by matching the database a lot of companies are excluded. The last limitation of this research is that Earnings management is measured on basis of discretionary accruals. Therefore only the accrual based earnings management is tested. In further research it is interesting if CSR does have a significant negative influence on real earnings management. This research is quantitative therefore it is difficult to measure other related matters. It would be interesting in further research to do a qualitative research to see the influence of CSR. Also in this research are the different sectors not taken into account. Maybe the influence of CSR is not significant negative on earnings management overall the sectors but it is in a single sector. This research is done in a very troubled period because of the financial crisis in 2008. This results in fluctuating financial information for measuring the earnings management. A fact is that CSR is upcoming and earnings management has to be minimalized. CSR reporting and detection and punishment of earnings management both have to go through a lot of developments. Therefore this is an interesting subject for further research and looking further to significant relations.. 36.

(37) References Carnegie, G. D., & Napier, C. J. (2002). Exploring comparative international accounting history. Accounting, Auditing & Accountability Journal, 15(5), 689-718. Carroll, A. B. (1979). A three-dimensional conceptual model of corporate performance. Academy of Management Review, 4(4), 497-505. Chan, M. C., Watson, J., & Woodliff, D. (2014). Corporate Governance Quality and CSR Disclosures. Journal of Business Ethics, 125(1), 59-73. Chih, H. L., Shen, C. H., & Kang, F. C. (2008). Corporate social responsibility, investor protection, and earnings management: Some international evidence.Journal of Business Ethics, 79(1-2), 179-198. Christensen, B. E., Glover, S. M., & Wolfe, C. J. (2014). Do Critical Audit Matter Paragraphs in the Audit Report Change Nonprofessional Investors' Decision to Invest?. Auditing: A Journal of Practice & Theory, 33(4), 71-93. Cohen, D. A., Dey, A., & Lys, T. Z. (2008). Real and accrual-based earnings management in the pre-and post-Sarbanes-Oxley periods. The accounting review, 83(3), 757-787. Cohen, W. A. (2009). What Drucker taught us about social responsibility. Leader to Leader, 2009(51), 29-34. Dawkins, C., & Ngunjiri, F. W. (2008). Corporate Social Responsibility Reporting in South Africa A Descriptive and Comparative Analysis. Journal of Business Communication, 45(3), 286-307. Dechow, P. M., Hutton, A. P., Kim, J. H., & Sloan, R. G. (2012). Detecting earnings management: A new approach. Journal of Accounting Research,50(2), 275334. Deegan, C. (2002). Introduction: the legitimising effect of social and environmental disclosures-a theoretical foundation. Accounting, Auditing & Accountability Journal, 15(3), 282-311. Flower, J. (2015). The international integrated reporting council: a story of failure. Critical Perspectives on Accounting, 27, 1-17. Healy, P. M. (1985). The effect of bonus schemes on accounting decisions. Journal of Accounting and Economics, 7(1), 85-107.. 37.

(38) Holder-Webb, L., Cohen, J. R., Nath, L., & Wood, D. (2009). The supply of corporate social responsibility disclosures among US firms. Journal of Business Ethics, 84(4), 497-527. Hong, Y., & Andersen, M. L. (2011). The relationship between corporate social responsibility and earnings management: An exploratory study. Journal of Business Ethics, 104(4), 461-471. Jiraporn, P., Miller, G. A., Yoon, S. S., & Kim, Y. S. (2008). Is earnings management opportunistic or beneficial? An agency theory perspective. International Review of Financial Analysis, 17(3), 622-634. Jones, T. M. (1995). Instrumental stakeholder theory: A synthesis of ethics and economics. Academy of Management Review, 20(2), 404-437. Kolk, A. (2004). A decade of sustainability reporting: developments and significance. International Journal of Environment and Sustainable Development, 3(1), 51-64. Kim, Y., Park, M. S., & Wier, B. (2012). Is earnings quality associated with corporate social responsibility?. The Accounting Review, 87(3), 761-796. Van der Laan Smith, J., Adhikari, A., & Tondkar, R. H. (2005). Exploring differences in social disclosures internationally: A stakeholder perspective. Journal of Accounting and Public Policy, 24(2), 123-151. Lindblom, C. K. (1994, June). The implications of organizational legitimacy for corporate social performance and disclosure. Critical Perspectives on Accounting conference, New York. Lucey, B. M., & Zhang, Q. (2011). Financial integration and emerging markets capital structure. Journal of Banking & Finance, 35(5), 1228-1238. Milne, Markus J., Helen Tregidga, and Sara Walton. "Words not actions! The ideological role of sustainable development reporting." Accounting, Auditing & Accountability Journal 22.8 (2009): 1211-1257. Murphy, P. E., & Schlegelmilch, B. B. (2013). Corporate social responsibility and corporate social irresponsibility: Introduction to a special topic section. Journal of Business Research, 66(10), 1807-1813. Pearce, D., & Stacchetti, E. (1988). The Interaction of Implicit and Explicit Contracts. Yale University.. 38.

(39) Pfeffer, J., & Salancik, G. R. (2003). The external control of organizations: A resource dependence perspective. Stanford University Press. Porter, M. E., & Kramer, M. R. (2006). The link between competitive advantage and corporate social responsibility. Harvard Business Review, 84(12), 78-92. Rinaldi, L., Unerman, J., & Tilt, C. (2014). The role of stakeholder engagement and dialogue within the sustainability accounting and reporting process. Sustainability Accounting and Accountability (2nd ed.), Routledge, Abbingdon, 86-107. Roychowdhury, S. (2006). Earnings management through real activities manipulation. Journal of Accounting and Economics, 42(3), 335-370. Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative science quarterly, 224-253. Scholtens, B., & Kang, F. C. (2013). Corporate social responsibility and earnings management: Evidence from Asian economies. Corporate Social Responsibility and Environmental Management, 20(2), 95-112. Sloan, R. (1996). Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?(Digest Summary). Accounting Review,71(3), 289-315. Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20(3), 571-610. Tschopp, D., & Huefner, R. J. (2014). Comparing the Evolution of CSR Reporting to that of Financial Reporting. Journal of Business Ethics, 1-13 de Villiers, C., Naiker, V., & van Staden, C. J. (2011). The effect of board characteristics on firm environmental performance. Journal of Management, 0149206311411506.. 39.

(40)

No results found