University of Amsterdam
Faculty Economics and Business
Master Business Administration
The Relationship between CSR, Earnings Management
and Mediating Effect of the Industry
Name
Max van Egmond
Student number
10001618
Version
22 juni 2015
2 ABSTRACT: This study examines the relationship between the level CSR performance, the extent of earnings management and the mediating role of the industry, that a company is active in. To find this relation, discretionary accruals are estimated, using the Modified Jones model, and used as proxy for earnings management. CSR performance is captured by using the KLD database. The sample consists of 5,494 observations of U.S. listed companies with available financial data of one or more of the fiscal years 2003 until 2009. This study finds, that there is no relation between earnings management and CSR, indicating that CSR is not used as concealment tool by managers or transfers ethical behavior to reporting earnings. Further, this study finds that the industry of a company can influence both earnings management and CSR. Industry effects the structure of the company, and this could affect the extent of earnings management. Industry influences CSR, due to that different societal expectations are placed upon corporations within specific social contexts. This study more specially finds that industries that are under large pressure by society and perform socially responsibility, show higher CSR performance and lower earnings management. Those findings suggests that no direct relation between earnings management and CSR exist.
3 Content
1. Introduction ... 4
2. Literature Review ... 8
2.1 Corporate Social Responsibility ... 8
2.2 Earnings management ... 12
2.3 CSR & Earnings Management ... 16
2.4 Industry ... 20
3. Research Design & Methodology ... 23
3.1 Earnings Management ... 24 3.2 CSR ... 26 3.3 Industry ... 28 3.4 Model ... 30 3.5 Sample ... 33 4. Results ... 34 4.1 Descriptive statistics ... 34 4.2 Regression Analyses... 40 4.3 Robustness check ... 43
Discussion and Conclusion ... 45
7. Appendix 1 ... 50
4
1. Introduction
The level of awareness on environmental issues has increased significantly in recent years. Together with the increase of awareness of the environment, the role that companies play in affecting the environment is becoming more prominent. In 2010 a major unpublished study for the United Nations found that the costs of pollution and other damage to the natural environment, caused by the world’s (3000) biggest companies, would wipe out more than one-third of their profits, if they were held financially accountable, indicating an estimated 2.2 trillion dollar of environmental damage alone in 2008 (Jowit, 2010). In 2010 PUMA created an environmental profit & loss account, where they concluded that 78 percent of their profit would by evaporated, if they would discount all their environmental costs (Puma, 2011).
The increased awareness of the environment and the role company’s play in influencing the environment has led to demands of investors, customers and other stakeholders, to let firms show their responsibility towards social issues. This has caused that corporate social responsibility is increasingly becoming more important to firms, has lead that companies have to adopt social policies and show their commitment to social issues, such as environment, employees and acting socially and ethically. Corporate social responsibility (CSR) can be defined as treating the stakeholders of the firm ethically or in a responsible manner (Dahlsrud, 2008). The area defined by advocates of CSR increasingly cover a wide range of issues such as employee relations, human rights, corporate ethics, community relations and the environment (Moir, 2001) .
CSR transforms from a fringe activity carried out by a few earnest companies, like The Body Shop and Ben & Jerry’s, to a highly visible priority for traditional corporations (Doane & Abasta-Vilaplana, 2005). Research has shown that companies participating in CSR activities are positively changing their company image among their stakeholders. This helps the company build community ties and establish reputation capital, therefore improving its ability to negotiate more favorable contracts with governments and suppliers, to charge premium prices for goods and services, and to reduce the cost of
5
capital (Fombrun, Gardberg, & Barnett, 2000). Next to the external changes that focusing on CSR activities bring, it can also have an impact on internal factors, like corporate governance or the quality of the financial statement ((Kim, Park, & Wier, 2012; Sacconi, 2006; Orlitzky et al., 2003).
The disclosure of reliable financial earnings information is important for CSR. Reliable financial information is significant, because it provides stakeholder with a basis of trust and confidence regarding the company’s claims and operations. This is especially the case for socially responsible firms, since CSR, related to ethics and moral issues, concerning corporate decision-making and behavior, goes beyond what is required by law and regulation (Heal, 2008). Earnings quality can be influenced by managing earnings, for example to influence the short term results of a firm (Vinten, Sevin, & Schroeder, 2005). Earnings management occurs when managers use judgement in financial reporting and structuring transactions, to alter financial reports, to either mislead some stakeholders about the underlying economic performance of the company or to influence contractual outcomes, that depend on reported accounting numbers (Healy & Wahlen, 1999). Therefore, earnings management has a negative influence on the quality of financial information, because it gives a false image of the company’s earnings towards different stakeholders (Prior et al., 2008).
Kim, Park and Wier (2012) and Chic, Shen and Kang (2008) hypothesized that firms expending effort and resources in implementing CSR practices to meet ethical expectations of society, are likely to provide more transparent financial information, because managers will try to behave more ethically. Those managers do not pursuit CSR, because of their self-interest or to conceal effects of corporate misconduct.
However, the paper of Prior, Surroca and Tribo (2008) suggested another relationship. It is suggested that a company that expends effort and resources in implementing CSR practices, can have a high level of earnings management, and therefore a lower earnings quality, because managers can try to disguise low quality of their financial statement by implementing CSR. This deflects the attention of the
6
stakeholders from the poor quality of the earnings of the company, towards the CSR performances of the company. Thus, by focusing more strongly on CSR activities, managers can keep legitimizing the company or themselves.
The relationship between earnings management and CSR has been examined by a selected number of papers. Papers that researched this relationship are those of Kim et al. (2012), Prior et al. (2008) and Chic et al. (2008). However these studies yielded different and inconclusive results as mentioned previously. Prior et al. found that CSR increases the level of earnings management by studying discretionary accrual and income smoothing. However, Kim et al. (2012) found a negative relation between earnings management and CSR, by researching discretionary accrual, GAAP violations and real earnings management.
Those papers do not extensively account for the motives of managers to act socially responsible, when determining the relationship between CSR and earnings management. Ellen, Webb and Mohr (2006) claims that consumers judge the motivation for performing CSR practice. Furthermore, Campbell (2007) found that the reasons for performing CSR activities are mediated by several institutional conditions: public and private regulation. Those institutional conditions can differ between countries and industries. Gjolberg (2009) found striking differences between the environment of a company; he suggested that depending on the environment a company is faced with a certain set of barriers and opportunities in engaging in CSR activities. Further, Prior et al. (2008) mentioned that the environment of the firm can influence earnings management through corporate governance, because managerial cultures and investor protection vary across environment settings. Thus CSR practice, performance and earnings management apparently are determined by more than ethics alone. It is suggested that the environment, and more specifically the industry in which the company competes, can influence CSR performance and earnings management (Chand et al. 2006; Dechow et al. 1995)
7
As previous mentioned the relation between earnings management and CSR have been examined by a few papers who provide conflicting results. Because the papers written on this topic are not that common and they are relatively new, this study investigates, and fail to capture the effect of the environment on the described relation. The research question, therefore states: “Does Corporate Social Responsibility influence the extent of earnings management of a company and does the industry of a company mediate this relationship?” To test the hypothesis this study uses the Modified Jones model, to estimate discretionary accruals. To measure CSR, the KLD database is used.
The literature on this topic is relatively new and includes only a very select number of papers, thus this study adds to the relative limited information on the relationship between earnings management and CSR performance. Furthermore, the current papers did not account for the effect that the environment, and more specifically the industry (pressure) could have on the relation between CSR and earnings management. Further most research covering this topic had a relative small time interval, this research will use a timer interval of 7 years, therefore, being able to find more generalizable results.
Prior research both found a positive and negative relation between earnings management and CSR. The inconclusiveness can have severe effects on the impression stakeholders have of a company or decisions that stakeholders will make. Findings of previous performed studies could give stakeholders the impression, when companies perform socially responsible, this is the result of managers trying to conceal earnings management or other improper behavior. Stakeholders can also get the impression when managers act socially responsible, that this results in higher quality of the reported earnings, because of the ethical values of managers. The practical contribution of this study lays in giving stakeholders a better indication of the implications of the relationship between earnings management and CSR, precluding stakeholders drawing wrong conclusions. This study further provides a better view of the motives of managers to perform socially responsible and if this effects CSR performance of a company, thus if a
8
company that is pressured to perform socially responsible actually acts more socially responsible or just provides more symbolic disclosures about CSR related topic
The rest of the article is organized as follows. First, a literature review section will be provided; this paragraph contains an overview of important subjects surrounding this research and hypothesizes are stated. Second, the research methodology, regression formulas and sample are given. Next, the results are presented. Finally, a discussion about the findings will be included and conclusion will be stated.
2. Literature Review
This chapter will start with defining CSR and will look at the reasons CSR is implemented. Next, this chapter will focus on earning management and the stakeholder theory. Further, the relation between CSR and earnings management and the underlying thought of the mediating role of the industry of the firm will be discussed.
2.1 Corporate Social Responsibility
The concept of corporate social responsibility (CSR) has a long and varied history (Carroll, 1999). The work of Howard R. Bowen (1953) is seen by many as the beginning of the modern period of literature on the subject of CSR. Bowen (1953) defined social responsibility as the obligation of businessmen to pursue those policies, to make those decisions, or to follow those lines of action which are desirable in terms of the objectives and values of our society; meaning that companies are responsible for the consequences of their actions in a sphere wider than their profit-and-loss statements.
In the seventies Keith & Blomstrom (1975) conceptualized CSR as “the managerial obligation to take action to protect and improve both the welfare of society as a whole and the interest of organizations”. In more recent years multiple studies tried to find a general definition of CSR, due to the many existing conceptualizations of CSR (Dahlsrud, 2008).
9
Moir (2001) analyzed different conceptualizations of CSR and created an overall framework, determining when a company can be seen as socially responsible. This involves meeting the needs of all stakeholders and not just shareholders against some form of ethical basis. Thus, when a company threats its employees fairly and equitably, operates ethically and with integrity, respects basic human rights, sustains the environment for further generations and is a caring neighbor, the company can be truly seen as a socially responsible firm.
Monteil (2008) showed that conceptualizations and measures of CSR and Corporate Sustainability (CS) are largely converging, even though there are some small points of differences between the two constructs. Both CSR and CS refer to social and environmental management issues, therefore literature or motivations of CSR and CS can also be largely applied to each other, due to the common ground between the two (Zink, 2005).
CSR is not a new idea. However, CSR has never been more prominent on the corporate agenda than it is today (Smith, 2003). It has become a major focus of interest not only for corporate managers, but also within the multilateral and bilateral development agencies. One of the primary methods to indicate if a corporation is performing socially responsible is CSR reporting. Therefore, with the rise of CSR, CSR reporting grows consistent with CSR.
Only a few years ago, CSR reporting was regarded as a niche activity that was only practiced by a few companies in select sectors (Quinn, 2013). Currently, it is a mainstream business focus across all sectors of industries, spanning companies of all sizes. According to research presented at the 7th Annual CR Reporting and Communications Summit in London, the overall number of CSR reports continues to increase rapidly year by year.
Today, more companies include CSR information in annual reports, integrated reports, and dedicated websites, rather than exclusively in stand-alone CSR reports. CSR reporting has developed organically, adding layers of issues to reflect the expectations of
10
a company’s stakeholders and, more widely, of society. Reports engage with topics such as supply chains, gender issues, ethics, and human rights, in addition to the original environmental issues that continue to form the core of reporting.
Many firms find the business case for greater attention to CSR to be compelling, particularly given reputational risk and other pressures of the contemporary business environment. However NGOs have, for the most part, been extremely critical of the voluntary initiatives undertaken by the corporate sector (Jenkins, 2005).
One of the reasons for this could be why corporations perform socially responsible. The main purpose of a corporations is to make profit. Thus, if being socially responsibility helps making more profit, corporations may implement it.
Therefore, different researches have been performed to test the relation between being socially responsible and financial performance. Different studies show that consumers are willing to pay premium prices for products that are more sustainable, thus acting social responsibility affects consumer behavior in a positive way (De Pelsmacker, Driesen, & Rayp, 2005; Kang, Stein, Heo, & Lee, 2012). However, it should be defined how this correspondents with the financial performance of a firm. To find this, multiple studies have been performed, to conclude if there exists a relation between performing socially responsible and the financial performance of a company. Orlitzky, Smmidt and Rynes (2003) found by performing a meta-analyses that there exists a positive association between corporate social performance (CSP) and corporate financial performance, however some operationalization’s of CSP and corporate financial performance moderate positive association. Waddock (2003) also found that corporate social performance (CSP) has a positively association with prior financial performance. CSP is also found to be positively associated with future financial performance, supporting the theory that good management and CSP are positively related. McWilliams and Siegel (2001) reported that most empirical studies about the relationship between CSR and profitability have been inconclusive, reporting positive, negative, and neutral results, due to flaws in the empirical analysis. When the model in empirical analysis is properly specified, the
11
authors found that CSR has a neutral impact on the financial performance of an organization.
If it is not sure if CSR has a positive association with the financial performance, what reasons could corporations have to perform CSR? The study of Garriga & Melé (2004) charted the different theories that explain why corporations implement CSR. These theories can be classified in four main groups; the instrumental-, the political-, the integrative- and the ethical theories. Each of the four groups exists of different theories with the same underlying core.
The instrumental theories group exists of theories like maximization of shareholder value and cause related marketing. In this group CSR is only seen as a strategic tool to achieve economic objectives and, ultimately, wealth creation. Its social activities are only means to achieve economic results. The second group political theories primary focusses on the interaction and connection between business and society, as well as on the power and position of business and its inherent responsibility. One of the main theories of this group is the social contract theory. This theory implies that there exist some indirect obligations that corporations have towards society. Thus, this group sees corporations as an entity that is part of society and therefor has its responsibilities. The third group integrative theories looks at how corporations integrate social demands. This theories also implies that corporations depend their existence on society. This group primary focuses on the detection and scanning of and response to the social demands that achieve social legitimacy, greater social acceptance and prestige. The most known theory in this group is the stakeholder theory. The stakeholder theory describes a firm as a series of connections of stakeholders, that managers of the firm attempt to manage. In light of the stakeholder theory, organizations tend the responds to the most ‘powerful’ stakeholders. In terms of CSR corporations act socially responsible of it copes with the most powerful stakeholders (Moir, 2001). Thus, this group of theories sees acting socially responsible as a demand of society that they need to fulfill. The last group ethical theories sees CSR as doing the right thing to achieve a good society (Garriga & Melé, 2004). Thus,
12
companies aim to act in a way that considers the present and future generations, to serve the common good of society.
To test some of these theories, multiple studies have been performed, to test what reasons a corporation could have to implement CSR. Campbell (2007) studied why corporations behave in socially responsible ways. He found that that economic conditions, specifically the relative health of corporations and the economy and the level of competition to which corporations are exposed, affect the probability that corporations will act in socially responsible ways. However, a variety of institutional conditions mediate these basic economic relationships, like state regulation, NGOs that monitor them etc. Moreover, socially responsible corporate behavior is more likely to occur to the extent that firms belong to industrial or employee associations and engage in institutionalized dialogue with stakeholders. These findings mostly comply with the stakeholder theory.
Overall CSR has seen a rise in the beginning of the twenty-first century. Companies have different motives to perform socially responsible. However, most of these motives originate from the purpose of a corporation; making a profit.
2.2 Earnings management
The job of the financial statements is to provide information about the financial position, performance and changes in financial position of an enterprise that is useful to a wide range of users in making economic decisions. Financial information is used by different stakeholders. Owners and managers need financial information to make important decisions that affect the company’s operations.Employees use financial reports to make collective bargaining agreements with the management. Prospective investors use financial statements to assess the viability of investing in a business, and financial institutions use them to decide whether to grant a company capital. If the financial information is incorrect, different stakeholders can make a wrong decision based on this
13
information. Thus, financial information is important, because it provides stakeholders with a basis of trust and confidence regarding the company’s claims and operations.
However, the information of the financial report does not always have to be reliable. Earnings quality can be seen as the coherence between the information the financial report gives about the performance of the company and the actual performance of the company. The quality of the earnings depends on coherence between the actual and reported performance (Dechow, Ge, & Schrand, 2010). This coherence depends on whether extent actual material wealth changes are effected in the reported financial performance of the company.
The extent of earnings management is seen as one of the determinants, that effect earnings quality. One definition of earnings management is to which extent manager’s exercise their discretion over the accounting numbers (Watts & Zimmerman, 1978). Another concept of earnings management, defined by Healy and Wahlen (1999), is that earnings management happens when managers exercise their discretion over the accounting numbers, aiming to alter financial reports, to mislead multiple stakeholder, about the underlying financial performance of the company or influence the consequences of contracts that depend on the reported financial numbers.
Research found evidence that earnings decreases and losses are frequently managed away. Firms with a small pre-managed earnings decrease, exercise discretion, to report earnings increases. Similar, firms with slightly negative pre-managed earnings exercise discretion to report positive earnings (Burgstahler & Dichev, 1997). A reason implicated by Burghstahler & Dichev for reporting earnings increases is that managers avoid reporting earnings decreases and losses, to decrease the costs imposed on the firm in transactions with stakeholders.
Thus, earnings management is a strategy that can used by the management of a company to deliberately manipulate the company's earnings, so that the figures match a pre-determined target. This practice is carried out for the purpose of income smoothing.
14
Thus, rather than having years of exceptionally good or bad earnings, companies will try to keep the figures relatively stable by adding and removing cash from reserve accounts.
Healy & Wahlen (1999) provided a broader explanation what reasons managers could have to perform earnings management. Three primary motivations are distinguished: capital market; contracting and regulatory motivation. Capital market motivations assume that accounting information is used by investors and financial analysts to help value stocks, this can create an incentive for managers to manipulate earnings in an attempt to influence short-term stock price performance.They found that firms can use earnings management to influence the expectations of specific types of investors.
Financial information was used to help monitor and regulate contracts between the firm and its stakeholders. Compensation contracts were used to align the incentives of management and external stakeholders. Lending contracts were used to limit mangers ’actions that benefit the firm’s stockholders at the expense of the creditors. However, these contracts create motives for management to manage earnings.
Regulate motivation originates from industry regulation and anti-trust regulation. Healy & Wahlen found strong support that accounting discretion is used to manage industry-specific and anti-trust regulatory constraints. All industries are regulated to some degree, and in certain industries constraints are laid up. These regulatory constraints create incentives to manage the income statement and balance sheet variables, to keep off interest of regulators. Similar, anti-trust regulation causes that managers of firms vulnerable to an anti-trust investigation or other adverse political consequences have incentives to manage earnings to appear less profitable.
There are two main approaches managers can adopt to manipulate earnings, giving external users a false image of the company, accrued based earnings management and real earnings management. Both of these methods of earnings manipulation are opportunistic in nature; not only do they disrupt a company’s intrinsic value, but they are
15
also detrimental to the company’s future performance. Accrued based earnings means that managers change the accrual part of earnings while not bringing real economic consequences (Dechow et al., 2010). Thus, they only change the accounting numbers but do not induce real economic changes. Real earnings management means that companies modify their business actions, thus bringing real economic consequences (Roychowdhury, 2006).
As mentioned before earnings management dilutes the reported financial earnings information, and therefore creates a gap between the reported financial performance and the actual financial performance. Earnings management can be seen as a method where managers use their discretion in financial numbers to increase their own benefits at the company and stake- and shareholders. Thus, this can be seen as an agency dilemma, because managers (agent) are able to make decisions on behalf of the company (principal), and those decisions can sometimes be based on the managers own best interest rather than those of the shareholders. Thus, earnings management creates agency-costs, because shareholders can be diluted by the reported financial information and therefore, make sub-optimal financial and operational decisions, what can impact the profit they make on decision. The signaling theory is useful for describing this behavior. When two parties (individuals or organizations) have access to different information, what is the case between managers and shareholder or stakeholders. Typically, one party, the sender, must choose whether and how to communicate (or signal) that information. The other party, the receiver, must choose how to interpret the signal. In either case, the fact is that actions that convey information lead to people altering their behavior (Connelly, Certo, Ireland, & Reutzel, 2011).
Different studies have been performed to study the stakeholder – agency costs of earnings management. Those studies found that earnings management has long-term negative consequences for the company and the shareholders (Roychowdhury, 2006; Sloan, 1996) Next to the negative consequences of earnings management for the company
16
and shareholders, different stakeholder groups like employees also incur negative costs due to earning management (D'Souza, Jacob, & Ramesh, 2000).
2.3 CSR & Earnings Management
The relation between CSR and earnings management can largely be explained through the signaling theory. As previous mentioned when two parties have access to different information, decisions need to be made how or whether to communicate information and how to interpret the information. Lys, Naughton and Wang (2015) found that CSR expenditures are not charity nor do they improve future financial performance, but is used as part of a signaling strategy. Firms undertake CSR expenditures in the current period, when they anticipate stronger future financial performance, thus signaling their financial performance through CSR. Lys et al. concludes that CSR disclosures are another channel by which firms convey private information about company related topic to outsiders. Next to communicating the real intentions of the company, managers can also convey people with sending information contradictory to the real course of events in the company. One example is greenwashing, were environmental or social benefits are added to a product, but those claims are misleading, and convey people to interpret the products more positively
Linking this towards the relation between earnings management and CSR performance and the literate on that topic. Two main streams of literate about the relation of earnings management and CSR performance can be divined. The first stream predicts a decrease in earnings management, when a firm strongly focusses on CSR. The second stream predicts a decrease in CSR performance when the extent of earnings management is high.
The stream that predicts a decrease in earnings management, when a firm strongly focusses on CSR bases this relation on the believe that managers are committed to the ethical values they proclaim; meaning that those managers will act more ethically when
17
producing the financial number, because CSR is seen as a moral responsibility to sustain to the ethical values they proclaim (Porter & Kramer, 2006). An important part of being socially responsible is business ethics, and earnings management is generally seen as an unethical practice (Kaplan, 2001). By performing earnings management managers manipulate the performance a firm, and ethical managers could be hesitant to perform this kind of behavior (Kim et al., 2012).
Another reason that CSR could decrease earnings management is the relationship between corporate governance, CSR and earnings management. Jo and Harjoto (2011) found that an effective corporate governance structure improves CSR engagement of a firm. Johnson and Greening (1999) also found a positive relation between the governance structure and the CSR performance of the firm. Beltratti (2005) studied the relationship between corporate governance and CSR and suggests that when a corporation has a good corporate governance system it would decrease the chance stakeholders would be negatively affected by illegal actions. An effective CSR mechanism will refrain actions that are legal but inappropriate, because of the negative consequences on stakeholders, therefore corporate governance and CSR are supplementary in behaving more ethically. This indicates that CSR and corporate governance both decrease the chance of inappropriate actions, what indicates a lower extent of earning management. The literature of earnings management and corporate governance found that there exists a positive relation between corporate governance and the earnings quality, thus less earnings management (Dechow, Sloan, & Sweeney, 1996).
Furthermore, the disclosure of earnings management can have more severe negative consequence for the reputation of a company that is seen as socially responsible. This can be explained by the fact that CSR firms proclaim ethical and social values, therefore this behavior would be seen as more inappropriate for CSR firms then other firms (Linthicum, Reitenga, & Sanchez, 2010). The severe negative consequences for CSR-firms, can lead that managers are more likely to refrain from earnings management. Based on the previous stated literature the first hypothesis of this research is:
18 H1: The level of CSR performance has a negative effect on the extent of
earnings management.
The other stream of literature suggests a positive relationship between earnings management and CSR. This literature builds on the thought that stakeholders view CSR as a positive phenomenon. Du, Bhattacharya and Sen (2010) found that by performing socially responsible, companies can not only generate favorable stakeholder attitudes and better support behaviors, but also, over the long run, build corporate image, strengthen stakeholder–company relationships, and enhance stakeholders’ advocacy behavior. Indicating that associating with CSR could positive influence the relation between the company and its stakeholders.
Practices involving misusing the positive associations of CSR to pursue own company or manager gain is skyrocketing the last couple of years; more and more firms combine poor environmental performance with positive communication about environmental performance (Delmas & Cuerel Burbano, 2011). Managers use CSR and more importantly CSR communication to capture the positive association towards CSR and transfer those positive associations towards the company. Nowadays CSR and advertising are seen as strategic complements. However, if a claim about the environmental or social benefits of a product is unsubstantiated or misleading, this practice is known as greenwashing. Greenwashing is a form of propaganda in which green PR or green marketing is deceptively used to promote the perception that an organization's products, aims or policies are environmentally friendly.
Next to greenwashing, CSR can also be used as a concealment tool by managers with the purpose of reducing the chance of being scrutinized by stakeholders. Meaning that CSR a disguise for performing earnings manipulation. Thus, managers can use CSR per their own personal gain, to pursue their own motives instead of the motives of the shareholders or help their career (Carroll, 1979). Therefore, CSR can be a tool for a firm to project a socially–friendly image as well as gain and maintain legitimacy from
19
stakeholders by deflect the attention of stakeholder and improve a firm’s reputation. This can be explained by the legitimacy theory. The legitimacy theory states that companies have a social contract with society because society provides organizations with a social license to operate if they act according to the bounds and norms society sets (Brown & Deegan, 1998). Legitimacy is a condition or status that exists when an entity’s value system is congruent with the value system of the larger social system of which the entity is part (Lindblom, 1994). Legitimacy is necessary for the survival of the company. Legitimacy is linked with CSR by that companies receive a license to operate, thus by performing CSR activities the company stakeholder satisfaction, but also benefits from the positive effects on its reputation and brand name among stakeholders. Thus to deflect the attention of the stakeholders, managers can perform CSR activities.
Another theory that could provide an explanations for a higher level of CSR performance and lower earnings quality is stakeholder theory. This theory is quite similar to the legitimacy theory, in the sense that corporations need to perform activities desired by stakeholders. However the stakeholder theory, states that corporations only have to act in a certain way when it copes with the most powerful stakeholders (Donaldson & Preston, 1995). Therefore, due that in the most recent years the pressure of stakeholders to perform socially responsible, has deviated the attention of stakeholders from earning quality. Thus, by performing socially responsible, if this is the focus of the most important group of stakeholder a corporation could neglect the quality of earnings.
Thus to deflect the attention of the stakeholders, managers can perform CSR activities. Therefor the second hypothesis is:
H2:The extent of earning management has a positive effect on the level of CSR performance.
20 2.4 Industry
It is suggested that the environment that the company competes in, could influence CSR performance and earnings management. A big part of the environment of a company is dependent on the industry a corporation is active in. Chand and Frasier (2006) studied the relationship between CSR performance, overall financial performance and the influence of the industry. They found that the relationship between CSP and financial performance differ greatly across different industries. Indicating that industry plays an important role as moderator in that relationship. Different industries create differences in stakeholder activism and interest across industries, which influences the relationship between CSP and financial performance of the firm.
Hrasky (2011) researched if between 2005 and 2008 Australian companies changed disclosures concerning a sustainability related topic, carbon-footprint, due to the increased awareness for sustainability. She found that Australian largest firms are disclosing more information about their carbon footprints in financial reports. Further, they attempt to draw more attention from stakeholders to those disclosures.
Next to that corporations changed their disclosures about sustainability topics due to increased awareness and/or pressure by stakeholders, there exists a difference between more and less carbon-intensive firms, in terms of disclosing. Hrasky found that less carbon-intensive sectors appear to move towards a symbolic disclosure strategy (pragmatic legitimacy) and more carbon-intensive sectors shift towards behavioral disclosure strategy (moral legitimacy). Where pragmatic legitimacy involves engaging in self-interested behavior to gather support of the stakeholders of a firm, by communicating disclosures thatare more of a indicative or rhetoric suggestive nature, but not reflecting any substantive action. Moral legitimacy involves in engaging in behavior that is positively evaluated by the organization’s stakeholders, by communicating an actual strategy that is undertaken to mitigate to act more socially responsibly. Thus,
Wanderley, Kucian, Frarache, Filho and Milton (2008) found that both country origin and industry sector have a significant influence over CSR. By investigating internet
21
websites, one of the main tool for CSR information disclosure, of corporations emerging from different countries and industry sectors they found that that different societal expectations are placed upon corporations within specific social contexts. Which creates different levels of pressure to perform socially responsible. Therefore, the third hypotheses states:
H3: The industry of a company influences the level of CSR performance
Next to the effect the industry has on the social responsibility, the industry of a firm can possible have a relation with the amount of earnings management. Dechow, Sloan and Sweeney (1995a) found that the amount of earnings management is correlated between companies in the same industries. This indicates that the industry influences the extent of earnings management. The role of a firm's industry as a determinant of earnings management has also been tested by Belkaoui and Picur (1984)and Albrecht and Richardson (1990). Both studies found that industry is a determinant of earnings management. They found that firms in the periphery sector have more opportunities and predisposition to report smoothed earnings numbers than core sector firms. Kinnunen, Kasanen and Niskanen (1995) also found a relation between earnings management and the industry of a company. However, contrary he found that both the opportunities and the predisposition of earnings management behavior are significantly larger in the core sector firms than in firms which are operating in the more peripheral sector of the economy.
Furthermore, the extent of earnings management depends on the structure of the company (Siregar & Utama, 2008). Companies in different industries perform other tasks and activities to manage the company. A retail company has a completely different structure compared to a service company, therefore companies in a certain industry could higher or lower extent of earnings management due to the characteristics of the industry. Therefore, the fourth hypotheses is:
22 H4: The industry of a company influences the extent of earnings management.
The reason why managers choose could to perform CSR activities could be an important factor in determining the relationship between CSR and earnings management. Reasons corporations perform CSR, can be derived from the stakeholder theory, legitimacy theory and increased financial performance (Campbell, 2007; Garriga & Melé, 2004)
Ellen et al.(2006) found that consumers judge the motivation for performing CSR practices and react negatively towards self-centered motivation to perform CSR activities. Lennox and Pittman found that the motivation for assuring better financial information can affect the credit rating of a firm. They found that when companies perform those activities by free will and are not obligated by law or pressured to do so, this has a positive effect.
Deegan, Rankin and Voght (2002) found a consistent pattern, by using media attention as a proxy for public concern. Deegan reported that when media attention on the company increased, so did the level of social and environmental disclosure in the annual report. Deegan, Rankin and Tobin (2000) and Patten (1992) investigated the level of environmental disclosure, surrounding firms related to a certain industry where an environment incident took place. They found that the amount of disclosures increased significant after the environment incident. This evidence that firms respond to situations that have the potential to threaten their legitimacy (Hrasky, 2011). This suggests that the motivation why to perform certain CSR activities, is an important factor for different stakeholder and can possibly affect the legitimacy of a firm.
Therefore it could be suggested that the environment of the company could pressure firms to perform CSR activities. The benefits of performing CSR activities like asking a price premium or positively chancing the perceived behavior of the firm, could
23
be less effective, when the environment demands socially responsible behavior (De Pelsmacker et al., 2005).
Another way that the motivation of performing CSR practices can influence the relation between CSR and earnings management are the ethical values of managers. It is believed that that managers who are committed to the ethical values they proclaim, also act more ethically when producing the financial number (Porter & Kramer, 2006). However if the company is obligated or pressured to act socially responsible, managers do not have to have a moral compass or ethical values, so does not have to act more ethically when producing financial numbers. Both explanations suspect that when a company is pressured or obligated to perform CSR there is a lower negative relation between CSR and earnings management. Thus industries that are under pressure to perform a higher level of CSR, have a higher extent of earnings management, therefore hypotheses five is:
H5: Pressure on the industry to perform socially responsible has a positive influences on the level of CSR performance and extent of earnings management
3. Research Design & Methodology
This chapter will outline the research method used to examine the effect of CSR on earnings management, and the moderating role of industry. The hypotheses stated in the prior section will be tested using a regression mode. The first paragraph will explain the sample selection. The second paragraph will show how earnings management will be captured, by using discretionary accruals and how CSR performance will be captured. The third paragraph will explain how the industry of firm is measured. The fourth paragraph describe the model used to test the hypotheses. The last paragraph will describe how the sample was created.
24 3.1 Earnings Management
Two types of earnings management can be distinguished, accrued based earnings management and real earnings management. This research will only use accrual based earnings management to test the relationship between earnings management and CSR performance. To measure earnings management, discretionary accruals are used, because by measuring discretionary accruals the degree of bias infused into the financial statement by the management will become known (Hoitash, Markelevich, & Barragato, 2007). To estimate the degree of earnings management the Modified Jones model will be used as stated by Dechow et al. (1995b). Dechow tested different accrual-based models to find which model has the most power in detecting earnings management. The model with the most power to detect earnings management was the Jones model (1991). However Dechow adjusted the model, by taking the changes in receivables into account when looking at the change in revenues. The modified Jones model assumes that all of the changes in credit sales during the sample period are the result of earnings management. This assumption is based on the suggestion that earnings management is easier to exercise over recognition of revenue on credit sales than it is to exercise earnings management over the recognition of revenue from cash sales. They further found that the Modified Jones model exhibits the most power in detecting earnings management. This model is much used in studies that want to find the quality of earnings (Gul, Chen, & Tsui, 2003; Hoitash et al., 2007; Velury, 2003). To calculate earnings management, the total amount of accruals needs to be known, before the Modified-Jones model can be used. Total accruals is calculated using income before extraordinary items and cash flows from operation form the cash flow statement.
𝑇𝐴𝑡= 𝐸𝑋𝐵𝐼𝑡− 𝐶𝐹𝑂𝑡 𝑇𝐴𝑡 = 𝑇𝑜𝑡𝑎𝑙 𝐴𝑐𝑐𝑟𝑢𝑎𝑙𝑠 (𝑖𝑛 𝑦𝑒𝑎𝑟 𝑡)
𝐸𝑋𝐵𝐼𝑡= 𝐼𝑛𝑐𝑜𝑚𝑒 𝐵𝑒𝑓𝑜𝑟𝑒 𝐸𝑥𝑡𝑟𝑎𝑜𝑟𝑑𝑖𝑛𝑎𝑟𝑦 𝐼𝑡𝑒𝑚𝑠 (𝑖𝑛 𝑦𝑒𝑎𝑟 𝑡) 𝐶𝐹𝑂𝑡= 𝐶𝑎𝑠ℎ 𝐹𝑙𝑜𝑤𝑠 𝑓𝑟𝑜𝑚 𝑂𝑝𝑒𝑟𝑎𝑡𝑖𝑜𝑛𝑠 (𝑖𝑛 𝑦𝑒𝑎𝑟 𝑡)
25
The next step after the total amount of accruals is calculated, is subtracting cash flows from operations from income before extraordinary items, to estimate the parameters. The parameters are used to find the amount of non-discretionary accruals. To find the parameters the equation below is used.
𝑇𝐴𝑡 = 𝛽1,𝑡[ 1 𝐴𝑡−1] + 𝛽2,𝑡[ ∆𝑅𝐸𝑉𝑡 𝐴𝑡−1 ] + 𝛽3,𝑡[ ∆𝑃𝑃𝐸𝑡 𝐴𝑡−1 ] + 𝜀 𝐴𝑡−1= 𝐴𝑠𝑠𝑒𝑡𝑠 (𝑖𝑛 𝑦𝑒𝑎𝑟 𝑡 − 1) ∆𝑅𝐸𝑉𝑡= 𝐶ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 (𝑦𝑒𝑎𝑟 𝑡 − 1 𝑡𝑜 𝑦𝑒𝑎𝑟 𝑡) ∆𝑃𝑃𝐸𝑡= 𝐶ℎ𝑎𝑛𝑔𝑒 𝑔𝑟𝑜𝑠𝑠 𝑝𝑟𝑜𝑝𝑒𝑟𝑡𝑦, 𝑝𝑙𝑎𝑛𝑡 𝑎𝑛𝑑 𝑒𝑞𝑢𝑖𝑝𝑚𝑒𝑛𝑡 (𝑦𝑒𝑎𝑟 𝑡 − 1 𝑡𝑜 𝑦𝑒𝑎𝑟 𝑡) 𝛽𝑡 = 𝑃𝑎𝑟𝑎𝑚𝑒𝑡𝑒𝑟𝑠 𝜀 = 𝐸𝑟𝑟𝑜𝑟 𝑜𝑓 𝐸𝑠𝑡𝑖𝑚𝑎𝑡𝑒
After the parameters are estimated in the previous equation, non-discretionary accruals can be found by filling out the equation below, with the parameter found in the previous equation, and the other stated variables.
𝑁𝐴𝑡= 𝛽1,𝑡[ 1 𝐴𝑡−1 ] + 𝛽2,𝑡[∆𝑅𝐸𝑉𝑡− ∆𝑅𝐸𝐶𝑡 𝐴𝑡−1 ] + 𝛽3,𝑡[ ∆𝑃𝑃𝐸𝑡 𝐴𝑡−1 ] + 𝜀 ∆𝑅𝐸𝐶𝑡 = 𝐶ℎ𝑎𝑛𝑔𝑒 𝑖𝑛 𝑅𝑒𝑐𝑒𝑖𝑣𝑎𝑏𝑙𝑒𝑠 (𝑦𝑒𝑎𝑟 𝑡 − 1 𝑡𝑜 𝑦𝑒𝑎𝑟 𝑡)
Now the total amount of accruals and the amount of non-discretionary accruals is known, discretionary accruals can be calculated by subtracting non-discretionary accruals from total accruals.
𝐷𝐴𝑡 = 𝑇𝐴𝑡− 𝑁𝐴𝑡
𝐷𝐴𝑡 = 𝐷𝑖𝑠𝑐𝑟𝑒𝑡𝑖𝑜𝑛𝑎𝑟𝑦 𝐴𝑐𝑐𝑟𝑢𝑎𝑙𝑠 (𝑖𝑛 𝑦𝑒𝑎𝑟 𝑡) 𝑁𝐴𝑡 = 𝑁𝑜𝑛 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑖𝑜𝑛𝑎𝑟𝑦 𝐴𝑐𝑐𝑟𝑢𝑎𝑙𝑠 (𝑖𝑛 𝑦𝑒𝑎𝑟 𝑡)
26
This thesis does not differentiate between positive or negative discretionary accruals, therefore all negative discretionary accruals are made positive, only the amount and not the direction of the accruals are taken into account.
3.2 CSR
This research will use the KLD database, short for the Kinder, Lydenberg, Domino & Co database, to measure CSR. The KLD database offers institutional investors easy-to-use, comprehensive and accurate social information about U.S. companies. The KLD-database is used in various research papers to measure CSR (Kim et al., 2012; Turban & Greening, 1997). Waddock (2003) states that the KLD database has proven itself to be factual, reliable, broad ranging, and maintained with consistency and transparency over the past decade, and that this database can be seen as the ‘de facto research standard at the moment’. Mattingly and Berman (2006) tested the KLD database for its usefulness, they found that the KLD database successfully measures specific aspects of corporate social action. Other researchers (Chatterji, Levine, & Toffel, 2009; Rahman & Post, 2012) also discussed and found that the performances scores of the KLD database correspond to the actual CSR performance of the firm.
This thesis will use the KLD database to differentiate between CSR positive and neutral or negative firms. This will be done by measuring the score on certain CSR strengths and concerns of firms in different categories, to grade the performance of a company on CSR. Those different categories evaluate the responsiveness of the firm towards the stakeholders. The KLD database differentiates between seven different categories: Community; Corporate Governance; Diversity; Employee Relations; Environment; Human Rights and Product. By measuring the CSR strengths and concerns on these different categories, the KLD database captures the responsiveness of five different stakeholder groups: consumers; employees; the community; diversity related stakeholder; environmental action groups. Strength and concern (positive and negative indicators) covers approximately 80 indicators in seven major qualitative issues areas
27
(categories). Thus, the strength and concerns indicates how a certain firm scores on one of the seven categories.
All seven categories will be used in this thesis. The strengths and concerns that cover the seven categories are elaborated in appendix 1. The main categories are explained in table 1 (Bird, Hall, Momentè, & Reggiani, 2007):
Table 1
KLD Categories
Categories Strengths Concerns
Community Strengths measures various contributions that the company makes to the community such as charitable contributions and support for the disadvantaged.
Concerns measure activities that are judged to have had a negative economic impact on the community and/or possibly mobilized community opposition.
Corporate Governance
Strengths are present when activities such as limited compensation for the management and the company has multiple ownership strengths.
Concerns are present when managers receive high compensation and there is low reporting quality.
Diversity Strengths measures the activities of the company in such areas as
providing employment
opportunities for minorities and providing working conditions that meet the special needs of minorities.
Concerns measure things like the non-representation of minorities in senior positions within the company and major controversies on affirmative action issues.
Employee
Relations Strengths are practice such as strong worker involvement within the company, generous profit sharing across the majority of employees, good retirement benefits and/or a good safety record.
Concerns arise when ac company might have bad union’s relations, a poor safety record and/or a poorly funded pension plan.
Environment Strengths are a result when a company performs environmentally sound practices such as pollution prevention, and recycling.
Concerns will arise when practices such as producing hazardous waste and/or environmentally unfriendly products are present.
Human
Rights Strengths are present when the company has a set of high quality of labor rights or follows human right policies imposed by society.
Concerns are measured by activities such as support for controversial regimes and low quality labor rights.
28 Products Strengths measures activities such
as high product quality, high innovation and the development of products to meet the special needs of the disadvantaged.
Concerns are present when the company has low product safety, controversies over how it advertises its products and other product-related community concerns.
Overall there are two different methods how CSR performance is measured. Firstly, CSR performance is measured by taking the total of strengths in the seven categories and subtracting the total of concerns from the total strengths. Following this method companies can either be CSR positive, CSR neutral or CSR negative. This method is used by Kim et al. (2012) who studied the relation between earnings management and CSR. In the second method companies are divided into two separate groups. The first group exists out companies, who overall have more strengths than concerns and the second group exists out firms with the same number or less strengths then concerns. The second method limits the separation of the CSR performance, because a company with a score of ten is put in the same group as a company with a score of one. To be able to better differentiate between CSR performance between different companies the first method will be used
3.3 Industry
This study aims to find if the environment, and more specifically the industry of a company have an effect on the relation of earnings management and CSR. However to be able to find this relation a separation between industries need to be made. To make this separation SIC codes are collected from the COMPUSTAT database. Standard Industrial Classification (SIC) codes are four digit numerical codes, these codes are assigned by the U.S. government to business establishments to identify the primary business of the establishment. The SIC system arrays the economy into 11 divisions, that are divided into 83 2-digit major groups, that are further subdivided into 416 3-digit industry groups, and finally disaggregated into 1,005 4-digit industries. This study will
29
only look at the 2-digit major groups, just like Kim et al. (2012).To measure the effect of industry, a separation between the industries will be made. Literature suggests that stakeholders are becoming more aware of sustainability topics, and are pressuring industries to be socially responsible (Hrasky, 2011). Therefore, companies that are active in industries that are perceived to have a negative effect on the environment, will face higher pressure to act socially responsible and to minimalize their impact on the environment. Research found that those industries that were point of criticism regarding their environmental and social impacts have achieved good ratings in terms of their sustainability reporting (oekom research, 2009). Therefore, this study uses a separation of industries based on the study of Hrasky (2011).
Table 1
Classification industries
Sectors SIC Codes
Large Environmental Impact
Agriculture, Forestry, Fishing 01-09
Mining 10-14
Construction 15-17
Manufacturing 20-39
Transportation & Public Utilities 40-49
Smaller Environmental Impact
Telecommunication 48
Wholesale Trade 50-51
Retail Trade 52-59
Financials, Insurance, Real Estate 60-67
Services 70-89
Public Administration 91-99
In order to differentiate between the two industries classifications a dummy variable will be used. A dummy variable is a numerical variable used in regression analysis to represent subgroups of the sample in your study. To distinguish between two industries classifications, dummy variables ‘one’ and ‘zero’ are created. ‘One’ is linked to industries that have a large impact on the environment and ‘zero’ to industries that have a smaller impact on the environment. See table 1 for more details on the separation.
30 3.4 Model
This study wants to research the relation between earnings management, CSR performance, and the mediating role of the industry, however next to these three variables, there are multiple factors that could potentially influence the relationship this study wants to test. Therefore, several control variables are included in order to test if the relation is statistically significant or not.
The first control variable that will be used is the size of the firm, firms size is measured by taking the natural log of the total assets. There are different views, what effect firm size can have on earnings management. Capital market pressures are greater for larger firms, because their performance is the focus of the analyst community, which spurs those firms to adopt aggressive accounting policies. In other words, larger firms have a greater incentive to manage earnings (Richardson, Tuna, & Wu, 2002). The opposite view is that firm size can be used as a proxy for information asymmetry. Larger firms which are often subject to closer scrutiny by outsiders and are required to disclose their information, and hence, there is a lower probability that they manage earnings. Small firms, however are able to withhold their private information more easily than large firms (Lee & Choi, 2002). Therefore, it can be suggested that the size of the firm can potentially explain significant variation in earnings management (Roychowdhury, 2006). Next to the effect firm size can have on earnings management, prior studies (McWilliams & Siegel, 2000; Prior et al., 2008; Waddock, 2003)) also show that firm size is correlated with CSR performance.
Return on assets (ROA) will be used to control for profitability and performance of the firm (Kim et al., 2012; Prior et al., 2008). Cochran & Wood (1984) found that there exists a correlation between CSR and the financial performance of the firm. Further Dechow et al. (1995) and Orlitzky et al. (2003) found that measures such as ROA are sensitive to manipulation by the management of the firm. Managers can make use of real and accrual earnings management to positively affect return on assets.
31
Market-to-Book ratio will be used to measure to perception of the market toward future growth. Literature suggest that growth stocks are particularly sensitive to stock price, and that the market reacts negatively to firms that break their string of consecutive earnings increases (Barth, Elliott, & Finn, 1999; Skinner & Sloan, 2002). Therefore, market-to-book ratio explains if a firm is under great pressure to adopt aggressive accounting policies, to increase earnings. The same principle counts for the sales growth.
Sales growth, measured as sales growth, divided by sales of the previous year. Firms that are experiencing strong growth, are subject of more media attention. Lee & Choi (2002) found that when a company are under closer scrutiny, there is a higher probability that they manage earnings.
Another control variable will be leverage, this will capture the impact of debt contracting on earning management. The relation between earnings management and leverage is subject to two posing empirical findings. Sweeney (1994) and Press and Weintrop (1990) suggested that high leverage firms tend to manage earnings aggressively. They find that firms response to debt contracting by reporting discretionary accruals (Becker, DeFond, Jiambalvo, & Subramanyam, 1998; Richardson et al., 2002). On the other hand, Dechow and Skinnder (2000), report that firm with high leverage are less likely to report small increases in earning. However, Chung and Kallapur (2003) do not find a relation between abnormal accruals and leverage at al. Meaning that the relation between leverage and earning management is uncertain.
A similar study (Kim et al., 2012) used R&D intensity as control variable, measured by R&D expense by net sales. McWilliams and Siegel (2000) found that R&D investment and CSR are correlated, because both are associated with product and process innovation. Therefore, an equation that includes CSR performance, should control for the upwardly biased estimates of CSR performance, by including a variable for R&D investments.
Auditor size will be used as control variable, in form of a dummy variable. Auditor offices are separated between being a Big-Four office (E&Y, Deloitte, KPMG, and PWC)
32
or not. Research found that big auditing offices can draw on expertise of their international network, that will result in higher audit quality (Carson, 2009; Francis & Wang, 2008). Further, research indicates that Big-Four auditors possess more reputational incentives than smaller auditors, therefore, Big-Four offices are less inclined to impair their independence for one client (Francis, 2004). Becker, Defond, Jiambalvo and Subramanyam (1998) studied the difference between being audited by the biggest audit offices and smaller audit offices, and their relation with earnings management. Becker et al. (1998) found that smaller offices report discretionary accruals that are, on average, 1.5-2.1 percent of total assets higher than the discretionary accruals reported by clients of big auditing firms.
All variables used in this study to find a relation between earnings management and CSR performance are described. The first formula to test this relation is stated below. The model presented below will be used to test hypotheses one and four.
𝑬𝑴 = 𝜶 + 𝜷𝟏 𝑪𝑺𝑹 + 𝜷𝟐 𝑰𝑵𝑫𝑼𝑺 + 𝜷𝟑𝑺𝑰𝑪 + 𝜷𝟑 𝑺𝑰𝒁𝑬 + 𝜷𝟒 𝑹𝑶𝑨 + 𝜷𝟓 𝑳𝑬𝑽 + 𝜷𝟔 𝑮𝑹𝑶𝑾𝑻𝑯 + 𝜷𝟕 𝑴𝑻𝑩 + 𝜷𝟖 𝑹&𝑫 + 𝜷𝟗 𝑩𝑰𝑮𝟒 + 𝜺
This model will test hypothesis two and three.
𝑪𝑺𝑹 = 𝜶 + 𝜷𝟏 𝑬𝑴 + 𝜷𝟐 𝑰𝑵𝑫𝑼𝑺 + 𝜷𝟑𝑺𝑰𝑪 + 𝜷𝟑 𝑺𝑰𝒁𝑬 + 𝜷𝟒 𝑹𝑶𝑨 + 𝜷𝟓 𝑳𝑬𝑽 + 𝜷𝟔 𝑮𝑹𝑶𝑾𝑻𝑯 + 𝜷𝟕 𝑴𝑻𝑩 + 𝜷𝟖 𝑹&𝑫 + 𝜷𝟗 𝑩𝑰𝑮𝟒 + 𝜺
Where:
EM = earnings management CSR = CSR performance
SIC = Standard Industrial Classification code
INDUS = dummy variable ‘one’ for CSR extensive industries, otherwise ‘zero’ SIZE = natural logarithm of total assets
ROA = return on assets (net income divided by total assets) LEV = total debt divided by total assets
33
MTB = market to book equity ratio (market value of equity divided by the book value of equity) R&D = R&D expense divided by the net sales
BIG4 = dummy variable ‘one’ if audited by Big 4 audit firm, otherwise ‘zero’
3.5 Sample
The sample will be based on archival data and needs to be selected in a way that the observations of the sample do not allow other factors to disrupt the relationship between earnings management, CSR and the industry of a firm. Therefore, the sample can only consist of companies that have the same economic and political environment. For this reason this study will only include companies based in the U.S. . Furthermore, the U.S. is one of the biggest industry and legislation regarding reporting of financial information is well developed. This study will only include observations from 2003-2009 to look at the relationship between earnings management and CSR. However to calculate discretionary accruals financial information of 2002 is also needed. This sample interval is chosen, because due to the changed regulation. The Sarbanes-Oxley Act of 2002, created new or enhanced standards for all U.S. public company boards, management and public accounting firms. The new and enhanced standards of SOX (Sarbanes-Oxley Act) had a significant effect on accrual-based earnings management (Cohen, Dey, & Lys, 2008). Due this strong difference in pre- and post-SOX accrual-based earnings management, only post-SOX years are taken into account. Further, most variables collected through the KLD database only go as far as 2009. Therefore, this study will focus on the years 2003 until 2009.
The sample first started with the different datasets. The COMPUSTAT dataset included over 76,000 observation and the KLD database over 20,000 observations. After merger the two dataset more than 11,000 observations remained. The first sample included U.S. corporations between the years 2003-2009. Then there was dealt with missing values by excluding cases list wise. This means that only observations that had no missing data in any variable are analyzed, the amount of observations with missing
34
values where 6,567. By performing a frequently test, the frequency of all variables were checked, no exceptional frequencies were found except the R&D variable. Due to this cumulating of data, 968 observations were deleted. Furthermore, 105 outliers were deleted, what resulted in a final sample of 5,494. The 5,494 observations will be used to conduct the examination of the two research models to test the hypothesis.
4. Results
This chapter will provide information about the sample and test the hypothesis using the models. The first paragraph will exists out descriptive statistics that will provide basic information about the sample, and an examination on the data in the sample. The second paragraph will provide an examination of two research models, by performing a regression analyses. The last paragraph will provide a robustness check, as extra measure to test the results found in the previous paragraph.
4.1 Descriptive statistics
Table 2 shows the industry specification for the firm-year observation for the sample. The sample is sorted by a two-digit SIC code. The frequency of industries is strongly focused on five industries: Chemicals; Allied Industries; Industrial and Commercial Machinery and Computer Equipment; Electronic and other Electrical Equipment and Components, except computer Equipment; Measuring, Analyzing, And Controlling Instruments; Photographic, Medical And Optical Goods; Watches And Clocks and Business Services, with respectively 18.97%, 13.09%, 17,18%, 14.45 and 15.25% of the sample.
35
Table 2
Frequency Industry Specifications
Industry SIC Frequency Percentage Cumulative
Metal Mining; 10 4 0.07 0.07
Coal Mining; 12 5 0.09 0.16
Oil And Gas Extraction; 13 38 0.69 0.86
Building Construction General
Contractors And Operative Builders; 15 5 0.09 0.95
Food And Kindred Products; 20 101 1.84 2.78
Tobacco Products; 21 3 0.05 2.84
Apparel And Other Finished Products
Made From Fabrics And Similar Materials; 23 3 0.05 2.89 Lumber And Wood Products, Except
Furniture;
24 7 0.13 3.02
Furniture And Fixtures; 25 36 0.66 3.68
Paper And Allied Products; 26 95 1.73 5.41
Printing, Publishing, And Allied Industries;
27 12 0.22 5.62
Chemicals And Allied Products; 28 1,042 18.97 24.59 Petroleum Refining And Related
Industries; 29 31 0.56 25.15
Rubber And Miscellaneous Plastics
Products; 30 65 1.18 26.34
Leather And Leather Products; 31 14 0.25 26.59 Stone, Clay, Glass, And Concrete Products; 32 44 0.80 27.39
Primary Metal Industries; 33 50 0.91 28.30
Fabricated Metal Products, Except Machinery And Transportation Equipment;
34 98 1.78 30.09
Industrial And Commercial Machinery And Computer Equipment;
35 719 13.09 43.17
Electronic And Other Electrical Equipment And Components, Except Computer Equipment;
36 944 17.18 60.36
Transportation Equipment; 37 251 4.57 64.93
Measuring, Analyzing, And Controlling Instruments; Photographic, Medical And Optical Goods; Watches And Clocks;
38 794 14.45 79.38
Miscellaneous Manufacturing Industries; 39 61 1.11 80.49
Transportation Services; 47 1 0.02 80.51
Communications; 48 53 0.96 81.47
Electric, Gas, And Sanitary Services; 49 3 0.05 81.53 Wholesale Trade-durable Goods; 50 13 0.24 81.76 Wholesale Trade-non-durable Goods; 51 11 0.20 81.96
Eating And Drinking Places; 58 7 0.13 82.09
