Master Thesis, Accountancy
The Relationship between Corporate
Social Responsibility Performance and
Public Country-by-Country tax
Reporting by Multinationals in Europe
Bjorn Lucas
S1887343
Bremstraat 49
9741 EB Groningen
+31612442394
b.lucas@student.rug.nl
Content
Abstract 3 1. Introduction 4 1.1. Scientific Contribution 7 1.2. Research Question 8 1.3. Research Structure 9 2. Theoretical Framework 9 2.1. Tax Transparency 9 2.1.1. CbC Reporting 102.2. Corporate Social Responsibility 11
2.2.1. CSR Performance 13
2.2.2. Stakeholder Theory 13
2.2.3. Legitimacy theory 14
2.3. Other factors 15
2.3.1. Extractive industry classification 15
2.3.2. Organizational size 16 2.4. Conceptual Model 17 3. Research Methods 17 3.1. Data sample 17 3.2. Variables 18 3.2.1. Dependent variable 18 3.2.2. Independent variable 19 3.2.3. Moderating variables 20 3.3. Methodology 21 4. Results 22 4.1. Descriptive statistics 22 4.2. Correlation analysis 23 4.3. Regression analysis 24
5. Results and Discussion 25
5.1. Conclusion 25
5.2. Limitations and recommendations 27
References 29
Appendix A: FTSEurofirst 100 datasample 39
Abstract
This study has examined whether CSR performance has a positive impact on public CbC Reporting. CSR performance is divided into three characteristics, namely environmental, social, and governance. These characteristics were merged in this study into an integral score; the ESG score. In addition, it has been investigated whether organizational size or extractive industry classification has an enhanced moderating effect on the relationships just described. Three hypotheses were tested to understand the relationship between CSR performance and public CbC Reporting of an organization. Conclusion is that: 1) There is no significant connection between CSR performance and public CbC Reporting; 2) There is statistical evidence to show that extractive industry classification has a significant effect on the relationship between CSR performance and public CbC Reporting; 3) Organizational size has no significant effect on the relationship between CSR performance and public CbC Reporting.
1.
Introduction
In April 2016 the world was shocked by the biggest data leak in history, called the Panama Papers. The Panama Papers are documents from the internal administration of a legal consultancy firm in Panama: Mossack Fonseca. They set up companies for customers in locations where their assets or properties are hardly taxed, so-called off-shore companies (The Guardian, 2016). Off-shore companies are legal, but they are also known to have little to no publicly available (tax) information. The Panama Papers have thus far uncovered the secret dealings of more than 140 prominent figures, and crimes that include bribery, arms deals, tax evasion, financial fraud, drug trafficking, and sanction busting (TUAC, 2016). This research focuses mainly on tax.
The Organisation for Economic Co-operation and Development (hereafter: ‘OECD’) is an association of 35 countries that study and coordinate social and economic policies, including for example the United States, United Kingdom and most of the European Union member states. The mission of the OECD states: “The OECD wants to promote policies that will improve the economic and social well-being of people around the world”. The OECD amongst others publishes standards regarding a wide range of topics, including tax standards. In addition, the G20 authorised the OECD to develop actions to which the aligned countries must comply (OECD, 2016).
The Trade Union Advisory Committee (hereafter: ‘TUAC’) to the OECD quickly responded to the revelations of the Panama Papers. The TUAC gave the advice that the OECD should, amongst others, put a much greater focus on corporate tax accountability and on how tax is treated as a risk in key OECD instruments on responsible business conduct, on corporate governance and on long term investment. The OECD should engage a discussion on the public disclosure of a Country-by-Country (hereafter: ‘CbC’) Reporting Framework (TUAC, 2016).
In 2013, the OECD and G20 countries published an initial report on its activities in connection with base erosion and profit shifting (hereafter: ‘BEPS’). The report contained fifteen action plans to address a range of issues including tax transparency, accountability, information exchange and other potential changes to international taxation (OECD, 2015). BEPS action 13 of this report addresses tax transparency of multinational enterprises (hereafter: ‘MNEs’) by providing revenue authorities with information to perform transfer pricing risk assessments (Longhorn et al., 2016). The main topic of BEPS action 13 is CbC Reporting. According to BEPS Action 13, MNEs should annually file, in whole or in part, their CbC report with local tax authorities for taxable periods commencing in 2016
(Lowell and Harrington, 2016). According to the OECD (2013), CbC Reporting means that “MNEs need to annually report for each tax jurisdiction where they do business in accordance with the laws of its jurisdiction of tax residence and with the information required to be reported under such laws”. The EU (2017) stated that tax transparency could be further improved by providing publicly available information relating to tax paid at the location where it is actually made. The European Union (hereafter: ‘EU’) (2017) defines this as public CbC Reporting; “the publication of a defined set of facts and figures by MNEs, thereby providing the public with a global picture of the taxes MNEs pay on their corporate income”.
CbC Reporting can only be applied to MNEs, which is defined by Murphy (2009) as an organization that operates in more than one (tax) jurisdiction. The OECD (2008) defines MNEs as “companies or other entities established in more than one country and so linked that they may co-ordinate their operations in various ways”. CbC Reporting usually applies to organizations that operate as groups; “organizations that structure their business through a number of organizations all of which are wholly, or partially owned by a holding or parent company”. These subsidiary organizations are located in different countries to manage the operations of the group in those different locations (Murphy, 2009). The relations between government, organizations and civil society could be improved by using CbC Reporting. For example, when the information of what MNEs do in every country becomes public, it could improve the oversight of both the government and civil society over the operations of a MNE (Wójcik, 2012). Although the previous proposes a major advantage of implementing CbC Reporting, business Europe is heavily opposed to public CbC Reporting (The Guardian, 2016). This gives rise to the question if organization’s stakeholders will accept minimal tax transparency after the scandals that derived from, for example, the Panama Papers.
In April 2016 the European Commission (hereafter: ‘EC’) initiated a proposal regarding the increase of tax transparency. This proposal amends the Accounting Directive (Directive 2013/34/EU of the European Parliament and the Council). The main goal of this proposal is that tax authorities will share information. This would require that multinationals that operate in the EU, and with global revenues of 750 million euro or more in a year, “publish key information on where they make their profits and where they pay their tax in the EU on a CbC basis”. This proposal concerns public CbC Reporting and should increase the accountability of large multinationals regarding tax matters. CbC Reporting is also part of the EU anti-tax avoidance package, presented on January, 28th 2016. The package was accepted on June, 21st 2016 and contains that MNEs in the EU must disclose CbC Reporting
information to tax authorities within the EU (EC, 2016). On January, 12th 2017 the European Parliament published an proposal for the implementation of public CbC Reporting (EU, 2017). The aim of the proposal is to add geographical information about corporate taxes in the countries where organizations actually make profits.
CSR Europe, which is the leading European business network for Corporate Social Responsibility (hereafter: ‘CSR’) announced a new project on tax transparency and responsible tax behaviour in November 2016. One of the deliverables of the project was to “support companies to move from tax transparency towards responsible tax behaviour through a learning network which can give inspiration to businesses through a wide array of experiences, knowledge and best practices from front-runners companies, policy makers and civil society actors”. According to CSR Europe (2016) a special focus is introduced to prepare organizations for public CbC Reporting. The reaction of CSR Europe (2016) on the proposal of the EC of April, 12th 2016 regarding Accounting Directive (Directive 2013/34/EU of the European Parliament and the Council) shows that there is a possible relationship between CSR and CbC Reporting.
Over the last thirty years, CSR continued to grow in significance and importance. Friedman (1970) found that is “the social responsibility of business to make profit”, but this seems outdated and too narrow. Carroll (1979) found four basic expectations that should reflect the view of social responsibility; the economic, legal, ethical, and discretionary categories of business performance. These responsibilities should address every business expectation towards society, which is already a broader perspective. The responsibilities mentioned by Carroll (1979) are further explained in the next chapter. Organizations have other responsibilities beyond making profits for shareholders (Carroll and Shabana, 2010). Dahlsrud (2006) expanded the responsibilities to five; the environmental, social, economic, stakeholder, and voluntariness responsibility. The stakeholder responsibility is introduced as a new concept in relation to CSR. Garriga and Mele (2004) state that “a socially responsible firm requires simultaneous attention to the legitimate interests of all appropriate stakeholders and has to balance such a multiplicity of interests and not only the interests of the firm’s stockholders”. Stakeholders are the shareholders, creditors, employees, suppliers, and the communities at large (Campbell, 2007; Clarkson, 1995). The stakeholder theory points out that organizations have a social responsibility that requires them to consider the interests of every party affected by their actions (Branco and Rodrigues, 2007) and offers a new way to organize thinking about organizational responsibilities (Jamali, 2008).
The question arises whether paying taxes should be seen as a CSR issue. There is research on whether tax planning or tax avoidance can be seen as issues of CSR (Avi-Yonah, 2008; Hasseldine and Morris, 2013; Sikka, 2010, 2013; Ylönen and Laine, 2015). Avi-Yonah (2008) answers this question affirmatively. He points out that if an organization engages in CSR as a legitimate corporate function it is expected that it pays its taxes to strengthen society as part of their adoption of CSR. In addition, Sikka (2010) found that taxation should be positioned within a broader socio-political perspective of ethics, power and authority. Hasseldine and Morris (2013), however, found that taxation should be seen from a legal point of view and that taxation is essentially a technical matter, instead of a social matter. They state that tax avoidance or tax evasion could not exist without the existence of regulations and tax codes (Hasseldine and Morris, 2013).
This research will examine the CSR performance of organizations, embedded in the Financial Times Stock Exchange Eurofirst (hereafter: ‘FTSEurofirst’) 100, and whether it will improve tax transparency.
1.1. Scientific Contribution
Several researchers studied the relation between CSR and tax. Some studies looked at the relation between different CSR activities and tax avoidance (Hoi et al., 2013; Huseynov and Klamm, 2012; Lanis and Richardson, 2015). Hoi et al. (2013) found that when firms engage in irresponsible CSR activities, there is a higher possibility that these firms engage in tax-sheltering activities and have greater discretionary/permanent book-tax differences. Huseynov and Klamm (2012) examined the effect of corporate governance, community and diversity on tax avoidance in firms with auditor provided tax services. They found that CSR affects tax avoidance in that the effective tax rate is closer to the statutory tax rate of organizations. Lanis and Richardson (2015) examined whether CSR performance is associated with corporate tax avoidance. Stainer (2006) defines performance on CSR as the ability to achieve goals in a resourceful and consistent manner. By assessing CSR performance an organization has the opportunity to “identify its strength and weaknesses, modify their strategies, and identify opportunities for improvement” (Giannarakis and Litinas, 2011). Lanis and Richardson (2015) found that when the level of CSR performance is higher, the possibility of tax avoidance will decrease.
Lanis and Richardson (2013) tested the legitimacy theory empirically by comparing CSR disclosures of tax aggressive companies. Legitimacy theory indicates that when there is a difference between
business actions and societal expectations, the management uses particular disclosure reports as to help to dispel concerns in the community (Lanis and Richardson, 2013). The results of Lanis and Richardson (2013) show a positive and statistically significant association between corporate tax aggressiveness and CSR disclosure, which means that tax aggressive organizations have more CSR disclosure. The legitimacy theory in the context of corporate tax aggressiveness was therefore confirmed (Lanis and Richardson, 2013). Zeng (2016) indicates that firms which act socially responsible will not engage in aggressive tax activities, while firms that are not interested in acting socially responsible are more likely to engage in aggressive tax activities. Fatma et al. (2015) states that the payment of corporate taxes can only be associated with CSR if it indeed has implications for the wider community, but confirms in the same research that paying taxes does have community and societal implications because it forms the vital function of helping to fund the provision of public goods in society. Both CSR and tax transparency are affiliated with voluntary disclosure.
This research contributes to the existing literature on the topic of the relation between CSR and tax issues by using a dataset of ASSET4 ESG, which is part of Datastream and contains economic, environmental, societal and governance rating scores of organizations. As mentioned before, this study will use organizations of the FTSE 100. There is no empirical evidence found yet in other studies on whether CSR performance affects the CbC Reporting of these organizations.
1.2. Research Question
This research will focus on whether CSR performance affects the CbC Reporting of organizations which are obtained in the FTSEurofirst 100. CSR performance will be measured by the social, environmental, and governance performance of these organizations. The first is focused on maintaining and improving the ecosystem and sustainability. The second has a stakeholder orientation and the last has institutional theory effectiveness as a goal (Hartmann and Uhlenbruck, 2015; Visser, 2014). The data for CSR performance is collected from the ASSET4 ESG database of Datastream and the data regarding CbC Reporting is obtained from the annual reports of the organizations embedded in the FTSEurofirst 100.The main research question of this study is:
To what extent does CSR performance affect the public CbC reporting of organizations of the FTSEurofirst 100?
1.3. Research Structure
This research is set up as follows. The next chapter elaborates on the theory, as well as the development of hypotheses. Chapter three discusses the research methods. Chapter four will show the results of this research and chapter five concludes the research with conclusions of the study, limitations, and some recommendations for future research.
2.
Theoretical Framework
In this section the following topics will be discussed: tax transparency, CbC Reporting, CSR, CSR performance, stakeholder theory, legitimacy theory, as well as the relationship between the variables, hypotheses and factors that might influence this relationship. A conceptual model will conclude this chapter.
2.1. Tax Transparency
Transparency implies that organizations will go the extra mile to ensure stakeholders are well-informed. Transparency has the potential to benefit an organization’s employees, customers, and partners, as well as entire societies. Trust is the most important benefit for transparent organizations (Parris et al., 2016). Fiscal transparency is defined as openness towards the public about functions, fiscal policy intentions, sector accounts, and projections. It involves ready access to reliable, comprehensive, timely, understandable, and internationally comparable information (Ellis and Fender, 2006). Ellis and Fender (2006) address internationally comparable information, which is also the main topic of BEPS Action plan 13 of the OECD: CbC Reporting. CbC Reporting mainly demands that MNEs “publish a profit and loss account and limited balance sheet and cash flow information for every jurisdiction in which they trade as part of their annual financial statements (Murphy, 2016). The OECD recommends that the first CbC reports should be required from January, 1st 2016 and it should apply to MNEs. In its final report on Action 13 the OECD sets the following necessary conditions underpinning the obtaining and the use of the CbC Reporting; confidentiality, consistency, and appropriate use. The OECD (2017) published a list on May, 24th 2017 of participating countries regarding CbC reporting to tax administrations; Argentina, Australia, Austria, Belgium, Bermuda, Brazil, Canada, Chile, China, Costa Rica, Curaçao, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Gabon, Georgia, Germany, Greece, Guernsey, Hungary, Iceland, India, Indonesia, Ireland, Isle of Man, Israel, Italy, Japan, Jersey, Korea (Rep.), Latvia, Liechtenstein, Lithuania, Luxembourg, Malaysia, Malta, Mauritius, Mexico, Netherlands, New Zealand, Nigeria, Norway, Poland, Portugal, Russia, Senegal, Slovakia, Senegal, South Africa, Spain, Sweden, Switzerland, United Kingdom, and Uruguay.
This study will show if MNEs, who are embedded in the FTSEurofirst 100, use CbC Reporting in their annual reports of 2011-2015 and whether it is affected by CSR performance and other factors.
2.1.1. CbC Reporting
In 2013 the G20 had a meeting about the problems concerning tax avoidance. The G20 countries agreed to start a project that should tackle the BEPS problems and the OECD was assigned to come up with this plan. The OECD published an action plan containing fifteen actions. Action 13 was designed to tackle the BEPS problems and one of the main topics is CbC Reporting. Action 13 contains three components. The first component is a master file with standardized information. The second component is a local file regarding transaction with regard to local taxpayers and the third and final component is a CbC report containing information about group income and tax payments and certain indicators of the location of economic activity (OECD, 2014). It is important to recognize that the information will remain confidential and does not have to be disclosed in a public report of a MNE. CbC became mandatory to tax administrations since January, 1st 2016 for MNEs in some countries with a turnover of €750 million or higher (PwC, 2016). The European Commission launched Council Directive (EU) 2016/881 to amend Directive 2011/16/EU, which was unanimously adopted on May, 25th 2016 and regards the exchange of tax information between tax administrations. On April, 12th 2016 the European commission launched a proposal to amend Directive 2013/34/EU. This proposal suggested to make CbC Reporting of MNEs public. It requires that MNEs publish information about their operations in EU Member States. According to chapter 10 of Directive 2013/34/EU it is required that organizations operating in the extractive and logging industries have to publish their payments to governments on a CbC basis. Public CbC Reporting is also required from EU credit institutions.
Although disclosure to tax administrations was not mandatory before 2016, organizations could have some reasons to disclose information about taxation in their annual reports. Managers could have strategic intentions by releasing information about the organization (Ferreira and Rezende, 2007; Holland, 2005; Lev, 1992). Lev (1992) finds that disclosure of information affects company activities. Disclosure firstly affects the perceptions of stakeholders about an organization and these perceptions of stakeholders will then affect the organization’s decisions and performance (Lev, 1992). Ferreira and Rezende (2007) found that public announcements of management are credible because managers are concerned about their reputation. This means that managers have a positive strategic incentive regarding the release of public information.
There are critics regarding the use of CbC Reporting (Evers et al., 2014). Evers et al. (2014) discussed whether CbC Reporting prevents MNEs profit shifting behaviour. The possible benefits of CbC Reporting are highly uncertain because MNEs can still legally exploit gaps and loopholes in national and international tax laws. While there is criticism on CbC Reporting, more organizations voluntarily implemented CbC Reporting more over the years. Transparency and reduced information asymmetry can be seen as key arguments for voluntary disclosure (Broberg et al., 2009).
2.2. Corporate Social Responsibility
CSR has emerged in the first decade of this century as both an important academic ideal and a pressing item on the corporate agenda. More and more companies are embracing CSR and feel the need to give some disclosure about it (Matin et al., 2009). CSR has a significant influence on consumer behaviour where consumers are demanding much more from a company than a low price and good quality product. Prior studies have found that CSR activities are one of the strategies that businesses employ when attempting to improve financial performance (Fatma et al., 2015).
Managers of organizations continually encounter demands from multiple stakeholder groups to devote resources to CSR. The definition of CSR is not always clear (McWilliams and Siegel, 2001). Carroll (1979) started with four basic expectations regarding social responsibilities; economic, legal, ethical, and discretionary (philanthropic) categories of business performance. Carroll (1991) depicted these components as a pyramid, wherein the economic performance underpins everything else. The legal component includes the expectation of an organization to comply with all laws and regulations. “The law is society’s codification of acceptable and unacceptable behavior”. The ethical responsibilities consist of the expectation to what is right, just, and fair. It is also important to avoid or to minimize harm to stakeholders. Finally, the philanthropic responsibility expects organizations to be good corporate citizens (Carroll, 1991). In general, the basic expectations of Carroll (1979) should address every business expectations towards stakeholders and society. McWilliams and Siegel (2001) defined CSR as actions that appear to further some social good, beyond the interests of the firm and that which is required by law. So an organization supporting CSR accepts ethical expectations above and beyond the law (Gribnau, 2015). Dahlsrud (2006) states that the definition of CSR is biased, because of differences in specific interests and expertise. He categorized all the different CSR definitions into five different expectations; the environmental, social, economic, stakeholder, and voluntariness expectations. The definition of the European Commission (2001) covers all these expectations. It defines CSR as “a concept whereby companies integrate social and environmental concerns in their
business operations and in their interactions with their stakeholders on a voluntary basis”. The definition of Gamerschlag et al. (2011) is much more limited. They define CSR disclosure as “the information that a company discloses about its environmental impact and its relationship with its stakeholders by means of relevant communication”. Fuente et al. (2017) define disclosure of information about the exercise of CSR as what on average is used by most companies to facilitate understanding of the social and environmental performance of an organization and to improve relationships with stakeholders.
It varies per country whether the disclosure is voluntary. In some countries there are some specific guidelines or requirements for the provision of CSR information (Kolk et al., 2001). The question then arises as to why an organization would voluntarily disclose CSR information (Dhaliwal et al., 2011). Lyon and Wood (2008) found that organizations with poor reputations will fully disclose, since they could gain from an announced success and lose little from an announced failure. On the other hand, organizations with an excellent reputation will disclose nothing, since they could lose a lot by disclosing a failure and win little from an announced success (Lyon & Wood, 2008). Cao et al. (2015), more recently, found that organizations with a high reputation provide more and better voluntary disclosures, because this is in line with their commitment to protecting stakeholder interests.
According to Bhattacharya and Sen (2004) it is important for organizations to understand the reactions of consumers to CSR, because the organization can develop CSR strategies that are optimal from both a normative perspective, as a business perspective. This turns out to be complicated, since people do not seem honest about the importance of CSR activities. According to Bhattacharya and Sen (2004) people say, when asked directly, that they take CSR matters into account in their purchasing decisions, but statistics mask the actual nature of consumer response to CSR activities. Bhattacharya and Sen (2004), having a U.S. perspective, mapped out six basic domains in which a company can engage in CSR activities; community support, diversity, employee support, environment, non-U.S. operations, and product.
As mentioned in the introduction, research has been done on the relationship between CSR activities and tax issues (Hoi et al., 2013; Huseynov and Klamm, 2012). While these studies examined a possible relationship regarding specific CSR activities like corporate governance or diversity, Lanis and Richardson (2015) examined CSR performance. The next section will elaborate on CSR performance.
2.2.1. CSR Performance
As with the definition of CSR, there could also arise some difficulties in defining CSR performance. The difficulties of making subjective measurement of CSR performance could probably lead to different perceptions of outcomes. Companies should identify their strength and weaknesses, modify their strategies and define opportunities for improvement when assessing CSR performance (Giannarakis et al., 2011). Visser (2014) has another view about CSR, since he focuses on the DNA of CSR. Visser (2014) states that the DNA of CSR nowadays contains the following elements; good governance, societal contribution and environmental integrity. Kock et al. (2012) found in 1998 and 2000 that there was scarce attention aimed at corporate governance mechanisms between managers and stakeholders and it was not yet rightfully recognized or integrated in CSR (Kock et al., 2012). Visser (2014) points out that good governance should have institutional effectiveness as a goal, and that the societal contribution should focus on the stakeholders. Environmental integrity should focus on maintaining and improving ecosystem sustainability.
To ensure that a company is socially responsible, and thus in order to measure CSR performance, it is important to express principles of CSR in measurable variables (Escrig-Olmedo et al., 2010). Independent agencies, such as Kinder, Lydenberg and Domini, Bloomberg, and Thomson Reuters, rate and rank organizations based on their CSR performance (Ioannou and Serafeim, 2012). With Thomson Reuters it is possible to obtain economic, environmental, societal and governance (herafter: ‘ESG’) metrics, which is in line with the characteristics that Visser (2014) mentioned about the DNA of CSR.
Lanis and Richardson (2015) found that organizations with a high level of CSR performance are less likely to engage in tax avoidance practices. In view of the lack of prior literature on the relationship between CSR performance and CbC Reporting, this research investigates the following hypothesis:
Hypothesis 1: CSR performance will positively influence public CbC Reporting.
2.2.2. Stakeholder Theory
In the 1990s, a major theme arose in addition to the existing CSR aspects; the stakeholder theory (Carroll, 1999). Campbell (2007) states that an organization is acting socially responsible when the organization does not harm its stakeholders on purpose. If the organization causes harm to their stakeholder, they must rectify the harm immediately when it is discovered. It is possible that the
stakeholder theory gives a better insight into the question of why organizations act socially responsible.
Freeman (1984) defines the stakeholder theory as a combination of groups and/or individuals that are affected or influenced by the activities of an organization (Freeman, 1984). These groups and/or individuals are usually shareholders, investors, consumers, employees, governments, media, local communities and the natural environment (Neville et al., 2005). Stakeholder theory is based on the “fiduciary duty of managers to promote the interests of a wide range of parties who are affected by the activities of their business” (Norman, 2008). An et al. (2011) describes the stakeholder theory as one which is “concerned with the relationships of an organization with a variety of stakeholders in society”. Organizations should try to meet multiple goals of a wide range of stakeholder, instead of just those of shareholders (An et al., 2011).
Clarkson (1995) defines stakeholders as those who placed something at risk in relationship with an organization. Clarkson (1995) distinguished two types of stakeholders; primary and secondary stakeholders. Primary stakeholders are those “without whose continuing participation the corporation cannot survive as a going concern”. Primary stakeholders are the shareholders, investors, employees, customers, suppliers, government and communities. Secondary stakeholders are described as “those who influence or affect, or are influenced or affected by the organization, but are not engaged in transactions with the organization and are not essential for its survival”. This distinction is important since there is a great level of interdependence between the organization and its primary stakeholders, and an organization is in its survival not dependable of secondary stakeholders. This study therefore will focus on the primary stakeholders; shareholders, investors, employees, customers (Campbell, 2007; Clarkson, 1995; Neville et al., 2005).
2.2.3. Legitimacy theory
The legitimacy theory is another theory which concerns the relationship between the organization and the society at large (An et al., 2011). Panwar et al. (2014) found that CSR is a manner through which organizations can attain legitimacy. CSR activities can help organizations to build a better reputation in society and help organizations to align with sociocultural norms of their environment. Organizations that derive legitimacy through CSR can improve confidence and consumer protection, provide greater attraction towards existing and potential employees and improve their appeal towards investors (Panwar et al., 2014). The legitimacy theory can be seen as an expansion of the stakeholder theory, since it not only holds the stakeholders’ perspective in mind but of the society as a whole. According to Deegan (2002) this results in “a poor resolution given that society is clearly made up of
various groups having unequal power or ability to influence the activities of other groups”. Deegan (2002) states that the stakeholder theory “explicitly accepts that different groups have different views about how organizations should conduct their operations, and have different abilities to affect an organization”. Clarkson (1995) supports this by stating that organizations manage relationships with their stakeholders and not with the society as a whole.
2.3. Other factors
The relationship between CSR performance and CbC Reporting is expected to be influenced by other factors. In line with research of Lanis and Richardson (2015) about CSR performance, this study will use the following factor that could moderate the relationship: extractive industry classification. Lanis and Richardson (2015) also used firm size to moderate a relationship. For this study this may seem unnecessary since this study is about CbC Reporting of MNEs, which means large companies. Within the FTSE 100 a distinction can be made between small, medium, and large organizations, so organization size will also be taken into account in this study. The variables will be discussed below.
2.3.1. Extractive industry classification
Previous research has recognized that CSR disclosure is different in certain industries (Cormier and Magnan, 2003; Gao et al., 2005; Roberts, 1992). Cormier and Magnan (2003) and Gao et al. (2005) acknowledged that industries differ in characteristics, which could relate to risks regarding the society, competition, opportunities and government interference. Roberts (1992) finds that organizations that operate in so-called high profile industries have higher levels of CSR disclosures. Gamerschlag et al. (2011) also recognized the fact that organizations with a great environmental impact have more incentives to disclose CSR information. For example, chemical companies are more environmentally sensitive and thus are more likely to disclose CSR information to the public than organizations in other industries (Meek et al., 1995). It is therefore expected that organizations that operate in an environmentally sensitive industry will disclose more information. This is supported by developments in tax. In 2002, the Extractive Industries Transparency Initiative (hereafter: ‘EITI’) was created, which is “a public-private partnership designed to help resource-rich countries avoid corruption in the management of extractive industry revenues” (Aaronson, 2011). Evers et al. (2014) states that the creation of EITI was not because of tax reasons, but because of a high risk of corruption in this industry. The aim of EITI is to have greater transparency in financial transactions within extractive industries (EITI, 2017). “This should improve the use of natural resources and contribute to sustainable development” (Wójcik, 2012). EITI is a voluntary initiative, but partners need to take specific actions. “First, governments must require extractive firms operating within their territory to
‘‘publish what they pay’’ policymakers for the right to explore and extract energy or minerals. Second, government officials must record the revenues they receive and entrust an independent administrator to compare extractive sales and revenues. Then, governments must create a multi-stakeholder group which includes civil society representatives. The multi-multi-stakeholder group is tasked to evaluate the information provided by business and government and review the independent evaluation. Finally, an outside organization checks or validates the reports” (Aaronson, 2011). Implementation of EITI is the responsibility of governments. After implementation, organizations will disclose payments to governments and governments will disclose what they receive (EY, 2013). This study investigates the relations between CSR performance and CbC Reporting about tax information, which includes the disclosure of tax information. Based on prior literature and the developments regarding EITI, the following hypothesis has been prepared:
Hypothesis 2: The relationship between CSR performance and public CbC Reporting will be stronger
for organizations that operate in the extractive industry.
2.3.2. Organizational size
Hackston and Milne (1996) find a relationship between organizational size and disclosure, based on the agency theory and the legitimacy theory. Since larger companies have a greater impact on society and have more shareholders, they tend to communicate information through their annual reports (Hackston and Milne, 1996). Gamerschlag et al. (2011) found that large organizations disclose more than small organizations. Gao et al. (2005) concluded that organizational size is positively related to CSR reporting. Thereby, “large firms tend to be more visible to relevant publics and so tend to be subject to greater political and regulatory pressure from external interests”. This is an incentive for large firms to make disclosures to show the public that their actions are legitimate (Brammer and Pavelin, 2006).
There is a lack of literature regarding tax disclosure, or CbC Reporting, in relation to organizational size. CSR reporting can be a voluntary action and public CbC Reporting in most countries is voluntary as well. Exception regarding CbC Reporting is the United States, where tax should be reported CbC by MNEs with a revenue of 850 million dollars or higher. Based on prior literature, the following hypothesis has been stated:
Hypothesis 3: Organizational size positively influences the relationship between CSR performance
2.4. Conceptual Model
The previously described concepts give rise to the following conceptual model with public CbC Reporting as dependent variable and CSR performance as independent variable. The other factors that could influence the relationship between CSR performance and public CbC Reporting consist of the extractive industry and organizational size.
Figure 1: Conceptual model
3.
Research Methods
This section will discuss the research methods being used, the data sample, the variables and the descriptive statistics being used.
The aim of this study is to research the relationship between CSR performance and CbC Reporting. The dependent and independent variables consist of ordinal data, so a multiple regression method is an appropriate method for this study. Multiple regression analysis is run with ordinary least squares (hereafter: ‘OLS’) using statistics of SPSS. The research is of a quantitative nature and is suitable to answer the research question. The variables and other factors that might moderate the relationship between these variables are discussed below.
3.1. Data sample
This study first wanted to investigate organizations embedded in the Dutch Association of Investors for Sustainable Development (hereafter: ‘VBDO’) and their tax transparency. The VBDO is committed to a sustainable capital market which, in addition to financial criteria, also takes into account social and environmental criteria. In order to create a sustainable capital market, VBDO composed a Tax Transparency Benchmark (VBDO, 2016). The Tax Transparency Benchmark
consists of 64 Dutch listed companies, which are ranked on the transparency that they provide into their responsible tax policy. This seemed like a good match with the variables of this study. Information regarding CSR performance was hard to obtain for the organizations embedded in the VBDO. This is the reason why this study chose another data sample.
The sample consists of 88 organizations embedded in the FTSE 100. “The FTSE 100 is a market capitalization weighted index of UK-listed companies. It is designed to measure the performance of the 100 largest companies traded on the London Stock Exchange that pass screening for size and liquidity”. (FTSE Russell, 2017). A list of the sample is set out in appendix A. A few organizations were eliminated since there was no information available about the CSR performance in 2015 or due to an outlier in CSR performance.
Data is obtained for a five-year period (2011-2015), which will mean that the sample has 445 observations. The CbC Reporting regarding tax information of organizations is obtained from the annual reports. The sample period ends in 2015, since the annual reports of 2016 will not be accessible in time for this study. To get the values for CbC Reporting, this study used a dummy variable, with value 1 for CbC Reporting in the annual reports and value 0 for no CbC Reporting regarding tax information. The CSR performance was obtained from the ASSET4 ESG database by Thomson Reuters, which is part of Datastream (Escrig-Olmedo et al., 2010). ASSET4 collected data and rated organizations on ESG principles since 2002. The data they collected must be publicly available and objective (Ribando and Bonne, 2010).
3.2. Variables
3.2.1. Dependent variable
As mentioned before, CbC Reporting became mandatory as of January, 1st 2016. So during the period 2011-2015 the CbC Reporting was voluntary. The disclosure was recorded in the annual reports, country reports, or tax reports and data will be collected from these reports. To research the CbC Reporting a dummy variable is used, with value 1 when an organization disclosed its tax information in all years for each country and value 0 if an organization does not disclose its tax information. A value between 0 and 1 is also possible if an organization disclosed CbC Reporting in some years but not in all. The weighted average is taken to measure the CbC Reporting for an organization.
3.2.2. Independent variable
The independent variable in this study is CSR performance. In previous research, there has not been a unambiguous method used to measure CSR performance. According to Giannarakis et al. (2011) there are five different approaches to assess CSR performance: measurements based on the analysis of the contents of the annual reports, pollution indices, perceptual measurements derived from questionnaire based surveys, corporate reputation indicators and data produced by measurement organizations.
Some studies used the KLD database when measuring the CSR performance (Lanis and Richardson, 2015; Lioui and Sharma, 2012). The KLD database is an independent rating service that concentrates on assessing CSR performance across a range of dimensions which are related to stakeholder concerns (Waddock and Graves, 1997). The database measures strengths and concerns of seven major categories: community relations, corporate governance, diversity, employee relations, environment, human rights, and products (Lanis and Richardson, 2015). They found evidence that the KLD database is well-correlated with measures of CSR. The ASSET4 ESG database on the other hand does not distinguish strengths and concerns of CSR, but focuses on the environmental, social and governance performance of an organization. In order to measure ESG scores, this study used the equally weighted average of the environmental, social and governance scores over the years 2011-2015. The environmental performance measures the emission reduction, product innovation and resource reduction of an organization, which are all factors that could impact living and non-living natural systems. This performance shows how an organization uses “best management practices to avoid environmental risks and capitalize on environmental opportunities in order to generate long term shareholder value (Thomson Reuters, 2017). The social performance focuses on the organization gaining trust and loyalty with their employees and its society. This reflects the reputation and health of its license to operate of an organization. These are necessary factors in order to generate long term shareholder value. Categories to measure the social performance are; product responsibility, community, human rights, diversity and opportunity, employment quality, health and safety, and training and development (Thomson Reuters, 2017). The governance performance measures the systems and processes of an organization, which should ensure that the board members and executives are acting in the best interest of its long term shareholders. The governance performance takes the following categories into account; board functions, board structure, compensation policy, and vision and strategy (Thomson Reuters, 2017). In table 1 an overview of the environmental, social, and governance performance and its categories is provided. The average, or integrated rating, of the scores on environmental, social, and governance factors will be used to measure the CSR performance, which is the logarithm of a score out of 100 (Thomson Reuters, 2017).
Table 1: Performance indicators
Environmental performance Social Performance Corporate governance performance
Resource Reduction Employment Quality Board Structure Emission Reduction Health & Safety Compensation Policy Product Innovation Training & Development Board Functions
Diversity Shareholder Rights
Human Rights Vision & Strategy Community
Product Responsibility
Although the database has not been used that often in previous research, the choice for this database is grounded in the fact that it is appropriate for how CSR performance is defined in this study. Visser (2014) focuses on the DNA of CSR to measure the CSR performance. The DNA of CSR contains good governance, societal contribution and environmental integrity (Visser, 2014). This is in line with the performance scores measured by ASSET4 ESG.
This study used Z-scores on CSR performance to identify potential outliers in the data sample. In accordance with Miller (1991) this study eliminated organizations that had an outlier of 3 or higher/lower. This reduces the effect of the extreme value in order to protect the accuracy of this research. One organization was eliminated from the data sample, due to outliers.
3.2.3. Moderating variables
The moderating variables consist of extractive industry classification, and organizational size. For the extractive industry classification a dummy variable is used, with value 1 for an organization that operates in the extractive industry and value 0 for an organization that operates in another industry.
The FTSEurofirst 100 sample in this study consists of 88 organizations, all with their own index weight as a percentage of the total. An organization with an index higher than 2% is considered a large organization. An organization with an index between 1% and 2% is considered a medium-sized organization and an organization below 1% is considered a small organization. A large organization will have value 1, a medium-sized organization value 0,5 and a small organization value 0. This properly refutes any possible differences.
3.3. Methodology
In order to obtain the results of the research, a linear regression analysis is used. The dependent variable, or public CbC Reporting, is the result of this analysis. When a significant difference occurs, it can be concluded whether the independent variables affect the dependent variable. This study uses coefficients, namely ß0 to ß4. These coefficients relate to the independent variable and the
moderators. Table 2 shows the outcomes and coefficients with the corresponding descriptions. Table 2: Description of the variables and moderators
Variable Description
Dependent variable: CbCr Public Country-by-Country reporting: weighted average of CbC Reporting in public reports (score between 0 and 1)
Independent variable: CSRPerf Environmental, social, and governance score: average of these pillars measured through the ASSET4 ESG database (logarithm score out of 100)
Moderator: ExtInd Extractive industry classification: dummy variable, with value 1 for organization in extractive industry and value 0 for an organization in another industry.
Moderator: OrgSize Organizational size: 1 for large, 0,5 for medium and 0 for small
These variables led to the following formula: 𝑃𝑢𝑏𝑙𝑖𝑐 𝐶𝑜𝑢𝑛𝑡𝑟𝑦 − 𝑏𝑦 − 𝐶𝑜𝑢𝑛𝑡𝑟𝑦 𝑟𝑒𝑝𝑜𝑟𝑡𝑖𝑛𝑔
= 𝛽0 + 𝛽1 ∗ 𝐶𝑆𝑅𝑃𝑒𝑟𝑓 + 𝛽2 ∗ 𝐸𝑥𝑡𝐼𝑛𝑑 + 𝛽3 ∗ 𝑂𝑟𝑔𝑆𝑖𝑧𝑒 + 𝛽4 ∗ 𝐶𝑆𝑅𝑃𝑒𝑟𝑓 ∗ 𝐸𝑥𝑡𝐼𝑛𝑑 ∗ 𝑂𝑟𝑔𝑆𝑖𝑧𝑒 + ℇ
Βt = the coefficients, ℇt = error term and t indicates years. Definition of the variables can be found in table 2.
4.
Results
In this chapter the research results will be presented and explained. First, the descriptive statistics will be addressed. Next, a correlation analysis will be described in which multicollinearity is tested. Finally, the results of the regression analysis will be discussed.
4.1. Descriptive statistics
The descriptive statistics consist of the minimum, maximum, average, and standard deviation of the dependent variable, independent variable, and moderators. Table 3 below shows the results of the descriptive statistics.
To investigate the descriptive statistics, Table 3 is used to compare the values of the independent variable and moderators with the values of the dependent variable. When the dataset is being investigated using the descriptive statistics, there is no conclusion as to whether CSR performance has a positive impact on public CbC Reporting. With a CSR performance score above the mean, public CbC Reporting is not always applied by organizations. This could mean that a high CSR performance has no influence on whether organizations apply public CbC Reporting. On the other hand, it is often the case that when organizations have applied public CbC Reporting, they are above the mean of CSR performance. This could mean that higher CSR performance results in the application of public CbC Reporting. Organizations that are active in the extractive industry apply public CbC Reporting to a greater or lesser extent. This could mean that the extractive industry classification positively influences the relationship between CSR performance and public CbC Reporting. Thereby, the organizations that are classified as large do not seem to apply public CbC Reporting to a greater extent than medium or small sized organizations. This could mean that organizational size does not influence the relationship between CSR performance and public CbC Reporting.
It can be concluded from Table 3 that on average, 20% of the observations have public CbC Reporting, but that the variation between organizations with or without public CbC Reporting is high since the standard deviation is higher than the mean. The mean of CSR performance is quite high and it has a relatively low standard deviation, which means that the variation between organizations in CSR performance of the observations is low. Both moderators have a higher standard deviation than the mean. This means that the variation is high with the organizational sizes, and extractive industry classification of the organizations.
Table 3: Descriptive statistics N=88 Description Minimu m Maximum Mean Std. Deviation Dependent variable Public CbC Reporting 0 1 0,2090 0,353 Independent variable CSR performance 0,632 0,934 0,853 0,061
Moderators Extractive industry classification
0 1 0,16 0,366
Organizational size 0 1 0,243 0,354
N= number of observations
4.2. Correlation analysis
Table 4 represents the correlation matrix in which the correlations are noted between the specific variables. The correlation matrix will prove if there is multicollinearity that is “an interdependency condition that can exist quite apart from the nature, or even the existence, of dependence between variables” (Farrar and Glauber, 1967). When the value is greater than 0.8, or less than -0.8 in the correlation matrix, then multicollinearity can be an issue (Brooks, 2014). The correlation analysis show no multicollinearity (all values above -0,8 and below 0,8) and the regression analysis can be executed.
During the regression analysis, the Variance Inflation Factor (hereafter: ‘VIF’) value is also considered to exclude multicollinearity. Multicollinearity can be an issue when the VIF score is higher than around 10. If the VIF value exceeds this value, it might be necessary to eliminate a variable or combining two variables (O’Brien, 2007). The highest VIF scores per model are shown in table 4. The VIF scores do not exceed the value of 1,023 and thus it can be concluded that multicollinearity will not be a problem.
Table 4: Correlation matrix Variable (1) (2) (3) (4) 1. CbCr 1 2. CSRPerf -0,029 1 3. ExtInd 0,218** -0,149 1 4. OrgSize -0,074 0,055 0,097 1 ***: Significant on 0,01 level **: Significant on 0,05 level *: Significant on 0,10 level
Correlations are based on 88 observations
4.3. Regression analysis
The regression analysis has been used to test the hypotheses of this research. Table 5 summarizes the main findings of the regression analysis. For all models, a significance level of 0,05 is used unless otherwise indicated. If the level of significance is below 0,05, a hypothesis can be accepted. If the significance level exceeds 0,05, the hypothesis will not be accepted. Model 1 contains the dependent variable and the independent variable and tests hypothesis 1. The regression analysis is negative but not significant. The relationship is negative, which is opposed to hypothesis 1. The values are stated in table 5 (β = -0,17, p = 0,79). Therefore it cannot be concluded that CSR performance influences public CbC Reporting.
Model 2 is used to statistically test hypothesis 2. Model 2 contains the dependent variable, the independent variable, and the extractive industry classification moderator. The regression analysis is positive and significant. The values are stated in table 5 (β = 0,21, p = 0,04). Therefore it can be concluded that the relationship between CSR performance and public CbC Reporting is stronger for organizations that operate in the extractive industry. Article 42 of Directive 2013/34/EU states: “Member States shall require large undertakings and all public-interest entities active in the extractive industry or the logging of primary forests to prepare and make public a report on payments made to governments on an annual basis”. This is in line with the results of model 2. Hypothesis 2 is accepted. Model 3 is used to statistically test hypothesis 3. Model 3 contains the dependent variable, the independent variable, and the organizational size moderator. The regression analysis is negative, but not significant. The relation is negative, opposed to what hypothesis 3 described. The values are stated in table 5 (β = 0,14, p = 0,47). Based on these results, hypothesis 3 cannot be accepted. So,
organizational size does not have a significant influence on the relationship between CSR performance and public CbC Reporting.
The results will be further discussed in the next chapter.
Table 5: Regression analysis
Model (1) (2) (3) Intercept 0,353*** (0.535) 0,160*** (0,535) 0,351*** (0,537) Independent variable: CSR Performance -0,166 (0,626) 0,021 (0,622) -0,141 (0,511) Moderators:
Extractive Industry Classification Organizational Size 0,208 (0,103) -0,078 (0,108) R² 0,001 0,046 0,007 Adjusted R² -0,011 0,024 -0,017 F-waarde 0,070 2,068 0,293 Hoogste VIF-waarde 1,000 1,023 1,003
5.
Results and Discussion
This chapter will discuss the results from chapter four, mention the limitations of this study, and give recommendations for future research.
5.1. Conclusion
This study has examined whether CSR performance has a positive impact on public CbC Reporting. CSR performance is divided into three characteristics, namely environmental, social, and governance. These characteristics were merged into an integral score; ESG score. In addition, it has been investigated whether organizational size or extractive industry classification has an enhanced moderating effect on the relationships just described.
Following the debate between CSR and tax practices (Hoi et al., 2013; Huseynov and Klamm, 2012; Lanis and Richardson, 2015) this study was set out to give more insight into how CSR performance influences public CbC reporting of organizations. Three hypotheses were tested to understand the relationship between CSR performance and public CbC Reporting of an organization. Based on the regression analysis, it can be concluded that there is no statistical reason to accept hypothesis 1 and 3 of this research. Firstly, this means that there is no significant connection between CSR performance and public CbC Reporting. Secondly, it can be concluded that organizational size has no significant
effect on the relationship between CSR performance and public CbC Reporting. The second hypothesis of this study can be accepted, which means that there is statistical evidence to show that extractive industry classification has a significant effect on the relationship between CSR performance and public CbC Reporting. Further discussion is required based upon these results. The rejected hypotheses will be discussed below.
The first hypothesis focused primarily on the relationship between CSR performance and public CbC Reporting. Based on previous literature on CSR activities and tax practices, it was expected that there would be a positive relation. However, the regression analysis showed a negative, non-significant, relationship. This means that organizations with a high CSR performance score during the period 2011-2015 did not implement, or scarcely implement CbC Reporting. The stakeholder theory might explain this result. According to An et al. (2011) an organization should try to meet multiple goals of a wide range of stakeholders. An organization with a high CSR performance score might believe that it satisfies the needs of stakeholder without implementing CbC Reporting, since the organization already discloses details about CSR. Another explanation can be found when reading statements in the annual reports or websites of the organizations. Not a lot of MNEs have implemented public CbC Reporting, but some organizations gave explanations. The CFO of Reckitt Benckiser recognized the interests of society in public CbC Reporting, but could not see how it would benefit the organization itself. Pearson stated in its annual report 2015: “Work has been performed to prepare systems for the increased tax disclosure requirements relating to CbC Reporting”. The head of tax of Nokia said about CbC Reporting: “However, the push towards public disclosure of information will make the whole tax audit environment more complicated. We think the incomprehension will cause reputational issues. There will undoubtedly be more press coverage of the affairs of big groups, which is not based on expert interpretation”. The reasons that organizations are giving for not implementing public CbC Reporting include not seeing how it would benefit the organization itself and potential reputational damage. Some organizations are still getting ready to implement it in the coming years. Appendix B shows an overview of organizations that implemented CbC Reporting and from what year.
The third hypothesis focused on the potential influence of organizational size on the relationship between CSR performance and CbC Reporting. Based on prior literature (Gao et al., 2005; Hackston and Milne 1996; Brammer and Pavelin, 2006), it was expected that there would be a positive relation. The regression analysis showed a negative not significant relation. The organizational size moderator used in this study are the hundred largest European organizations of the London Stock Exchange. Since the organizations embedded in the FTSEurofirst 100 do not differ much in size, it might be possible that this is an explanation on why organizational size does not have a significant effect. This
could mean that the FTSEurofirst 100 does not provide a good measure. Therefore, manual operations have been performed to see if there are any connections between organizational size and public CbC Reporting. Twelve organizations have been recognized as a ‘large’ organization and it stands out that ten of those organizations implemented CbC Reporting between 2011-2015. These organizations are mentioned in table 6. So, it might be possible that large organizations have a positive influence on the relationship between CSR performance and CbC Reporting, but the scale of the FTSEurofirst 100 index could be a flawed measurement.
Table 6: ‘Large’ organizations with CbC Reporting
Anglo American Repsol
BHP Billiton Rio Tinto
BP Royal Dutch Shell
Eni Total
Glencore Volkswagen Pfd
5.2. Limitations and recommendations
In this study, some limitations can be recognized. The first limitation concerns the data collection of public CbC Reporting. The interpretation of public CbC Reporting can be seen as a subjective process, since every researcher can differ in his or her opinion about public CbC Reporting. One person might see public CbC Reporting as a large overview of all the countries to which an organization must pay taxes. Another person might recognize public CbC Reporting when an organization discloses tax payments to two or more countries. The latter view is used in this study, because the number of organizations with CbC Reporting is hereby increased. If these study would not have followed this view, the number of organizations with CbC Reporting would have been too low. The rise of public CbC Reporting, and the aforementioned intention of organizations to implement it, could make it interesting to have more research on the possible intentions of implementation of public CbC Reporting. This study only contains voluntary public disclosure on CbC Reporting. If, in a few years, it will be mandatory for every MNE to publish public CbC Reporting it would be interesting to see whether this has effects on, for example, stakeholder trust or financial performance.
Another limitation is the amount of moderators used in this study. Future research might use more variables or moderators to further search for possible connections regarding public CbC Reporting. One specific variable could be media attention. Brown and Deegan (1998) have found that there are arguments stating that it is possible for the media to influence how interested society is in issues like the environment. A theory aligned to this is the media agenda setting theory. The media agenda setting theory states that “the media shapes public awareness, with the media agenda preceding public concern for particular issues” (Islam and Deegan, 2010). An important connection in the theory is that increased media attention to a topic gives an increased degree of importance to society. Media attention refers to the awareness of an organization by the media “usually gauged by the sheer volume of stories or space dedicated to topics in newspapers, television news and so on” (Kiousis, 2004). Zyglidopoulos et al., (2012) state that the media has a positive influence on CSR by organizations. They found that organizations rely on the reports of the media because the media are “the main legitimate source of information asymmetry reduction for many stakeholders”.
Dawkins and Fraas (2011) mentioned that media visibility is often overlooked with regard to voluntary disclosure. Two studies confirmed that media attention, in particular negative media attention, through its perceived impact on legitimacy, does evoke a disclosure response from organizations (Deegan et al., 2002; O’Donovan, 1999). Gamerschlag et al. (2011) found that CSR disclosure is positively associated with more media attention. Since public CbC Reporting can be seen as an disclosure response, media attention could have an effect on public CbC Reporting. It could be interesting to research these possible effects.
References
Aaronson, S. A. (2011). Limited partnership: Business, government, civil society, and the public in
the Extractive Industries Transparency Initiative (EITI). Public Administration and
Development, 31(1), 50-63.
An, Y., Davey, H., & Eggleton, I.R.C. (2011). Towards a comprehensive theoretical framework for voluntary IC disclosure. Journal of Intellectual Capital, 12(4), 571-585.
Avi-Yonah, R.S. (2006). Corporate Social Responsibility and Strategic Tax Behavior, Working paper.
Bhattacharya, C.B., & Sen, S. (2004). Doing Better at Doing Good: When, Why, and How Consumers Respond to Corporate Social Initiatives. California Management Review, 47(1), 9-24.
Brammer, S., & Pavelin, S. (2006). Voluntary environmental disclosures by large UK companies. Journal of Business Finance & Accounting, 33(7‐8), 1168-1188.
Branco, M.C., & Rodrigues, L.L. (2007). Positioning Stakeholder Theory within Debate on Corporate social Responsibility. Electronic Journal of Business Ethics and Organization Studies, 12(1), 5-15.
Brooks, C. (2014). Introductory econometrics for finance. Cambridge university press.
Brown, N., & Deegan, C. (1998). The public disclosure of environmental performance information—a dual test of media agenda setting theory and legitimacy theory. Accounting and business
research, 29(1), 21-41.
Campbell, J.L. (2007). Why Would Corporations Behave in Socially Responsible Ways? An Institution Theory of Corporate Social Responsibility. The Academy of Management Review, 32(3), 946-967.
Cao, Y., Cassell, C. A., Myers, L. A., & Omer, T. C. (2015). Does Company Reputation Matter for Voluntary Disclosure Quality? Evidence from Management Earnings Forecasts.
Carroll, A.B. (1979). A Three-Dimensional Conceptual Model of Corporate Performance. The
Academy of Management Review, 4(4), 497-505.
Carroll, A.B. (1991). The Pyramid of Corporate Social Responsibility: Toward the Moral of Organizational Stakeholders. Business Horizons, 39-48.
Carroll, A.B. (1999). Corporate social responsibility: Evolution of a definitional construct. Business &
Society, 38(3): 268-295.
Carroll, A.B. & Shabana, K.M. (2010). The Business Case for Corporate Social Responsibility: A Review of Concepts, Research and Practice. International Journal of Management Reviews, 12(1), 85-105.
Clarkson, M.B.E. (1995). A Stakeholder Framework for Analysing and Evaluating Corporate Social Performance. Academy of Management Review, 20(1), 92-117.
Cormier, D., & Magnan, M. (2003). Environmental reporting management: A continental European perspective. Journal of Accounting and Public Policy, 22(2), 43-62.
CSR Europe. CSR Europe to launch new project on tax transparency and responsible tax behaviour,
http://www.csreurope.org/csr-europe-launch-new-project-tax-transparency-and-responsible-tax-behaviour#.WK7k6O8zXq4
Dahlsrud, A. (2008). How Corporate Social Responsibility is Defined: an Analysis of 37 Definitions.
Corporate Social Responsibility and Environmental Management, 15(1), 1-13.
Dawkins, C., & Fraas, J. W. (2011). Coming clean: The impact of environmental performance and visibility on corporate climate change disclosure. Journal of business ethics, 100(2), 303-322.
Deegan, C. (2002). The legitimising effect of social and environmental disclosures – a theoretical foundation. Accounting, Auditing & Accountability Journal, 15(3), 282-311.
