Does CSR performance influence CSR transparency? An
empirical analysis of Dutch and French firms.
Abstract: This research investigates the influence of CSR performance on CSR transparency of firms including firm size, industry sensitivity, year and country as moderating variables. The empirical analysis was conducted using 113 publicly listed
firms domiciled in the Netherlands and France. Primarily, it finds support for a positive relationship between these variables at all aspects of CSR. Furthermore, it
finds evidence that for the total CSR aspects industry sensitivity weakens this relationship. For the environmental aspects industry sensitivity and firm size
strengthen this relationship. For the social aspects firm size strengthens this relationship. Ultimately, for corporate governance aspects, firm size weakens this
relationship and country strengthens it.
Key words: CSR performance, CSR transparency, ASSET 4 ESG
s1853112
Hadewich de Roo van Alderwerelt MSc International Financial Management
Faculty of Economics & Business University of Groningen
1
Table of Contents
1 Introduction 2
2 Theoretical Background 3
2.1 Corporate Social Responsibility 3
2.2 CSR performance 6
2.3 CSR transparency 7
2.4 CSR performance and CSR transparency 8
2.5 Factors influencing the relationship between CSR performance and
transparency 8 2.5.1 Firm size 9 2.5.2 Time effect 9 2.5.3 Country effect 9 2.5.4 Industry effect 11 2.6 Conceptual model 11 3 Methodology 12
3.1 Research method & outline 12
3.2 Data sample 13 3.3 Variables 13 3.3.1 Independent variable 13 3.3.2 Dependent variable 14 3.3.3 Moderating variables 15 3.4 Regression models 15 3.4.1 Endogeneity 16
3.4.2 Robustness of the model 16
3.5 Descriptive statistics and correlation 18
3.5.1 Descriptive statistics 19
3.5.2 Multicollinearity 21
4 Results 22
4.1 Regression analysis 22
4.1.1 Hypothesis 1: the relationship between CSR performance and
transparency 24
4.1.2 Hypothesis 2: the influence of firm size 25
4.1.3 Hypothesis 3: the time effect 25
4.1.4 Hypothesis 4: the country effect 25
4.1.5 Hypothesis 5: the industry effect 26
5 Discussion & Conclusion 27
6 References 30
2
1 Introduction
In 1971, a landmark contribution to the concept of CSR occurred when the Committee for Economic Development of the Conference Board which is based in Washington D.C. published about CSR (Carrol, 1999). Since then, the evolution of CSR has risen exponentially (Cochran, 2007). It became a topic not only discussed by academics but also by investors, business, NGO’s and national and international institutions. Nowadays, governments hand out awards for firms with the best sustainability report and they organize sustainability days. Furthermore, investors’ reliance on environmental, social and other financial indicators has grown. Based on a survey conducted by EY and International Integrated
Reporting Council that was answered by more than 200 institutional investors, the percentage of investors that consider mandatory board oversight of nonfinancial performance reporting as essential has grown from 36% in 2014 to 80% in 2015. (EY& IIRC, 2015). Previous research (Fifka and Drabble, 2012) concludes that the quality of CSR Reporting differs between countries. Moreover, the importance of literature on sustainability reports for detecting a firm’s CSR commitment and its stakeholder orientation is increasing (Perrini, 2010; Dawkins and Ngunjiri, 2008). Using stakeholder theory, several authors (Calabrese, Costa, Menichini and Rosati, 2013; Becker-Olsen, Cudmore and Hill, 2006) stress the importance of the stakeholder perception and show ‘that business returns from CSR activities depend greatly on how stakeholders perceive the company CSR commitment’ (Costa and Menichini, 2013:158).
3 influence CSR transparency as CSR reporting is, according to Bouten (2011), often considered to be an appropriate marketing tool. Hence, firms would only be transparent on issues when they perform well. The relationship between CSR performance and transparency is useful for practitioners to see whether or not firms are only transparent about the CSR aspects on which they perform well. This is specifically relevant for investors trying to make an appropriate decision when they would like to invest socially responsibly. When CSR performance significantly positively influences CSR transparency, investors could assume that when firms are transparent about their CSR performance it means that they perform well and are attractive to invest in. Another reason why it is relevant for practitioners is that when firms perform well but are not transparent, obliging firms to be transparent might be less effective when governments are looking at the question of how to motivate firms to act sustainably.
The remainder of this paper is structured as follows. Section 2 provides a theoretical background. Section 3 discusses the methodology. Section 4 presents the results and section 5 motivates the discussion and conclusion.
2 Theoretical Background
In the theoretical background the following topics will be discussed: CSR, CSR performance, transparency, the relationship between previous variables, factors that might moderate this relationship and lastly, the conceptual model.
2.1 Corporate Social Responsibility
4 environmental integrity’. Based on Carrol, (1979), the European Commission (2002) and Visser (2014), the definition that will be used during this research will focus on the firm’s capacity to generate long-term shareholders value by managing aspects in terms of corporate governance, social and environmental aspects.
5 that they are more likely to be transparent about this benefit. Consequently, CSR has been and still is often emphasized by academics, governments and business leaders (Bénabou and Tirole, 2010).
Additionally, there are several visions on CSR. According to some CSR enthusiasts, good corporate behaviour is perceived as being a financially profitable approach, while others believe that acting in name of the social good will sacrifice some profit. Some see CSR as an investment into innovation and others see it as pure form of corporate expenditure (Bénabou and Tirole, 2010). With these
interpretations in mind they describe three visions of CSR: ‘Doing well by doing good, delegated philanthropy and insider-initiated corporate philanthropy. (I) Doing well by doing good is according to Cochran (2002) the most appropriate vision of CSR. It is a strategic profit-maximizing use of CSR. Furthermore, Porter and Kramer (2002) suggest that firms should exploit the synergies that can be created when the social and the economic are combined. When a firm does not have a competitive advantage in a given philanthropic area, investments in those fields will have no impact on the long-term. Falck and Heblich (2007) conclude that when a firm has a long-term profit maximization strategy, the firm can do well by doing good as they are expected to treat society well and that society will return the favour. (II) The delegated philanthropic vision implies that stakeholders are ready to forfeit money for social objectives. It sees the firm as a channel that states the principles of a citizen. Moreover, Walton (1967) identified a concept of social responsibility whereby top managers have to keep the relation between the firm and society in mind. A crucial aspect includes the degree of voluntarism, which means that it might not be possible to see any direct measurable economic returns. In contrast with the vision mentioned above, this type of CSR does not raise particular corporate governance issues. The last vision, (III) insider-initiated corporate philanthropy, simply means doing charity with other owners’ capital. This has been criticised by the right and left side of the political spectrum (Bénabou and Tirole, 2010).
6 2.2 CSR performance
To explain CSR performance, it must be illuminated, which aspects should be included when measuring CSR and what the limitations are when including particular aspects in CSR performance. To be more specific, it should be noted that in previous literature some of those limitations have been ignored. For example, using Carrol’s definition of CSR. Due to the multifaceted nature of these aspects, Clarkson (1995) concludes, while reviewing a 10-year project to quantify and review corporate performance based on the four types mentioned in the definition by Carrol (1979), that they do not advance themselves to empirical testing. They describe that two of these types are challenging for empirical testing. First, the legal dimension is problematic, and as a result most existing CSR definitions stress the voluntary character of CSR, such as the ones endorsed by the EU Commission, the World Business Council for Sustainable Development, the International Chamber of Commerce and the UN Global Compact. Consequently, one voluntary action in one country might be regulatory compliance in another country. This is also the argument made by Matten and Moon (2008). Secondly, neither is the ethical element
supportive, as long as what is ethical is not identified (Clarkson, 1995). It could be reasoned that ‘ethical responsibilities’ relate to the voluntary element referred to above, or more specifically, movements beyond those authorised by law (Gjølberg, 2009). Therefore, as mentioned before, this view of CSR performance is not
appropriate for this project as the legal framework between the two countries used in the sample are not the same.
However, according to Visser (2014) there are several aspects that should be considered when looking at CSR, one of them is its DNA. Visser (2014) claims that the DNA of CSR nowadays consists out of the following elements: (I) good
7 features of CSR. This includes fair labour practise, stakeholder engagement, supply chain integrity and community participation. According to Tang and Tang (2012) these are the most serious fundamentals of CSR. As a last remark, the aim of
corporate governance is as important as the other aspects as when the institution fails, it undermines all the other items that CSR tries to achieve. It includes the control of a firm’s rights and responsibilities through encouragement. Hence, these are the aspects that will be focused on when looking at CSR performance.
2.3 CSR transparency
There are, according to Deegan (2002), two types of stakeholder theories. There is an ethical type, which offers recommendations in terms of how organizations should treat their stakeholders, which accentuates the duties of organisations (Hasnas, 1998; Donaldson, Preston, 1995; Freeman and Reed, 1983). As it is a theory that offers explanation of duties, it does not play an active role in forecasting managerial
performance. The theory that will be applied in this research is the managerial type of stakeholder theory (Deegan, 2002). It focusses on the necessity to cope with particular stakeholders, specifically, the powerful stakeholders, which are able to regulate assets that are essential to the processes of the firm. More energy will be utilised in
supervising the relationship when a stakeholder is more important. ‘Information is a major element that can be employed by the organisation to manage the stakeholder to gain support and approval or to distract their opposition and disapproval.’ (Gray et al., 1996: 45). In this research the information on CSR aspects available to all
stakeholders will in this research be defined as CSR transparency. The more relevant information they publish the more transparent they are.
Consequently, an important aspect of CSR transparency is to create awareness by the stakeholders of the CSR performance of the firm. CSR reporting is a tool for firms to deliver evidence for different stakeholders concerning social and environmental matters. (Golob and Bartlett, 2007). Therefore, it could be assumed that this applies to corporate governance matters as well. Only the information that is available to all stakeholders is taken into account when assessing firm’s transparency in CSR. CSR reporting is an aspect of stakeholder management as it focusses on transparency of CSR to the stakeholders of the firm. According to Deegan (2002) motivations of management reporting includes the desire to comply with legal requirements,
8 investment funds. In line with CSR performance, environmental, social and corporate governance aspects are included in CSR transparency.
Furthermore, when looking at the world’s premium standard setter when it comes to sustainability reporting, the Global Reporting Initiative guidelines, the amount of material aspects that a firm should report is increasing. Additionally, these guidelines suggest that firms should not only report on their positive results but also on negative results when reporting on CSR (GRI, 2015). Nevertheless, this is still only a small part of the guidelines, and this could mean that firms are providing more balanced information when reporting on CSR.
2.4 CSR performance and CSR transparency
According to Gamerschlag, Möller and Verbeeten (2011) CSR disclosure relates to a firms’ social and environmental performance as it provides information on those issues and firms spend a large effort on CSR disclosures, which is consistent with political cost theory. Furthermore, the findings of Bouten (2011) reveal that firms still favour to report only on the information that ‘looks good in the market’, even when firms voluntarily report on CSR. This is because most firms consider the annual report an important marketing tool. Consequently, this would mean that when a firm scores high on its performance, it would also be likely to score high on CSR transparency. Accordingly, the CSR performance and the CSR transparency of a firm are expected to closely relate to each other. Furthermore, CSR performance and CSR transparency are not only assumed to correlate with each other. Therefore it is
expected that CSR performance will influence CSR transparency as firms decide which aspects they are reporting about and are expected to report when they are performing well. Thus, the following hypothesis is presented:
Hypothesis 1: CSR performance will positively influence CSR transparency
2.5 Factors influencing the relationship between CSR performance and transparency
9 2.5.1 Firm size
The firm size-disclosure relationship is supported by legitimacy theory and agency theory (Hackston and Milne, 2011). Most, but not all, empirical studies find a relationship between firm size and CSR transparency (Belkaoui and Karpik, 1989; Cowen et al., 1987; Kelly, 1981; Pang, 1982; Patten, 1991, 1992; Trotman and Bradley, 1981; Gao et al., 2005). For example, Gao et al. (2005) found evidence that firm size is positively related to the level of CSR reporting. Additionally, larger firms have greater visibility and a larger operational impact and are expected to have to spend more on the transparency of their CSR performance which would cause them to be more transparent than smaller firms (Barnea and Rubin, 2010). Therefore, it is expected that firm size will reinforce the positive relationship between CSR
performance and transparency. Consequently, the following hypothesis was formed:
Hypothesis 2: Firm size reinforces the positive relationship between CSR
performance and CRS transparency. 2.5.2 Time effect
Not only the importance and ‘appropriate behaviour’ of CSR have evolved in the last couple of years, the amount of sustainability reports has increased as well over the years. The last couple of years, several countries have enforced laws that oblige firm to be transparent about their CSR performance. Furthermore, the CSR
disclosures may also change over time when different stakeholder groups lose or gain influence (Gamerschlag, Möller and Verbeeten, 2010). Also the most recent GRI guidelines suggest that firms should not only report about their positive results but also on negative results when reporting on CSR (GRI, 2015). Though, this is still only a small part of the guidelines, this could mean that the balance of negative and
positive results in CSR reporting is increasing. Therefore it is expected that there will be a time fixed effect in the investigated relationship.
Hypothesis 3: The relationship between CSR performance and CSR
transparency will contain a time fixed effect.
2.5.3 Country effect
Fifka and Drabble (2012) identify the existence of certain regulations in the country to investigate whether the quantity of CSR disclosures differs in countries. When
10 and social information (Renneboog, 2008). Since May 2012, France enforces the Grenelle II act in which it is stated that sustainability reporting is mandatory for large listed firms as of 2012 and small companies up to 500 employees and have a total assets or net annual sales of €100 million euro by 2014. Furthermore, they are also obliged to have a verification of the results of a 3rd party. (Barbu, Dumontier, Feleaga and Feleaga, 2014). Table 2 shows the topics, which should be included in the CSR report.
Table 1: List of topics firms should Report on according to the Grenelle II Act
Environment Social
General environmental policy Pollution and Waste Management Sustainable use of resources Climate change
Protection of biodiversity
Territorial impact, economic and social of the company's activity
Relationships with persons or
organizations interested in the company's activity
Subcontracting and suppliers Fair practices
The actions undertaken to prevent corruption
Human rights Employment Work organization Social relations Health and security Training
Diversity and equal opportunity / equal treatment
Promotion and respect for fundamental provisions of ILO Conventions relating
11 have a strong business case for CSR as they have a large need to be perceived as responsible. Considering regulation, listed firms domiciled in France are obliged to be more transparent about their CSR performance than Dutch firms.
Hence, as the culture of France firms will be more transparent, they might report on their performance in spite of how they perform. This indicates that the relationship between CSR performance and CSR transparency will be stronger for Dutch firms. Therefore, the following hypothesis arise.
Hypothesis 4: The relationship between CSR performance and transparency
will be stronger for Dutch firms.
2.5.4 Industry effect
According to several studies (Gao et al., 2005; Kelly, 1981; Cowen et al., 1987), the sector that the firm is operating in relates to the amount, the content and the location of the Corporate Social and Environmental disclosure. Furthermore, Adams and Kuasirikun (2000) found that transparency of CSR performance can be indicated by industry government relations, culture, responses of industry associations and political pressure groups. However, as by law, firms are not obliged to be transparent about all indicators that are used to measure their performance, it is expected that the amount they disclose will still heavily depend on how they perform on these
indicators. Campbell (2003) and Gamerschlag et al. (2011) found that firms that are more environmentally sensitive will disclose more. Therefore, it is expected that for firms operating in an environmental sensitive industry sensitivity will have stronger influence on the relationship. Thus, this moderation of that influence on CSR transparency will also be tested. These arguments give rise to the following hypothesis:
Hypothesis 5: The relationship between CSR performance and transparency will
be stronger for firms operating in an environmental sensitive industry sensitivity.
2.6 Conceptual model
The concepts in the previous paragraphs give rise to the following conceptual model with CSR transparency as dependent variable and CSR performance as
independent variables. The factors that previous literature has identified as factors that could influence this relationship are the year of observations, the firm size, the
12
Figure 1: Conceptual model
3 Methodology
This section will discuss the research method & outline of the data sample, the variables, the regression models and the descriptive statistics of this research.
3.1 Research method & outline
The aim of this research is to look at the relationship between CSR transparency and CSR performance and to investigate which variables might moderate this
influence. As mentioned before quantitative research is appropriate to answer the research questions. Due to the fact that most independent and dependent variables are ordinal data, a linear multiple regression analysis is a suitable method. Therefore, to test the formulated hypotheses, multivariate regression equations are run with ordinary least squares (OLS) using panel data in EViews. The variables used are discussed below. To achieve the research goal, the following research question is formulated: Does CSR performance influence CSR transparency? In analysing this
13 research question, the factors that might moderate this relationship will be considered. Those factors will be firm size, the industry effect, the country effect and the time effect. For firm size also pooled data will be used for the regression as a singularity matrix occurs when the variable time is included in panel data.
3.2 Data sample
The sample consists of 113 firms. Results from 2006 to 2014 were used as those years were the only ones including more than 30 observations. Consequently the used sample varies per model between 792 and 849 observations. The regression analysis uses firm-level data. The CSR performance, CSR transparency, the industry and the country data are acquired by the data from the ASSET4 ESG (Environmental, social and governance) database, which is part of Datastream. This database attempts to provide access to in-depth, objective and comparable data. It is based on only publicly available sources of nearly 5,000 publicly listed companies. The firm size variables are acquired from ORBIS of Bureau van Dijk. This database contains data concerning the most important companies, worldwide.
3.3 Variables
This section will discuss the methodology of the independent, the dependent and moderating variables.
3.3.1 Independent variable
As independent variable CSR performance is used. In previous literature there does not appear to be one agreed upon method in how to measure CSR performance. Lioui and Sharma (2012) and Chatterji et al. (2009) look at CSR strengths and CSR concerns from the KLD database when identifying CSR performance. They find evidence that the current measurements are appropriate when activists would like to target firms that are historically performing low on environmental issues. However, for investors interested in eco-efficiency of firms, these ratings do not distinct which firms are performing low environmentally adjusted by firm size. Their most important critique is that more weight should be put on historical performances of firms. Thus, it remains for investors a challenge to ‘identify valid measures environmental
14 concerns and strength but looks at the environmental, social and corporate governance performance of a firm. This appears to be the most recent data of CSR indicators. Even though not many published articles have used this score so far, it will be used in this research as it is appropriate for how CSR is defined in this research.
The environmental pillar measures ‘a company's impact on living and non-living natural systems, including the air, land and water, as well as complete
ecosystems. It reflects how well a company uses best management practices to avoid environmental risks and capitalize on environmental opportunities to generate long term shareholder value’ (Thomson Reuters, 2015. Web). For environmental
performance it looks at emission reduction, resource reduction and product innovation. The social pillar measures ‘a company's capacity to generate trust and loyalty with its workforce, customers and society, through its use of best management practices. It is a reflection of the company's reputation and the health of its license to operate, which are key factors in determining its ability to generate long term
shareholder value’ (Thomson Reuters, 2015. Web). For a firm’s social performance, it considers employment quality, health &safety, training & development, diversity, human rights, community and products responsibility for social performance. The corporate governance pillar measures ‘a company's systems and processes, which ensure that its board members and executives act in the best interests of its long term shareholders. It reflects a company's capacity, through its use of best management practices, to direct and control its rights and responsibilities through the creation of incentives, as well as checks and balances to generate long term shareholder value’ (Thomson Reuters, 2015. Web). Board structure, board function, compensation policy, shareholder rights and vision & strategy are considered for corporate governance. An extensive overview of the indicators are provided in Appendix A. Thus, the equal weighted-average of these scores will be used to measure CSR performance, which is measured as the logarithm of a score out of 100.
3.3.2 Dependent variable
15 reports. This is similar to in the performance variables, divided into social,
environmental and corporate governance aspects. The score is also reported as the logarithm of a score out of 100.
3.3.3 Moderating variables
Firms size will be measured as the natural logarithm of turnover of firm i in thousand euros in year t.For industry sensitivity, a dummy variable is used where a number is appointed to the industry they are operating in. For industry sensitiveness also a dummy variable is used, 1 when a firm is operating in an environmental sensible industry (primary industry and industrial production) and 0 for other
industries. (Fifka and Drabble, 2012). All the firms in the sample have the same type of ownership as they are all publicly listed firms. This means that there will be less heterogeneity of the type of firms in terms’ of expectation of stakeholders and
regulations. For country a dummy variable will be used 0 when a firm is domiciled in the Netherlands and 1 when a firm is domiciled in France. These are according to previous literature (Gao et al., 2005; Hackston & Mille, 1996; Gamerschlag et al., 2011) the most important variables that could influence this relationship.
3.4 Regression models
The models that will be used are:
𝐶𝑆𝑅_𝑇!,! = 𝐵! + 𝐵!CSR_P!,!!!+ 𝐸!,!
As suggested by Brooks (2014) logarithmic values of CSR performance and transparency are used to anticipate on the potential issues of heteroskedasticity. Therefore, the model will be transformed into:
(1) 𝑙𝑜𝑔𝐶𝑆𝑅_𝑇!,! = 𝐵! + 𝐵!logCSR_P!,!!!+ 𝐸!,!
16 CSR_P!,!!! = the average score of the social, environmental and corporate governance performance for a particular firm in the previous year measured as a percentage.
CSR_Ti,t= B0 + B1CSRPi,t-‐1+B2 FSi,t-‐1+B3COUi +B4 INDi+B5(CSRPi,t-‐1 X FSi,t-‐1)
+B6(CSRPi,t-‐1X COUi)+B7(CSRPi,t-‐1X INDi) +Ei,t
𝐹𝑆!,!!! = firm size measured as the turnover in thousand euros for a firm in
the previous year
𝐶𝑂𝑈! = Country of domicile of a particular firm.
𝐼𝑁𝐷! = Industry either measured as industry sensitiveness.
In align with the previous model; this model has been transformed into:
(2) log (𝐶𝑆𝑅_𝑇!,!) = 𝐵! + 𝐵!log (CSR!!,!!!) + 𝐵!𝑙𝑜𝑔 (𝐹𝑆!,!!!) + 𝐵!𝐶𝑂𝑈! + 𝐵! 𝐼𝑁𝐷!+ 𝐵!(log (𝐶𝑆𝑅!!,!!! )𝑋 log (𝐹𝑆!,!!!)) +
𝐵!(log (𝐶𝑆𝑅!!,!!!)𝑋 𝐶𝑂𝑈!) + 𝐵!(log (𝐶𝑆𝑅!!,!!!)𝑋 𝐼𝑁𝐷!) + 𝐸!,!
3.4.1 Endogeneity
When looking at the causal relation of CSR performance on CSR transparency, the concern that arises is the potential endogeneity and omitted variables bias, which could cloud the interpretation of this relationship. As this relationship is not
investigated before this causality is based on the following argument.
Due to regulations, firms could be obliged to be transparent about their CSR performance and therefore, could makes sure that they also perform well. To avoid potential endogeneity problems and reverse causality effects (Oikonomou et al. 2012), the independent variables that are time specific are lagged by one year.
3.4.2 Robustness of the model
17 transparency. Gamerschlag, Möller & Verbeeten (2011)’s results propose that there are several dissimilarities in the disclosure dimensions environmental and social. The firm characteristics that drive environmental disclosures seem to be different than those that drive social disclosures. Since Entine (2003) and Godfrey et al. (2009) indicate that, because agglomeration abolishes the information value contained in the more comprehensive measures, data can become insincere when individual item scores are combined. Therefore, both overall and individual scores are included in the models to be able to distinguish between the impacts of individual CSR performance and transparency aspects, to avoid abolishing valuable information and to be able to see whether the overall aggregated CSR score reduces the information provided by the individual scores. The following regression models will be used for the
environmental, social and corporate governance measures. 3.4.2.1 Environmental
Similar to how models 1 and 2 are developed, the models for the environmental aspects also include logarithmic values:
(3) log (𝐸_𝑇!,!) = 𝐵! + 𝐵!log (𝐸_𝑃!,!!!) + 𝐸!,!
(4) log (𝐸_𝑇!,!) = 𝐵! + 𝐵!log (E!!,!!!) + 𝐵!𝑙𝑜𝑔 (𝐹𝑆!,!!!) + 𝐵!𝐶𝑂𝑈! + 𝐵! 𝐼𝑁𝐷!+ 𝐵!(log (𝐸!!,!!!)𝑋 log (𝐹𝑆!,!!!)) + 𝐵!(log (𝐸!!,!!!)𝑋 𝐶𝑂𝑈!) + 𝐵!(log (𝐸!!,!!!)𝑋 𝐼𝑁𝐷!) + 𝐸!,!
𝐸_𝑇!,! = the environmental transparency for a particular firm in a specific year measured as a percentage.
𝐸_𝑃!,!!! = the score of the environmental performance for a particular firm in
the previous year measured as a percentage.
3.4.2.2 Social
Similar to how models 1 and 2 are developed, the models for the environmental aspects also include logarithmic values:
18 (6) log (𝑆_𝑇!,!) = 𝐵! + 𝐵!log (S!!,!!!) + 𝐵!𝑙𝑜𝑔 (𝐹𝑆!,!!!) + 𝐵!𝐶𝑂𝑈! +
𝐵! 𝐼𝑁𝐷!+ 𝐵!(log (𝑆!!,!!!)𝑋 log (𝐹𝑆!,!!!)) + 𝐵!(log (𝑆!!,!!!)𝑋 𝐶𝑂𝑈!) + 𝐵!(log (𝑆!!,!!!) 𝑋 𝐼𝑁𝐷!) + 𝐸!,!
𝑆_𝑇!,! = the social transparency for a particular firm in a specific year measured as a percentage.
𝑆_𝑃!,!!! = the score of the social performance for a particular firm in the
previous year measured as a percentage.
3.4.2.3 Corporate Governance
Similar to how models 1 and 2 are developed, the models for the environmental aspects also include logarithmic values:
(7) log (𝐶𝐺_𝑇!,!) = 𝐵! + 𝐵!log (𝐶𝐺!!,!!!) + 𝐸!,!
(8) log (𝐶𝐺_𝑇!,!) = 𝐵! + 𝐵!log (CG_P!,!!!) + 𝐵!𝐿𝑜𝑔 (𝐹𝑆!,!!!) + 𝐵!𝐶𝑂𝑈! + 𝐵! 𝐼𝑁𝐷!+ 𝐵!(log (𝐶𝐺!!,!!!)𝑋 log (𝐹𝑆!,!!!)) +
𝐵!(log (𝐶𝐺!!,!!!) 𝑋 𝐶𝑂𝑈!) + 𝐵!(log (𝐶𝐺!!,!!!) 𝑋 𝐼𝑁𝐷!) + 𝐸!,! 𝐶𝐺_𝑇!,! = the Corporate Governance transparency for a particular firm in a specific year measured as a percentage.
CG_P!,!!! = the score of the Corporate Governance performance for a particular
firm in the previous year measured as a percentage.
Furthermore, for the first hypothesis another robustness check will be done. The first hypothesis concerns the relationship between CSR performance in models 1, 3, 5 and 7. These models will be compared with entity fixed effects and no entity fixed effects. Entity fixed effects are not included in the original models 1 tot 8 as this is not possible when including the moderating variables as these include variables that do not change over time.
3.5 Descriptive statistics and correlation
19 3.5.1 Descriptive statistics
Table 3 shows the descriptive statistics of the variables firm size, industry sensitiveness, country, total performance of the year before, environmental
performance of the year before, social performance of the year before and corporate governance performance of the year before, total transparency environmental transparency, social transparency and corporate governance transparency.
Table 2: Descriptive statistics
Observations Mean Median Maximum Minimum Std. Dev.
CSR_P 829 76,409 % 77,254 % 90,722 % 55,877 % 6,382 % E_P 829 62,670 % 63,830 % 87,234 % 40,426 % 11,761 % S_P 829 75,308 % 75,000 % 90,000 % 55,000 % 6,972 % CG_P 829 91,251 % 94,118 % 97,059 % 61,765 % 6,049 % CSR_T 829 71,271 % 76,637 % 96,173 % 7,870 % 19,186 % E_T 829 77,707 % 88,170 % 96,760 % 10,300 % 21,689 % S_T 829 78,525 % 87,950 % 98,530 % 5,780 % 21,523 % CG_T 829 57,580 % 60,640 % 96,640 % 2,160 % 24,659 % COU 829 0,811 1 1 0 0,392 IND 829 0,668 1 1 0 0,471 FS 829 30,615,996 7,501,800 5,854,000,000 0 265,990,264
This table reports descriptive statistics of the main variables. The sample consists out of 113 firms from 2 countries of 8 years in the sample period 2006 -2014. CSR_P is the firm’s total CSR performance score of the year before. E_P is the environmental performance of the firm of the year before. S_P is the social performance score of the year before. CG_P is the corporate governance performance score of the year before. CSR_T is a firm’s total CSR transparency score. E_T is the environmental transparency score. S_T is the social transparency score. CG_T is the corporate governance
transparency score. COU is the country of domicile of the firm. IND is whether the firm is operating in an environmental sensitive firms and FS is the firm size measured as the operating turnover in thousand euros of the previous year.
20 firms is 100, it can be concluded that no firm scores the maximum score on all
aspects. As none of the minimum of these firms is for both 0, it can be concluded that no firm scores the minimum score on all aspects.
Table 4shows the descriptive statistics per country. I will only elaborate on the remarkable descriptive statistics. It shows that the statistics do not vary that much per country. Firms domiciled in France score slightly higher on the environmental and social measurements of performance and transparency than firms domiciled in the Netherlands. For the corporate governance performance and transparency, firms domiciled in the Netherlands score higher. Moreover, the relative amount of firms an environmental sensitive industry are the same for both countries. The average firm size is much higher for firms domiciled in the Netherlands.
Appendix B shows the descriptive statistics per year. I will only elaborate on the remarkable descriptive statistics. It shows that the average CSR performance is from 2006 until 2012 increasing from 65,931 to 76,094 and in 2013 it lightly decreased to 75,922. From 2007 to 2008 the average transparency decreased a bit from 74,449 to 70,129, then it increases until 2012 to 78,452 and afterwards it stabilizes. Moreover, the average firm size increases exponentially from 2011 to 2012 from 18,6 billion euros to 64,9 billion euros.
These models are regressed based on panel data. As mentioned before, there are variables that do not change over time. Therefore, entity effects are not applicable here. However, I investigated whether time fixed effects are applicable when the average value of the variables changes over time but not per firm. This was tested by first applying the Redundant fixed effects to see whether the data can be pooled or whether fixed approach is more effective. For all models, the p-value was less than 1 % indicating that pooled data is not appropriate and a fixed effect model is preferred. Then, the Hausman test was applied to see random effects or fixed effects are
applicable here. This test also showed for all models a p-value smaller than 1%, which indicates that the random effects model is not applicable and the fixed effects
21
Table 3: Descriptive statistics per country
Observations Mean Median Maximum Minimum Std. Dev.
Netherlands CSR_P 87 74,438 % 80,407 % 94,027 % 15,063 % 17,950 % E_P 87 75,983 % 81,430 % 94,810 % 13,940 % 19,666 % S_P 87 77,462 % 86,300 % 96,530 % 5,780 % 22,369 % CG_P 87 69,870 % 69,850 % 94,910 % 21,830 % 17,868 % CSR_T 87 75,445 % 76,231 % 84,552 % 59,112 % 5,928 % E_T 87 61,262 % 61,702 % 78,723 % 40,426 % 10,713 % S_T 87 72,443 % 72,500 % 87,500 % 57,500 % 6,766 % CG_T 87 92,630 % 94,118 % 97,059 % 79,412 % 3,585 % IND 87 0,655 1 1 0 0,478 FS 87 147,019,740 7,041,598 5,854,000,000 155 811,686,299 France CSR_P 672 70,959 % 76,065 % 96,173 % 7,870 % 19,345 % E_P 672 78,632 % 89,135 % 96,760 % 10,480 % 21,534 % S_P 672 78,747 % 88,375 % 98,530 % 6,020 % 21,478 % CG_P 672 55,497 % 57,250 % 96,640 % 2,160 % 25,208 % CSR_T 672 76,847 % 77,626 % 90,722 % 55,877 % 6,305 % E_T 672 63,450 % 63,830 % 87,234 % 40,426 % 11,828 % S_T 672 76,112 % 77,500 % 90,000 % 55,000 % 6,652 % CG_T 672 90,980 % 94,118 % 97,059 % 61,765 % 6,356 % IND 672 0,667 1 1 0 0,472 FS 672 16,279,805 7,512,472 182,000,000 0 23,332,339
This table reports descriptive statistics of the main variables per country. The sample consists out of 24 firms domiciled in the Netherlands and 89 firms domiciled in France of 8 years in the sample period 2006 -2014. CSR_P is the firm’s total CSR performance score of the year before. E_P is the environmental performance of the firm of the year before. S_P is the social performance score of the year before. CG_P is the corporate governance performance score of the year before. CSR_T is a firm’s total CSR transparency score. E_T is the environmental transparency score. S_T is the social transparency score. CG_T is the corporate governance transparency score. COU is the country of domicile of the firm. IND is whether the firm is operating in an environmental sensitive firms and FS is the firm size measured as the natural logarithm of its operating turnover in thousand euros of the year before.
3.5.2 Multicollinearity
To test whether the variables are correlated with each other a Pearson
correlation test was done. Table 5 shows the Pearson correlation test and is used to assess the potential collinearity problems with the variables firm size, industry sensitiveness, country, total performance, environmental performance, social
22 correlation of 0.8 or higher indicates serious multicollinearity between the variables (Grewal, Cote and Baumgartner, 2004). This is not the case for most variables. Only CSR performance with social, environmental and corporate governance performance and CSR transparency with social, environmental and corporate governance
transparency have a high correlation value. However, these variables are used in separate models. Therefore, this is not an issue. For the other variables the matrix indicates that each variable captures a different component.
Table 4: Correlation Matrix
CSR_T S_T E_T CG_T CSR_P S_P E_P CG_P FS COU IND CSR_T 1,000 S_T 0,723 1,000 E_T 0,889 0,468 1,000 CG_T 0,602 0,227 0,332 1,000 CSR_P 0,672 0,447 0,541 0,561 1,000 S_P 0,605 0,482 0,488 0,410 0,879 1,000 E_P 0,643 0,455 0,560 0,425 0,857 0,726 1,000 CG_P 0,476 0,224 0,345 0,577 0,813 0,540 0,486 1,000 FS 0,043 0,105 -0,009 0,035 0,064 0,051 0,043 0,067 1,000 COU 0,142 0,239 0,137 -0,093 -0,034 0,021 0,088 -0,175 -0,112 1,000 IND 0,152 -0,006 0,245 0,012 0,100 0,057 0,196 0,012 0,048 -0,007 1,000 This table reports descriptive statistics of the main variables. The sample consists out of 113 firms from 2 countries of 8 years in the sample period 2006-2014. CSR_P is the firm’s total CSR performance score. E-P is the environmental performance of the firm. S_P is the social performance score. CG_P is the corporate governance performance. CSR_T is a firm’s total CSR transparency score. E_T is the environmental transparency score. S_T is the social transparency score. CG_T is the corporate governance transparency score. COU is the country of domicile of the firm. IND is whether the firm is operating in an environmental sensitive firms and FS is the firm size measured as the operating turnover.
4 Results
4.1 Regression analysis
23 Table 5:Regression results
Dependent variable CSR_T CSR_T E_T E_T S_T S_T CG_T CG_T
Model (1) (2) (3) (4) (5) (6) (7) (8) Intercept 3.654 (0.024)*** 4.046 (0.288)*** 3.035 (0.055)*** 4.669 (0.722)*** 3.870 (0.026)*** 4.682 (0.339)*** 4.171 (0.0111)*** 3.359 (0.126)*** CSR_P 0.158 (0.006)*** 0.017 (0.067) E-P 0.247 (0.013)*** -0.253 (0.163) S_P 0.102 (0.006)*** -0.125 (0.076) * CG_P 0.084 (0.003)*** 0.265 (0.032)*** COU 0.020 (0.067) -0.019 (0.137) 0.116 (0.075) -0.095 (0.045)** IND -0.194 (0.046)*** -0.484 (0.107)*** -0.029 (0.054) -0.001 (0.022) FS -0.1036 (0.050) -0.188 (0.122) -0.135 (0.061)** 0.141 (0.020)*** P* FS 0.015 (0.011) 0.058 (0.027)** 0.036 (0.014)*** -0.031 (0.005)*** P *IND -0.049 (0.011)*** 0.127 (0.025)*** 0.004 (0.012) -0.001 (0.006) P*COU 0.002 (0.016) 0.014 (0.032) -0.014 (0.017) 0.022 (0.011)** Adjusted R- squared 0.565 0.614 0.373 0.454 0.313 0.366 0.570 0.617 F-statistic 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Number of observations 962 935 962 935 960 933 962 935 Number of firms 113 113 113 113 113 113 113 113 This table presents the regression analysis with panel data of CSR performance, CSR transparency, Firm size, Country and
24 4.1.1 Hypothesis 1: the relationship between CSR performance and transparency
As mentioned before Hypothesis 1 was described as follows: ‘CSR performance will positive influence CSR transparency’, where performance and transparency include all CSR aspects, I find a significant positive relationship in model 1.
Table 6: Difference in regression with entity fixed effects Dependent variable CSR_T E_T S_T CG_T Model (1) (9) (3) (10) (5) (11) (7) (12) Intercept 3.654 (0.024)*** 3.953 (0.028)*** (0.055)*** 3.035 3.866 (0.056)*** (0.026)*** 3.870 4.094 (0.031)*** 4.171 (0.0111)*** 4.198 (0.018)*** CSR_P 0.158 (0.006) 0.087 (0.007)*** E-P 0.247 (0.013)*** 0.054 (0.013)*** S_P 0.102 (0.006)*** 0.050 (0.007)*** CG_P (0.003)*** 0.084 (0.005)*** 0.078 Adjusted R- squared 0.565 0.845 0.373 0.824 0.313 0.732 0.570 0.738 Number of observations 962 962 962 962 960 960 962 962 F-statistic 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Number of firms 113 113 113 113 113 113 113 113
This table presents the regression analysis with panel data to show the difference between with and without entity fixed effects for the variables time of CSR performance and CSR transparency, CSR_P is the logarithm of the firm’s total CSR performance score of the year before. E_P is the logarithm of the environmental performance of the firm of the year before. S_P is the logarithm of the social performance score of the year before. CG_P is the logarithm of the corporate governance performance score of the year before. CSR_T is the logarithm of a firm’s total CSR transparency score. E_T is the logarithm of the environmental transparency score. S_T is the logarithm of the social transparency score. CG_T is t the logarithm of he corporate governance transparency score. The significance at the 10%, 5% and 1% level are indicated by *,** and ***, respectively.
25 4.1.2 Hypothesis 2: the influence of firm size
When looking solely at the influence of firm size on CSR transparency, I find an insignificant negative linear log-relationship in model 2. For H2: ‘Firm size reinforces the positive relationship between CSR performance and CRS transparency’, I find an insignificant positive relationship of the moderator variable with CSR transparency in model 2.
Looking at the influence of firm size on environmental transparency, I find an insignificant negative log -log-relationship relationship in model 4. For H2, in which performance and transparency are both measured on only environmental aspects, I find a significant positive log-log relationship in model 4. When looking at the influence of firm size on social transparency, I find an insignificant negative log log-relationship in model 6. For H2, in which performance and transparency are both measured on only social aspects, I find a significant positive log-log relationship in model 6. When investigating the influence of firm size on corporate governance transparency, I find a significant positive log log-relationship in model 6. For H2, in which performance and transparency are both measured on only corporate governance aspects, find a significant negative log-log relationship in model 8. Therefore,
hypothesis 2 is supported in model 4 and 6.
4.1.3 Hypothesis 3: the time effect
To test Hypothesis 3: ‘The relationship between CSR performance and CSR transparency will contain a time fixed effect’, the Hausman and Redundant fixed effects test were done to test whether there is a time effect in the investigated models. The test showed a p-value smaller than 0.001. Therefore, it can be concluded that the model contains a time fixed effect. Furthermore, the amount of data explained increased in all models and the models are more significant when fixed effects are included
4.1.4 Hypothesis 4: the country effect
26 Looking at the influence of country on environmental transparency, I find an insignificant negative relationship in model 4. For H4, in which performance and transparency are both measured on only environmental aspects, I find an insignificant positive relationship of the moderator variable with environmental transparency in model 4. When investigating the influence of country on social transparency, I find an insignificant positive relationship in model 6. For H4, in which performance and transparency are both measured on only social aspects, I find an insignificant negative relationship of the moderator variable with social transparency in model 6. When looking solely at the influence of country on corporate governance transparency, I find a significant negative relationship in model 8. The coefficient implies that firms operating in the Netherlands will be more transparent. For H4, in which performance and transparency are both measured on only corporate governance aspects, I find a significant positive relationship of the moderator variable with corporate governance transparency in model 8. Therefore, the moderator reinforces the relationship between corporate governance performance and transparency. This means that for French firms the relationship between CSR performance and CSR transparency will be stronger than for Dutch firms. Thus, hypothesis 4 is rejected in all models. 4.1.5 Hypothesis 5: the industry effect
When looking solely at the influence of the industry sensitiveness on CSR transparency, I find a significant negative relationship in model 2. The coefficient implies that firms operating in a non-environmental sensitive industry will be more transparent. For H5: ‘The relationship between CSR performance and transparency will be stronger for firms operating in an environmental sensitive industry sensitivity’. I find that the moderator environmental sensitivity weakens the relationship between CSR performance and CSR transparency in model 2. This means that for firms operating in an environmental sensitive industry the relationship between CSR
performance will be weaker than for firms operating in a non-environmental sensitive industry.
Looking at the influence of the industry sensitiveness on environmental
27 means that for firms operating in an environmental sensitive industry the relationship between performance and transparency will be stronger than for firms operating in a non-environmental sensitive industry. When investigating the influence of the industry sensitiveness on social transparency, I find an insignificant negative relationship in model 6. For H5, in which performance and transparency are both measured on only social aspects, I find an insignificant positive relationship of the moderator variable with transparency in model 6. When looking at the influence of the industry sensitiveness on corporate governance transparency, I find an
insignificant negative relationship in model 8. For H5, in which performance and transparency are both measured on only corporate governance aspects, I find an insignificant negative relationship of the moderator variable with transparency in model 8. Therefore, hypothesis 5 is only supported in model 4.
5 Discussion & Conclusion
The aim of this research was to investigate the relationship between CSR performance and CSR transparency. This was done by examining 113 publicly listed firms, which are domiciled in the Netherlands or France from 2006 until 2014. To test this relationship a multivariate regression analysis was performed.
This research found support for a positive relationship between performance and transparency. Consequently, this research does support the findings of
Gamerschlag et al. (2011) and Bouten (2011). I find support for a positive effect of CSR performance on CSR Transparency. Remarkably, the percentage variance
explained is highest when CSR is measured with corporate governance indicators. For models including the moderating variables, the amount of variance explained
increases however; in the models including environmental and social aspects the variable performance loses its significance. The reason that the corporate governance model supports the theoretical background best could be due to the fact that it
28 Additionally, when entity fixed effects are included, the firm’s scores are related to its own scores, which are likely to explain more of the variance.
The relationship between CSR transparency and performance is useful for practitioners to see whether firms are only transparent about the CSR aspects on which they perform well. In specific, this is relevant for investors trying to make an appropriate decision when they would like to invest socially responsible. The results show that the total CSR and its individual aspects significantly positively influences transparency. Therefore, it can be assumed that the better firms perform on CSR, the more transparent they will be and the more attractive they are to invest in. Another reason why it is relevant for practitioners is that because when firms perform well but are not transparent, obliging firms to be transparent should be less effective when governments are looking at how to motivate firms to act sustainable. Thus, due to the results of this research, it could be argued that it is effective.
Furthermore, this research looks at how country, industry sensitiveness, firm size and time moderate this relationship. This research finds support that firm size reinforces the influence of environmental and social performance on environmental and social transparency. This only partially supports Barnea and Rubin’s (2010) results. However, it does not find support that this is the case for the total CSR aspects. It finds that for corporate governance firm size weakens the relationship between performance and transparency. This could be due to the strategic aspect of corporate governance as well. Future research is necessary to investigate this possible explanation. Next, in all models, hypothesis 3 is supported. Hence, there can be concluded that this research supports previous literature (Gamerschlag et al., 2010) on whether there is a time effect. Future research could investigate whether time
29 sensitiveness weakens the relationship. This means that for firms operating in an environmental sensitive industry the relationship between performance and
transparency will be slightly weaker than for firms operating in a non-environmental sensitive industry. This could be due to the fact that the stakeholders of firms
operating in an environmental sensitivity industry focus less on the performance of those firms on corporate governance and social factors. Consequently, these firms can report on these issues in spite of how they perform. For the environmental aspects, industry sensitiveness slightly strengthens the relationship. This means that for firms operating in an environmental sensitive industry the relationship between
performance and transparency will be slightly stronger than for firms operating in a non-environmental sensitive industry. This supports the results of Campbell (2003) and Gamerschlag et al. (2011) for the environmental aspects but not for the total CSR performance and transparency. As firms in environmental sensitive industries are more watched by their stakeholder and thus, are likely to only report on the environmental aspects they perform well on.
As mentioned before there have been some limitations in measuring CSR performance in past research. The most important limitation discussed before was to measure CSR performance as a voluntary aspect. This is again not completely
30 mentioned when looking at the limitations of data used to measure performance in previous research, these ratings do not show how firms perform, in terms of CSR, adjusted by their firm size. However, I this variable was controlled for in in the models. Likewise, Chatterji et al.’s (2009) most important critique, that more weight should be put on historical performance, is only partially corrected for as this research uses only performance of the previous year.
6 References
- Adams, C.A., Kuasirikun, N. 2000. A comparative analysis of corporate reporting on ethical issues by UK and German chemical and pharmaceutical companies. European Accounting Review 9(1), 53–79.
- Arago´n-Correa, J. A., Sharma, S. 2003. A contingent resource-based view of proactive corporate environmental strategy. Academy of Management
Review, 28(1), 71–88.
- Barbu, E.M., Dumontier, P., Feleaga, N., Feleaga, L. 2014 Mandatory Environmental Disclosures by companies complying with IASs/IFRSs: The cases of France, Germany, UK. The international Journal of Accounting. 49(2), 231-247.
- Barnea, A., Rubin, A. 2010. Corporate social responsibility as a conflict between shareholders. Journal of business ethics, 97(1), 71-86.
- Baron, D., 2001. Private politics, corporate social responsibility and integrated strategy. Journal of Economics and Management Strategy 10, 7-45.
- Bartolomeo, M., Bennett, M., Bouma, J.J., Heydkamp, P., James, P., Wolters, T. 2000. Environmental management accounting in Europe: current practice and future potential. European Accounting Review 9(1), 31–52
- Becker-Olsen, K. L., Cudmore, B. A., Hill, R. P., 2006. The impact of perceived corporate social responsibility on consumer behaviour. Journal of Business Research, 59, 46–53.
- Belkaoui, A., Karpik, P.G. 1989. Determinants of the corporate decision to disclose social information. Accounting, Auditing & Accountability Journal. 2(1), 36-51.
31 - Bouten, L., 2011. CSR reporting: The mastery of the internal dynamics.
Unpublished working paper. University of Gent.
- Brooks, C. 2014 Introductory econometrics for finance. Cambridge University press. Cambridge
- Burrell G., Morgan G. 1979. Sociological Paradigms and Organisational Analysis. Heinemann, London, U.K.
- Business and Human Rights Research Centre, 2015. http://business-
humanrights.org/en/un-guiding-principles/implementation-tools- examples/implementation-by-governments/by-type-of-initiative/national-action-plans [Accessed 19 Augustus 2015].
- Carroll, A. B. 1979. A three-dimensional conceptual model of corporate social performance. Academy of Management Review 4: 497-505.
- Calabrese, A., Costa R., Menichini, T., Rosati, F., 2013. Does Corporate Social responsibility hit the mark? A stakeholder oriented methodology for CSR assessment. Knowledge and Process Management. 20(2), 77-89. - Campbell, D. 2003. Intra- and intersectoral effects in environmental
disclosures: evidence for legitimacy theory? Business Strategy and the Environment 12(6), 357–371.
- Chatterji , A.K., Levine, D.I., Toffel, M.W., 2009. How well do social ratings actually measure corporate social responsibility? Journal of Economics and Management Strategy, 18, 125-169.
- Clarkson, M. B. E. 1995. A stakeholder framework for analysing and
evaluating corporate social performance. Academy of Management Review, 20(1), 92—117.
- Cochran, P.L. 2007. The evolution of Corporate Social Responsibility Business Horizons 50, 449–454.
- Cohen, J.R., Pant, L.W. and Sharp, D.J. 1996. A methodological note on cross-cultural accounting ethics research, International Journal of Accounting, 31(1): 55-66.
32 - Cowen, S.S., Ferreri, L.B., Parker, L.D. 1987.The impact of corporate
characteristics on social responsibility disclosure: a typology and frequency-based analysis. Accounting, Organisations and Society. 12(2), 111-22.
- Debusschere, L., Popelier, E. 2011 Onderzoek naar het nut en het gebruik van de Sociale Balans, Universiteit van Gent, Working Paper
- Deegan, C., 2002 The Legitimizing Effect of Social and Environmental Disclosure – a Theoretical Foundation, Accounting, Auditing &
Accountability Journal 15(3), 282-311.
- Deegan, C., Rankin, M. and Tobin, J. 2000 Firms’ disclosure reactions to major social incidents: Australian evidence Accounting forum 24(1), 101-130 - Donaldson, T. and Presont, L. 1995 The stakeholder theory of the corporation – concepts, evidence, and implications academy of management review 20(1), 65-92
- Eisenbeiss, S.A., 2012. Re-thinking ethical leadership: an interdisciplinary integrative approach. Leadership Quarterly 23(5), 791-808.
- El Ghoul, S., Guedhami, O., Kwok, C.C.Y., Mishra, D.R., 2011. Does corporate social responsibility affect the cost of capital? Journal of Banking and Finance, 35, 2388-2406.
- Entine, J. 2003.The myth of social investing: a critique of its practice and consequences for corporate social performance research. Organization Environment 16, 352-368.
- European Commission. 2015. Corporate Social Responsibility
http://ec.europa.eu/growth/industry/corporate-social-responsibility [accessed 2 november 2015]
- EY& International Integrating Reporting Committee. 2015. Tomorrow’s investment rules 2.0: Emerging risk and stranded assets have investors looking for more from nonfinancial reporting.
- Falk, O., Heblich, P. 2007. Corporate Social Responsibility: doing well by doing good Business Horizons 50, 247–254
- Frankental, P., 2001. Corporate social responsibility—a PR invention? Corporate Communications 6(1), 18–23.
33 - Freeman, R. , Reed, D. 1983 Stockholders and stakeholders: a new perspective
on corporate governance Californian Management Review 25(2), 88-106 - Gamerschlag, R. Möller, K., Verbeeten, F. 2010 Determinants of voluntary
CSR discloser: empirical evidence from Germany Review of managerial science 5(2), 233-262
- Gao, S.S., Heravi, S., Zezheng Xiao, J., 2005. Determinants of corporate social and environmental reporting in Hong Kong: a research note. Accounting Forum. 29(2), 233-242
- Gjølberg, M., 2009. Measuring the immeasurable? Constructing an index of CSR practises and CSR performance in 20 countries. Scandinavian Journal of management. 25, 10-22
- Global Reporting Initiative 2015
https://www.globalreporting.org/Pages/default.aspx [Accessed 10 November 2015]
- Godfrey, P.C., Merrill, C.B., Hansen, J.M. 2009. The relationship between corporate social responsibility and shareholder value: an empirical test of the risk management hypothesis. Strategic management journal 30, 424-445. - Golob, U., Bartlett, J.L., 2002. Communicating about corporate social
responsibility: A comparative study of CSR reporting in Australia and Slovenia. Public Relation Review 33, 1-9
- Gray, J.S., 1988. Towards a theory of cultural influence on the development of accounting systems internationally. Abacus 24(1), 1-15.
- Gray, r., Owen, D. & Adams, C., 1996. Some theories for social accounting?: A review essay and a tentative pedagogic categorisation of theorisations around social accounting, 1-54.
- Greening, D. W., Turban, D. B., 2000. Corporate social performance as a competitive advantage in attracting a quality workforce. Business and Society, 39(3), 254-280.
- Grewal, R., Cote, J.A., Baumgartner, H. 2004. Multicollinearity and
Measurement Error in Structural Equation Models: Implications for Theory Testing. Marketing Science 23(4), 519-529
34 - Guthrie, J., Parker, L.D. Ghauri, P., 2004. Designing and conducting case
studies in international business research. In:. Marschan-Piekkari, R., Welch, C. (Ed.), Handbook of qualitative research methods for international business. Cheltenham, UK: Edward Elgar. pp.109-124.
- Hackston, D., Milne, M.J. 1996. Some determinants of social and
environmental disclosures in New Zealand companies. Accounting, Auditing & Accountability Journal. 9(1), 77-108
- Hasnas, J. 1998. The normative theories of business ethics: a guide for the perplexed. Business ethics quarterly. 8(1), 19-42
- Hofstede, G., 2015. National culture. http://geert-hofstede.com/national-culture.html [accessed 19 August 2015].
- Hossain, M.M. , Alam, M. , Azizul Islam, M. , Hecimovic, A., 2004. Do stakeholders or social obligations drive corporate social and environmental responsibility reporting? Managerial views from a developing country Qualitative Research in Accounting & Management 12(3), 287-314 - Johnson, P., Buehring, A., Cassell, C., Symon, G., 2006. Evaluating
qualitative management research: towards a contingent criteriology. International Journal of Management Reviews, 8(3), 131-156.
- Kelly, G.J. 1981. Australian social responsibility disclosure: some insights into contemporary measurement. Accounting and Finance, 21(2), 97-104. - Kitzmueller, M., Shimshack, J., 2012. Economic perspectives on corporate
social responsibility.Journal of Economic Literature 50, 51-84.
- Lioui, A., Sharma, Z., 2012. Environmental corporate social responsibility and financial performance: Disentangling direct and indirect effects, Ecological Economics. 78, 100-111.
- Louche, C., Van Liedekerke, L., Everaert, P., Leroy, D., Rossy, A., D’Huart, M. 2007. CSR in Belgium: The Institutional Context and Practices, Vlerick Leuven Gent Management School – Working Paper
- Matten, D., & Moon, J. 2008. Implicit’’ and ‘‘Explicit’’ CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review, 33(2), 404-424.
35 - McWilliams, A., Siegel, D. S., Wright, P. M., 2006. Corporate Social
Responsibility: Strategic Implications. Journal of Management Studies, 43, 1– 18.
- Morgan, G. & Smircich, L., 1980. The case for qualitative research. Academy of Management Review, 5(4), 491-500.
- MSCI, 2015. MSCI ESG ratings. https://www.msci.com/esg-integration [accessed 19 August 2015].
- O’Donovan, G., 2002 Environmental disclosure in the annual report: Extending the applicability and Predictive. Accounting. Auditing & Accountability Journal. 15(3), 344- 371.
- Pang, Y.H. 1982. Disclosures of corporate social responsibility. The Chartered Accountant in Australia: 32-34.
- Patten, D.M. 1991. Exposure, legitimacy, and social disclosure. Journal of Accounting and Public Policy. 10, 297-308.
- Patten, D.M. 1992. Intra-industry environmental disclosures in response to the Alaskan oil spill: a note on legitimacy theory. Accounting, Organisations and Society. 17(5), 471-5.
- Peterson, D. K., 2004. The relationship between perceptions of corporate citizenship and organizational commitment. Business and Society, 43(3), 296–319.
- Pigou, A. 1920. The Economics of Welfare. London: Macmillan. - Porter, M. E., & Kramer, M. R. 2002. The competitive advantage of
corporate philanthropy. Harvard Business Review, 80(12), 56-68.
- Radebaugh, L.H., Gray, S.J., 2001 International Accounting and Multinational Enterprises, 5th edition; New York, Wiley
- Renneboog, L., Horst, J. ter & Zhang, C., 2008. Socially responsible
investments: Institutional aspects, performance, and investor behavior, Journal of Banking and Finance, 32, 1723-1742.
- Russo, M. V., Fouts, P. A. 1997. A resource-based perspective on corporate environmental risk management and profitability. Academy of Management Journal, 40(3), 534-559.
36 - Tang, Z., Tang, J. 2012 Stakeholder–firm power difference, stakeholders’
CSR orientation, and SMEs’ environmental performance in China. Journal of Business Venture 27(4), 436–455
- Thomson Reuters 2015. ASSET 4 ESG: Data fact sheet
- Thomson Reuters 2015. http://thomsonreuters.com/en/about-us/corporate-responsibility-inclusion/esg-performance.html [accessed 9 December 2015] - Tice, D.M., Baumeister, R.F. 2004, Masculinity inhibits helping in
emergencies: personality does predict the bystander effect. Journal of Personality and Social Psychology. 49(2), 420-428.
- Trotman, K., Bradley, G.W. 1981. Associations between social responsibility disclosure and characteristics of companies, Accounting, Organisations and Society, 6(4), 355-62.
- Turban, D. B., & Greening, D. W., 1996. Corporate social performance and organizational attractiveness to prospective employees. Academy of
Management Journal, 40(3), 658–672.
- Visser, W., 2014. CSR 2.0 Transforming Corporate Sustainability and
Responsibility, SpringerBriefs in Organisational Studies, Heidelberg Springer, Nijmegen, the Netherlands.
- Walton, C. C. 1967. Corporate social responsibilities. Belmont, CA: Wadsworth.
37
Appendix A: Indicators for performance scores
Environmental Performance Score
Social Performance Score Corporate Governance Performance Score Emissions Reduction Employment
Quality Board Structure
Emissions Reduction Policy Policy Board Structure/ Policy CO2 Equivalents Emission Total
(tonnes)
Employment Satisfaction
Experience / Average years serving on Board
CO2 Equivalents Emission Direct
(tonnes) Salaries
% Non-Executive Board Members
CO2 Equivalents Emission Indirect (tonnes)
Salaries
Distribution % Independent Board Members CO2 Equivalent Indirect Emissions,
Scope Three (tonnes)
Bonus Plan for
Employees CEO-Chairman Separation Commercial Risks and/or
Opportunities Due to Climate Change
Generous Fringe
Benefits Background and Skills CO2 Reduction
Employment Awards
Size of Board (Number of Board Members)
Ozone-Depleting Substances Reduction
Trade Union Representation
Board Diversity (% Women on Board)
Employees Leaving NOx and SOx Emissions Reduction
Turnover of
Employees Board Function
NOx Emissions (tonnes)
% Audit Committee Independence SOx Emissions (tonnes)
Health &
Safety % Audit Committee Management Independence VOC Emissions Reduction Policy Audit Committee Expertise VOC Emissions (tonnes)
Total Injury Rate
% Compensation Committee Independence
Lost Time Injury Rate
% Compensation Committee Management Independence Waste Total (tonnes) Lost Days
% Nomination Committee Independence
Non-Hazardous Waste (tonnes) HIV-AIDS Programme % Nomination Committee Management Independence Hazardous Waste (tonnes) Number of Board Meetings Waste Recycling Ratio
Training &
Development % Board Meeting Attendance Average Water Pollutant Emissions (tonnes) Policy
Waste Reduction Initiatives
Average Training Hours
Per Employee Compensation Policy
Training Costs
Total Compensation Policy Environmental Management System
Certified Percent
Internal
Promotion Highest Remuneration Package Sustainable Transportation
Management Training
Total Board Member Compensation
