Do financial expert CEOs engage less in Corporate
Social Responsibility decoupling?
Thesis Master Accountancy
The University of Groningen, Faculty of Economics and Business
18 January 2021 Lars Venema Student number: 4170474 Email: l.venema.5@student.rug.nl Supervisor: N. Hussain Word count: 11,947 Abstract
There is a lack of understanding about how managerial characteristics influence the firm’s susceptibility to institutional pressures. Therefore, this study examines this phenomenon and provides new insights concerning how the CEO’s expertise and background play an essential role in tackling the institutional pressures that the firm faces and how it affects the firm’s CSR decoupling. Therefore, this study contributes to the institutional- and upper echelon theory. In this study, I hypothesise that financial expertise CEOs influences the firm’s CSR strategy. Based on a worldwide sample of 10,982 observations from 2,513 firms operating in 29 countries from 2006 to 2017 and by using the fixed-effects model to eliminate alternative explanations of the results, the study notes that financial expertise CEOs reduce the gap between the actual CSR performance and reported CSR performance. The additional test shows that financial expert CEOs only affect the environmental dimension since stakeholders see this as the main dimension of CSR and care the most about this dimension. The results implicate that financial expertise matters for improving the implementation of CSR and for increasing the CSR reporting quality. This reduces the information asymmetry between the firm and its stakeholders.
Keywords
Financial expert CEOs ∙ CSR decoupling ∙ Institutional theory ∙ Upper echelon theory ∙ Corporate Governance
Table of content
1. Introduction ... 2
2. Theoretical background and hypotheses development ... 6
2.1. Institutional influences and CSR decoupling ... 6
2.2. CEO’s characteristics and CSR decoupling ... 7
2.3. Corporate governance and CSR decoupling ... 8
2.4. Hypothesis development ... 13
2.4.1. Financial expert CEOs and CSR decoupling ... 13
2.4.2. The moderating role of board independence ... 14
2.4.3. The moderating role of female directors ... 15
3. Research design ... 17
3.1. Data collection and sample selection ... 17
3.2. Variables measurement ... 17
3.3. Empirical model ... 20
4. Results ... 21
4.1. Descriptive statistics and Pearson’s matrix ... 21
4.2. Regression analysis ... 23
4.3. Additional analyses ... 25
4.3.1. Alternative financial expert CEO measurement ... 25
4.3.2. CSR decoupling decomposed ... 25
5. Discussion and conclusion ... 29
5.1. Discussion ... 29
5.2. Conclusion ... 30
5.3. Practical implications, limitations and future research directions ... 31
2
1. Introduction
Corporate Social Responsibility (CSR) reporting has become more and more important for firms in the last decade (Hinze & Sump, 2019). It reflects the social initiatives firms engage in, in the demand of firms’ stakeholders (Tang et al., 2015). With a CSR report, firms can show to the outside world their environmental engagement. However, Boiral (2013) concluded that 90 per cent of the firms that had received an A or A+ from the Global Reporting Initiative did not report all significant negative events in their CSR reports. Firms tend to highlight their positive CSR performances and obfuscate negative ones (Diouf & Boiral, 2017). A recent well-known CSR scandal is the Dieselgate scandal in which Volkswagen (VW) is involved. VW installed a defeat device in its cars, which enabled the cars to detect when they were being tested. As a result, the cars emitted up to 40 times more nitrogen oxide (NOx)
in daily traffic than during laboratory tests (Clemente & Gabbioneta, 2017; Siano et al., 2017). Thus, VW operated less sustainable than they claimed to the outside world. When VW’s stakeholders found out about this scandal, they wanted to penalise VW with calls for boycotts, governmental penalty fines and other forms of stakeholder activism (Bouzzine & Lueg, 2020). This had major financial consequences for VW. VW’s stock price collapsed by 22 per cent in one day, and this was the worst financial meltdown since 2008 (Siano et al., 2017).
Because CSR reports do not represent firms’ real CSR performance, CSR reporting is used as impression management (Diouf & Boiral, 2017). This results in a difference between the actual CSR performance and the reported CSR performance (Sauerwald & Su, 2019). “The gap between how firms communicate about CSR and what firms do in terms of CSR is called CSR decoupling” (Sauerwald & Su, 2019, p. 283). CSR decoupling reduces a firm’s transparency regarding its CSR performance. Without reliable reports, society cannot evaluate and monitor the firm's CSR activities, and therefore they cannot correct the firms (Cho et al., 2015). Moreover, stakeholders are misinformed by the unreliable reports, and when CSR decoupling is detected by the firm’s stakeholders, it can have major (financial) consequences for the firm, such as a reduction in legitimacy and reputation damage (Seele & Gatti, 2017). Since firms face institutional pressures from their stakeholders, it can cause them to decouple CSR (Graafland & Smid, 2019; Hyatt & Berente, 2017; Luo et al., 2017; Meyer & Rowan, 1977). Scholars have recognised that the manager’s preferences, expertise and background might play an essential role in how the firm responds to the institutional pressures it faces (Bansal & Roth, 2000; Cordano & Frieze, 2000; Delmas & Toffel, 2008). This study answers the question whether the CEO’s financial background and expertise matters with respect to a firm’s CSR decoupling decisions.
CSR decoupling can occur in two ways. The first way of CSR decoupling is called a communication gap. In this way of CSR decoupling, a firm invests in sustainable activities but communicates an understated CSR performance to the outside world (Wickert et al., 2016). Firms can struggle to inform stakeholders about their actual CSR performance, resulting in a communication gap (Kim & Lyon, 2015). In case of a communication gap, firms bear the CSR costs, but they do not benefit from the sustainable investments (Bosgra, 2019), such as a better financial performance (Wang et al., 2016) or insurance like-protection from disasters or legal actions (Blacconiere & Patten, 1994; Godfrey et al., 2009). Firms can also create a communication gap when their financial performances decline, so that investors do not think that their money is being wasted on unnecessary CSR activities (Kim & Lyon, 2015). The second way in which CSR decoupling can occur, is by creating an implementation gap. In this way of CSR decoupling, a firm communicates an optimistic CSR performance, but in reality the firm invests less in sustainable activities (Wickert et al., 2016). An implementation gap is caused by increased pressure from stakeholders because stakeholders expect that firms behave socially responsible (Crilly et al., 2012; Hawn & Ioannou, 2016). Firms want to engage in CSR activities to gain social legitimacy from their stakeholders. In the meantime, firms also face pressure to remain efficient
3 internally. Engaging in the stakeholders' demands can result in less internal efficiency, and thus firms can create an implementation gap to make it look like they meet the demands of the stakeholders (Meyer & Rowan, 1977). Moreover, CSR investments are expensive and therefore firms need a lot of financial resources to engage in CSR. When firms lack financial resources to answer to the pressure of stakeholders, firms tend to publish over-optimistic CSR performances to meet stakeholders’ demands since the CSR communication costs are low (Wickert et al., 2016). A wider CSR gap results in a higher information asymmetry between the firm and its analysts. Consequently, analysts and investors become sceptical and price-protective against the firm, which results in higher costs of capital and reduces the firm’s access to finance (García-Sánchez et al., 2020; Reverte, 2012).
Much has been studied about what can cause CSR decoupling and most studies used an institutional lens to examine the CSR decoupling phenomenon (Bromley & Powell, 2012; Crilly et al., 2012; Delmas & Burbano, 2011; Graafland & Smid, 2019; Tashman et al., 2019). However, the majority of these studies did not examine how the firm’s management perceives institutional pressures and how management’s characteristics can influence the decision-making regarding CSR decoupling. Scholars have recognised that the manager’s preferences and background might play an essential role in how the firm responds to the institutional pressures and the demand of CSR performances (Bansal & Roth, 2000; Cordano & Frieze, 2000; Delmas & Toffel, 2008; Lewis et al., 2014). However, there is little understanding about how the manager’s preferences, characteristics and backgrounds affect the responses to the institutional pressures.
An interesting development in the appointment of CEOs is that more firms hire CEOs with financial expertise (Cullinan & Roush, 2011). Firms within the Fortune 100 that have a CEO who was former CFO increased from 12 per cent to 20 per cent over the past ten years (Durfee, 2005). Moreover, out of 264 CEO appointments in publicly traded firms from 2001 to 2004, 15 per cent of the appointed CEOs before the introduction of SOX had financial or accounting experience, while 33 per cent of the CEOs appointed after the introduction of SOX had this financial experience (Cullinan & Roush, 2011). This development also continues outside the United States. In Chinese firms in 1995, only 0.9 per cent of the CEOs had financial experience, which increased to five per cent in 2002 (Jiang et al., 2013). This percentage remained above five per cent from 2003 to 2010 (Jiang et al., 2013). Financial expert CEOs can play an important role in determining the quality of the firms’ disclosures (Gounopoulos & Pham, 2018). Financial expert CEOs make different decisions than CEOs who do not have this background, and therefore it is interesting to examine how this affects the firm’s strategic decision-making regarding CSR (Finkelstein et al., 2009).
Research regarding the managerial characteristics is usually based on the upper echelon theory. This theory argues that the CEO’s strategic decisions are affected by the CEO’s characteristics (Hambrick & Mason, 1984). According to this theory, the decisions regarding CSR performance and CSR reporting are strategic decisions that can be affected by the CEO (Ahn et al., 2020; Hambrick & Mason, 1984; Tang et al., 2015). The costs and benefits of the CSR- disclosure and investments are uncertain, and therefore CSR related decisions can be subject to managerial interpretation (Clarkson et al., 2008; George et al., 2006; Lewis et al., 2014). The managerial interpretation is affected by the CEO’s functional background and experience (Clarkson et al., 2008; George et al., 2006; Y. Li et al., 1997). Financial expert CEOs have built financial expertise throughout their financial careers. This financial expertise allows them to have a deeper understanding of disclosure issues where they may draw upon to make better disclosure decisions and improve the firms’ reporting process (Francis et al., 2008; Gounopoulos & Pham, 2018). Furthermore, they have a better understanding of the investors’ information demands and understand that they have to publish high-quality information to reduce information asymmetry (Gounopoulos & Pham, 2018). In addition, they understand the disadvantages that arise from higher information asymmetry, such as price-protective analysts and investors, which
4 increase the costs of capital (García-Sánchez et al., 2020; Gounopoulos & Pham, 2018). Therefore, financial expert CEOs have more incentives and knowledge to provide high-quality reporting (Gounopoulos & Pham, 2018). Financial expert CEOs have also gained knowledge and expertise on how to manage the firm’s financial resources efficiently (Custódio & Metzger, 2014). Since the CSR investments are expensive and firms can lack financial resources to invest in CSR (Wickert et al., 2016), it is likely that financial expert CEOs can manage the financial resources more efficiently and invest it in CSR activities to meet the stakeholders’ demands. However, it is still unknown how financial expert CEOs affect CSR decoupling.
If the CEO’s characteristics influence the firm’s decision-making, it is important to monitor the CEO’s actions to protect the value of the firm and maintain the relationships with the stakeholders (Kim & Lyon, 2015; Sauerwald & Su, 2019). Within a firm, the board has the monitoring function and has to reduce conflicts between the CEO and the firm’s stakeholders (Daily et al., 2003; de Villiers et al., 2011; Hussain et al., 2016). How efficient the board executes its monitoring function depends on the composition of the board (de Villiers et al., 2011). This study focuses on two board characteristics, namely board independence and board gender diversity.
Independent directors increase the board’s monitoring effectiveness because the CEO has less power over independent directors. Therefore, the independent directors can objectively judge the CEO’s performance which improves the board’s monitoring effectiveness (de Villiers et al., 2011; Jizi, 2017). Furthermore, the interests of independent directors are more aligned with the stakeholders’ interests and they provide various inputs into strategic decision-making for a broader set of stakeholders (Dunn & Sainty, 2009; Jensen & Meckling, 1976). Therefore, independent directors can better execute their oversight function for a broad group of stakeholders. Moreover, independent directors are more sensitive to social demands and promote socially responsible behaviour within the firm (Ibrahim & Angelidis, 1995; O’Neill et al., 1989).
Another board characteristic that is widely studied in CSR research is gender diversity (Jizi, 2017; Vitolla et al., 2020). The board needs different capabilities, experiences, and backgrounds as it brings various resources to the board (Hillman & Dalziel, 2003; Jizi, 2017). Therefore, the firm’s challenges can be managed more efficiently and can help the CEO to better understand stakeholders’ demands (Bear et al., 2010; Boyd, 1990). Female directors are more sensitive than male directors (Nielsen & Huse, 2010), tend to care more about the firm’s environmental performance (J. Li et al., 2017), and increases the extent of CSR disclosures (Giannarakis, 2014). A greater female ratio on the board will create a solid firm environment in which the firm will be more committed towards CSR activities (Bear et al., 2010; Hussain et al., 2016). However, research also suggests that greater gender diversity on the board can bring disadvantages to the firm or do not have any value at all (Darmadi, 2011). Moreover, board gender diversity can be a response to the outside pressure to achieve diversity quotas (Ahern & Dittmar, 2012; Farrell & Hersch, 2005). Therefore, female directors are not appointed because of their background and expertise but to achieve gender quotas (Ahern & Dittmar, 2012; Farrell & Hersch, 2005). In addition, female directors have less experience or talent in effectively monitoring the CEO (Ahern & Dittmar, 2012; Farrell & Hersch, 2005). Studies did not examine how the composition of the board influences CSR decoupling. Therefore, it is still unknown how these two board characteristics affect the CEO’s decision-making regarding CSR decoupling and whether it adds any value to the board regarding CSR decoupling.
Therefore, this study focuses on the CEO’s financial background and expertise related to CSR decoupling and the effect of board independence and board gender diversity on this relationship. This results in the following research question: “What is the effect of financial expert CEOs on the extent of CSR decoupling, and how does the composition of the board influence this relationship?”
5 This study makes several contributions. First, this study contributes to the upper echelon theory (Hambrick & Mason, 1984) and contributes to the call of Delmas and Toffel (2008), who argue that a lack exists in understanding how managerial characteristics influence the firm’s susceptibility to institutional pressures. This study extends current literature about the firm’s strategic responses to institutional pressures and how the CEO responds to these pressures (Lewis et al., 2014; Sharma, 2000; Sharma et al., 1999). Additionally, the study provides new insights into how the CEO’s financial background and expertise affect the firm's CSR strategy. Second, this study contributes to the agency theory (Fama & Jensen, 1983). Board independence and board gender diversity are already widely studied in the CSR literature (Dunn & Sainty, 2009; Jizi, 2017; Vitolla et al., 2020; Walls et al., 2012). However, studies regarding board characteristics related to CSR decoupling are limited. Therefore, this study extends the existing literature by examining how board independence and board gender diversity affects the CEO’s decision-making regarding CSR decoupling. Third, most of the studies that examined CSR decoupling determinants used a sample based on one or several countries (Kim & Lyon, 2015; Sauerwald & Su, 2019; Wickert et al., 2016). This study examined CSR decoupling determinants based on a worldwide sample, making the results better generalisable. Finally, this study provides new insights and argumentations to standard setters, regulatory bodies and firms about the discussion whether financial expertise CEOs and the board’s composition matters for reducing CSR decoupling.
This study used a worldwide sample of listed firms between 2006 and 2017. The sample consists of 10,982 firm-year observations of 2,513 firms operating in 29 different countries. The analyses were performed using the fixed-effects model. Thereby, this study accounts for time-invariant unobserved factors, which eliminates several alternative explanations of the results (Custódio et al., 2019). The results from the fixed-effects model show that financial expert CEOs reduce CSR decoupling. An additional test is conducted where CSR is decomposed into the three different dimensions of CSR. The test shows that financial expert CEOs only reduce the gap between environmental disclosure and environmental performance. No evidence was found that financial expert CEOs affect the social- and governmental dimension. In examining the board characteristics, no evidence was found that board independence and board gender diversity affect the CEO’s decision-making regarding CSR decoupling. Moreover, a robustness test is conducted where the financial expert CEO measurement is extended and also classified CEOs with an MBA degree as financial experts. The robustness test shows no significant effect on CSR decoupling.
The structure of this paper is as follows: First, the theoretical background and hypothesis development are presented (section 2). Hereafter, the research design is explained, where the sample selection and methodology of the study are described (section 3). Furthermore, the results of the statistical tests are described (section 4). Finally, the last section provides the discussion, conclusion and the limitations of this study (section 5).
6
2. Theoretical background and hypotheses development
This chapter provides the theoretical background to get a better understanding of how the CEO’s characteristics play an important role in how institutional pressure is perceived and how the characteristics are related to CSR decoupling. Hereafter, the corporate governance literature will be discussed, and it will be explained how this affects the CEO’s decision-making related to CSR decoupling. Finally, the hypotheses are formulated.
2.1. Institutional influences and CSR decoupling
The literature on the subject of CSR decoupling or greenwashing is still limited, but it is becoming a more popular phenomenon that is being investigated in studies. “Greenwashing is a phenomenon related to decoupling. Whereas decoupling refers to the combination of promising policy statements and poor implementation of programs and impact, greenwashing is defined as the intersection of positive communication about performance (e.g., through reporting) and poor performance (Graafland & Smid, 2019, p. 236). Table 1 shows a literature overview with studies that examined CSR decoupling and greenwashing. According to the literature overview, many studies have been conducted to examine the determinants of CSR decoupling and greenwashing. Most of the studies used an institutional lens. “The institutional theory examines organisational forms and explains the reason for having homogenous characteristics or forms in organisations which are within the same organisational field“ (Fernando & Lawrence, 2014, p. 164). This theory argues that firms operate within a social framework that consists of norms, values and assumptions about what acceptable behaviour is (Carpenter & Feroz, 2001). The institutional theory links the firm’s practices, such as CSR practices, to the social framework of the values and norms (Deegan, 2009). Society drives a firm to gain and maintain legitimacy (Fernando & Lawrence, 2014; Scott, 1995). Firms want to protect their legitimacy, and therefore they conform to the stakeholders’ expectations and institutions (Aldrich & Fiol, 1994; DiMaggio & Powell, 1983; Scott, 1995). Through isomorphic processes, firms adopt institutional practices to conform to the stakeholders’ expectations (Dillard et al., 2004).
There are two dimensions within the institutional theory: isomorphism and decoupling (Fernando & Lawrence, 2014). In case of isomorphism, firms are forced to resemble other firms that face the same environmental conditions, resulting in homogenisation of firms (DiMaggio & Powell, 1983). Decoupling is related to the separation between the firm's external image and its actual practices (Fernando & Lawrence, 2014). In the case of CSR decoupling, a firm can use social and environmental disclosure to report its image, which differs from the firm’s actual performance (Deegan, 2009).
In the literature, CSR decoupling is formulated as “the gap between how firms communicate about CSR and what firms do in terms of CSR” (Sauerwald & Su, 2019, p. 283). Wickert et al. (2016) argue that firms can decouple CSR in two ways. The first way is creating a communication gap. A communication gap means that the firm communicates an understated CSR performance to the outside world (Wickert et al., 2016). Thus, a firm invests a lot in CSR activities but communicates less about its CSR investments to stakeholders. Firms can struggle to inform the stakeholders about their actual CSR performance, resulting in a communication gap (Kim & Lyon, 2015). As a result, firms bear the CSR costs but do not benefit from the CSR investments (Bosgra, 2019), such as a higher financial performance (Wang et al., 2016) or insurance like-protection from disasters or legal actions (Blacconiere & Patten, 1994; Godfrey et al., 2009). Firms can also create a communication gap when their financial performances decline, so that investors do not think that their money is being wasted on unnecessary CSR activities (Kim & Lyon, 2015).
The second way of CSR decoupling is creating an implementation gap. In this case, firms communicate an optimistic CSR engagement but they invest less in the CSR activities (Wickert et al.,
7 2016). This way of CSR decoupling results from increased pressure from stakeholders because they expect that firms behave socially responsible (Crilly et al., 2012; Hawn & Ioannou, 2016). CSR investments are expensive and therefore firms need a lot of financial resources to implement the CSR strategies within their core business. Firms can have a lack of financial resources to answer to the stakeholders’ pressure. Because the CSR communication costs are low, it motivates firms to report over-optimistic CSR performances to meet stakeholders’ demands (Wickert et al., 2016). Firms want to show that they meet the stakeholders’ CSR demands, but firms cannot uphold these CSR policies and stated commitments in practice, which results in CSR decoupling (Graafland & Smid, 2019).
Moreover, according to the literature review, there are various determinants that explain why firms decouple CSR. Cho et al. (2015) argue that institutional pressure requires firms to engage in hypocrisy and develop facades. The institutional pressure leads to CSR decoupling and limits the prospects reported in the CSR reports. Crilly et al. (2012) argue that different stakeholders expectations make firms to engage in CSR decoupling. Internal stakeholders demand positive financial statements, but external stakeholders care more about the environment and care about a positive environmental performance (Crilly et al., 2012). Firms want to meet the demand of both stakeholders but they are not able to meet both expectations and therefore firms decouple CSR. Additionally, governments’ conflicting demand can also result in CSR decoupling, namely when the central- and local governments have conflicting demands, and the firm cannot meet both demands (Luo et al., 2017). Furthermore, the firm size also affects the firm’s CSR implementation and CSR disclosure. Wickert et al. (2016) argue that larger firms are more likely to have an implementation gap because for larger firms it is relatively costly to implement the CSR practices in their core business, compared to the costs of CSR communication. On the other hand, smaller firms are more likely to have a communication gap, because it is relatively expensive for smaller firms to communicate CSR information to the stakeholders, in contrast to the implementation costs of the CSR practices in their core business.
Thus, according to the institutional theory, institutional pressure affects firms’ CSR practices and CSR reporting. These studies have examined why different firms have similar environmental strategies. However, research suggests that firms respond heterogeneously, even when they face similar institutional pressures. Various studies argue that the role of the CEO is very important in how the firm responds to institutional pressures. These studies examined how managerial values and characteristics affect the firm’s response to its stakeholders (Lewis et al., 2014). However, there are limited studies that examined how the CEO’s characteristics influence the firm’s response to institutional pressures (Eesley & Lenox, 2006; Lewis et al., 2014). Therefore, there is a lack of understanding about how managerial characteristics influence the firm’s susceptibility to institutional pressures. Delmas and Toffel (2008) have called for further research regarding this topic.
2.2. CEO’s characteristics and CSR decoupling
Research regarding the managerial characteristics is meanwhile based on the upper echelon theory. This theory explains that individuals have different skills, expertise and backgrounds, which affects their decision-making and the organisational outcomes (Hambrick & Mason, 1984). Moreover, managerial characteristics affect how the firm’s institutional pressures are perceived and interpreted (George et al., 2006), and it is the reason why different executives make different decisions (Hambrick & Mason, 1984; Sharma, 2000). Within the top management, the CEO has the most power and ability to influence the firm's outcomes and decisions on the implementation of environmental initiatives (George et al., 2006). Thus, the CEO affects how the firm perceives external pressures and influences whether it results in CSR decoupling. Moreover, the costs and benefits regarding the CSR disclosure and investments are uncertain (George et al., 2006; Lewis et al., 2014). Due to this uncertainty, the CEO faces difficulties in assessing the costs and benefits of CSR disclosure (Clarkson et al., 2008).
8 Additionally, because of this ambiguity, the CEO relies on its own knowledge, expertise, and interpretation for decision-making regarding CSR disclosure and investments (Delmas & Toffel, 2008), which is affected by the CEO’s functional background and expertise (Clarkson et al., 2008; George et al., 2006; Y. Li et al., 1997).
Given this reliance on individual interpretation, I argue that the CEO’s decisions regarding CSR performance and CSR disclosure are influenced by its background and experience (Lewis et al., 2014). These different characteristics and backgrounds can cause some CEOs to see the request for voluntary disclosure as an opportunity or as a threat (Sharma et al., 1999). Crilly et al. (2012) argue that firms with identical pressure decouple in different ways and argue that it depends on how the CEO perceives the different stakeholders’ interests. CEOs imprint their values and cognitive styles upon the firm, and it affects their decision-making (Lewis et al., 2014). Drawing on the upper echelon theory (Hambrick & Mason, 1984), I examine how the CEO‘s financial expertise and background influence strategic decisions about the firms’ CSR decoupling.
2.3. Corporate governance and CSR decoupling
If the CEO’s characteristics directly influence CSR decoupling, it is important to monitor the CEO’s actions to protect the value of the firm and maintain the relationships with the stakeholders (Kim & Lyon, 2015; Sauerwald & Su, 2019). The agency theory explains that the firm’s board has to monitor and discipline the firm’s management to reduce conflicts between the stakeholders and the CEO (de Villiers et al., 2011; Hussain et al., 2016; Jensen & Meckling, 1976). The board has to mitigate problems, such as corporate misconduct (Fama & Jensen, 1983). The firm’s strategic choices are also affected by characteristics of the board and therefore, the board matters to the firm’s decision-making (Hitt & Tyler, 1991). Moreover, boards have become more inclined to invest in CSR activities and report about these activities to engage in the demands of the stakeholders (Rowe et al., 2014; Xie & Hayase, 2007) Studies provide evidence that the composition of the board affects the board’s monitoring effectiveness (Hillman & Dalziel, 2003; Jizi, 2017). Effective boards have the interests of the directors aligned with the interests of the stakeholders “and are more successful in engaging and reporting on firm’s sustainable development activities as a business strategy to achieve the acceptance of stakeholders and develop sustainable competitive advantage” (Jizi, 2017, p. 643). Moreover, the composition of the board also affects the CSR activities where the firm engages in (de Villiers et al., 2011; Harjoto & Jo, 2011; Walls et al., 2012), and affects the CSR disclosure of the firm (Dunn & Sainty, 2009; Eng & Mak, 2003; Liao et al., 2015). However, a research gap exists in how the composition of the board affects the CEO’s decisions regarding CSR decoupling.
9 Table 1: Literature review determinants of CSR decoupling and greenwashing
Study Independent variable(s)
(Results in parenthesis)
Dependent variable(s)
Decoupling/greenwashing definition
Data source Theory applied Country Firms Baumann-Pauly et al. (2013) Firm size (+) CSR decoupling
Gap between external communication and internal implementation of CSR.
Interviews Stakeholder theory
Switzerland Five MNCs and seven SMEs. 15 interviews conducted. Boiral (2013) External stakeholder
pressure (+)
Greenwashing The extent of the information and images in sustainability reports can be considered a simulacrum that is disconnected from the real impacts of the organisation’s activities. Content analysis Legitimacy- and voluntary theory
International 23 firms that received an A or A + on their sustainability reports. Bromley & Powell (2012) Stakeholder pressure on organisations (+) CSR Decoupling
A gap between policy and practice. Literature review Institutional theory n/a n/a Cho et al. (2015) Contradictory societal and institutional pressures (+) CSR decoupling
The gap between corporate sustainability talk and practice.
Content analysis Signalling- and legitimacy theory United States
Chevron and Conoco Phillips. Crilly et al. (2012) Competing stakeholder expectations (+) CSR decoupling
Decoupling policy from practice. Interviews and archival data Institutional theory International 17 multinational corporations. 359 interviews conducted. Delmas & Burbano (2011) Nonmarket external drivers (+), market external drivers (+), organisational drivers (+), individual psychological drivers (+)
Greenwashing The act of misleading consumers regarding the environmental practices of organisations (firm-level greenwashing) or the environmental benefits of a product or service (product-level greenwashing). Literature review Institutional theory n/a n/a
10 Table 1: Literature review CSR decoupling/greenwashing (continued)
Study Independent variable(s)
(Results in parenthesis)
Dependent variable(s)
Decoupling/greenwashing definition
Data source Theory applied Country Firms Delmas & Montes-Sancho (2009)
Weak political pressure (+), strong relationships with government agencies (-), strong relationships industry trade associations (-), firm visibility (-)
Greenwashing Degree of participation by firms in collective corporate political strategies that aim to shape government policy Department of energy database Institutional theory United States 132 major investor-owned electric utilities.
García-Sánchez et al. (2020)
Analyst coverage (-) CSR decoupling
The gap between CSR disclosure and CSR performance. KLD and Bloomberg Stakeholder theory United States 7,681 firm-year observations. Graafland & Smid (2019) CSR responsibility at the board (-) CSR decoupling
A condition of full divergence among policies, programs, and impacts amounting to purely ceremonial CSR.
Sustainalytics Institutional theory
International About 1,000 large firms.
Hyatt & Berente (2017)
Internal- and external normative stakeholder pressure (+)
CSR decoupling
The distinction between substantive- and symbolic environmental strategies. Survey Institutional theory United States 214 surveys involving supply chain managers.
Jamali et al. (2017)
Stakeholder pressure (+) CSR decoupling
Decoupling of core labour and humanitarian issues from the CSR agenda.
Interviews Institutional theory
India 37 CEOs of SME football manufactures.
Kim & Lyon (2015) Growing firms (+) Deregulated environment (-), lower profits in deregulated environments (-) Greenwashing Brownwashing
Difference between reported and actual emissions reductions.
Department of energy database Legitimacy theory United States 54 investor-owned electric utilities.
11 Table 1: Literature review CSR decoupling/greenwashing (continued)
Study Independent variable(s)
(Results in parenthesis)
Dependent variable(s)
Decoupling/greenwashing definition
Data source Theory applied
Country Firms
Kim & Lyon (2011)
Regulatory pressure (+), pressure from
environmental groups (-)
Greenwashing Selective disclosure of favourable information while withholding unfavourable information. Department of energy database n/a United States 98 investor-owned electric utilities. Lyon & Maxwell (2011)
Activist auditing (-) Greenwashing The selective disclosure of positive information about a company’s environmental or social performance, while withholding negative information on these dimensions.
An economic model of greenwashing.
n/a n/a n/a
Lyon & Montgomery (2013)
Social media (-) Greenwashing Decouple environmental activities from practice to greenwash. Literature review Institutional theory n/a n/a Luo et al. (2017) Conflicting demands from central- and local government (+)
CSR decoupling
Issuing low-quality reports. Content analysis
Institutional theory
China 2,028 listed firms on the Shenzhen or Shanghai Stock Exchanges. Marquis et al. (2016) Firms headquartered in countries that are more connected to global society (-), civil and liberties and political rights (-), strong monitoring activists (-)
Greenwashing Disclose positive environmental actions while concealing negative ones.
Trucost database
Institutional theory
International 4,750 large publicly traded firms. Marquis & Qian (2014) Governmental monitoring (-) CSR decoupling Extent to which CSR
communications are symbolically decoupled from substantive CSR activities.
Content analysis
Institutional theory
China 1,600 publicly listed firms.
12 Table 1: Literature review CSR decoupling/greenwashing (continued)
Study Independent variable(s)
(Results in parenthesis)
Dependent variable(s)
Decoupling/greenwashing definition
Data source Theory applied
Country Firms
Ramus & Montiel (2005)
Firms in service industry (+)
Greenwashing The extent to which environmental policies are implemented. CEP- and employee database Institutional theory
International 188 firms from 20 countries and 586 surveys involving employees from 10 leading-edge European firms. Sauerwald & Su (2019) CEO overconfidence (+) CSR decoupling
The gap between how firms communicate about CSR and what firms do in terms of CSR.
Content analysis Upper echelon theory United States
273 firms from the S&P 500.
Tashman et al. (2019)
Pervasiveness of EM-MNEs' home country institutional voids (-), EM-MNEs’
Internationalisation (-)
CSR decoupling
The degree of misalignment between a firm’s CSR reporting and CSR performance.
MSCI IVA Neo-institutional theory International 93 non-financial emerging market multinational enterprises. Wickert et al. (2016) Firm size (+) CSR decoupling
The gap between the firm’s CSR communication and the
implementation of it into its core structures and procedures.
Literature review Contingency- and agency theory n/a n/a
13
2.4. Hypothesis development
2.4.1. Financial expert CEOs and CSR decoupling
The CEO’s skills, expertise and background, affect its decision-making and also affects how the firm responds to the institutional pressures it faces (Hambrick & Mason, 1984; Lewis et al., 2014). This study focuses on the CEO’s financial expertise because it is likely that its financial- background and expertise influences the firm’s decision-making choices (Custódio & Metzger, 2014; Jiang et al., 2013; Matsunaga et al., 2013).
Financial expert CEOs have built expertise throughout their financial career and therefore they have more expertise in accounting issues and have a deeper understanding of disclosure issues (Matsunaga & Yeung, 2008). This expertise enables the CEO to better monitor the firm’s disclosure policies and this will improve the firm’s reporting quality (Francis et al., 2008; Gounopoulos & Pham, 2018; Matsunaga & Yeung, 2008). Also, financial expert CEOs have an understanding of the capital market participants and their informational needs (Jiang et al., 2013). They understand that investors demand high-quality information to reduce information asymmetry, and they have a better understanding of how to communicate the information effectively to their stakeholders (Matsunaga et al., 2013). Because financial expert CEOs have this understanding, they are more likely to publish high-quality information that meets the investors’ information demands, which reduces the information asymmetry (Gounopoulos & Pham, 2018). Financial expert CEOs also appreciate the implications of disclosure quality on the firm value. They establish the relationship between higher information quality and the cost of capital for the firm and the firm value, in contrast to CEOs who do not have this expertise. Because financial expert CEOs establish this relationship, they have more incentives to provide high-quality reporting to the market to reduce the costs of capital (Matsunaga et al., 2013).
Most of the time, financial expert CEOs hold qualifications that require them to adhere to ethical codes of conduct. Because they have to adhere to ethical codes, it affects their risks attitudes towards a greater conservatism in reporting (Gounopoulos & Pham, 2018; Matsunaga et al., 2013). And if financial expert CEOs commit misconduct and the misconduct is detected, it will negatively affect their reputation (Gounopoulos & Pham, 2018). The reputational concerns may restrain financial expert CEOs from CSR decoupling. Furthermore, in education nowadays, more emphasis is placed on the importance of the firm’s sustainability investments. Business education enhances students’ belief that sustainability is a crucial element of the firm’s performance, which increases the firm’s CSR engagement (Neubaum et al., 2009; Slater & Dixon-Fowler, 2010).
CSR investments cost firms a lot of financial resources, and sometimes firms do not have the financial resources to answer to the pressure from stakeholders to invest in CSR (Graafland & Smid, 2019). Or the firm’s shareholders and managers view CSR investments as deleterious to financial performance and do not want to invest too much in CSR (Crilly et al., 2012). Because financial expert CEOs have more expertise and knowledge on how to manage the firm’s financial resources efficiently, they can increase the firm’s CSR investments efficiently to meet the demands of both stakeholder groups (Custódio & Metzger, 2014). Additionally, financial expert CEOs are more skilled in strategic decision-making and take advantage of CSR opportunities that increase the value of the firm (Geletkanycz & Black, 2001). Therefore, it is more likely that financial expert CEOs manage financial resources more efficiently to invest in CSR activities to meet the stakeholders’ demands.
Prior research found evidence that CEOs who were former CFOs engage less in earnings management and provide more information in the financial disclosures (Matsunaga et al., 2013). Furthermore, CEOs who were former CFOs have more conservative accruals, provide more precise earnings guidance to analysts and improve the quality of financial- and voluntary disclosures
14 (Matsunaga & Yeung, 2008). Bamber et al. (2010) found evidence that CEOs with a financial or accounting background have a more precise disclosure style which improves the information quality. Additionally, Jiang et al. (2013) found evidence that financial expert CEOs tend to engage less in real earnings management, which improves the disclosure quality. Moreover, Gounopoulos & Pham (2018) found evidence that newly listed firms with financial expert CEOs are less likely to engage in accrual-based and real earnings management in the offering year and therefore they improve the disclosure quality. They also concluded that CEOs who have a professional qualification in finance or accounting are also less likely to engage in accrual-based earnings management and therefore provide reports of higher quality. Custódio & Metzger (2014) found evidence that firms with financial expert CEOs manage their financial policies more actively and their firm investments are less sensitive to cash flows. Moreover, they concluded that financial expert CEOs can more easily raise external funds, even when the firm’s credit conditions are tight. Therefore, it is more likely that firms with financial expert CEOs have more financial resources available, and this can be managed more efficiently to invest in CSR-related activities.
Concluding, financial expert CEOs understand that the markets demand high-quality information to reduce information asymmetry, have more expertise in managing the financial resources more efficiently and report more precisely. Moreover, financial expert CEOs recognise and understand the importance of high-quality reports. Hence, it is likely to expect that financial expert CEOs increase the firm’s CSR engagement and improve the CSR reporting quality, resulting in the following hypothesis:
Hypothesis 1 (H1): Financial expert CEOs will reduce the gap between CSR disclosure and CSR performance.
2.4.2. The moderating role of board independence
Directors on the board have the function to monitor the incentives of the CEO (Fama & Jensen, 1983). The CEO holds less power over independent directors, and this enables them to objectively judge the CEO’s performance, resulting in a higher level of monitoring effectiveness (de Villiers et al., 2011; Jizi, 2017). Independent directors are more likely to “objectively questioning and evaluating management and firm’s performance” (de Villiers et al., 2011, p. 1641; Kesner & Johnson, 1990). Within a firm, the board has an oversight function. Independent directors can better execute this function since their interests are more aligned with the stakeholders’ interests and because they provide various inputs into strategic decision-making for a broader set of stakeholders (Dunn & Sainty, 2009; Jensen & Meckling, 1976).
Independent directors are more sensitive to social demands and therefore promote socially responsible behaviour within the firm (Ibrahim & Angelidis, 1995; O’Neill et al., 1989). Independent directors also have higher incentives to engage in social-environmental innovations because they are more likely to be aware of how corporate social performance improves the firm’s standing with stakeholders (de Villiers et al., 2011; Johnson & Greening, 1999). Additionally, firms with more independent boards are more involved in CSR activities and are more likely to disclose CSR activities (Arora & Dharwadkar, 2011). This disclosure of CSR activities leads to a higher level of transparency and promotes the firm's long-term performance (Donnelly & Mulcahy, 2008; Ibrahim et al., 2003; Jizi, 2017). Unlike inside directors, independent directors do not have compensation plans that are related to the firm’s short-term performance (Jizi, 2017). This has as a result that independent directors focus more on the firm’s long-term performance which benefits the firm’s CSR activities and transparent disclosure since CSR activities aim at the long-term performance of the firm (Dunn & Sainty, 2009).
Prior research regarding CSR suggests that greater board independence results in a higher level of voluntary disclosure (Cheng & Courtenay, 2006). Moreover, a greater level of independent directors
15 increases the quality of the firm’s CSR reports (Jizi, 2017; Jizi et al., 2014; Vitolla et al., 2020), and it improves the firm’s environmental performance (de Villiers et al., 2011; Dunn & Sainty, 2009). Additionally, independent directors have to protect their reputation to ensure continued director appointment (de Villiers et al., 2011), which can be enchanted by reducing the firm’s CSR gap.
Concluding, a greater concentration of independent directors increases the monitoring effectiveness and increases the firm’s transparency. It is likely to expect that independent directors encourage the CEO to increase the firm’s transparency and take actions to improve the firm’s long-term performance. Hence, it is likely to expect that greater board independence results in a higher CSR engagement and higher CSR reporting quality, resulting in the following hypothesis:
Hypothesis 2 (H2): A greater concentration of independent directors on the board will moderate the relationship between financial expert CEO and CSR decoupling.
2.4.3. The moderating role of female directors
The board needs different capabilities, experiences, and backgrounds because it brings various resources to the board (Hillman & Dalziel, 2003; Jizi, 2017). A board with more diversity can increase the monitoring effectiveness because directors have different points of view, skills and expertise (Amorelli & García‐Sánchez, 2020; Jizi, 2017). Research suggests that female directors have other resources than male directors, and they make different decisions (Adams & Ferreira, 2009). The firm’s challenges can be managed more efficiently due to these various resources, and it can help the CEO to better understand the demands of stakeholders (Bear et al., 2010; Boyd, 1990). The different resources between male- and female directors is related to their fundamental background, which differs between male and female (Williams, 2003).
Compared to male directors, female directors are more sensitive, which can positively influence the firm’s CSR performances and disclosures (Nielsen & Huse, 2010). They tend to care more about the firm’s environmental performance and increases the extent of CSR disclosures (Giannarakis, 2014; J. Li et al., 2017). Prior research mainly found evidence that more gender diverse boards increase the firm’s CSR commitment and transparency (Amorelli & García‐Sánchez, 2020; Bear et al., 2010; Fernandez-Feijoo et al., 2014; Fernández-Gago et al., 2016; Zhang et al., 2020). This supports the idea that females care more about social concerns (Giannarakis, 2014). Furthermore, firms with more female directors have higher earnings quality and more public disclosure, suggesting that female directors encourage transparency and are better in monitoring the CEO’s activities (Gul et al., 2011; Srinidhi et al., 2011). Additionally, greater gender diverse boards reduce the risk of impression management in sustainability disclosures (García-Sánchez et al., 2019). Regarding their risk appetite, females are less risk-taking than men and tend to avoid risky situations, while men tend to participate in risky situations (Arch, 1993; Croson & Gneezy, 2009). CSR decoupling is risky because when stakeholders detect the misconduct, it can reduce the firm’s ability to obtain resources and legitimacy (Kim & Lyon, 2015). Therefore, female directors are more likely to discourage CSR decoupling. On addition, females are more trustworthy than men, and therefore they are less likely to manipulate disclosures (Heminway, 2007)
On the other hand, Darmadi (2011) argue that greater gender diversity on the board can bring disadvantages to the firm or do not have any value at all. Namely, gender diversity can be a response to the outside pressure to achieve gender diversity quotas (Ahern & Dittmar, 2012; Farrell & Hersch, 2005). Therefore, female directors are not appointed because of their background and expertise but to achieve gender quotas (Ahern & Dittmar, 2012; Farrell & Hersch, 2005). The appointed female directors are younger and have less experience or talent in effectively monitoring the CEO, and therefore they bring less valuable resources to the board (Ahern & Dittmar, 2012; Farrell & Hersch, 2005). Additionally, gender diversity can bring disadvantages to the board such as a greater likelihood
16 of conflicts between the directors, which can slow down the decision-making process and reduces the monitoring effectiveness (Dwyer et al., 2003; Hambrick et al., 1996; Joshi et al., 2006).
It is unclear whether greater gender diversity can bring any value to the board and whether it increases the board’s monitoring effectiveness. If female directors add value to the board, they may reduce CSR decoupling. However, if female directors do not add any value to the board because they have less experience and talent in monitoring the CEO, it is likely to expect that greater gender diverse boards will increase CSR decoupling. From this discussion, the following hypothesis is formulated:
Hypothesis 3 (H3): A greater concentration of female directors on the board may moderate the relationship between financial expert CEO and CSR decoupling.
17
3. Research design
In the third chapter, the research design will be discussed. First, the data collection and the sample selection will be described. Hereafter, the measurement of all the variables will be discussed. Finally, the empirical model will be discussed.
3.1. Data collection and sample selection
The data in this research is archival data and is collected from different databases. Data regarding the CEOs’ characteristics to identify whether CEOs have financial expertise is obtained from the BoardEx database. The BoardEx database contains profiles of more than 420,000 managers and directors around the world. In addition, the CSR disclosure scores are obtained from the Bloomberg database. This database process Environmental, Social, and Governance (ESG) disclosure scores which imply the firm’s CSR reporting quality. This CSR disclosure data is widely used in other studies that investigated CSR decoupling (García-Sánchez et al., 2020; Tashman et al., 2019). The ESG disclosure score measures the transparency of CSR reports. Stakeholders can use this score to evaluate how transparent and accountable firms are regarding CSR issues (García-Sánchez et al., 2020). Moreover, the data regarding the firm’s CSR performance is obtained from the Asset4 database. The Asset4 database is one of the largest databases that provides CSR performance information. It collects information from publicly available sources (Schons & Steinmeier, 2016). This information is widely used in other CSR performance studies (Graafland & Smid, 2019; Hawn & Ioannou, 2016; Schons & Steinmeier, 2016). The Asset4 database provides information regarding the ESG dimensions of CSR which implies the firm’s CSR performance scores (Hawn & Ioannou, 2016). The governance characteristics of the firms are also collected from the Asset4 database. Finally, the financial accounting information is collected from the Worldscope database.
This study focuses on the period from 2006 to 2017. The reason for this period is that ESG disclosure quality data became available from 2006 onwards. Observations are omitted if the necessary information required for the analysis was missing. Due to a CEO replacement, a firm could have multiple CEOs in a certain year. If this was the case, the observation for the new CEO is omitted because a newly appointed CEO does not have much power in the first years and does not affect much of the firm’s investments, in contrast to a CEO with longer tenure (Cannella et al., 2008). Some countries contained a small number of observations. Observations from countries with less than 25 firm-year observations are omitted to avoid representativeness issues (Piets, 2019). After merging the data from the three different databases and deleting missing values, the sample consists of 10,982 firm-year observations for 2,513 firms operating in 29 different countries. The data is unbalanced panel data since some firms are not observed in every year.
3.2. Variables measurement
Dependent variable
In the literature, CSR decoupling is formulated as “the gap between how firms communicate about CSR and what firms do in terms of CSR” (Sauerwald & Su, 2019, p. 283). This study relies on the CSR decoupling measurement of Tashman et al. (2019) and Hawn & Ioannou (2016). Tashman et al. (2019) measured CSR decoupling by subtracting the values of CSR performances from the value of the intensity of CSR reporting. This CSR decoupling measurement is in line with the CSR decoupling measurement of Hawn & Ioannou (2016), which measured CSR decoupling as the difference between the current external actions and prior internal actions. In this study, the external actions are the CSR disclosure scores, and internal actions are the CSR performance scores. Hawn & Ioannou (2016) argue that it takes at least a year for firms to translate their CSR performance into the CSR disclosure. Therefore, this study measures CSR decoupling as the absolute difference between the current CSR
18 disclosure score and the lagged CSR performance score. A negative CSR gap suggests that firms perform better than they disclose (García‐Sánchez et al., 2020).
Independent variable
The independent variable in this study is financial expert CEO, which is measured as a dummy variable taking the value of “1” if the CEO is classified as a financial expert CEO and “0” otherwise. Prior studies classified a CEO as a financial expert if the CEO has held a CFO position (Matsunaga & Yeung, 2008). Jiang et al. (2013) used a broader definition. They classified a CEO as a financial expert when the CEO has been employed in a position of a: “CFO, Chief accounting officer or vice-CEO in charge of the finance or accounting department” (Baatwah et al., 2015, p. 1007). Aier et al. (2005) argue that a CEO with a qualification in accounting or finance also have an understanding of reporting issues. In line with Baatwah (2015), this study used a broader definition of financial expert CEO and classified a CEO as a financial expert when the CEO holds an accounting or finance qualification or has financial work experience in such as a CFO or controller.
Moderating variables
Board independence is measured as the percentage of independent directors on the board (de Villiers et al., 2011). Independent directors are “non-executive directors who are not involved in a (financial) relationship with the firm” (de Villiers et al., 2011, p. 1649). The gender diversity of the board is measured as the percentage of female directors on the board.
Control variables
CEO age is measured as the current age of the CEO in a certain year. Older CEOs can be more interested in improving their legacy (Matta & Beamish, 2008), and may result in more positive CSR activities (Sauerwald & Su, 2019).
Firm size is measured as the natural logarithm of the total assets of the firm (Vitolla et al., 2020). Prior studies argue that larger firms face more pressure regarding their environmental impact (Christmann & Taylor, 2006). Furthermore, larger firms are more likely to have a CSR implementation gap due to the high costs of implementation and relatively lower costs of communication. Smaller firms are more likely to have a communication gap because it is relatively costly for smaller firms to communicate their CSR information compared to the lower costs of CSR implementation (Wickert et al., 2016). CEO duality is a dummy variable taking the value of “1” if the CEO is also chairman of the board and “0” otherwise. In the case of CEO duality, the CEO is also chairman of the board. Therefore, the CEO has more power to influence the firm, which can affect CSR reporting quality or CSR investments (Sauerwald & Su, 2019).
Board size is measured as the total number of directors on the board. Larger boards often have free-rider issues and high coordination costs. Therefore, a larger board size can reduce the monitoring effectiveness of the board (Sauerwald & Su, 2019).
Board Meetings are measured as the number of board meetings during the year. A higher amount of board meetings suggest a higher monitoring intensity (Laksmana, 2008).
Analyst coverage is measured as the number of analysts that cover a firm in a year (Hawn & Ioannou, 2016). Prior research found evidence that higher analyst coverage reduces the CSR gap (García-Sánchez et al., 2020).
Leverage is measured as the total debt divided by total shareholder equity since firms with a higher debt ratio have fewer resources to invest in CSR practices (Tashman & Rivera, 2010).
Capital intensity is measured as the ratio of assets to sales since it affects how firms use their assets for the CSR performance (Russo & Fouts, 1997).
Return on Assets is measured as earnings before interest and taxes divided by average total assets, lagged by one year (Liao et al., 2015). It is used as a proxy for profitability since firms with a higher profit have more resources to invest in their CSR policies (Graafland & Smid, 2019).
19 Sales Growth is measured the percentage change in total sales with respect to the total sales of the previous year since a higher increase in sales give firms more opportunities to invest in CSR activities (Hawn & Ioannou, 2016; Hussain et al., 2016).
R&D intensity is measured as the ratio of R&D expenditures to sales since firms with higher R&D intensity are more innovative and may be better able to improve their CSR investments (McWilliams & Siegel, 2000).
Table 2 provides an overview of all the variables that are used in this study. It shows the mnemonics of the variable, the role and the variable’s measurement.
Table 2: variable measurements
Variable Mnemonics Role Measurement variable
CSR decoupling CSR_Decoupling Dependent The current CSR disclosure score minus prior year CSR performance score.
CEO Financial expertise
CEOFex Independent Dummy variable taking the value of “1” if the CEO is classified as a financial expert CEO and “0” otherwise. Board
independence
BoardInd Moderator The percentage of female directors on the board.
Board diversity BoardDiv Moderator The percentage of independent directors on the board.
CEO Age CEOAge Control The current age of the CEO in a certain year.
Firm size FirmSize Control The natural logarithm of total assets of the firm.
CEO duality CeoDua Control Dummy variable taking the value of “1” if the CEO is
also chairman of the board, and “0” otherwise.
Board size BoardSize Control The total number of directors on the board.
Board Meetings NumBoardMeet Control The number of board meetings during the year.
Analyst coverage AnalystCov Control The number of analysts that cover a firm in a certain year.
Leverage Leverage Control The total debt divided by total shareholder equity.
Capital intensity CapInt Control The ratio of assets to sales.
Return on Assets RoA Control The earnings before interest and taxes divided by
average total assets lagged by one year
Sales Growth Change_Sales Control The percentage change in total sales with respect to the total sales to the previous year.
20
3.3. Empirical model
The fixed-effects model is used in analysing the empirical model. The fixed-effects model controls for all variables that do not change in time and that are not included in the regression model. Moreover, this model controls for the time-invariant variables that cannot be measured. Thus by using the fixed- effects model, it eliminates the probability that these time-invariant characteristics impact the regression results. Furthermore, the fixed-effects model helps to eliminate several alternative explanations, such as reverse causality and endogeneity “because it solely relies on within-firm–CEO variation” (Custódio et al., 2019, p. 461). Thus, by using the fixed-effects model, the reliability of the results increase. The empirical model to examine the effect of financial expert CEOs on CSR decoupling is as follows:
CSRDecouplingi,t
= β0+ β1CEOFexi,t+ β2BoardIndi,t+ β3BoardDivi,t+ β4CEOAgei,t+ β5FirmSizei,t
+ β6CEODuai,t+ β7BoardSizei,t+ β8NumBoardMeeti,t+ β9AnalystCovi,t+ β10Leveragei,t
+ β11CapInti,t+ β12RoAi,t+ β13Change_Revenuei,t+ β14RnDInti,t+αi,t+εi,t
21
4. Results
In chapter four, the results of the study will be presented. First, the descriptive statistics and Pearson’s correlation matrix are described. Hereafter, the results of the regression analysis, robustness test and additional tests are discussed.
4.1. Descriptive statistics and Pearson’s matrix
Table 3 shows the descriptive statistics of the used variables. Interestingly, CSR decoupling has a positive mean of 2.635. The positive CSR decoupling average indicates that firms engage in CSR decoupling in a way that they have a higher disclosure score than CSR performance score. Thus, firms have a better CSR disclosure than their real CSR performance, and therefore firms are overstating their CSR disclosure. The financial expert CEO average is 23.6 per cent which means that 23.6 per cent of the firms have a financial expert as a CEO. The board independence has a mean of 71.3 per cent, which indicates that a big part of the directors that serve on boards are independent. However, some boards do not have independent directors at all, and other boards consist of directors who are all independent. The mean of board diversity is 15.5 per cent, which means that on average 15.5 per cent of the board members in the sample are female. Moreover, some boards do not have females serving on the board at all.
The variance inflation factors (VIF) are calculated to test for multicollinearity. In the case of multicollinearity, the variables measure the same concept. The maximum variance inflation factor in this study is 1.19, which is far under the acceptable threshold of 10 (O’brien, 2007) and therefore multicollinearity does not occur in this study.
Table 4 provides information regarding the Pearson’s correlation. Within the Pearson´s correlation, multicollinearity exists when variables are correlated more than 0.7, which does not occur in the sample. Therefore, multicollinearity is not an issue in the sample.
Table 3: Descriptive statistics
Variable Mean SD Min Max VIF
CSR Decoupling 2.635 30.898 -90.660 74.120 -
Financial expert CEO 0.236 0.424 0.000 1.000 1.08
Board independence 71.279 20.097 0.000 100.000 1.19 Board diversity 15.453 11.060 0.000 75.000 1.09 CEO age 56.401 6.987 29.000 87.000 1.05 Firm size 15.648 2.989 0.000 27.811 1.19 CEO duality 0.476 0.499 0.000 1.000 1.09 Board size 10.486 2.971 1.000 35.000 1.10 Board meetings 8.377 3.819 1.000 52.000 1.04 Analysts coverage 9.006 8.468 0.000 53.000 1.18 Leverage 25.356 18.347 0.000 90.330 1.05 Capital intensity 13.070 27.140 0.000 262.140 1.11 Return on Assets 5.595 8.328 -53.850 34.480 1.05 Change sales 8.296 23.882 -55.085 240.724 1.02 R&D intensity 0.994 0.179 0.000 2.149 1.05
The sample consists of 10,982 firm-year observations of 2,513 firms.
The accounting variables (Leverage, Capital intensity, Return on Assets, Change sales and R&D intensity) are winsorised by year at 1 per cent and 99 per cent.
22
Table 4: Pearson’s Correlation matrix
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. CSR_ Decoupling 1.000 2. CEOFex -0.048*** 1.000 3. BoardInd -0.128*** 0.239*** 1.000 4. BoardDIv 0.108*** 0.035*** 0.179*** 1.000 5. CEOAge 0.014 0.013 0.056*** -0.012 1.000 6. FirmSize -0.157*** 0.026*** 0.070*** -0.012 0.019** 1.000 7. CEODua 0.081*** -0.120*** -0.159*** -0.021** -0.187*** -0.023** 1.000 8. BoardSize 0.161*** 0.022** -0.123*** 0.162*** 0.064*** -0.018* -0.061*** 1.000 9. NumBoard Meet 0.082*** -0.037*** -0.012 0.029*** 0.001 0.012 0.095*** 0.044*** 1.000 10.AnalystCov -0.106*** 0.011 0.027*** -0.002 0.013 0.390*** -0.013 -0.010 -0.004 1.000 11. Leverage -0.009 0.068*** 0.015 0.006 -0.010 0.036*** -0.019** -0.010 0.054*** 0.016 1.000 12. CapInt -0.006 0.022** 0.007 0.134*** -0.008 -0.001 0.019** -0.157*** 0.005 -0.022** 0.182*** 1.000 13. RoA 0.025*** 0.035*** -0.024** 0.028*** -0.016* -0.003 -0.041*** -0.0140 -0.125*** 0.003 -0.045*** -0.155*** 1.000 14. Change_ Sales -0.068*** -0.039*** -0.031*** -0.078*** -0.053*** -0.001 0.031*** -0.082*** -0.022** -0.010 -0.033*** 0.079*** 0.015 1.000 15. RnDInt -0.004 -0.066*** -0.165*** -0.061*** -0.068*** -0.041*** 0.020** -0.072*** -0.067*** 0.009 0.026*** -0.031*** -0.020** 0.006 1.000 The sample contains 10,982 observations for 2,513 firms.
The accounting variables (Leverage, CapInt, RoA, Change_Sales and RnDInt) are winsorised by year at 1 per cent and 99 per cent. The coefficients are significant at: * p<0.10, ** p<0.05, *** p<0.01.
