The moderating effect of industry regulation on the relationship between corporate social performance and corporate financial performance : is this relationship explained by R&D?

(1)

The moderating effect of industry regulation on the relationship between

corporate social performance and corporate financial performance

Is this relationship explained by R&D?

Master Thesis

Author: Laura Termeer Student No.: 10664974

Supervisor: Pushpika Vishwanathan

MSc. in Business Administration – Strategy Track University of Amsterdam

(2)

Statement of Originality

This document is written by Laura Termeer who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

Governments increasingly encourage and promote corporate social responsibility (CSR) in order to influence companies to consider environmental and societal problems. The industry-specific regulations and possible sanctions forced by the government affect the engagement of firms in CSR activities and therefore affect their profitability. This study provides information on the moderating effect of industry regulations on the corporate social performance (CSP) and corporate financial performance (CFP) relationship. Conceptual papers suggest that this effect is positively explained by investments in research and development (R&D). This study not only attempts to uncover the moderating effect of industry regulations, but also contributes to empirically verifying these theoretical concepts. The relationship is tested by using three third-party databases and using a sample, which consists of 467 companies from the S&P 500 from years 2012-2013. Findings show that CSP positively affects the financial performance variable’s return on equity (ROE) and return on sales (ROS). The other results were inconsistent with the hypotheses, which implies that no significant relationship was found between the moderating effect of industry regulations and the CSP-CFP relationship. The appearance of CSR as a norm in an industry due to regulations can be considered as a reason for the lack of answer of the research question. This indicates that the rejected predictions ask for future research.

(6)

I. Introduction

Nowadays, with a growing population, ever-rising demand for resources and awareness for the consequence of climate change, interesting innovation in policies and business practices is growing. Recently, worldwide initiatives have been established to help small and medium-size enterprises to move to circular models. This year, the Swedish government introduced tax breaks to tackle the ‘throwaway culture,’ thus accelerating the transition towards a circular economy (European Commission, 2017).

In the early 21st

century, corporate social responsibility (CSR) has become a topic of priority on government agendas. Governments increasingly encourage and promote CSR to act as a mediator between businesses and NGOs (Albareda et al., 2008). This enables governments to influence companies regarding actual environmental and societal problems (Albareda et al., 2007). In general, CSR is seen as a business-driven approach in which companies pay much attention to the development of CSR initiatives (Albareda et al., 2008). However, stakeholders, such as governments, have taken on a relevant role as drivers of CSR recent years (Moon, 2004).

With the rising debate regarding the role of governments in CSR, many studies have focused on this relation (Zadek, 2001; Aaronson & Reeves, 2002; Fox et al., 2002; Lepoutre et al., 2004; Bell, 2005; Albareda et al., 2008). Zadek (2001), a pioneer is this field, explained the third generation of corporate citizenship that goes beyond individual activities and where the government and civil society organizations form partnerships. Zadek (2001) argued that by recognizing the impact of businesses on economic and social aspects, governments need to connect the corporate citizenship debate to their public policy goals. One of the most well-known classifications was published by Fox et al., (2002). They identified four different roles the public sector could play when creating an environment where companies are encouraged to minimize social and environmental costs while gaining financial benefits: mandatory

(7)

(legislative), facilitation (guidelines on content), partnering (engagement with multi-stakeholder processes) and endorsing tools (publicity) (Fox et al., 2002).

In this research, there is a focus on the mandating role of governments in industries insofar as they set minimum standards for businesses performances, by imposing regulations. Industries are encouraged by regulations to invest in CSR, which is defined as a voluntary concept that guides responsible business activities (European Commission, 2011). The threat of regulatory pressure may motivate industries to voluntarily engage in CSR activities, which could result in better compliance with current and future regulations (Uchida & Ferraro, 2007).

Previous studies have shown the involvement of government in CSR and the role of regulations as a predictor of CSR (Aguinis & Glavas, 2012). However, there has been less research on the impact of a company’s financial performance due to the increase in CSR caused by industry regulations. Regulations as a moderator in the CSP-CFP relationship has previously only studied on a conceptual level based on literature studies (Campbell, 2007).

The costs of complying with government regulations vary widely across industries (Leone, 1986). Industry-specific regulations and possible industry-specific sanctions forced by the government affect the engagement of a firm in CSR activities and therefore affect its profitability (Campbell, 2006; Waddock & Gravas, 1997). Porter and Van der Linde (1995) have suggested that environmental regulations from governments could enhance innovation within a company. Firms do not always make optimal choices, and therefore regulations could result in an increase of a company’s efficiency, which can lead to financial benefits. Prior literature has found that innovation is associated with CSP (Hull & Rothenberg, 2008; McWilliams & Siegel, 2000; Surroca et al. 2010; Berrone et al. 2007). Jaffe and Palmer (1997) have also shown a positive link between innovative efforts as a response to regulation and broader research and development (R&D) expenses. To test if industry regulations lead to

(8)

innovation in the context of CSP, this study also investigates the impact of industry regulations on the relationship between CSP and R&D.

This study aims to touch upon the role of industry regulations in the relationship between CSP and CFP. This research investigates if innovation is an explanation for a positive effect. Based on the discussion above, the following research question is formulated:

What is the moderating effect of industry regulations on the relationship between corporate social performance and corporate financial performance? What is the role of R&D in this relationship?

The relationship was empirical tested by using three third-party databases and a sample that consists of companies 467 from the S&P 500 in years 2012-2013. For CSP and CFP, a lag-time of one year was used to take into account an indirect effect. To test the hypothesis, regression analyses were conducted for the different variables. This includes a moderation analysis, which was employed to test the effect of industry regulations. The results of this study reflect that there is no evidence for the moderating effect of industry regulation on the CSP-CFP nexus. Furthermore, no evidence was found for the role of R&D in this relationship. This indicates that the rejected predictions require future research.

This study contributes to CSR research in different ways. As mentioned before, previous studies have focused on regulations as a predictor of CSR (Aguinis & Glavas, 2012). This study not only attempts to uncover the moderating effect of industry regulations on the CSP-CFP nexus, but also contributes to empirically verifying theoretical concepts. The increase of innovation in response to regulations is empirically tested, which contributes to the conceptual literature from Porter and Van der Linde (1995). The empirical investigation is important to prove these theories and advise governments on how their regulations can affect industries and thereby the economic growth.

(9)

This thesis is structured as follows. First, the existing relevant literature on the CSP-CFP relationship and the relevant theoretical literature on industry regulations is discussed. This is followed by the defined research question, proposed hypotheses and conceptual model. Afterward, the methodology is explained. Next, the empirical results are presented. Subsequently, the discussion is given, which is proceeded by the contributions of the paper to both theory and practice with suggestions for future research. Finally, conclusions from this research are drawn.

(10)

II. Literature review

This chapter will review the theories been used for this paper. In the first sections, the concepts of corporate social responsibility, corporate social performance, and corporate financial performance will be discussed. Next, an overview of the previous empirical studies on the CSP-CFP relationship will be given. Furthermore, the concept of industry regulation will be discussed, as well as its possible effects on the CSP-CFP relationship. The last section will explain the conceptual framework, and hypotheses will be presented.

2.1 Corporate social performance and financial performance

Corporate social responsibility

The concept of corporate social responsibility (CSR) originated in the 1950s when the economist Howard R. Bowen had the Social Responsibilities of the businessman published. This book was the first comprehensive discussion of business ethics and social responsibility. Bowen (1953) defined CSR as follows: “it refers to the obligation of businessman to pursue those policies, to make those decisions, or to follow those lines of action which are desirable in terms of the objectives and values of our society” (p. 6). The definition of corporate social responsibility increased in use with the rise of multinational corporations in the early 1970s. Other synonyms, such as sustainable responsible business and corporate citizenship, have been used more frequently in recent years (Carroll, 2000). To give a complete definition of CSR, multiple dimensions should be considered (Carroll, 1979). The idea behind CSR is based on the “triple bottom line” concept, created to measure a company’s full economic value, including social, environmental, and financial responsibility, which shapes a firm’s corporate responsibility (Elkington, 1994). These three dimensions are highly correlated (Deegan, 2000). Many researchers tried to give an all-encompassing definition of CSR, which

(11)

is unbiased and robust. Currently, no consensus has been reached on a general accepted definition of CSR due to the different interests and perceptions of stakeholders. Dahlsrud (2008) studied 37 definitions of CSR in the period between 1980 and 2003 and found significant similarities between them, all referring to five dimensions, categorized as voluntariness, stakeholder, social, environmental, and economic. The definition of CSR from the European Commission (2001, 2002, 2006) is the most frequently used in academic literature and included all five dimensions. The Commission describes CSR as “a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholder on a voluntary basis” (European Commission 2001, p. 6). In this definition, there are two important mechanisms of CSR included. First, it refers to the presence of stakeholders. When engaging in CSR, public interests are considered in the corporate decision-making process. Second, the definition implies CSR is multidimensional when creating value. Dahlsrud (2008) argued the lack of one universally accepted CSR definition is overestimated due to all the overlaps between the definitions. Some studies use the concept without giving a specific definition (McWilliams et al., 2006).

Corporate social performance

Corporate social performance (CSP) is a way of making CSR applicable because CSR is not a variable, which could be measured (Van Beurden & Gössling, 2008). Some scholars tried to offer categories for evaluating CSP (Sethi, 1979). However, like CSR, this concept also has no generally accepted definition mainly because both are broad meta-constructs consisting of multiple dimensions: economic, environmental, and social. Wartick and Cochran (1985) see CSP as “the underlying interacting among the principles of social responsibility, the process of social responsiveness, and the policies developed to address social issues” (p. 758). In many studies, corporate social performance is used to measure CSR. The multidimensional aspect of CSP, comprising internal and external measures, causes difficulties in measuring the

(12)

construct. Different measurements could be used, which needs to be considered due to their influence on outcomes (Orlitzky et al., 2003). Beurden and Gössling (2008) distinguished three different types of measurement: first, the degree of social disclosure about matters of social concern (Wu, 2006); second, the measurement of corporate action, such as the social activities and programs companies execute; and third, the measurement of corporate reputation ratings, such as MSCI and Fortune (Waddock & Graves, 1997; Hull & Rothenberg, 2008). All measurements have some constraints. For example, the Fortune rating focuses more on overall management performance. Therefore, some studies are based on more than one measure to mitigate the constraints of a single measurement (Griffin & Mahon, 1997).

Corporate financial performance

A company’s financial performance can be described as the company’s ability to generate revenues from its core activities (Brealey et al., 2011). By comparing these values of related benchmark firms, an entity can assess how well it is performing in the market. There are several ways of measuring corporate financial performance. Griffin and Mahon (1997) performed a wide analysis of 80 CFP indicators studying methodological inconsistencies, where 70% were used only once. The limited overlap between measures makes it difficult to assess the reliability and validity of these measures. According to Orlitzky et al. (2003), there are three ways to measure it: Market-based, accounting-based, and perceptual measures. This research focuses on accounting-based measures, which will be further discussed in the methodology.

Motivations to engage in CSR

There is wide disagreement about the paybacks of CSR investments. Some scholars argue CSR leads to several benefits, and others argue CSR can diminish financial performance if used incorrectly. This means the use of CSR needs to fit with the company’s strategy and culture; otherwise, it will incur indirect costs (Margolis & Walsh, 2003).

(13)

There are different motivations for firms to engage in CSR. From an altruistic view, firms invest in CSR to behave as good citizens. According to Lantos (2001) “Altruistic CSR includes action that morality does not mandate but which are beneficial for the firm’s constituencies although not necessarily for the company” (p.38). From this perspective, the engagement in CSR has a positive effect on the society as a whole.

Another motivation to invest in CSR is to attain higher financial benefits. According to the European Commission (2008), there are several important determinants that could cause positive economic effects when engaging in CSR: human resources, differentiation, innovation, risk management, and reputation. These determinants increase a firm’s financial performance in a direct or indirect way.

Human resources

By adopting CSR, firms can use it as an instrument to motivate their employees, attract new talent, and retain highly skilled personnel. A higher quality of workforce can lead to a better financial performance (Waddock & Graves, 1997).

Differentiation

Another motivation to invest in CSR for companies is to differentiate themselves. Mackey and Barney (2007) argued that the higher level of CSR, the more firms can differentiate. In addition, Porter (1980, 1996) argued that differentiated firms achieve above-average returns. Differentiation is for example possible through investments in innovation (Miller, 1986).

Innovation

The development of an innovation in a company can cause a more efficient business process or product. This can lead to increased efficiency within the company. Firms can turn the increasing environmental and societal risks into new business opportunities by using CSR (Grayson & Hodges, 2004). In addition, innovation is seen as a collaborative exercise, which

(14)

can results in constructive relationships with a new range of stakeholders (Mendibil et al., 2007).

Risk management

There is increasing awareness of environmental and social problems. The firm’s environment determines its internal and external pressures (Aguilera et al., 2007). This causes an increasing pressure on firms to become more transparent to their stakeholders and engage in more socially responsible activities. Some firms face increasing pressure to engage in CSR from stakeholders, such as the government, media, activists, and shareholders. This stakeholder pressure leads companies to go green (Kassinis & Vafeas, 2006). By adopting CSR, firms will reduce risk, which will lead to a reduction in costs (Orlitzky & Benjamin, 2001).

Reputation

Studies found CSR can be used to strengthen a company’s brand, which can lead to an increase in customer loyalty and a better relationship between company and stakeholders in general. An increased reputation is also related to a firm’s attractiveness to future employees (Turban & Greening, 1997). A positive reputation is an important determinant of improved financial performance (Orlitzky et al, 2003; Waddock & Graves, 1997).

There are different approaches companies can choose regarding CSR: inactive, reactive, and proactive (Van Tulder et al., 2008). An inactive approach indicates that companies only have the responsibility to create profits. This inward-looking perspective is related to Friedman’s (1970) view, which is discussed in the next section. This approach is focused on the economic responsibility and is wealth oriented. Proactive CSR goes further than the rules and regulations set by law, focusing on the three dimensions of the triple bottom line and society’s long-term concerns (Aguilera et al., 2007; Groza et al., 2011), possibly to anticipate forthcoming future regulations and strengthen the company’s

(15)

relationship with the government (McWilliams et al., 2006). This approach is focused on social responsibility and is welfare oriented. In contrast, a reactive strategy is used to react to changing environments, comply with government restriction, and mitigate harm when reputation could be damaged (Chang, 2015; Murray & Vogel, 1997). In general, consumers view proactive CSR more positively than reactive CSR because they perceive the investment as strategic-driven, as it is planned in advance (Lee et al., 2009).

The CSP-CFP relationship

There are contrasting opinions on whether engagement in CSR results in financial benefits to firms. With his proactive statement on the role of business in society, the economist Milton Friedman (1970) raised the question of whether directors can act in any way to increase profits. In his view, the only purpose of the firm is to generate profit for shareholders. Furthermore, companies adopting a socially responsible behavior will be less competitive because they are confronted with binding constraints, which distracts from their core obligation and causes indirect costs. Based on the agency theory, Friedman (1970) also argued CSR initiatives could cause self-interested behavior in managers at the expense of shareholders (McWilliams & Siegel, 2001). In relation to CSR, Friedman (1970) stated companies should only invest in CSR if shareholders’ value will be maximized. However, social issues are in general the responsibility of governments. On the contrary, according to Freeman’s (1984) stakeholder theory, companies should focus on all their stakeholders, and not only on their shareholders’ wealth. This theory states that focusing on all stakeholders’ interests is vital for the firm’s financial performance. Another commonly used theory to explain the positive effect of CSR in a firm is the resource-based view (RBV), which was founded by Wernerfelt (1984) and further developed by Barney (1986, 1991). This theory postulates companies can gain a sustained competitive advantage if they have resources that are valuable, rare, inimitable, and non-substitutable, called VRIN-resources (Barney, 1991).

(16)

McWilliams and Siegel (2001) used the RBV to receive optimal investments in CSR, when applying a differentiation strategy. The RBV and stakeholder theory are closely related. Hart (1995) argued that if firms have strong relationships with their stakeholders, they will develop valuable resources and capabilities at the same time, including intangible resources, such as innovation and organization culture (Wang et al., 2016).

Previous studies tried to determine the value of CSR. This is not possible by examining the concept alone; therefore, prior studies investigated the relationship between CSP and CFP. Bragdon and Marlin (1972) conducted the first study on the CSP-CFP relationship, followed by many studies on this topic by other researchers. Research conducted has not yet led to conclusive results on the correlation between this relationship (Hull & Rothenberg, 2008). The advantages and drawbacks of CSR and the importance for management are still ambiguous. The positive, negative, and neutral correlations can be explained by inconsistencies between research frameworks (Griffin & Mahon, 1997; Waddock & Graves, 1997). Varying methodologies exist due to the multidimensional nature of CSR (Wu, 2006) and the impact of managerial decisions that are part of CSP (Wood, 1991a). Other reasons for inconclusive results are varying financial performance measures (Margolis et al., 2007; Waddock & Graves, 1997), unsuitable sampling techniques (Davidson & Worrell, 1990), and the lack of theoretical foundation (Ruf et al., 2001).

Most scholars found empirical evidence for a positive relation due to strengthening of stakeholder relations while preventing costly stakeholder conflicts (Schuler & Cording, 2006). According to Wang and Choi (2013), good stakeholder relationships are a source of competitive advantage. Surroca et al. (2010) and Waddock and Graves (1997) found a bidirectional relationship between CSP and CFP based on the resource-based view. Barnett (2007) argued CSR cannot universally produce favorable returns for all firms all the time, so favorable findings will never be replicable across all data sets. However, the positive effects

(17)

of CSR activities generally outweigh the negative, which results in net neutral or positive economic returns.

Various meta-studies were conducted, which found positive relationships between CSP and CFP. Margolis and Walsh (2007) collected 167 published studies in the period 1972 to 2002. Almost half the studies found a positive relationship and only seven studies found a negative relationship. Margolis and Walsh (2007) concluded the effect of the relationship is complex, small, and mildly positive. Orlitzky et al. (2003) conduced a meta-analysis of 52 studies on the CSP-CFP relationship in the period of 1972 to 2002. The results show a “bidirectional and simultaneous” (Orlitzky, et al., 2003, p. 427) relationship where CSR was related to financial performance in the long term.

Other studies found a negative relationship between CSP and CFP (McGuire et al., 1988). As mentioned before, Friedman (1970) argued the relationship is negative due to an increase in costs. In addition, Aupperle et al. (1985) argued firms that perform responsibly are at a competitive disadvantage with unresponsive firms. López et al. (2007) empirically studied the relationship on the short-term and concluded a negative relationship during the first years of CSR engagement.

Some scholars found no relationship between CSP and CFP (Alexander & Buchholz, 1978; McWilliams & Siegel, 2001), which could be caused by the presence of too many influencing variables (Ullman, 1985). Aupperle et al. (1985) also found no relationship for both long-term and short-term financial performance. Furthermore, a mixed relationship was found, showing a U-shape (Barnett & Salomon, 2012; Bowman & Haire, 1975). The varied outcomes have called for more studies on the contingencies influencing the CSP-CFP relationship (Barnett, 2007).

(18)

2.2 Stakeholder theory

The concept of the “stakeholder theory” was used first by Ansoff (1965), who tried to determine the purpose of a firm. This objective was to create harmony between the conflicting interests of the firm’s stakeholders (Roberts, 1992). According to Freeman’s (1984) stakeholder theory, firms view their stakeholders as part of an environment that must be managed to assure revenues, profits, and returns to shareholders. Stakeholders have the ability to positively or negatively affect an organization’s activities (Murray & Vogel, 1997). The main responsibility of a firm is to maximize wealth for all their stakeholders, not only to enhance shareholders’ wealth. A reason for this is that firms are open systems and interact with their external environments (Freeman, 1984). It is worth mentioning stakeholders and shareholders are not synonyms. Shareholders hold stocks of a firm, and therefore, they benefit from a positive financial condition of the firm (Clarkson, 1995). One of the broadest definitions in literature is from Freeman (1984), where stakeholders are defined as “any group or individual who can affect or is affected by the achievement of the firm’s objectives” (p. 46). In this definition, the word “affect” plays a central role, which is explained by two principles. First, the principle of corporate rights entails companies should not contravene the rights of others. Second, according to the principle of corporate effect, companies should take full responsibility for the consequences to stakeholders as a result of their activities (Daub et al., 2004). This theory focuses on various stakeholder groups and the role of management to satisfy these groups. Freeman (1984) made a distinction between internal (e.g., employees) and external stakeholders (e.g., governments). Clarkson (1995) created two different categories of stakeholders, primary and secondary, to integrate the important aspect of interdependency. Primary stakeholders are essential for the firm’s survival, so a company strongly depends on them. This includes public stakeholders, such as government and regulatory bodies. Secondary stakeholders are not directly engaged in transactions with the

(19)

firm; however, they can cause significant damage to a company. Also, the company’s performance will be less positive when ignoring the interests of secondary stakeholders (Bird et al., 2007).

If a firm wants to manage its stakeholders, it has to determine stakeholders’ importance. According to Post et al. (2002), the ability to generate value over time is determined by its relation with critical stakeholders. In addition, Freeman (1983) argued the degree of stakeholder power enhances the probability of meeting the stakeholders’ demands. Building on Freeman’s (1983) thoughts, Ullmann (1985) developed a three-dimensional model, including stakeholder power, firms’ strategy, and economic performance. Those dimensions can result in companies’ increased social responsibility activity and disclosures. Mitchell et al. (1997) focused on the stakeholder salience theory, which suggests managers are more responsive to “salient” stakeholders (Mitchell et al., 1997). Salience is the degree to which managers give priority to competing stakeholder claims. Salient stakeholders are described as stakeholders with power, legitimacy, and urgency. Legitimacy indicates the demands comply with the prevailing norms and beliefs. Mitchell et al. (1997) described legitimacy as “the socially accepted and expected structures of behavior that is often implicitly linked with power to evaluate the nature of relationships in society” (p. 38). A stakeholder has power when it can impose its will on the other party and urgency implies the desire for immediate attention. David et al. (2007) found managers are more likely to settle proposals filed by “salient” shareholders. Governments are also an important stakeholder with power, legitimacy, and urgency. Gago and Antolin (2004) indicated governments have the highest salience in environmental issues compared to other stakeholders. They have the power to permit new laws and punish firms who fail to comply. Also, it is the government’s function to take care of the environment because it is considered a public good.

(20)

As mentioned, companies have to deal with different stakeholders. However, those different stakeholders could have conflicting interests. This causes companies to only satisfy the stakeholders that have the strongest influence to achieve the most important objectives of the firm (Freeman, 1984). This contradicts the essence of the stakeholder theory. Jensen (2001) found an answer to this specification problem with the enlightened stakeholder theory. This theory posits that managers should strive to maximize the long-term value of the firm, which enables them to make trade-offs among different stakeholders. This theory makes an important contribution to the investments in CSR of firms, since CSR initiatives are likely to increase the long-term value of the firm by actively managing all key stakeholder relationships (Bird et al., 2007). Surroca and Tribó (2008) investigated the suggested relationship between CSR and the stakeholder theory. By using a sample of 358 firms, they found that maximization of CSR has a positive correlation to the compliance of stakeholders’ interests.

The stakeholder theory is often used as a cornerstone for the conceptual explanation of the positive relationship between CSP and CFP. A perspective of the stakeholder theory is the instrumental stakeholder theory, which posits that CSR and CFP are related because the satisfaction of stakeholder groups is instrumental for a firm’s financial performance (Clarkson, 1995; Donaldson & Preston, 1995; Mitchell et al., 1997). From this perspective, Jones (1995) showed interactions between a firm and its stakeholders should be based on mutual trust and cooperation to gain competitive sustainable advantage because productivity could increase by enhancing ties with stakeholders that provide important resources for the company. This paper has important implication for the CSP-CFP literature due to the conclusion that firms’ social attributes are related to financial benefits. In addition, Waddock and Gravas (1997) highlighted the way in which CSP yields goodwill from various stakeholders, which results in financial benefits.

(21)

2.3 Industry regulation

Governments act in several ways. To raise money to operate, governments impose tariffs and tax their citizens on their income. An important governmental task is setting regulations to influence the behaviors of industries, firms, and individuals (Viscusi et al., 2002). According to Regdata, there is a general growth of all kinds of regulations in U.S. states on industries in recent years (Allen & Shonnard, 2001) (appendix A). Cane and Tushnet (2003) reasoned that regulations are methods to assure direction in economic activities and influence moral behavior. Two main theories explain the existence of regulations, imposed by governments. Pigou (1938) was the first developer of the Public interest theory, which implies that regulation occurs in markets with failures as demand from the public to correct these inefficiencies. This means public interest underlies the incentive of regulation (Levine & Forrence, 1990). If markets are unregulated, they can exhibit multiple failures, varying from monopoly power to externalities. Price and entry regulations are often established to prevent industries with monopoly power. Positive and negative externalities are either countered by setting taxes or stimulated by subsidies (Levine & Forrence, 1990). Viscusi et al., (2002) divided this into two different types of government regulation: economic and social. Governments can impose economic regulations to control for a certain degree of competitiveness in the industry via, for example, mechanisms of setting prices. Also, governments can establish social regulations to control for industries’ social behavior. This includes health, safety, and environmental quality of employees as well as customers and, in general, the society as a whole. In an ideal world, every industry, firm, and individual would internalize its effects so no negative externalities exist (Viscusi et al., 2002). Related to the public interest theory, the capture theory assumes regulations arise to serve the interests of the industry involved (Stigler, 1971). Regulatory capture occurs when a regulatory agency acts in favor of an industry, instead of benefiting the public. The public interest theory and capture

(22)

theory both imply regulations are determined by supply from governments and demand from interest groups.

Industry regulation related to CSR

Regulations in the context of CSR are often divided into two types: self-regulation (Carroll, 1979) and prescriptive regulation (Newell, 2001). Regulation is not always the responsibility of the government. Many industries create regulatory mechanisms by themselves, to encourage socially responsible behavior, which is known as self-regulation (Campbell, 2006). Industries may be motivated to adopt such standards by external pressure from various stakeholders (King & Lenox, 2000). In addition, initiatives of self-regulations could also arise to reduce corrupt business practices (Porter & Kramer, 2003). According to Sarre (2002), “firm self-regulations are the non-legally, voluntary initiatives to increase the firm’s standard of behavior, ethics and responsibly, and includes the engagement to control its own standard beyond requirements of government regulation” (p. 6). In addition to the operations alongside governments, companies can also complement governmental initiatives (Matten & Moon, 2008). Self-regulation by industries is often related to the state. Industries could use self-regulations as a precaution to avoid state regulatory interventions because industries want to keep control over their standards and protect themselves against scandals. They become increasingly important political actors in the society (Matten & Crane, 2005). In addition, industries’ self-regulation is only powerful when there is strong enforcement and control, which should be reinforced by the state to succeed (Campbell, 2006). According to Ihugba (2012), “Prescriptive regulation refers to mandatory statutory laws or judicial decisions” (p. 108). Governments and companies show irresponsible behavior if an independent party does not control them (Pritchard, 2003). Therefore, regulation is an approach to encourage responsibility and transparency and to recover public trust. There is a difference between “soft

(23)

law” and “hard law” to encourage CSR. Soft law does not always correspond to coercion and punishment (Ayres & Braithwaite, 1992). For example, governments can make a company’s disclosure of CSR impacts mandatory without setting targets. Advocates of prescriptive regulation argue it is inexpensive if companies integrate regulation in their core businesses and incorporate it in current frameworks for CSR activities (Kolk et al., 1999)

As mentioned, risk management is, for companies, a motivation to invest in CSR. To reduce risk, they have to comply with the demands from stakeholders who exert the most pressure (Aguilera et al., 2007). Consequently, regulatory pressure will lead to more investments in CSR activities. Moreover, as discussed, governments are salient stakeholders, so companies are likely to respond to them. Furthermore, the increase in CSR is not only the effect of single regulations. More important is the result of the threat of regulatory pressure, which encourages industries to adopt a proactive CSR approach through initiatives of self-regulation. This initiative will strengthen the relationship with the government and stimulate the government’s goodwill. It could also result in better compliance with current regulations and a prevention of future regulations (Uchida & Ferraro, 2007), which can lower the compliance costs (Christmann, 2000). From this literature, it is expected industry regulation has a positive effect on CSR. As stated, this study tries to investigate the effect of industry regulation on the relationship between CSP and CFP.

Industry regulation on the CSP-CFP relationship

Some studies have argued that strict regulations lead to increased costs for firms. Consequently, there is a trade-off between benefits and costs to the firms (Walley & Whitehead, 1994). On the other hand, studies showed positive outcomes from the emergence of regulations. The views of Porter and Van der Linde (1995) have become very important in the interdisciplinary field of business and the environment (Ambec et al., 2013). They claimed that strict environmental regulations have a broad and positive impact on the innovation rate

(24)

of an industry. This statement is known as the “Porter hypothesis,” which has encouraged debate in environmental regulations literature (Ambec et al., 2013). Environmental regulations entail, for example, technological standards, environmental taxes and tradable emissions permits. Regulations can trigger innovation and offset or partially offset the additional the costs of complying with the regulations, which is possible through both product and process offsets. An example of a product offset is the reduction of product costs by the elimination or replacement of pricey materials, while process offsets could include a more efficient utilization of resources within a process. In addition, a firm can receive a first-mover advantage, which means it took initiative to comply with environmental regulations early (Lieberman & Montgomery, 1988).

From the viewpoint of Porter and Van der Linde (1995), regulations are necessary because firms do not always make optimal choices, which is often a result of market inefficiencies, bounded rationality, incomplete information and organizational barriers. Some opponents of Porter’s Hypothesis have contended that companies always strive for profit maximization. Orlitzky (2011) argued that if additional CSR activities resulted in competitive advantages, firms would more immediately adopt and execute those social and environmental activities. However, Porter and Van der Linde (1995) maintained that the contemporary world does not fit into the Panglossian belief, which very optimistically implies that the world is the “best of all possible worlds.” Therefore, firms do not view all profitable opportunities for innovation in the absence of regulation. For example, pollution indicates the inefficiency of a production system because there is a loss of resources. Porter and Van der Linde (1995) claimed that it is important to emphasize that regulation is needed for companies to commit to invest in these environmental innovations and thereby decrease their environmental impact. Merely making companies aware of these issues without imposing sanctions will not impact them to make optimal innovation investments (Short & Toffel, 2010).

(25)

Porter and Van der Linde (1995) discussed several reasons for the purpose of regulations. First, they allow firms to become aware of their resource inefficiencies and potential improvements. Second, the research within the firm that results from regulations could increase corporate awareness. Regulations also reduce uncertainty when engaging in environmental activities and enhance the incentive to innovate. Finally, regulations level the playing field, which ensures that firms cannot behave opportunistically by avoiding CSR initiatives.

There is much confusion in the literature about Porter’s Hypothesis (Walley & Whitehead, 1994). The hypothesis does not claim that all regulations lead to innovation; rather, in order to receive positive outcomes, regulations need to be designed well. In addition, the hypothesis does not argue that innovation always offset the costs of regulation. Porter and Van der Linde (1995) explained, “We readily admit that innovation cannot always completely offset the cost of compliance, especially in the short term before learning can reduce the cost of innovation-based solutions” (Porter & Van der Linde, 1995, p. 100). This quotation implies that the probability of financial benefits obtained by innovation is higher in the long term.

There are two ways through which regulation can enhance a company’s performance. First, regulations put pressure on firms to optimize processes and become more efficient (Porter & Van der Linde, 1995). Furthermore, governments can stimulate a technology push by setting regulations on the environment, which therefore force companies to become more innovative. For example, regulations create turbulence in the environment and therefore change the status quo, and this turbulence can cause a reduction in the value of existing knowledge. This situation forces companies to create new knowledge to deal with the changed environment (Posen & Levinthal, 2012). Companies could change from a more exploitative approach of knowledge use to a more explorative way in which new knowledge

(26)

is created. This change could be completed by radical innovation, which encompasses higher order innovation that could lead to the diversification of products or operating in new markets. Another way this could be done is through incremental changes that encourage the status quo, such as applying inventions for applications other than their current uses (Tushman & Romanelli (1985). Firms can improve their cost savings and profitability and at the same time reduce their environmental impact, which lead to a win-win outcome (Karagozoglu & Lindell, 2000).

Previous scholars separated Porter’s Hypothesis into different hypotheses to test the theory with empirical evidence. First, the “weak” version states that environmental regulations trigger innovation in a company. Porter’s Hypothesis does not mention the effect of innovation in a company, and therefore it is called “weak.” The “strong” version states that environmental regulation can lead to an increase in the competitiveness of a firm. Ambec et al., (2013) expressed support for the positive link between environmental regulations and innovation. Concerning the “strong” version, the most recent studies found evidence of the increase in productivity regarding the pressure of environmental regulations (Alpay et al., 2002; Berman & Bui, 2001; Segerstrom, 1991). The degree of innovativeness of a company is often indicated by the expenses of the research and development (R&D) team or their number of patents. Although the strength of the relationship varies, most of the previous empirical studies found a positive relationship between environmental regulations and the level of innovation in a company (Arimura et al., 2007; Johnstone et al., 2010; Lanjouw & Moby, 1996; Lanoie et al., 2011; Popp, 2003). Jaffe and Palmer (1997) have also shown a positive link between innovative efforts as a response to regulations and the broader link between R&D expenses and actual invention capabilities of a company (Lanjouw & Moby, 1993; Taylor, 2005).

(27)

described. In general, R&D activities are intended to improve and create new products and processes to meet the changing requirements of the environment. Deloitte (2011, p. 3) defined R&D as “the process of discovering new knowledge about products and services and application of such knowledge to create new and improved products and processes to meet market requirements.” The overall purposes for firms to engage in R&D activities are to produce additional assets and improve both product and process innovation (McWilliams & Siegel, 2000). Prior literature has found that innovation is associated with CSP (Berrone et al., 2007; Hull & Rothenberg, 2008; McWilliams & Siegel, 2000; Surroca et al., 2001).

Several previous studies found R&D to be an important determinant of a company’s financial performance (Griliches 1979). It has been proven that a firm’s high investment in R&D is positively related to an increase in the financial performance of the firm (Hull & Rothenberg, 2008). Furthermore, R&D activities can result in improved brand recognition and an active role in the market (O’Reilly & Tushman, 2004). In addition, several studies found a strong positive relationship between R&D and growth in total factor productivity, which can lead to strong long-term economic and financial performances (Hall & Mairesse, 1995; Lichtenberg & Siegel, 1991; Griliches, 1998; Guerard et al., 1987).

2.4 Hypotheses and conceptual model

Based on the abovementioned literature, a conceptual framework was developed. To analyze the effect of industry regulations on the CSP-CFP link, the following research question is proposed:

What is the moderating effect of industry regulation on the relationship between corporate social performance and corporate financial performance? What is the role of R&D in this relationship?

(28)

is no consensus on the CSP-CFP relationship, based on the stakeholder theory and most existing literature, it is expected that CSP has a positive effect on CFP (Orlitzky et al., 2003, Margolis et al., 2007). Therefore, the overall relationship is predicted to be positive. The following hypotheses is proposed:

H1: Corporate social performance positively affects corporate financial

performance.

This research investigates the moderating role of industry regulations on the CSP-CFP relationship. As stated before, a strong motivation for companies to engage in CSR is the threat of industry regulation, which can result in self-regulation, to predict for future regulation. This implies that all kinds of regulation can affect the engagement of CSR. Existing conceptual literature has shown positive results of the industry regulation role on the CSP-CFP nexus. Scholars argued that environmental regulations could lead to innovation, which results in financial benefits for a company (Porter & Van der Linde). Therefore, it is expected that industry regulation will positively moderate the CSP-CFP link. The second hypothesis is presented as follows:

H2: The level of industry regulation moderates the relationship between corporate

social performance and corporate financial performance. The greater the industry regulation, the greater the positive impact of corporate social performance on corporate financial performance.

The degree of innovativeness of a company is indicated by the R&D expense. Research shows that R&D and CSP are positively correlated. In addition, there is an indication that regulation increases R&D. From hypothesis 2, it is expected that regulation positively influences the CSP-CFP relationship due to an increase in innovation. Hypothesis 3 strengthens hypothesis 2 to provide a more in-depth investigation of the underlying mechanism of R&D in this relationship. The third hypothesis is defined as follows:

(29)

H3: The level of industry regulation moderates the relationship between corporate

social performance and R&D. The stricter the industry regulation, the greater the positive impact of corporate social performance on research and development.

Figure 1 shows the three visualized hypotheses. The two dependent variables, corporate financial performance and R&D, are presented. One independent variable, corporate social performance, is shown. Furthermore, the moderated variable industry regulation is displayed. In addition, three control variables—size, leverage and industry—are included in the analysis.

H1

Figure 1 - Conceptual model

Corporate Social Performance Corporate Financial Performance Research & Development Industry Regulation H3 H2

(30)

III. Methodology

This chapter provides the research design of this study. To test the hypotheses, secondary data was gathered on CSP, CFP, R&D, industry regulation, firm size, firm leverage, and industry. The data was collected using three objective, third-party databases (MSCI, COMPUSTAT, and RegData). In the first section, the research sample will be discussed. In the second section, the measurement of corporate financial performance will be provided, followed by the second dependent variable, R&D. Next, the measurement of the independent variable, corporate social performance, will be discussed. Subsequently, several control variables and their measurements will be described. Finally, the research method will be provided to test the hypotheses.

3.1 Sample

The study’s sample consists of all companies listed in the S&P 500 index that were represented in the year 2013. The index is based on 500 leading U.S. companies, capturing 70% of available U.S. market capitalization. The U.S. stock market is the most developed capital market in the world and, therefore, of interest to this research (Kawaller et al., 1987). Furthermore, the concept of CSR originated in the US, which makes finding the concept and its impact here plausible. The S&P 500 adapts to changes when companies have been merged, acquired, or have gone private. The year 2013 was chosen because it enables full availability of data from COMPUSTAT. Also important is that the year enables the use of the most recent data to calculate industry regulations from the RegData database. For the final sample, data from COMPUSTAT was matched with data from MCSI, and missing data were deleted. This resulted in a sample of 457 companies, belonging to different industries. To identify the industries, the North American Industry Classification System (NAICS) was used. This

(31)

system is preferred over the standard industrial classification code (SIC) due to the specific and sophisticated classification of industries (appendix B).

3.2 Corporate financial performance

As mentioned, there are several ways to measure financial performance to test the relationship between CSP and CFP. Some studies argue there are several motivations behind the different dimensions of CSP, which result in specific consequences for financial performance measures (Brammer & Millington, 2008). Previous studies often use data on financial performance from the databases COMPUSTAT or AMADEUS (Margolis et al., 2007). Orlitzky et al, (2003) divided this measurement into three categories: market-based, accounting-based, and perceptual measures. The last form of measurement uses a survey, which is not suitable for large samples and is time consuming (Punch, 2005). Therefore, most previous studies concerning CSP and CFP used market-based and accounting-based measures. Market-based measures reflect the response of the market to corporate activities. They are calculated by stock performance and price per share. In contrast, accounting-based measures show a firm’s internal efficiency, decision-making capabilities, and managerial performance referring to profit and returns on assets. Previous studies show, in general, CSP is more highly correlated with accounting-based measures than with market-based measures (Margolis et al., 2009; McGuire et al., 1988; Orlitzky et al., 2003; Wu, 2006). According to Wu (2006), an accounting-based measure is a better predictor of CSP than market-based measures because market measurements report a smaller relationship between CSP and CFP. On the other hand, accounting measures are more reliable because market measures unavoidably incorporate market mispricing (McGuire et al., 2004.) Therefore, the focus in this study is on four types of accounting-based measures, return on assets (ROA), return on equity (ROE), return on investment (ROI), and return on sales (ROS), to measure the financial performance. Even

(32)

performance. ROA suggests how efficiently the firm generates income using its assets. This financial performance measure depends on size and scope and could only be compared within the same industry. ROE reflects the profitability of the firm by measuring the investors’ return, so this is important for investors. ROI indicates the valuable results investments had on the firm. Furthermore, ROS implies how efficiently profit is generated against sales (Griffin & Mahon, 1997). The measures were chosen based on the frequency of application in prior studies. More than one measure was also used to overcome the bias, which is caused by using a single measure. Prior studies have inconsistently used one measure to evaluate CFP (Griffin & Mahon, 1997; Davidson & Worrell, 1990; Kedia & Kuntz, 1981). The database COMPUSTAT North America, published by Wharton, was used to calculate the variables for the corporate financial performance. This is an extensive database of firm-level operational and performance data on over 30,000 publicly traded companies (Barnett & Salomon, 2012). It covers the firms included in the S&P 500 for the year 2013.

The measure of financial performance was calculated as follows:

• Return on assets (ROA) = the ROA was calculated as the net income divided by the total assets (Waddock & Graves, 1997).

• Return on equity (ROE) = the ROE was calculated as the net income divided by the common equity.

• Return on sales (ROS) = the ROS was calculated as the net income divided by the total sales (Griffin & Mahon, 1997).

• Return on investment (ROI) = the ROI was calculated as the net income divided by the total invested capital.

(33)

3.3 Research & development

R&D is intended to improve and create new products and processes to meet changing requirements of the environment (McWilliams & Siegel, 2000). In this study, the degree of innovativeness of a company is indicated by the R&D intensity. In the literature, different methods of R&D intensity measurements were applied. Hull and Rothenberg (2008) measured R&D intensity using a three-year average of R&D expenditures. Other research measured R&D intensity as the total R&D expenditures to total assets on a log scale (Berrone at al., 2007). The most common measurement is the total R&D expenditure divided by the total sales (McWilliams & Siegel, 2000; Prior et al., 2008). For this measurement, Bouquet and Deutsch (2007) used R&D reported by firms in their annual financial reports and collected by the S&P Research Insight database. In addition, they divided this number by the firms’ total sales, to adapt the firm size. For this research, the COMPUSTAT database was used to measure R&D for the year 2013. Based on the existing literature, the measure of this variable was calculated as follows:

• Research and development (R&D) = the total R&D expenditure divided by total assets.

3.4 Corporate social performance

Many of the previous studies on the CSP-CFP relationship used the Fortune Corporate Reputation Index and the MSCI social ratings index, which focus large companies (Van der Laan, 2008). To select the sample, MCSI (formerly KLD & GMI) was used to obtain the corporate social performance because it passed several tests of construct validity (Shafman, 1996). Most importantly, MCSI provides the largest dataset that covers many of the underlying dimensions of CSP (Van der Laan, 2008). The included firms in the MCSI index change every year, for example due to acquisitions, expansions of datasets, and lack of data.

(34)

Launched in 1990, the MSCI database has evolved as the most widely used form of measuring CSP. Studies from Waddock and Graves (1997), Sharfman (1996), and Hull and Rothenberg (2008) are prominent MSCI supporting studies. The construct of CSP is multidimensional, and therefore, different aspects of CSP will be considered and all have their own implications for financial performance (Brammer & Millington, 2008).

CSP is subdivided into three dimensions, environmental, social, and governance, which take both internal and external factors into account. MSCI offers ratings on these dimensions for more than 3000 U.S. firms, dating from 1991 (Tang et al., 2012). In addition, MSCI uses a combination of both internal and external sources: surveys, financial statements, articles on companies in the popular press, academic journals, and government reports to score firms (Waddock & Graves, 1997). This database includes seven major dimensions: Corporate Governance, Diversity, Employee Relations, Community, Environment, Human Rights, and Product Quality. These dimensions are also the most commonly researched in existing literature (Tang et al., 2012). Binary scores of 1 or 0 were given to the different indicators of the seven dimensions. Previous studies use different ways to measure the total social score. Waddock and Graves (1997) calculated the total social score by making aggregate CSP scores ranging all dimensions. They created a weighting scheme where companies score differently on each dimension varying from -2 (major concern) to +2 (major strength). In this research, for every category the total score was calculated, subtracting the concerns from the strengths (Kim & Statman, 2012). Afterward, the mean score was calculated for all dimensions, which formed the total CSP score. According to Waddock and Graves (1997), each dimension contributes equally to the firm’s CSR performance. This also applies to this research, where the MSCI social ratings from 2012 were used.

When testing the hypotheses, lagged variables were used. Most studies used measures of CSP and CSF in the same year while studying the effect between them (Schreck, 2011;

(35)

Wang & Bansal, 2012). Some studies argue that investments in CSP could not have an immediate impact on a firm’s financial performance (Busch & Hoffmann, 2011; Waddock & Gravas, 1997). This research also uses lagged variables with a lag-time of one year. Therefore, the dependent variables CSF and R&D are measured in 2013, and the independent variable CSP and control variables are measured in 2012.

3.5 Industry regulation

Baron and Kenny (1986) determined moderation as the impact of a predictor (independent variable) on a criterion (dependent variable) as a function of a third, moderating variable. Data for the moderator variable, industry regulation, was collected through the database RegData, developed by Mercatus Center from the George Mason University (Al-Ubaydli & McLanghilin, 2015). RegData annually quantifies federal regulations by industry and regulatory agency for all federal regulation from 1995 to 2014 (RegData 2.2). Previous studies have analyzed government regulations for decades to address market failure (Pigou, 1938). Those studies are often focused on the causal effect of single regulations, resulting in studies with limitations in scope (Greenstone, 2002). To measure government regulations, researchers counted pages and file sizes published in the Federal Register (Coffey et al., 2012; Mulligan & Shleifer, 2005). However, this method has some limitation. First, not all information in the Federal Registers is dedicated to regulation. Second, there is a risk of counting deregulation as an increase in regulation. Finally, pages of regulatory text can differ in content and importance, which causes a lack of understanding in regulation structure and consequences.

RegData improves on existing measures of regulation in two principal ways. First, it quantifies regulations based on the actual contact of regulatory text. This means the database was built using text analysis to count the number of binding constraints in the wording of

(36)

The use of algorithms to analyze the federal regulatory code makes it much more efficient. To illustrate, the code of federal regulation published in 2010 is 174,454 pages, which would take on average three years of full-time reading. Second, RegData quantifies regulation by industry, which allowsit to create industry-specific measures of regulations over time (Al-Ubaydli & McLanghilin, 2015). This point, in particular, is important for this research. RegData is the first dataset to quantify the level of regulation by 215 different industries. The data enables the examination of the growth of regulation relevant to a particular industry over time or the comparison of growth rates across industries. A limitation of RegData is there is no distinction made between regulation types to investigate the impact of different categories of regulations. Another limitation is the data is not used in highly cited articles published in well-known journals. However, RegData currently is the most sophisticated data on industry regulation.

RegData measures regulation for industries at the two-, three-, and four-digit levels of the North American Industry Classification System (NAICS). This study uses the three-digit level. Three-digit level data consists of more precise subdivisions of the 2-digit classifications. The 2-digit codes of industry regulation were not complete due to missing industries, such as 33-35, which corresponds to the manufacturing industry. This is a result of the NAICS description, which was ambiguous, resulting in search terms that were probably not used in the code of federal regulations (Al-Ubaydli & McLanghilin, 2015). Many of the firms included in the used sample are part of the manufacturing industry. For this reason, the 3-digit codes of the NAICS were used, which includes more complete data (appendix C). The RegData index measures the intensity of regulation of a particular industry and contains a measure of industry targeting. For this research, the growth of industry regulation between 2010 and 2013 was measured.

(37)

3.6 Control variables

It is important to use control variables because, otherwise, the results could be biased (Van

Beurden & Gössling, 2008). Those variables influence the dependent variable but are not part of the relationship. Prior studies have indicated important control variables are firm size, risk, and industry (McWilliams & Siegel, 2000; Ullman, 1985). For this study, risk was not used due to data unavailability. Therefore, firm leverage was used as control variable, which is also common (Ullmann, 1985). Firm size and firm leverage were measured on a continuous scale. The control variable industry is a categorical variable.

Size

Size is a relevant control variable because larger firms are more likely than smaller firms to engage in CSR because of their resource availability (Margolis & Elfenbein, 2007; Waddock & Graves, 1997). Also, larger companies are more vulnerable to negative publicity and face more pressure from their stakeholders (Watts & Zimmerman, 1978). Therefore, large companies need to respond more openly to their demands, which could improve the company’s investment in CSP (Burke et al., 1986). The source of data was the COMPUSTAT database. As a proxy for size, the logarithm of total assets was used (Van der Laan et al., 2008) to avoid the infringement of the normality distribution, which is required for regression analysis.

Firm leverage

Firm leverage is often used as a control variable in the CSP-CFP relationship (Jayanchandran, 2013; Schreck, 2011; Ullmann, 1985). This term refers to the debt or the amount of loaned funds used to finance a firm’s equity. Leverage measures emphasize the long-term and specify to which degree the company is using long-term debt (Brealey et al., 2001). If leverage it used to finance company’s assets, then the company’s risk will increase. In addition, firms with less risk are more likely to engage in CSR (Alexander & Buchholz,

(38)

1978). According to Roberts (1992), with a high the degree of leverage, more importance would management give to creditors instead of other stakeholders. Most studies found a negative relationship between leverage and financial performance (Blanco et al., 2013; Surroca et al., 2010; Waddock & Graves, 1997). The source of data is the COMPUSTAT database. Firm leverage is calculated by long-term debt divided by equity.

Industry

Previous studies show social performance is different among industries (Branco & Rodriques, 2006; Waddock & Graves, 1997). There are many reasons why these differences exist. McWilliams and Siegel (2002) argued the degree of competition in an industry could play a role. Furthermore, industries have different impacts on the environment. Industries with a high negative impact, such as the oil industry, are more likely to engage in CSR (Margolis et al., 2009). External pressure, such as regulation, media, and activism, also plays a role in the degree of CSR activities an industry executes (Campbell, 2007). In this study, the sample consists of firms from 15 different industries, according to the NAICS two-digit rating. To control for the different industries, 14 dummy variables were used.

3.7 Research method

This study applied a Pearson correlation analysis, which shows the relationship between two sets of data (Field, 2009). To test the hypotheses that were developed based on the conceptual model, the multiple regression analysis was used to test whether the dependent variable was affected by the independent variable. In total, five tables represent the five different dependent variables. For every dependent variable, four separated regressions were tested. At the same time, the variables firm size, firm leverage, and industry were controlled for, which could also have an influence on the dependent variable. The statistical software SPSS 24.0 (IBM Corp., 2016) was used in this research. For the moderation analysis, the PROCESS plugin for SPSS was used (Hayes, 2013)

The moderating effect of industry regulation on the relationship between corporate social performance and corporate financial performance : is this relationship explained by R&D?