An analysis of corporate social performance as a mediator of the board diversity – corporate reputation relationship

(1)

June 21^st, 2017

Does Board Diversity Contribute to Corporate Reputation?

An analysis of corporate social performance as a mediator of the board diversity – corporate reputation relationship

Kathrin Kappen Student number: S2966212

K. de Vriezestraat 38, 9741 AG Groningen k.kappen@student.rug.nl

Supervisor: Dr. R.W. De Vries Co-assessor : Dr. D. H. M. Akkermans

Master thesis - MSc, International Business & Management Faculty of Economics and Business, University of Groningen

(2)

2

ABSTRACT

Congruent with requests from both scholars and practitioners to brighten up black box processes, the purpose of this paper is to extend the existing literature by focusing on the following central research question: Does board diversity contribute to corporate reputation through corporate social performance as a mediator? In order to provide an answer, this paper collected data from 109 small to large sized UK companies over a period of three years. Although the mediation results were not found to be significant, this study contributes to research by unraveling new insights about the diversity - corporate social performance relationship. It was found that some diversity measures such as nationality, age, and functional background have a negative effect on corporate social performance. Thus, new insights are given into the importance of not only ensuring a diverse board but paying attention to the appropriate level of diversity. Additionally, light has been shed on considerable measurement challenges insinuat- ing major possible improvements leading to more relevant as well as more robust results in the future. Finally, it is worth mentioning that outside pressures for enhanced firm reputation, increased corporate social performance, and more diverse boards are likely to occur. Hence, this study on the relation between those three concepts will hopefully trigger further empirical research along the steps accentuated in this paper.

Keywords: Corporate social performance, board diversity, corporate governance, corporate reputation

(3)

3

LIST OF FIGURES

Figure 1: Conceptual Model... 24 Figure 2: Conceptual Model with Controls... 31 Figure 3: Diversity over Time ... 38

(6)

6

LIST OF TABLES

Table 1: Shapiro-Wilk Test, Skewness, Kurtosis... 35

Table 2: Descriptive Statistics ... 37

Table 3: Pearson-Zero Order Correlations, Means, Standard Deviations, 2012 ... 40

Table 4: Regression Results (Hayes Model 4) ... 44

Table 5: Mediated Effects of Board Diversity on Firm Reputation ... 45

(7)

7

LIST OF ABBREVIATIONS

CSP Corporate social performance

CR Corporate reputation

CSR Corporate social responsibility CFP Corporate financial performance

BD Board diversity

et al. et alii

n.s. not significant

a.o. among others

SPSS Statistical Package of the Social Sciences

UK United Kingdom

US United States

CEO Chief executive officer ROE Return on equity

(8)

8

1. Introduction

Research suggests that various benefits are related to a positive corporate reputation (a.o. Gatewood, Gowen, & Lautenschlager, 1993; Riordan, Gatewood, & Bill, 1997). Corpo- rate reputation (CR) is defined as “perceptual representation of a company’s past actions and future prospects that describe the firm’s overall appeal to all its key constituents when being compared to other leading rivals” (Fombrun, 1996: 72). A positive reputation improves organizational branding, enhances an organization’s ability to attract new employees, and influences employee retention as the workforce feels its firm is well regarded by the environment (Dowling, 2006). Moreover, a survey by Mercer Investment Consulting has shown that 46%

of all investors look at corporate, environmental, and social governance dimensions before deciding whether to invest in a company (Fombrun, 2006). Furthermore, retaining a good reputation has been argued to enhance a corporation’s chance to maintain an above average performance in the long term (Dowling, 2006). Various factors have been identified that contribute to a good reputation, such as media visibility, firm size, and demonstration of societal concerns. In addition, scholars identified corporate campaigns, customer satisfaction and corporate citizenship campaigns to aid the development of a positive CR (a.o. McCorkindale, 2008; Gardberg & Fombrun, 2006). With enhanced public interest in corporate governance and corporate boards, it is expected that the composition of the board, especially regarding its diversity, is likely to be another relevant factor related to CR (Bear, Rahman, & Post, 2010).

Organizational diversity has also attracted a considerable amount of attention in research with important strands focusing on issues regarding the management of a diverse workforce (Dass & Parker, 1999), the influence of diversity on team performance (Randel &

Jaussi, 2003), and the relation between diversity and firm performance and survival (Fila- totchev & Toms, 2003). Of specific interest in this study is the research on corporate board diversity (Bilimoria, 2000; Singh, Vinniecombe, & Johnson, 2001). Board diversity relates to

(9)

9

“board composition and the varied combination of attributes, characteristics and expertise contributed by individual board members” (Van der Walt & Ingley, 2003: 219). Based on this definition, a distinction between readily observable (demographic) and cognitive dimensions (unobservable) can be made (Milliken & Martins, 1996). Previous research focused mostly on readily demographic aspects and particularly on gender diversity. Consequently, little is known about other dimensions of the construct. Hence, mixed findings and a lack of scholarly understandings are concomitants when investigating diversity. Cross-cultural differences among different nationalities persist and result in varying values and norms across different countries (Hofstede, 1998). Although these differences are commonly known, scant research exists on nationality diversity and its relationship to corporate social performance (a.o. Oxel- heim & Randøy, 2013; Elstélyi & Nisar, 2016). Similarly, research investigating the relationship of functional background diversity and corporate social performance (CSP) is rare, although it is found to have some impact on processes and firm outcomes (Hafsi & Turgut, 2013). Moreover, research on the relationship between age diversity and CSP is inconsistent (Pelled, Eisenhardt & Xin, 1999; Bunderson & Sutcliffe, 2002). Finally, findings with regards to organizational tenure diversity and its connection to CSP are rare and lacking (Hafsi &

Turgut, 2013; Krüger, 2009). Hence, this study answers to the call of previous research to include other dimensions of board diversity (Boulouta, 2013).

Few scholars have attempted to empirically assess the positive effects of board diversity (BD) on board and consequently corporate effectiveness (Vinnicombe & Singh, 2003; van der Walt & Ingley, 2003). Corporate board diversity can be seen as a signal of corporate governance because of its expectations to improve a firm’s effectiveness (Van der Walt & Ingley, 2003), the delivery of beneficial resources to the board (Carver, 2002), or signals of sensitivity towards the concerns and needs of stakeholders (Bilimoria & Wheeler, 2000). However, whether and in which way BD impacts corporate reputation is insufficiently understood and

(10)

10 will thus be addressed in this paper. Likewise, the underlying mechanisms through which board diversity impacts corporate reputation will be investigated. In previous research concerning the direct link between board diversity and company outcomes (e.g. corporate reputation) it is commonly presumed that BD acts as a signal thereby directly impacting outside reviewers (e.g. investors). Nonetheless, it is also possible that board diversity influences CSP, which in turn, strengthens corporate reputation. “Corporate social performance refers to a business organization’s configuration of principles of social responsibility, processes of social responsiveness, and policies, programs and observable outcomes as they relate to the firm’s societal relationships” (Wood, 1991: 693). Due to the increased interest in CSP and the func- tioning mechanisms through which it applies to companies (Kakabadse, 2007) as well as evidence suggesting that board diversity enhances elements of CSP (Post, Rahman, & Rubow, 2011), this study addresses the following research question: Does board diversity contribute to corporate reputation through corporate social performance as a mediator?

By investigating the mediating role of CSP on the board diversity – corporate reputation relationship, this study contributes to and extends existing research on antecedents of CR and the underlying processes that enable the relationship. Besides contributing to the literature, this paper generates valuable implications for practitioners. Through the past decades, an increasing number of governments and organizations around the world have been trying to implement gender quotas or rules regarding the disclosure of diversity practices (Bøhren &

Strøm, 2010; SEC Release 33-9089, 2009). Thus, this study sheds light on possible extensions of previous practices. In the following section a literature review is conducted, in which the present scientific understanding of the concepts and their relationships as well the hypotheses are presented. Thereafter, the methodology of this study is outlined. Then, the results of the analysis are presented and discussed. Lastly, this paper concludes by addressing limitations, contributions, and propositions for future research.

(11)

11

2. Literature Review

The following chapter presents the current state of research in the field of board diversity, corporate social performance, and corporate reputation. Firstly, the concept of corporate reputation is investigated, followed by the introduction of CSP. In section 2.3 the existing literature with regard to the reputation-corporate social performance relationship is being discussed. Section 2.4 demonstrates the relevance of the board of directors in the context of board diversity. Thereafter, the relation between the different dimensions of board diversity (gender, nationality, age, functional background, and organizational tenure diversity) and CSP is examined. Following that, the mediating role of CSP in the relationship between board diversity and corporate reputation will be outlined. Lastly, the deducted hypotheses are exhibit- ed and the corresponding conceptual model is presented.

2.1 Corporate Reputation

The concept of corporate reputation gained popularity with the rise of limited liability of firms. Rules were introduced to govern a firm’s behavior (Dowling, 2016). This led to the establishment of standards with regards to what comprised good and bad firm behavior. Now- adays, the importance of corporate reputation has increased, as it can either facilitate or hinder economic transactions. Good reputation represents a source of trust, whereas bad reputation might signal precaution to other parties. From a scholarly perspective, CR is thought to function as a signal in the markets for stakeholders (Roberts & Dowling, 2002). It can therefore operate as a mechanism to attract and retain customers, employees, and investors. Although, the concept of corporate reputation has been studied in various fields and from different perspectives, no consensus has yet been reached for one agreed definition.

This study makes use of Fombrun’s (1996) corporate reputation definition as it has been widely used across different studies examining this construct (a.o. Gardberg & Fombrun, 2002; Roberts & Dowling, 2002). According to the author, corporate reputation is “a percep-

(12)

12 tual representation of a company’s past actions and future prospects that describe the firm’s overall appeal to all its key constituents when compared to other leading rivals” (Fombrun, 1996: 72). It is hypothesized that a firm’s reputation stands for the public opinion of the company over time and as such points towards a firm’s capability to meet public expectations (Fombrun & Shanley, 1990). A company’s reputation is created through multiple signals, which are emissioned either directly from the organization or indirectly through the stock market or the media. As different stakeholders hold different expectations regarding procedures, actions, and outcomes of the company, reputation is dependent on a match between public expectations and outgoing behavioral signals of a firm (Fombrun & Shanley, 1990).

This stakeholder view on firm reputation implies that different corporate bodies can have different reputations according to different stakeholders (Gardberg & Fombrun, 2002). Thus, a firm can be seen as having a great working environment from the employee perspective while simultaneously having the reputation of a trustworthy product maker among customers. How- ever, there is no guarantee that these different perceptions of a firm’s reputation are con- sistent. It might be the case, that a firm is perceived as positive from one stakeholder group (e.g. employees) while being perceived as negative from investors, another stakeholder group (Bromley, 2002). Hence, it becomes necessary to state the content of the reputation under study as well as to mention the stakeholder group within which that reputation is perceived.

This study relies on the judgment of senior managers in similar companies. This method has been widely used and brings about measures representing the reputation of a firm’s effectiveness (Fryxell & Wang, 1994).

The effects of corporate reputation on organizational outcomes have been studied before, however, little is known about its antecedents (Fombrun & Shanley, 1990). Building on this, this study expects board diversity to affect corporate reputation through corporate social performance. Scholars have found that the composition of the board of directors can operate

(13)

13 as a behavioral signal to stakeholders about the quality of the firm and its compliance to governance mechanisms (Beatty & Ritter, 1986; Albinger & Freeman, 2000). Backhaus, Stone, and Heiner (2002) extend this literature to the concept of diversity. The authors find that board diversity can function as an informational signal employed by, for example investors, to compare different organizations. In addition, they state that diversity issues within a firm mir- ror its working environment and thus impact the firm’s reputational view. Although the relation between overall board diversity and corporate reputation has been subject to research before, whether and how the distinct dimensions of diversity such as gender, nationality, age, functional background and organizational tenure diversity influence corporate reputation re- mains understudied. This study contributes to theory, by accounting for this flaw.

2.2 Corporate Social Performance

Throughout past decades, business has regularly been undergoing serious criticism from the public regarding its responsibilities to society. Therefore, the concept of corporate social performance has increased in popularity for both practitioners and researchers (Zhang, 2012). CSP is derived from the broader and less measurable concept of corporate social responsibility (CSR), which depicts an organization’s duty to take on responsibility not only for profit seeking motives, but also for its business environment and all its stakeholders involved (Beurden & Gosseling, 2008). Carroll (1979: 499) defines CSR as “the responsibility to fully address the entire range of obligations business has to society, including the economic, legal, ethical, and discretionary categories of business performance”. Building on this, CSP as a multilevel construct can be employed to measure the extent to which a firm complies with society’s expectations (Maron, 2006). It assesses an organization’s general stance regarding a range of social issues such as community relations, sustainable investments, and pollution (Graves & Waddock, 1999). For the purpose of this study, the stakeholder view on CSP is incorporated. According to this perspective, CSP is referred to as “a business organization’s

(14)

14 configuration of principles of social responsibility, processes of social responsiveness, and policies, programs and observable outcomes as they relate to the firm’s societal relationships”

(Wood, 1991: 693). According to this view, a socially responsible organization should con- sider all stakeholders in its processes and not constrain its decisions solely to the sharehold- er’s best interests. Overall, CSP can be seen as a construct incorporated into business while being abstracted from company operations to provide a deeper understanding of the relation between society and business (Wood, 1991).

2.3 Corporate Reputation and Corporate Social Performance

As mentioned earlier, corporate social performance represents an organization’s duty to hold responsibility to all stakeholders concerned with its actions. This multidimensional concept accounts for a variety of social, environmental, economic, and governance concerns, all resulting in different actions from companies. These actions were found to lead to enhanced corporate reputation in previous studies. For instance, Gardberg and Fombrun (2006) found that investments in corporate social performance conceive intangible assets that pro- mote firm reputation and assist organizations in expanding their business. Moreover, it is argued that CSR and with-it also CSP help companies to enhance their reputation regarding a broad spectrum of stakeholders such as investors, competitors, suppliers and customers (Branco & Rodrigues, 2006). Beyond that, actions such as donating for charity were found to be signals for a company’s consciousness regarding social issues and therefore enhanced corporate reputation (Fombrun & Shanley, 1990). Additionally, the presentation of social programs to the public enhances a firm’s credibility as well as reputation (Pfau, Haigh, Sims &

Wigley, 2008). In a Canadian study from 2008, Donker, Poff, & Zahir found that the integra- tion of CSP keywords such as respect, trust, and responsibility into codes of ethics increased the company’s market-to-book value. Lastly, it is argued that CSP can serve as a kind of safe-

(15)

15 ty net through which reputational damage can be lessened and risk mitigated (Fombrun &

Gardberg, 2000). From these theoretical points, the following hypothesis is derived:

H1: Corporate Social Performance positively influences Corporate Reputation

2.4 The Board of Directors

The board of director’s functions can be explained by means of two main research streams;

the agency theory and the resource dependence theory. The agency theory states that the interests of managers and shareholders diverge due to the separation of ownership and control and therefore need alignment (Fama & Jensen, 1983). Hence, the board of directors functions as an internal monitoring mechanism that ensures that the interests of shareholders are not at risk by opportunistic behavior from management. For that purpose, the board can call upon activities such as regulating CEO compensation and monitoring its performance while over- seeing the firm’s performance (Hillman & Dalziel, 2003). The resource dependence theory stresses the duty of the board to connect the company to its external environment as well as providing it with critical resources (Boyd, 1990). This perspective emphasizes the role of the board in elaborating an organizations external legitimacy and relations to stakeholders through the use of social networks. Thereby, every director acts as an intermediary between the external environment and the firm (Zhang, 2012). Furthermore, the board functions as an expert committee, thereby providing advice to the management while improving decision- making processes.

Both, agency and resource dependence theory characterize the board as internal governance instrument, influencing the governance of a firm through its direct relations to the external environment and the management of the company (Zalata & Clare, 2016). Previous studies confirm that the board of directors impacts both management and organizational performance (a.o. Dalton & Dalton, 2005; Choi, Choi, Jang, & Park, 2014). Hence, the board also has an impact on CSP decision-making as it is, among others, responsible for appointing,

(16)

16 firing, and advising executives who are in turn responsible for implementing CSP guidelines (Huse, Nielsen & Hagen, 2009). In this context, it is argued that different board characteristics and values should be taken into account. Whether directors exert a rather positive or negative attitude towards CSP and hence pressure management to invest in it, is dependent on their individual beliefs (Zhang, 2012). Thus, board diversity represents an important factor to account for when investigating the impact on CSP.

2.5 Board Diversity and Corporate Social Performance

Scholars commonly define diversity as a characteristic that mirrors the extent to which there are subjective and objective differences among group members (van Knippenberg &

Schippers, 2007; Carter, D’Souza, Simkins & Simpson, 2010). Diversity comprises various dimensions, ranging from sexuality to nationality, from educational background to religious background, and from relation specific skills to task specific skills. Researchers regularly classify these dimensions according to social category diversity and informational/functional diversity. Social category diversity refers to differences in readily observable characteristics (demographics) such as ethnicity, gender, nationality and age, whereas informational diversity means the differences in underlying characteristics that are less obvious such as educational, occupational, functional background, tenure, and industry experience (Milliken & Martins, 1996). Despite the distinction between cognitive and demographic dimensions of diversity in boards, most of the research so far focused on readily perceptible attributes of directors such as ethnicity, gender, or age (Ibrahim & Angelidis, 1991; Bernardi & Threadgill, 2010; Post et al., 2011). Here, the focus is generally laid on the relation between demographic diversity and corporate financial performance (Carter et al., 2010). Notwithstanding some inconclusive results, most research found a positive relation between these two concepts (a.o. van Knippen- berg & Schippers, 2007; Carter, Simkins & Simpson, 2003).

(17)

17 In former years, boards have regularly been seen as a rather homogenous body of elites incorporating similar views on appropriate corporate practices (Useem, 1986). Howev- er, in the last years, corporate governance as well as emerging norms have supported increased demographic diversity in organizational boards, thereby rising up against such an assumption (Westphal & Milton, 2000). For example, the presence of women on boards of directors has continuously increased (Daily & Dalton, 2003). Current research suggests that board diversity has become advisable for at least two reasons: outside and inside firm dynam- ics (Hafsi & Turgut, 2013). First, it is argued that board diversity increases a firm’s capability to recognize and act according to the needs and concerns of all stakeholder groups by generat- ing strategies that account for their differing interests (Luoma & Goodstein, 1999). Addition- ally, institutional investors could constrain firms to integrate diversity routines (Coffrey &

Fryxell, 1991) and customers expect diversity in boards as it is assimilated with the management’s sensitivity towards their concerns and preferences (Luoma & Goodstein, 1999; Hafsi

& Turgut, 2013). Second, diversity increases a firm’s capability to attract the best workforce from the labor market while at the same time avoiding biases of gender, age, or ethnicity (Berman, Wicks, Kotha & Jones, 1999). Also, board diversity is stated to be a source of new and diverse ideas, bearing innovation and thus leading to enhanced corporate performance (Powell, 1999; Zahra & Pearce, 1989). Thus, board diversity stresses the individual’s contributions to the corporation’s strategic management and is expected to improve relations with stakeholders as well as the receptivity to differences and concerns of society (Bear et al., 2010). On the contrary, homogeneity in the board might result in a myopic perspective due to similar ways of thinking. Such a high level of unity might enhance feelings of pressure towards conformity (Miller & del Carmen Triana, 2009).

Regardless of the fact that diversity in boards is widely recognized as valuable and seems to be related to CSP, not much research has been accomplished on the relationship be-

(18)

18 tween board diversity and corporate social performance (Zhang, 2012; Hafsi & Turgut, 2013).

Most studies on CSP and board diversity have investigated gender diversity and how differences in values among women and men influence CSP decision-making (Boulouta, 2013).

Other academics identified further board characteristics such as functional background, age, nationality, and tenure to possibly impact values concerning CSP (Hafsi & Turgut, 2013). As mentioned before, little is known about these concepts of board diversity. Hence, this study will investigate these dimensions and their relation to CSP.

Over the last years, gender diversity has been the most researched demographic regarding its relation to CSP. The reason for this excessive attention is that scholars commonly agree upon the fact that women and men respond to differing beliefs, norms, perspectives, and attitudes (Pelled et al., 1999). Thus, female board members are said to add different experience sets as well as values than their male counterparts. Within the firm, female employees are animated to reach for higher positions (Milliken & Martins, 1996). Outside of the firm, female representation on the board could have social impacts. Gender diversity can act as a signal to the outside of the company’s adherence to social justice, and thus impact the public’s vision of its social performance (Miller & del Carmen Triana, 2009). Likewise, gender diverse boards send signals regarding an organization’s working conditions, hinting towards the incorporation of diversity standards in firm values and norms (Albinger & Freeman, 2000).

Scholars have stated that women have a wider range of social network relations than men (a.o. Ibarra, 1993). These broader networks can facilitate a company’s ability in grasping and responding to its external environment and enhance its interactions with various stakeholders (Beckman & Haunschild, 2002). Since stakeholder relationships represent a major role in CSP, it stands to assume that excelling relations with stakeholders contribute to an enhanced CSP (Clarkson, 1995). Burgess and Tharenou (2002) found that women are more sensitive to corporate social performance than men and in favor of ethical issues. Finally, Bear et al.

(19)

19 (2010) indicated that female representation on corporate boards delivers new perspectives with regard to social strategies and can thus increase CSP. The theoretical points mentioned above contribute therefore to the expectation that gender diversity positively influences corporate social performance. Hence, the following hypothesis has been formulated:

H2: Gender Diversity positively influences Corporate Social Performance

Research on nationality diversity in boards is scant and only a few scholars have investigated its relationship with organizational performance (Oxelheim & Randøy, 2013; Ma- sulis, Wang & Xie, 2012; Elstélyi & Nisar, 2016). Those who did, find that nationality diversity positively influences organizational performance due to the inclusion of differing perspectives and broader networks (Zhang, 2012). Thus, Milliken and Martins (1996), propose, based on relational demography theory, that nationality diversity may be positively linked with greater quality ideas. It is stated that similarities can increase interpersonal appeal and bring about bias in appraisal decisions (Byrne, 1971). So far, no research has been conducted on the relationship between nationality diversity at board level and CSP. However, Hafsi &

Turgut (2013) investigated the impact of ethnicity on CSP. In their study, the authors hypothesized that ethnically diverse boards could produce more innovative ideas and thereby increase performance due to their differing understanding and access to information. They rea- soned that earlier findings depicted that Anglo-Saxon units were rather individualistic in con- trast to ethnic diverse teams who showed traits of collectivism thereby reducing decision bias (Zajac & Westphal, 1996). The authors stated that ethnic diverse board members could impact boards’ decisions on better responding to stakeholders’ needs that might usually be over- looked. The latter could in turn enhance the company’s image and relations with these stakeholders, and lastly its firm outcomes (Hafsi & Turgut, 2013). Nevertheless, the authors could not provide evidence for their assumption that ethnic diverse boards tend to score higher on corporate social performance. This study follows Hafsi’s and Turgut’s (2013) line of reason-

(20)

20 ing although extending it beyond the ethnic separation of Non-Caucasians and Caucasians as cultural differences might persist within these two groups. According to Hofstede (1998), countries vary in their cultural dimensions such as collectivism even though being at close distance from each other. Therefore, this study contends that nationality diversity on board level positively impacts corporate social performance, as different understandings and perspectives can lead to ideas and decisions of higher quality concerning CSP. From these theoretical points, the following hypothesis is drawn:

H3: Nationality Diversity positively influences Corporate Social Performance

Previous results on the relationship between age diversity and organizational outcomes like corporate financial or social performance were found to be inconsistent (Pelled et al., 1999; Bunderson & Sutcliffe, 2002). In most cases though, research found age diversity to positively influence firm outcomes (Kunze, Boehm & Bruch, 2011; Kearney, Gebert & Voel- pel, 2009) as the combination of different levels of life and work experiences from board directors contribute to more creative and founded decision-making (Schneid, Isidor, Steinmetz

& Kabst, 2014). On the one hand, Post et al. (2011) found that the competency for moral rea- soning develops over years. According to the authors, age represents a board member’s experience and is at the same time proof of their maturity in leading the business. Their results suggest that age has a significant impact on philanthropic decision-making and therefore could impact a firm’s social performance. Furthermore, Kets De Vries and Miller (1984) ar- gue that a board member’s generational behavior as well as his/her sensitivity towards society increases along with maturity. On the other hand, recent research suggests that younger gen- erations of board members are often perceived as more conscious and sensitive to ethical and environmental topics. This sensitivity and consciousness comes along with environmentally friendly as well as socially responsible behavior (Bekiroglu, Erdil & Alpkan, 2011), contributing to CSP decision-making. The previously presented theory suggests that the presence of

(21)

21 age diversity in boards is probable to enhance balanced decision-making with regards to a company’s responsibility to a greater group of stakeholders (Aguilera & Jackson, 2010), thereby enhancing CSP. Thus the following hypothesis is being proposed:

H4: Age Diversity positively influences Corporate Social Performance

A board’s set of skills is composed out of the board members’ individual expertise and experience gained throughout their careers. This proficiency includes business experts with know-how of organization strategy, specialists with legal knowledge, and corporate insiders with expertise of company operations (Bear et al., 2010). Although functional background diversity is expected to have some impact on processes and firm outcomes, research investigating its relation to CSP is rare (Hafsi & Turgut, 2013). Sutcliffe and Huber (1998), state that a board member’s industry background significantly influences a director’s sensitivity towards miscellaneous issues. Experience gained within one industry leads to more or less ex- posure to critical social concerns (McDonald & Westphal, 2003). This fact leads to different levels of capability in handling new or different social problems. Even though independent directors may be unfamiliar with a specific company, they might be aware of the industry specific customs due to their prior experience. These board members might therefore have a broader view which is still in consensus with the industry characteristics and social issues.

Likewise, their firm independent view might contribute new insights on practices unfamiliar to the company (Hillman & Dalziel, 2003). Therefore, directors with different functional backgrounds might be more available to changes as well as open to contributions from other stakeholders. Thus, research on top management teams has found expertise diversity to represent a significant asset. For example, Joshi and Roh (2009) show that functional background diversity improves team innovation via the conception of differing solutions. Hence, the more functional background diversity, the higher the aptitude for problem solving and understanding that can empower the board to answer to calls from the business environment and foster

(22)

22 CSP. From these theoretical points it can be derived that functional diversity is apt to increasing a board member’s attention towards a company’s social performance (Hafsi & Turgut, 2013). Hence the following hypothesis is stated:

H5: Functional Background Diversity positively influences Corporate Social Perfor- mance

Organizational tenure diversity means the differences between group members with respect to the quantity of time spent within a firm (Stewart, 2006). Findings with regard to this construct are rare and lacking (Hafsi & Turgut, 2013; Krüger, 2009). Two main theoretical approaches pertain: the management friendly hypothesis and the experience hypothesis.

Following the management friendly perspective, it is argued that as a board member’s organizational tenure increases, his/her familiarity with the firm with regards to management and strategy routines increases along. This may in turn result in directors becoming captive of management and thus restricted in their critical thinking, resulting in lower corporate governance mechanisms and thus in inferior corporate social performance (Krüger, 2009). Hence, directors with shorter tenure, who are less conscious of the company and its managers, may be more apt to raise their voice with regards to critical social issues. However, this advantage can engender problems with regards to defending a non contestable critical position. In con- trast, the experience approach states that highly tenured board members may positively contribute to a firm’s corporate governance. As tenure increases, a director’s willingness to con- front management with rather critical issues might increase as well. Hence, seasoned board members might use their knowledge and experience to diminish the risk of adverse effects for stakeholders, while increasing responsiveness to CSP. In addition, the combination of high and low tenured directors might lead to increased sensitivity with regards to CSP. It is thus expected that organizational tenure diversity creates a positive balance concerning corporate

(23)

23 social performance (Hafsi & Turgut, 2013). Hence, the following hypothesis is derived from the prior theoretical arguments:

H6: Tenure diversity positively influences Corporate Social Performance

2.6 CSP as a Mediator between Board Diversity and Corporate Reputation

So far, research focused on investigating a direct link between board diversity and company outcomes like for example reputation. For that purpose, it has been argued that board diversity acts as a signal thereby directly impacting outside reviewers (e.g. investors).

This study suggests that it is also possible that board diversity influences corporate social performance, which in turn, strengthens corporate reputation. Thus far, this paper has argued that board diversity, namely gender, nationality, age, functional background, and organizational tenure diversity contributes to a firm’s corporate social performance. Moreover, it has been stated that corporate social performance positively influences corporate reputation. Corre- spondingly, it is expected that the relationship between the distinct dimensions of board diversity and corporate reputation is mediated by corporate social performance. Thus, the following hypotheses have been derived:

Hypothesis 7a: The relationship between Gender Diversity and Corporate Reputation is positively mediated by Corporate Social Performance

Hypothesis 7b: The relationship between Nationality Diversity and Corporate Reputa- tion is positively mediated by Corporate Social Performance

Hypothesis 7c: The relationship between Age Diversity and Corporate Reputation is positively mediated by Corporate Social Performance

Hypothesis 7d: The relationship between Functional Background Diversity and Cor- porate Reputation is positively mediated by Corporate Social Perfor- mance

(24)

24 Hypothesis 7e: The relationship between Organizational Tenure Diversity and Corpo-

rate Reputation is positively mediated by Corporate Social Perfor- mance

2.7 Conceptual Model

In order to depict the previously stated hypotheses in a concise manner, a conceptual model has been drawn from the proposed theory.

Tenure Diversity

Corporate Reputation CorporateSocial

Performance

Functional Background Diversity Gender Diversity

Age Diversity Nationality Diversity

Figure 1: Conceptual Model

(25)

25

3.

Research Methodology

The following section introduces the research methodology. First, the methodology used is presented. Thereafter, the sampling and data collection methods are explained and the different variables are introduced. Lastly, the data analysis is outlined and methodological and data limitations are introduced.

3.1 Methodology

In order to develop and design an appropriate research strategy that fully covers all re- quired aspects, various stages of research must be acquainted for. A deductive research approach has been adopted to generate the hypotheses from the theory, which enables a strategy to determine how the selected variables coincide. The mono – method approach applies (Mar- tin, 1990) as the study solely relies on a quantitative methodology. Lastly, data for all independent variables as well as all control variables were gathered in the period of 2012 to 2014, while the dependent variable (CR) was gathered from the 2015 Management Today survey.

The independent variables were time lagged under the presumption that directors need to be in their positions for a certain period of time to have an impact on both corporate social performance and firm reputation. In the following section, more emphasis will be put on the techniques and procedures used in this research.

3.2 Data Collection

Data for the independent variables including gender, nationality, age, functional background, and organizational tenure diversity is gathered from the merging database of annual reports, Orbis database, Reuters, and corporate websites. Data for firm reputation, the dependent variable, is obtained from the ranking of “Britain’s most admired Companies” from the Management Today survey, 2015. Data for CSP, the mediator variable, is gathered from the Thomson Reuters database ASSET 4 (ESG). Data on CEO duality as well as board inde-

(26)

26 pendence, two control variables, are also gathered from the Thomson Reuters database AS- SET 4 (ESG). Lastly, data on board size, firm size, measured in number of employees, as well as return on equity, are obtained from the Orbis database.

3.3 Sample

The population for this study consists of firms chosen from the Management Today 2015 most admired companies List. This survey covers 25 industries including media, construction, banking, chemicals, and beverages resulting in 247 companies. There are several reasons why this study examines these firms. Firstly, these companies represent the 25 main sectors within the UK. Secondly, the sample consists of small, medium, and large firms, which increases its generalizability. Lastly, because of the companies being active in several sectors, the outcomes might also contribute to the generalizability of the results. Only those companies with complete data for all research variables and available in both the Management Today survey and the Thomson Reuters database ASSET 4 (ESG) have been employed resulting in a sample containing 109 companies and 1112, 1098, and 1124 board directors in 2012, 2013, and 2014 respectively. The sample of 2012 consists of 84.4% (N = 938) male board members and 15.6% (N = 174) female board directors. The age of the directors ranges from 33 to 87, with a mean age of 57.13 (SD = 7.91). On average directors remain on the board for 6.39 years (SD = 5.40), ranging from 1 year to 49 years. The board size varies be- tween 6 and 19 board members per company (M = 10.19, SD = 2.83). With 68%, most of the board members are from the United Kingdom, followed by United States citizens with 12.6%

and Irish citizens with 3.4%. Most directors have a functional background in finance (40.1%) and management (36.3%), whereas least members have a background in human resources (0.8%). The technology and engineering background account for 7.6%, consumer, sales, and marketing for 6.1%, consulting for 4.8%, and only 2.4% have a legal background.

(27)

27 In 2013, the sample consists of 1098 board members of which 17.5% (N = 192) are female. The age of the board members ranges from 34 years to 88 years, with a mean age of 57.29 (SD = 7.80). The average tenure of board members is 6.30 (SD = 5.30), ranging from 1 year to a maximum of 50 years. The board size varies between 5 and 14 members across the companies (M = 10.83, SD = 3.04). As in the year before, the commonest nationality on boards is British with 66.8%, followed again by United States Citizens with 12.9%, and Irish citizens with 4%. The finance and management background are most represented in the sample with 40.5% and 35.8%, respectively. Technology and engineering account for 7.6% of the sample, consumer, sales, and marketing for 5.9%, consulting for 4.8%, and only 2.7% have a background in the legal sector and 1.1% in the HR sector. The remaining 1.5% have a com- pletely different background.

The sample of 2014 consists of 1124 board directors of which 79.3% (N = 891) are male and 20.7% (N = 233) are female. The age of the directors ranges from 35 and 89 years, with a mean age of 57.65 years (SD = 7.66). On average directors remain on the board for 6.08 years (SD = 5.28), ranging from 1 year to 51 years. The board size varies between 6 and 18 board members per company (M = 11.01, SD = 2.84). With 66.2%, most of the board members are from the United Kingdom, followed by United States citizens with 13.6% and Irish citizens with 3.5%. Most directors have a functional background in finance (40.1%) and management (34.4%), followed by a technology and engineering background with 8.8%.

Consumer, sales, and marketing account for 6.6%, consulting for 4.2%, and 2.9% of the board members have a legal background. Least members have a background in human resources (1.2%).

3.4 Measures

The following section describes each variable as well as its measurement in more detail. First, corporate reputation as dependent variable is described, followed by the independ-

(28)

28 ent variables of board diversity. Thereafter, corporate social performance as a mediator is further explained. Finally, relevant control variables are introduced.

3.4.1 Dependent variable

Corporate reputation as the dependent variable is measured by means of the ranking from “Britain’s most admired Companies” from Management Today, 2015. The survey uses a comparable methodology to the one employed to obtain the Fortune index, which is often employed to measure US firms’ reputation (Fombrun & Shanley, 1990). For this purpose, the largest organizations of the United Kingdom as well as leading City investment companies’

analysts are asked to evaluate their peers (excluding their own firm) using 12 criteria: quality of management, financial soundness, quality of goods and services, ability to attract retain and develop top talent, value as long-term investment, innovation, quality of marketing, community and environmental responsibility, use of corporate assets, inspirational leadership, company competitiveness, and corporate governance. Each criterion is rated on a scale from zero to ten with zero being poor and ten being excellent. The evaluations of each company are then averaged across the different criteria and participants to finally produce a single reputational score.

3.4.2 Independent variables

Diversity in Boards as independent variable is assessed through the construction of five diversity indices. In order to construct the indices Blau’s index of heterogeneity is used (Blau, 1977). This index is calculated by means of the following formula:

1 − ∑ 𝑝²𝑖

𝑘

𝑖=1

where p is the proportion of directors in a category and i the number of categories. Once measured, Blau’s heterogeneity index ranges from zero to (k-1)/k. Hence, the maximum value

(29)

29 depends on the number of categories present in one dimension. When only one category within a construct is represented on the board, the index equals zero meaning that the board is ho- mogeneous. In turn, when the index is close to the maximum, it indicates a rather diverse group. To obtain the maximum, each category must comprise the same number of cases. The more categories are defined, the higher the possible maximum (Biemann & Kearney, 2010).

This study does not construct an overall diversity index as the particular effect of each dimension provides more in depth insights of the construct.

Gender is the index of diversity measured with two categories: 1 = male and 2 = wom- an. This data is investigated by considering the annual reports of each company where a list with names and titles of the current board members is provided. When the name did not provide guidance with regard to gender, pictures or pronouns in the description are assessed. Na- tionality diversity is assessed according to 28 distinct categories, each representing a different country (see Appendix A). Board members with two nationalities are labeled according to their foreign nationality since it might influence the board. Age diversity is measured according to five categories: less than 40-years old, 40 to 49, 50 to 59, 60 to 69, and 70-years old or more. Functional background diversity is measured according to eight distinct areas: consulting, management, finance, legal, human resources, consumer, marketing and other area (technology, research, etc.). Organizational tenure diversity is assessed in years accomplished on the current board of directors. This construct is measured according to six categories, each category representing a time frame of three years (1=less than 3 years, 2,3,4,5, and 6 = more than 15 years).

3.4.3 Mediator

Corporate social performance as mediator variable is measured using the Thomson Reuters database ASSET 4 (ESG) including over 4000 companies in various industries and across several countries. The dataset evaluates firms along 750 points according to 18 different cate-

(30)

30 gories. The latter are divided along four distinct pillars, namely environmental performance, social performance, economic performance, and corporate governance performance. The data is gathered among others, from company sustainability and annual reports of firms. The data has been collected from 2002 until today and is updated every year. For the purpose of this study, only the environmental as well as the social pillar are considered in order to obtain an adequate measure for corporate social performance. Therefore, both dimensions will be added up and then divided by two, thereby obtaining an equally weighted average for each pillar.

This approach has been used by previous research such as Cheng, Ioannou, and Serafeim (2013) and Lys, Naughton, and Wang (2015). The environmental pillar comprises measures such as product and process innovation, resource reduction, and emissions and emissions reduction. The social pillar constitutes dimensions such as employment quality, employee’s rights, community involvement, employee training, and human rights, health, and safety issues (Thomson Reuters, 2017).

3.4.4 Control variables

Control variables are employed in this study, as other external and internal determi- nants might have an impact on the results. Board size as well as board independence are stressed by scholars to be important variables to control for (Waddock & Graves, 1997; Hafsi

& Turgut, 2013). As stated in the literature, corporate boards represent a hub between the firm and its external environment. Larger boards are said to more efficiently be providing the link between a company and its environment (Pfeffer & Salancik, 1978). Moreover, external pressures for a better representativeness in the board are probable to push companies to extend their board size (Hillman & Keim, 2001). Therefore, board size is measured as a control variable and assessed in terms of the number of directors on the board of the firm. Furthermore, literature suggests, that an increased number of independent directors on the board increases its attention towards corporate social performance (Wang & Coffrey, 1992; Hafsi & Turgut,

(31)

31 2013). Hence, board independence is controlled for by assessing the percentage of directors that are not employees of the corporation. CEO duality is depicted from research as another important control variable. CEO duality refers to the fact that the same individual holds both the chairman as well as the chief executive position within a company. This fact is said to affect board monitoring with regards to effectiveness and is therefore controlled for by means of a dichotomous variable (1 = chairman is also CEO, and 0 = otherwise). In addition, re-

search has argued that corporate financial performance (CFP) is linked to corporate social performance (Weber & Gladstone, 2014; Beurden & Gosseling, 2008). It is stated that companies which are highly profitable will make available more funds for CSP than firms that are not as liquid (Hafsi & Turgut, 2013). Therefore, this variable is controlled for as well. CFP is measured by means of the ROE of the firm. Lastly, firm size in number of employees is oper- ated as a control as diversity is presumably more pertinent in large firms than medium sized

Tenure Diversity

Corporate Reputation CorporateSocial

Performance

Functional Background Diversity Gender Diversity

Age Diversity Nationality Diversity

CEO Duality Board Size Director Independence

Firm Size CFP

Figure 2: Conceptual Model with Controls

(32)

32 or small firms (Wang & Coffey, 1992). Since large companies are supposedly more often the center of attention, they need to be more transparent with regard to their actions including social responsibility (Waddock & Graves, 1997). From this section, the conceptual model from section 2.7 is extended to include the possible influenc ing variables presented above.

3.5 Data Analysis

The software package IBM SPSS is used to complete all statistical analyses needed.

To assess if the assumption of normal distribution of all variables is met, a Shapiro-Wilk test is computed. The data is tested for skewness, kurtosis and heteroscedasticity. The independence of errors as well as multicollinearity are assessed. Mediation is tested by use of Hayes model 4 process (Hayes, 2013) as it has advantages over Baron and Kenny’s (1986) tradition- al method. Hayes approach takes all relationships in the model into consideration including the relation between the predictor and the mediator. Since bootstrapping creates accurate es- timates of the standard error, it has been identified as being superior to regression analysis and is therefore used in this study (Field, 2009).

3.6 Methodological and Data Limitations

Although accounting for differences over time by using time lagged data over a period of three years, it might be advisable to obtain rather longitudinal data in order to study possible changes in the results. The use of a mixed-method approach by, for example, conducting qualitative interviews with board directors would be advisable for future research to examine the research question in more depth. Beyond that, the moderately large sample size and the fact that only UK firms are considered constrain the generalizability of the findings. Further limitations of this study are discussed in more detail in section 5.3.

(33)

33

4. Results

The following section presents the results of the study. First, the quality of the data is assessed. Then, descriptive statistics are presented and preliminary results by means of correlations are disclosed. Thereafter, the main analysis is conducted, testing the previously stated hypotheses. Finally, additional analyses are outlined and robustness tests are conducted.

4.1 Preliminary Analysis

Before testing the previously stated hypotheses presented in the literature review, the quality of the data is assessed. For that purpose, the assumptions of parametric tests are tested (Field, 2009).

4.1.1 Normality

The first assumption that is tested is the one of normality. For that purpose, the Shapiro-Wilk test for normality is computed and skewness and kurtosis are looked at. The results of the statistical analysis can be found in Table 1. The Shapiro-Wilk values for the year 2012 indicate that only three variables, namely board independence (D (109) = .329, p <

.05), firm reputation (D (109) = .201, p < .05), and functional background diversity (D (109)

= .256, p < .05), out of 12 are normally distributed. In 2013, board independence (D (109) = .391, p < .05) and firm reputation (D (109) = .201, p < .05) are normally distributed, and in 2014 board independence (D (109) = .689, p < .05) and firm reputation (D (109) = .201, p <

.05) are normally distributed. However, these scores have limitations as large sample sizes easily depict significant results already from small deviations of normality. Therefore, a significant result is not enough to flaw any statistical procedures applied to the data. In this case, one should take a look at skewness and kurtosis as well as the shape of the distribution (Field, 2009).

(34)

34 4.1.2 Skewness and kurtosis

According to Field (2009), values between +/- 1 are considered acceptable for skewness. Regarding kurtosis, values differ depending on the sample size: for medium sample sizes, a value smaller than +/- 2.58 is regarded as appropriate. When these rules are applied to the data, firm size depicts high levels of positive skewness as well as kurtosis whereas the overall CSP score and the individual environmental and social scores depict negative skewness. Tenure Diversity also shows strong negative skewness as well as high levels of kurtosis in 2012. In 2013, board size and firm size are positively skewed and firm size depicts a score higher than 2.58 for kurtosis. Tenure Diversity is again negatively skewed with high levels of kurtosis. In 2014, firm size is again positively skewed with high kurtosis, whereas both, age and tenure diversity are negatively skewed with high levels of kurtosis.

In order to deal with non-normality and unequal variances, transformation of the data can be applied. To maintain the comparability of the data, all variables need to be transformed the same way even though not all variables are non-normal. In this case, square root transformation is chosen to account for positively skewed data and unequal variances. To be able to apply this transformation to the negatively skewed variables, reverse score transformation is applied to these beforehand (Field, 2009). The exemplary results for the year 2012 of the transformation can be found in Appendix B. Although, skewness and kurtosis are now within the appropriate frame for every variable, the Shapiro-Wilk test for normality depicts now significant results for all variables, thereby confirming non-normality. Transforming data is a highly debated topic (Glass, Peckham, & Sanders, 1972; Games, 1984) as the consequences of data transformation are sometimes worse than analyzing the original data (Field, 2009).

An analysis of corporate social performance as a mediator of the board diversity – corporate reputation relationship

Does Board Diversity Contribute to Corporate Reputation?