Corporate Social Performance and
Corporate Reputation
Azza Wijnen 10206221 July 1st, 2016
MSc Business Administration - Strategy University of Amsterdam
Final Version Master Thesis Supervisor: Dr. Pushpika Vishwanathan
STATEMENT OF ORIGINALITY
This document is written by Azza Wijnen 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.
ABSTRACT
Because prior literature on boardroom diversity and CSP is limited and provides inconsistent results, and because the academic field did not yet address firms’ corporate reputation in the eyes of their financial as well as public stakeholders, this thesis examines the impact of boardroom diversity – in terms of ethnicity and outside director representation – on CSP, and, in turn, of CSP on corporate reputation among different stakeholder groups. Using a sample of the Fortune 500 firms, the findings suggest significantly higher public reputations for firms that have higher institutional CSP strength ratings, and significantly higher public as well as financial reputations for firms that have higher technical CSP strength ratings. Moreover, firms with ethnically diverse boards have significantly higher institutional CSP strength ratings, while firms with a higher proportion of outside directors on their boards have significantly higher institutional as well as technical CSP strength ratings. These increased CSP strength ratings, in turn, lead to significantly higher public reputations for the firms. The findings of this research thus imply that boardroom diversity can positively affect firms’ public reputation through improvement of their CSP. Building on the agency theory, the resource dependence theory, and the signaling theory, it is argued that these increased CSP strength ratings result from the superior monitoring positions, increased cognitive resource pool, and signals of norm adherence that boardroom diversity brings to the firm. The findings provide guidelines for managers on deliberate structuring of the composition of the board to enhance their firm’s CSP and corporate reputation. Also, the findings guide managers in choosing to emphasize certain CSP aspects, depending on the specific stakeholder group they want to address.
TABLE OF CONTENTS
1. INTRODUCTION ... 5
2. THEORETICAL FRAMEWORK ... 10
2.1 CSPANDCORPORATEREPUTATION ... 10
2.2 BOARDOFDIRECTORS ... 14
2.2.1 Functions ... 14
2.2.2 Board Composition ... 16
2.3 BOARDROOMDIVERSITYANDCSP ... 17
2.4 CSPASAMEDIATOROFTHERELATIONSHIPBETWEENBOARDROOM DIVERSITYANDCORPORATEREPUTATION ... 22
3. METHODS ... 25
3.1 SAMPLEANDDATACOLLECTION ... 25
3.2 MEASURES ... 26
3.2.1 Dependent variable: Corporate reputation ... 26
3.2.2 Mediating variable: Corporate social performance ... 27
3.2.3 Independent variable: Boardroom diversity ... 28
3.2.4 Control variables ... 28 3.3 STATISTICALANALYSES ... 29 4. RESULTS ... 31 4.1 DESCRIPTIVESTATISTICS ... 31 4.2 HYPOTHESESTESTING ... 36 4.3 SUPPLEMENTARYANALYSES ... 44 5. DISCUSSION ... 45 5.1 ACADEMICRELEVANCE ... 46 5.2 MANAGERIALIMPLICATIONS ... 51
5.3 LIMITATIONSANDSUGGESTIONSFORFUTURERESEARCH ... 53
6. CONCLUSION ... 55
ACKNOWLEDGEMENT ... 58
REFERENCES ... 59
LIST OF FIGURES FIGURE 1.CONCEPTUAL MODEL. ... 24
LIST OF TABLES TABLE 1.DESCRIPTIVE STATISTICS:MEANS, STANDARD DEVIATIONS, CORRELATIONS BETWEEN VARIABLES ... 33
TABLE 2.BREAKDOWN OF NUMBER OF MINORITY AND OUTSIDE DIRECTORS FOR SAMPLE FIRMS BY INDUSTRY. ... 35
TABLE 3.REGRESSION RESULTS OF RELATIONSHIPS OF CSP WITH PUBLIC REPUTATION. ... 37
TABLE 4.REGRESSION RESULTS OF RELATIONSHIPS OF CSP WITH FINANCIAL REPUTATION ... 37
TABLE 5.REGRESSION RESULTS OF RELATIONSHIPS WITH CSP ... 38
TABLE 6.REGRESSION RESULTS OF RELATIONSHIPS WITH FINANCIAL REPUTATION. ... 42
1. INTRODUCTION
“Lose money for the firm, and I will be understanding; lose a shred of reputation for the firm, and I will be ruthless.”
– W. Buffet (1991) In today’s hypercompetitive industries a positive corporate reputation seems to be critical for the success of firms (Roberts & Dowling, 2002). Corporate reputation is a social perception that reflects stakeholders’ judgment of what they think and how they feel about a firm (Fombrun & Shanley, 1990). A growing body of research argues that positive corporate reputations have strategic value for the firms that possess them (Dierickx & Cool, 1989; Fombrun & Shanley, 1990). It is widely recognized as an intangible asset that can be valuable, rare, and difficult to imitate or substitute, which may lead to sustained superior financial performance (Peteraf, 1993; Barney, 1991; Grant, 1991). Consequently, several studies have confirmed the expected benefits associated with a positive corporate reputation, such as attracting superior employees (Greening & Turban, 2000), increasing attractiveness to customers (Roberts & Dowling, 2002), and enhancing relationships with various stakeholders such as investors, bankers, and suppliers (Branco & Rodrigues, 2006).
Not surprisingly, a considerable amount of research has been devoted to finding factors that can contribute to a positive corporate reputation. According to Fombrun and Shanley (1990), stakeholders do not only rely on firms’ financial performance, but also on signals related to other organizational attributes to form judgments about firms’ reputation. Indeed, the ability to adopt governance practices that are perceived as being desirable is found to be a critical determinant of firms’ reputation (Musteen, Datta, & Kemmerer, 2010). Furthermore, in recent years public scrutiny around boards increased (Hymowitz, 2003) and firms have been pressured to increase diversity in their boards of directors to reflect the
changes in society (Van der Walt & Ingley, 2003; Ramirez, 2003). As such, boardroom diversity has become an important factor of influence for firms’ reputation.
Various studies that evaluate the direct relationship between these constructs usually assume that boardroom diversity can act as a signal (e.g., of firm quality and adherence to social laws and values), which directly influences the assessment of firms’ reputation by its stakeholders (Deutsch & Ross, 2003; Musteen, et al., 2010). However, it is also possible that boardroom diversity enhances certain firm outcomes, which, in turn, impact how the firm’s reputation is assessed. An example is the study on the relationship between gender diversity in the board of directors and corporate social performance (CSP), and, subsequently, of CSP on corporate reputation (Bear, Rahman & Post, 2010). In it’s simplest form, CSP can be seen as a corporation’s broader responsibility towards its numerous stakeholders and the natural environment (Waddock, 2004). Research has shown that good CSP provides information concerning a firm’s ability to manage and deal with important issues, and hence can serve as an effective means for establishing a good overall corporate reputation (Brammer & Pavelin, 2006; Orlitzky, 2011; Wang & Berens, 2015). In line with this research, this study will take CSP into account when examining the boardroom diversity-corporate reputation relationship. Even though it is generally assumed that diversity in the board of directors brings at least differentiated sensitivity to social issues, and therefore likely increases CSP (Bear, et al., 2010; Rao & Tilt, 2015), only a few empirical studies that addressed this relationship have provided some support (Bear, et al., 2010; Post, Rahman, & Rubow, 2011; Zhang, 2012; Zhang, Zhu, & Ding, 2013). Moreover, many existing studies appear to present inconsistent findings, partly due to differences in theoretical perspectives, research methodologies, or performance measures used (Zahra & Pearce, 1989). For example, some studies find that outside directors are positively related to CSP (Johnson & Greening, 1999; Zhang, et al., 2013), while others report negative or no effects (Coffey & Wang, 1998; Hafsi & Turgut,
2012). Using database research, a large sample of firms, and a valid operationalization of CSP (Mattingly & Berman, 2006), this study aims to clarify the relationship between two relevant boardroom diversity variables and CSP.
Accordingly, because CSP has been recognized as a critical factor for the evaluation of firms’ corporate reputation, and because of growing but inconsistent evidence that boardroom diversity is beneficial for firms’ CSP, the present study aims to contribute to the literature by examining the impact of boardroom diversity – in terms of ethnicity and outside director representation – on CSP, and, in turn, the impact of CSP on corporate reputation. The research question of this study will thus be: What is the influence of ethnic diversity and the outside director representation on firms’ CSP, and how, in turn, does CSP influence corporate reputation?
This study contributes to the CSP literature in several ways. First, this paper makes a contribution to the diversity and governance literature by providing a better understanding of how the relationships between boardroom diversity and corporate reputations operate. As previously mentioned, studies analyzing the relationship between diversity and reputation have produced general arguments but only limitedly addressed possible explaining mechanisms of this relationship. Accordingly, of great importance to governance researchers is an understanding of how boardroom diversity positively affects corporate reputation. Therefore, this study extends the literature by providing evidence of CSP as a mediating mechanism.
Moreover, this study makes a theoretical contribution by analysing the relationships between boardroom diversity, CSP, and corporate reputation within the framework of three major theories: the agency theory (Eisenhardt, 1989; Fama & Jensen, 1983), the resource dependence theory (Hillman & Dalziel, 2003), and the signalling theory (Certo, 2003; Deutsch & Ross, 2003). By building on these theories to form hypotheses and subsequently
testing their significance, this research adds to the legitimacy of these theories and provides researchers additional understanding of the nature of the link between boardroom diversity and CSP.
Third, in prior research, corporate reputation has often been measured with the use of Fortune’s MAC list (Walker, 2010; Walker & Dyke, 2014). Because the respondents of Fortune’s MAC List are mainly members of the business community, it is seen as an indicator of reputation among shareholders and investors of the firms (Brown & Perry, 1994; Musteen, et al., 2010; Wang & Berens, 2015). As such, previous studies have measured corporate reputation as financial reputation: the firm’s reputation among financial stakeholders (shareholders and investors). However, as CSP requires firms to manage the interests of other stakeholders as well, corporate reputation in this study will be assessed according to financial stakeholders as well as public stakeholders (Clarkson, 1995). Moreover, research on CSP as a mechanism through which boardroom diversity affects not only firms’ financial but also their public reputation has been virtually absent in previous work on CSP and might potentially open new avenues for future research.
Current study furthermore has important contributions to practice. As firms have direct influence on the composition of their board of directors, this research can be of particular importance to them. Increased knowledge of a board composition beneficial to CSP and corporate reputation can help firms appoint a suitable combination of directors to strengthen their CSP and reputation. Moreover, by examining the impact of different CSP activities on two types of corporate reputation, managers will gain insights on whether they will have to emphasize certain CSP aspects, depending on the specific stakeholder group they want to address.
This paper is structured as follows: first there is a theoretical framework that provides the theoretical background and reasoning for the relationships between the variables
examined in this paper including their corresponding hypotheses. Following that, the methods section will provide an explanation of the methods used to collect and analyze the data. Subsequently the results will be discussed and the paper will conclude with a review of the discussion and conclusion, including contributions, limitations and directions for future research.
2. THEORETICAL FRAMEWORK
In order to evaluate the indirect link between boardroom diversity – in terms of ethnic diversity and outside director representation – and corporate reputation through CSP, the relationship between CSP and corporate reputation will be examined first. Subsequently, the boardroom diversity variables and their relationship with CSP will be addressed. Finally, there will be examined whether CSP mediates the relationship between boardroom diversity and corporate reputation.
2.1 CSP AND CORPORATE REPUTATION
CSP has been used as an umbrella term to embrace corporate social responsibility (CSR), responsiveness, and the whole spectrum of socially beneficial activities of businesses (Carrol, 1991). Because there has been quite some disagreement about the meaning of CSP and about the kinds of activities that encompass it, scholars have experienced difficulty in establishing empirically valid measures of it (Carrol, 1991). As CSP has commonly been defined as “a discretionary allocation of corporate resources toward improving social welfare that serves as a means of enhancing relationships with key stakeholders” (Barnett, 2007, p.801), a frequently used measure focuses on CSP that targets these stakeholders, and thereby distinguishes between CSP toward primary or secondary stakeholders. Nevertheless, as current study aims to examine the effect of different CSP activities on corporate reputation among different stakeholder groups, conceptualizing CSP as targeted towards primary or secondary stakeholders will unduly influence these stakeholders' perceptions, and hence reputations, of the firm. Moreover, Griffin and Mahon (1997) point out that this heterogeneity among the measures of CSP suggests the need to address specific types of CSP. As such, it is more suitable for this study to focus on a distinction in classes of corporate social action that can be targeted towards all stakeholders of the firm.
Therefore, following Mattingly and Berman (2006) and their use of the Kinder, Lyndenberg, Domini (KLD) social ratings data, a distinction is made between two CSP constructs that reflect positive corporate social action patterns. The first construct, institutional strength, is composed of positive actions toward community and diversity issues. It reflects the way a firm deals with its institutional environment, represented by local communities and the public interest groups that are the source of normative expectations to which firms can respond or neglect (Mattingly & Berman, 2006). The second construct, technical strength, is composed of positive actions concerning products, corporate governance, and employees. This construct reflects the source of resource exchanges and concerns stakeholders that are associated with the technical environment of the firm.
Researchers have found that engagement in CSP can promote favorable relationships with key stakeholders by satisfying their concerns and expectations, subsequently leading to a better reputation for the firm (Brammer & Pavelin, 2006; Branco & Rodrigues, 2006). However, since a corporate reputation consists of the subjective perceptions of a specific group of stakeholders, it is likely that different stakeholder groups may have different beliefs or perceptions of how firms will behave in the future (Fombrun & Shanley, 1990; Wang & Berens, 2015). As Lange, Lee, and Dai (2011) state: “An organization’s external observers have varying interests and therefore are attuned to different valued organizational outcomes. The perceptions of an organization’s reputation by particular stakeholder groups such as environmental activists, shareholders, community members, and consumers may vary substantially” (p.158). As such, it seems important to make a distinction in corporate reputation among different stakeholder groups.
Building on Clarkson’s (1995) stakeholder framework, corporate reputation among stakeholders can be assessed according to two essential stakeholder groups. Financial stakeholders, which are the investors and shareholders of the firm, and public stakeholders,
which are the governments and communities in the environment of the firm (Clarkson, 1995). As financial stakeholders care about economic value creation, continuous growth, and proper governance of the firm (Mazzola, Ravasi, & Gabbioneta, 2006), and public stakeholders place great importance on business activities that conform to social norms (Wang & Berens, 2015), it seems likely that the distinct CSP action patterns will have differing effects on a firm’s public or financial reputation.
According to Mattingly and Berman (2006), institutional CSP, encompassing community and diversity strengths, mainly indicates “the extent to which firms respond to normative expectations that they will be inclusive in their employment practices and that they will give back to the communities that sustain them” (p.35). The community strength rating primarily indicates corporate philanthropy towards local community-development initiatives. As most governments are supportive of corporate philanthropy because it helps to lower governmental burdens (Wang & Qian, 2011), and local communities presumably are thankful for firms’ philanthropic support, institutional CSP will likely lead to a positive reputation among public stakeholders. Indeed, Turban and Greening (1997) found that when firms invest in social programs, their reputation tends to improve. As such, these expenditures are in the interest of public stakeholders and in line with social norms, and will therefore likely result in positive perceptions among these stakeholders. Furthermore, the diversity strength rating mainly focuses on positive treatment of minorities, women, disabled, and gay and lesbian stakeholders of the firm, which again is in line with social norms and in the interest of public stakeholders of today (Carter, et al., 2003; Van der Walt & Ingley, 2003). Therefore, the following is expected:
Following these arguments, the impact may be different for financial stakeholders, as their interests vary from those of public stakeholders (Lange, et al., 2011). According to economic theory, shareholders and investors expect firms to focus on maximizing their wealth (Margolis & Walsh, 2003). As such, financial stakeholders may regard institutional CSP as distracting attention away from a firm’s primary wealth creation activities, and thus not appreciate it. In the eyes of financial stakeholders these expenditures might divert valuable resources from a firm’s core business, leading to less favorable perceptions of the firm among the financial stakeholders. Hence:
Hypothesis 1b: Institutional CSP is negatively associated with financial reputation.
The technical CSP construct includes “positive corporate social action toward technical stakeholders, specifically consumers (product), stockholders (governance), and employees” (Mattingly & Berman, 2003, p.36). For instance, firms that actively and strongly manage product quality, have strong governance structures, demonstrate transparency, and encourage employee involvement, score high on technical strength ratings. These product, governance, and employee constructs can be seen as the essential concerns of a firm, as they contribute to the proper functioning of it. According to Carrol (1991), the essential requirements of firms by their stakeholders are to be profitable and obey the law. Managing proper product quality, strong governance structures and employee involvement can be seen as fundamental contributors to firms’ profitability. In addition, demonstrating transparency and managing legal governance of the firm are clear demonstrations of obeying the law. Moreover, as public stakeholders have certain expectations of firm’s products and their actions towards consumers, and financial stakeholders place great importance on proper governance, products and human capital maintenance in order to sustain value creation
(Mazzola, et al., 2006), technical CSP activities seem to address the demands of both stakeholder groups. Consequently, technical CSP is expected to be in the interest of both financial and public stakeholders. Therefore:
Hypothesis 1c: Technical CSP is positively associated with public reputation. Hypothesis 1d: Technical CSP is positively associated with financial reputation.
2.2 BOARD OF DIRECTORS
A board of directors is an elite group of elected or appointed members who jointly oversee and make decisions regarding the activities of a firm, facing complex tasks concerning strategic issue processing. It is the formal link between the shareholders of a firm and the managers in charge of the day to day functioning of the firm (Rivas, 2012). Furthermore, the board of directors “has the power to hire, fire, and compensate senior management teams, and serves to resolve conflicts of interest among decision-makers and residual risk bearers” (Baysinger & Butler, 1985, p.101). As such, they are capable of increasing shareholder value and serve to improve the functioning of the firm (Baysinger & Butler, 1985).
2.2.1 Functions
The board of directors is expected to perform multiple functions in a firm (Rivas, 2012). There are two dominant theoretical perspectives, the agency theory and the resource dependence theory, that explain these functions of the board. According to the first theory, the agency theory, there is a potential conflict of interest between shareholders and managers due to information asymmetries and lack of control (Eisenhardt, 1989). In order to solve these potential conflicts, shareholders install a board of directors as a monitoring and control
shareholders (Hillman & Dalziel, 2003). As such, the board of directors has the responsibility to actively control and monitor top management behaviors in order to safeguard shareholders’ interests.
The second theory, the resource dependence theory, provides the rationale for the board’s function as a provider of resources to the firm. Board capital, which consists of the human and relational capital of directors of the board, will influence the effectiveness of the provision of resources to the firm. According to this view, boards have the duty to effectively utilize their capital to link the firm to its external environment and bring critical resources to the firm (Boyd, 1990; Hillman & Dalziel, 2003).
In addition, during recent years the responsibilities of the board of directors have been extended from traditional shareholder-centric concerns to responsibilities that encompass various stakeholders (Rao & Tilt, 2012). According to Zhang et al. (2013), the board’s role is to identify salient stakeholder claims and subsequently augment firm resources to help address these important claims. As such, the effectiveness of the board of directors in creating value for stakeholders depends on how well the board combines and performs the aforementioned functions. However, because the board of directors is made up of different individuals, each particular director’s associations and experiences with stakeholders impact how the importance of various stakeholder claims are assessed, and how much effort will be spend to secure the resources necessary to address these claims (Zhang, et al., 2013). As such, different compositions of directors will differ in their effectiveness in identifying and managing stakeholder claims (Zhang, et al., 2013). The effectiveness of a board, and thereby different firm outcomes, has therefore often been associated with the composition of the board.
2.2.2 Board Composition
The composition of the board of directors reflects the formation of the board and encompasses various attributes. Board composition can refer to the number and type of directors. Regarding director type, a distinction is made between inside directors and outside directors. Inside (“non-independent”) directors consist of members who are executives of the firm, such as members of the top management team and employees of the firm or its subsidiaries (Pearce & Zahra, 1992). Outside (“independent”) directors on the other hand consist of members who are not executives of the firm and have no personal or professional relationship to the firm, other than their membership on the board (Johnson & Greening, 1999). Outside directors are considered important for the board’s ability to perform its watchdog responsibility by helping protect shareholders from management’s opportunistic behaviors (Jensen & Meckling, 1976). Due to their independence from the firm, they are expected to fulfill a superior monitoring role (Dunn & Sainty, 2009). However, as inside directors have superior knowledge about the firm, outside directors might have a disadvantage compared to them (Dunn & Sainty, 2009). Nevertheless, as outside directors appear less attached to economic performance (Ibrahim & Angelidis, 1995) and more concerned with external stakeholders (Johnson & Greening, 1999), they are likely to be important contributors to a firm’s CSP.
Board composition can furthermore refer to the characteristics of directors. According to Johnson, Schnatterly, and Hill (2013), these can fall in three general categories: demography, human capital, and social capital. Demography refers to director attributes such as age, education, gender, race, and ethnicity, which are all thought to affect board cognition and decision-making. Human capital refers to ‘the skills and experiences that individual directors bring to the decision-making process’ (Johnson, et al., 2013, p.240). Lastly, social
capital refers to directors’ social relationships, such as their ties to other firms, their personal relationships with firm managers, or their social standing (Johnson, et al., 2013).
According to the agency theory, the composition of the board arises from the choices that economic actors make in response to governance issues they face in the firm (Oxelheim, Gregoric, & Randøy, 2013). The resource dependency theory on the other hand views board composition as the outcome of a rational organizational response to the circumstances of the external environment (Hillman, Withers, & Collins, 2009). However, since today’s increasingly integrated word has changed the environment in which firms operate as well as the governance issues they face (Carter, 2003), it seems imperative that the composition of the board will have to change as well.
2.3 BOARDROOM DIVERSITY AND CSP
Boardroom diversity has been broadly defined as “variety in the composition of the board of directors” (Kang, Cheng, & Gray, 2007, p.195) and has been widely identified as a critical element in the ability of the board to impact firm outcomes. However, there is little consensus in the literature about how a board should look like and what kinds people it should be made up of (Johnson, et al., 2013).
One of the least studied demographic attributes of boardroom diversity is directors’ ethnicity (Johnson, et al., 2003). Ethnic diversity refers to heterogeneity in the ethnicity of board members. Because social issues vary by country and geographic locations, directors’ cultural backgrounds and experiences may shape their views on stakeholder issues (Post, et al., 2011). Furthermore, as the contemporary international marketplace requires firms to be culturally sensitive (Carter, et al., 2003), it seems plausible to assume that ethnic diversity is a critical element for the board of directors.
With regard to CSP, ethnic diversity in the board of directors has received very limited attention. There is evidence that all-Anglo groups tend to be individualistic, while
ethnically diverse groups tend to be more collectivistic in their social orientations (Cox, Lobel, & McLeod, 1991), which might result in better servicing of stakeholders’ needs. In fact, a study by Wang and Coffey (1992) shows that having more minority directors on board is positively related to corporate philanthropy, which can be considered a proxy for CSP. However, a prior study by Hafsi and Turgut (2013) finds no effect of ethnic diversity on firm’s CSP. Hence, the few studies that addressed this relationship have not showed any cohesion in whether or not ethnically diverse boards have a positive influence on firm’s CSP.
There are three theories – the agency theory, the resource dependence theory and the signaling theory – that provide underpinning for how diversity in the board of directors might influence firms’ CSP and corporate reputation. As stated before, the agency theory considers the board of directors to be responsible for control over management to protect the interests of stakeholders, and shareholders in particular (Eisenhardt, 1989). For effective exercise of this monitoring function, the board of directors needs an appropriate mix of capabilities and experience to assess management and evaluate the impact of business strategies on their CSP (Hillman & Dalziel, 2003). Having minority (i.e., non-Caucasian) board members can increase the independence of the board, as they do not come from traditional backgrounds (Arfken, Bellar, & Helms, 2004; Carter, et al., 2003; Kang, et al., 2007). The minority directors might raise different questions than directors from more traditional backgrounds, which in turn might lead to a more activist board. Furthermore, as this independence seems to be critical for boards to function in the best interests of shareholders, having an ethnically diverse board is likely to decrease agency problems and thus enhances managements’ governance. Moreover, the nontraditional backgrounds of minority directors might lead to a board that not only monitors management decision-making in terms of shareholders’ interests, but also in terms of other stakeholders. As such, an ethnically diverse board might help broaden the domain of corporate governance beyond shareholders to all stakeholders
and therefore lead to better management decision-making pertaining to CSP (Kang, et al., 2007). Accordingly, ethnic diversity in the board of directors might be a critical element for the ability of firms to implement effective CSP activities.
As noted above, the resource dependence theory considers the board to be responsible for the provision of resources to the firm by utilizing their board capital. An appropriate mix of board capital will therefore enable the board to provide critical resources to the firm, such as legitimacy, advice, and counsel (Hillman & Dalziel, 2003). Having an ethnically diverse board will likely bring unique resources to the firm from sources not easily accessible to an ethno-centered board (Hafsi & Turgut, 2013). An increase in ethnic diversity in the board tends to enrich individual viewpoints, produce unique approaches, and increase the supply of ideas and knowledge available within the board, which can enhance the quality of decision-making and performance (Van Veen, Sahib, & Aangeenbrug, 2013; Watson, et al., 1993; Zahra & Pearce, 1989). Campbell and Minguez-Vera (2008) have furthermore shown that minorities tend to have a wider social network and more ties with otherwise disconnected social networks. Subsequently, these broader social networks can increase the firm’s ability of understanding and reacting to the environment and can facilitate communication with different stakeholder groups (Beckman & Haunschild, 2002; Boyd, 1990). Additionally, minority directors seem to give more attention to social issues and the welfare of diverse stakeholder groups because of their social and economic background (Wang & Coffey, 1992). As such, having ethnically diverse directors provides the board with unique resources and broader social networks, which can contribute to the ability to address various stakeholders’ needs that would normally be overlooked (Hafsi & Turgut, 2013), and can hence be beneficial for decisions regarding effective CSP strategies.
Finally, signaling theory provides additional theoretical underpinning for the relationship between ethnic diversity and CSP. Signaling theory posits that firms can convey
relevant but not readily observable information through observable signals that are purposeful for the other party (Bear, et al., 2010; Certo, 2003; Deutsch & Ross, 2003). Ethnically diverse boards may signal to stakeholders that the firm pays attention to minorities and is culturally diverse, which bolsters the firm’s image and relationships with these stakeholders and indicates the inclusion of diversity in organizational norms and values (Zhang, 2012). As such, these symbols of the firm’s commitment to cultural and social justice can enhance perceptions of the firm’s social performance and hence contribute to the firm’s CSP.
In sum, minority directors, through their nontraditional backgrounds and independent monitoring position, may be better in servicing various stakeholders’ needs. In addition, minority directors may extend firm’s resources through their diverse perspectives, superior social networks, and heightened sensitivity to social issues. As such, their enhanced stakeholder management can contribute to decisions regarding CSP and they can signal diversity norm adherence that enhances the perceptions of their CSP practices. Hence:
Hypothesis 2a: The higher the ethnic diversity on the board, the better the firm’s CSP.
In contrast to ethnic diversity, outside director representation is one of the most thoroughly studied structural attributes of boardroom diversity (Dunn & Sainty, 2009; Ibrahim & Angelidis, 1995; Johnson & Greening, 1999; Post, et al., 2011). As previously mentioned, outside directors are the ‘independent’ directors on the board. The Sarbanes– Oxley Act of 2002 in the United States, which entails legislation that calls for more independence of the members on boards, has greatly increased the ratio of outside to total number of directors in boards (Zhang, et al., 2012). However, there have still remained conflicting views about the benefits of outside directors after the enactment of this act (Hafsi & Turgut, 2013). For instance, Hafsi and Turgut (2012) argue that the presence of outside
directors on boards has become a norm due to the Sarbanes-Oxley Act, and therefore makes no difference on CSP, or other firm outcomes, anymore. Zhang (2012) on the other hand argues that greater presence of outside directors can now only help mitigate bad ratings of CSP, but, by itself, is not adequate enough to enhance positive ratings of CSP.
Nevertheless, according to agency theory, having more outside directors on a board will aid the board in monitoring and controlling opportunistic behaviors of top management (Fama & Jensen, 1983). Since outside directors are independent of the firm and no members of the senior management team, they will not collude with inside directors or management and can thus contribute to minimizing agency problems (Dunn & Sainty, 2009). As such, boards with a higher proportion of outside directors will likely provide superior governance to the firm. Moreover, due to their independence of the firm, outside directors might not only monitor management to conform to shareholders’ interests, but also with regard to interests of other stakeholders of the firm. As such, they might broaden the domain of corporate governance beyond shareholders to all stakeholders and therefore cause better management decision-making pertaining to CSP (Kang, et al., 2007).
Furthermore, proponents of outside director representation reason that outside directors have important differences in CSP orientation compared to inside directors, and that this differed orientation can help the board in managing stakeholder claims by increasing their salience (Zhang, et al., 2013). For instance, Ibrahim and Angelidis (1995) found that outside directors tend to be more concerned with society’s needs and expectations of the various stakeholders than inside directors. In fact, prior studies have shown that firms with a higher proportion of outside directors have a higher level of charitable giving (Wang & Coffey, 1992; Williams, 2003), and more favorable work environments (Johnson & Greening, 1999). As Wang and Dewhirst (1992) state: “outside directors recognize that their
responsibility encompasses more than shareholders and are very conscious about the needs and expectations of the various constituencies of their firms” (p.12).
Additionally, in line with the resource dependence theory, outside directors are shown to have greater knowledge of the outside world, more interaction with external management teams, more diverse educational backgrounds, and broader viewpoints than their inside counterparts (Hafsi & Turgut, 2013; Williams, 2003), which can help the board in augmenting resources to manage their stakeholder claims.
In sum, outside directors likely mitigate agency problems, take the interests of various stakeholders into account, increase the salience of stakeholder claims through their broadened orientation, and provide diverse resources to the firm. As such, it is expected that a greater proportion of outside directors will aid the board in effective management of the firm’s CSP. Therefore:
Hypothesis 2b: The higher the proportion of outside directors on the board, the better the firm’s CSP.
2.4 CSP AS A MEDIATOR OF THE RELATIONSHIP BETWEEN BOARDROOM DIVERSITY AND CORPORATE REPUTATION
Previous research has shown that specific characteristics of board members can send signals to stakeholders to indicate firm quality, subsequently influencing a firm’s status and reputation (Certo, 2003; Spence, 1973). As such, in a marketplace where valuing diversity is the norm (Van der Walt & Ingley, 2003), boardroom diversity signals the firm’s dedication to advancing diversity at all levels, and thereby improves firms’ reputation. Yet, few studies empirically show how this process occurs. Accordingly, current study reasons that CSP can
function as a mediating variable, which transmits the effect of boardroom diversity to corporate reputation.
As reasoned above, increasing boardroom diversity can lead to better monitoring, more cognitive resources, and broader social networks, which positively contributes to firms' CSP. This enhanced CSP, in turn, will lead to an increase in firms’ reputation among its stakeholders. In addition, prior research shows that CSP is positively associated with overall corporate reputation (Brammer & Pavelin, 2006; Orlitzky, 2011; Wang & Berens, 2015). As such, it is expected that boardroom diversity positively affects corporate reputation through improvement of firms’ CSP.
Moreover, Miller and Triana (2009) argue that boardroom diversity can serve as a symbol of the firm’s commitment to social justice, and thereby influences stakeholders’ perception of the firm’s social performance. As such, a diverse board of directors likely conveys a strong commitment by the firm to social performance, which in turn creates a positive perception of their CSP, satisfying stakeholders’ concerns and expectations of the firm, and thereby subsequently leads to a better reputation for the firm. Boardroom diversity can thus serve as an important contributor to firms’ corporate reputation through enhancement of their CSP. Accordingly, it is expected that CSP mediates the relationship between the boardroom diversity variables and both forms of corporate reputation. Since CSP as well as corporate reputation consist of two constructs – institutional and technical CSP, and financial and public reputation – both boardroom diversity variables will have four corresponding mediation hypotheses. Hence:
Ethnic diversity:
Hypothesis 3a: Institutional CSP mediates the relationship between ethnic diversity and financial reputation.
Hypothesis 3b: Institutional CSP mediates the relationship between ethnic diversity and public reputation.
Hypothesis 3c: Technical CSP mediates the relationship between ethnic diversity and financial reputation.
Hypothesis 3d: Technical CSP mediates the relationship between ethnic diversity and public reputation.
Outside directors:
Hypothesis 4a: Institutional CSP mediates the relationship between outside directors and financial reputation.
Hypothesis 4b: Institutional CSP mediates the relationship between outside directors and public reputation.
Hypothesis 4c: Technical CSP mediates the relationship between outside directors and financial reputation.
Hypothesis 4d: Technical CSP mediates the relationship between outside directors and public reputation.
The previously discussed hypotheses result in the conceptual model as shown in Figure 1.
Figure 1. Conceptual model.
The previously discussed hypotheses result in the conceptual model as shown
Boardroom diversity: - Ethnic diversity - Outside director
representation
The previously discussed hypotheses result in the conceptual model as
The previously discussed hypotheses result in the conceptual model as shown in
CSP: Corporate reputation:
- Institutional CSP
3.1 SAMPLE AND DATA COLLECTION
This study uses a database research design to examine the effect of boardroom diversity on CSP, and, in turn, of CSP on corporate reputation. The initial sample included firms that were listed on the Fortune 500 list for the year 2015. The Fortune 500 is an annual list compiled and published by Fortune magazine that ranks 500 of the largest U.S. firms by total revenue for their respective fiscal years. Since the nationality of a firm matters to its CSP (Gjølberg, 2009), the Fortune 500 list was used in order to control for possible nationality differences as it only includes firms from the U.S. This however also creates a limitation of this study, as the results cannot be generalized to other countries. The resulting sample consists of the firms for which adequate information was available regarding financial reputation, public reputation, CSP, and their board of directors.
The firms within this sample belong to all sorts of industries. The use of this sample therefore offers great insight in boardroom diversity, CSP, and corporate reputation of firms in diverse industries. The firms in the remaining sample are situated in five different industries. Most firms operate in the manufacturing industry (37.6%), followed by the services industry (20.0%), the wholesale and retail industry (16.8%), the finance industry (13.6%), and the transportation industry (12.0%).
All the data was gathered in an Excel spreadsheet. After the removal of firms for which data was not available from the databases, the merged final data set existed of data for 125 firms. The Fortune and Pulse scores represent the financial and public reputation of firms in 2015. The KLD scores for institutional CSP and technical CSP refer to firm’s CSP in 2013. Furthermore, director data refers to directors on board in 2013, from the same period the CSP ratings were gathered. Data for the independent variables was lagged to ensure that
initiated CSP activities by the board would have sufficient time to be installed, noticed, and impact public and financial stakeholders.
In contrast to previous similar research, which used CSP and corporate reputation data that predates the global financial crisis (from 2007 or older) (Bear, et al., 2010; Post, et al., 2011; Wang & Berens, 2015; Zhang, 2012; Zhang, et al., 2013), this study’s data fully postdates the global financial crisis and might therefore provide interesting new insights to the literature.
3.2 MEASURES
3.2.1 Dependent variable: Corporate reputation
Ratings for corporate reputation among financial stakeholders were drawn from the Fortune’s 2016 Most Admired Companies (MAC) List, which reports the results of 2015 for 340 firms in 54 industries. To create the list, Fortune asks executives, directors, Wall Street investors and financial analysts to rate the firms in their own industry on nine criteria including innovation, people management, use of corporate assets, social responsibility, quality of management, financial soundness, long-term investment value, quality of products/services, and global competitiveness. These ratings are used to produce a list of the top-ranked firms in each industry and an overall score of total firm’s reputation from 1 to 10. Because the respondents of Fortune’s MAC List are mainly shareholders and investors from the business community, the MAC rating is seen as an appropriate indicator of financial reputation in this study (Brown & Perry, 1994; Musteen, et al., 2010; Wang & Berens, 2015). There has been a considerable amount of criticism on the Fortune MAC rating that stated that financial performance is a significant factor in the reputation rating (Brown & Perry, 1995; Fombrun & Shanley, 1990; Fryxell & Wang, 1994). However, a recent study by Flanagan, O’Shaughnessy and Palmer (2011) re-assessed the relationship between Fortune’s
variables on reputation is much weaker today than it was in the past” (p.13). Nevertheless, measures of financial performance were included as controls in the study. Furthermore, as MAC continues to be the most widely used measure of corporate reputation (Walker, 2010; Walker & Dyke, 2014), the use of MAC ratings in this study seems to be a valid measure of financial reputation.
Furthermore, for corporate reputation among the general public, the Pulse Score provided by the Reputation Institute (RI) was used. The RI conducts an online survey among 60,000 general public respondents that are familiar with the firms to measure the corporate reputation of firms in 29 countries. The Pulse Score is the mean of four underlying reputation dimensions: Trust, Feeling, Esteem, and Admire and Respect, which range from 0 to 100, with 100 as the most favorable rating. As Ponzi, Fombrun, and Gardberg (2011) have analyzed the Pulse scale and demonstrated its reliability and validity, the use of the Pulse Score is seen as a valid measure of corporate reputation among public stakeholders.
3.2.2 Mediating variable: Corporate social performance
Ratings for CSP were drawn from the Kinder, Lydenberg and Domini (KLD) database. The KLD ratings are based on ratings of seven broad categories: community, governance, diversity, employee relations, environment, human rights, and product, which are further categorized into either strengths or concerns. According to Mattingly and Berman (2006), strength and concern ratings are “qualitatively distinct types of social action” (p.38) and “should not be combined in future research” (p.20). Therefore, the focus is on the social performance strength constructs developed by them: institutional strength and technical strength. The institutional CSP strength construct is composed of community and diversity strengths, and the technical CSP strength construct of product, corporate governance, and employee relations’ strengths. KLD assigns a rating of 0, 1, or 2 for each strength area, and the overall constructs are composed of adding the strength area ratings.
The KLD ratings are “among the oldest and most influential, and, by far, the most widely analyzed by academics” (Chatterji, Levine, & Toffel, 2009, p. 125), and therefore a valid measure of CSP in this study. Additionaly, KLD ratings are shown to have the advantage of not being highly correlated with Fortune’s MAC ratings (Szwajkowski & Figlewicz, 1999), providing further reinforcement for the use of these ratings in this study.
3.2.3 Independent variable: Boardroom diversity
Data on the independent variables ethnic diversity and outside director representation were obtained from the ISS database, which is formerly known as RiskMetrics. The ISS database provides detailed data on individual directors, of which ethnicity and board affiliation data were obtained. Prior research finds ISS database a reliable source for academic research (Gompers, Ishii, & Metrick, 2003). Ethnic diversity was operationalized as the proportion of non-Caucasian directors, which was calculated by dividing the number of non-Caucasian directors on the board by total board size. The same was done for the proportion of outside directors.
3.2.4 Control variables
A number of control variables are included that have been shown in prior research to influence CSP or corporate reputation. CEO duality is defined as the presence of a single individual holding the dual role of Chairman and Chief Executive Officer (CEO). As research has shown that CEO duality is less common in socially responsible firms compared to non-socially responsibly firms (Webb, 2004), CEO duality is included as a control variable. A dummy variable was created with data from the ISS database (0 = CEO is not chairman of the board, 1 = CEO is chairman of the board).
Furthermore, financial performance was included to control for the possible impact of financial performance on reputation (Brown & Perry, 1994) and on CSP (Waddock &
Graves, 1997). Using Compustat data, financial performance was measured as the firm’s return on assets (RoA), by dividing the firms’ net income by their total assets.
In addition, prior research suggests that larger firms have to meet higher expectations of the public by demonstrating more socially responsible behaviors than smaller firms (Waddock & Graves, 1997). Moreover, Fombrun and Shanley (1990) argue that firm size is an indicator of firms’ access to resources and increases visibility, which in turn can impact corporate reputation. Consequently, firm size was controlled for, with data for firm size obtained from Compustat and measured as the two-year average of number of employees (in thousands). Additionally, prior research argues that board size might impact the effectiveness of board monitoring (Post, et al., 2011). However, since board size is a marker of the proportion of non-Caucasian directors (ethnic diversity) and the proportion of outside directors, board size is not used as a direct control in regression analyses, as the effects would be substantially overlapping with the effects of the boardroom diversity variables.
Finally, as firms in a particular industry may be more engaged in CSP than others simply due to the industry’s nature (Boutin-Defresne & Savaria, 2004), accounting for industry differences seems necessary. In addition, as Flanagan et al. (2011) suggest the use of industry dummy variables when modeling corporate reputation, industry dummies were included to account for these effects. The five industries included in the sample are defined as dummy variables that equal one if a firm belongs to the industry and zero otherwise.
3.3 STATISTICAL ANALYSES
Hierarchical multiple regression analyses were used to test the predicted relationships. Multiple regressions were performed in order to test the mediating effect of both forms of CSP on the relationships between the boardroom diversity variables and the two forms of corporate reputation. Since the variables have different measurement scales, they were first standardized into z-scores for easier comparison and interpretation.
To examine whether CSP mediates the relationship between boardroom diversity and corporate reputation first the relationship between the independent variable (e.g., ethnic diversity) and the predicted mediator (e.g., institutional CSP) and the relationship between the predicted mediator and the dependent variable (e.g., public reputation) are examined, as these relationships have to be present in order for mediation.
Subsequently, following the procedure by Baron and Kenny (1986), regression analyses are conducted to test mediation. First, the mediator is regressed on the independent variable; second, the dependent variable is regressed on the independent variable; third, the dependent variable is regressed on the mediator and on both the independent variable and the mediator. To confirm mediation, “the effect of the independent variable on the dependent variable must be less in the third step than in the second step” (Baron & Kenny, 1986, p.1177).
Furthermore, as a first step in all hierarchical multiple regressions the control variables – the industry dummies, CEO duality, RoA, and firm size – were entered so that the shared variability of these variables with the predictive variable can be controlled. As such, the observed effects of the predictor variables on the dependent variables are independent of the effect of these control variables. With regard to the industry dummies, the comparison dummy (manufacturing industry) was left out of the analysis in order to be able to control for dummy variables with more than two categories and to not fall into the dummy variable trap.
4. RESULTS
4.1 DESCRIPTIVE STATISTICS
Table 1 provides descriptive statistics for the sample firms. The average size (number of employees) of the firms in the sample is 99.261 thousand. In almost 69% of the firms, the chairman of the board is also the CEO. Furthermore, the average board is made-up of 11.344 directors, of whom, on average, 1.911 are minorities and 9.737 are outsiders. When looking at the descriptive values concerning the corporate reputation of firms it shows that the mean public reputation of the firms in the sample is 67.552 out of 100 with a standard deviation of 7.697 and their mean financial reputation is 6.794 out of 10 with a standard deviation of 0.752. This implicates that the firms in the sample have similar average reputations among their public and financial stakeholders. Furthermore, the average firm in the sample has an institutional strength rating of 2.460 and a technical strength rating of 3.460.
The correlations of the transformed z-scores of the merged data set are also presented in Table 1. It is interesting to note that some of the control variables are significantly related to some of the predictor variables in the model. Table 1 shows that CEO duality is significantly related to the proportion of outside directors (r = 0.250, p < 0.01), which likely indicates that the more a CEO holds the dual role of CEO and Chairman of the board, the more the board will be made up of outside directors. Furthermore, firm size and ethnic diversity are also significantly related (r = 0.184, p < 0.01). This likely indicates that the more employees a firm holds, the more ethnically diverse its board of directors will be. Nevertheless, as these correlations do not imply causation, the relationships can also be the other way around.
The variables are checked for multicollinearity by assessing the correlations between all the predictor variables. Multicollinearity can exist when there are high levels of correlation between predictor variables (Field, 2009). Table 1 shows that the two
operationalizations of CSP, institutional CSP and technical CSP, are significantly related (r = 0.325; p < 0.01), but not very highly (not above 0.7), which confirms that the concepts relate to the same construct. Nevertheless, their relation generates no issues as they are used in different models. Furthermore, even though using the correlation matrix to check for multicollinearity is a good start, it does not take into account indirect correlations and can therefore miss the more subtle forms. Accordingly, the tolerance levels and the variance inflation factors (VIF) are calculated. The tolerance level indicates how much of the variability of the predictor variables is not explained by other predictor variables in the model, and the VIF indicates the degree to which each predictor variable is linearly related to one another (Field, 2009). A tolerance level of less than 0.20 or a VIF value of 10 and above indicates a multicollinearity problem (Field, 2009). All the tolerance levels are around 0.9 and the VIF values associated with the predictors ranged between 1.148 and 1.539, indicating that multicollinearity is not a problem in this dataset.
Table 1. Descriptive statistics: Means, standard deviations, and correlations between variables𝒂𝒂. Variables Mean SD 1 2 3 4 5 6 7 8 9 1. Ethnic diversity 1.911 1.220 1 2. Outside directors 9.737 2.146 0.209* 1 3. Institutional CSP 2.460 1.495 0.399** 0.448** 1 4. Technical CSP 3.460 2.116 0.090 0.394** 0.325** 1 5. Public reputation 67.552 7.697 0.318** 0.226* 0.402** 0.317** 1 6. Financial reputation 6.794 0.752 0.030 0.148 0.056 0.352** 0.131 1 7. RoA 0.069 0.054 0.031 -0.156 -0.161 -0.060 0.011 0.175 1 8. CEO duality 0.688 0.465 0.005 0.250** 0.163 -0.043 0.075 0.054 0.044 1 9. Firm size 99.261 108.162 0.184* -0.082 0.056 -0.034 0.085 0.026 0.139 -0.113 1
𝑎𝑎Means and standard deviations of raw data, transformed data z-scores used for correlations and regression.
** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed)
Table 2 provides a breakdown of the minority and outside directors by industry. Panel A reports the distribution of minority directors and Panel B the distribution of outside directors. As can be seen in Panel A, 12% of the firms do not have a single minority director in their board. Most (28.8%) of the sample firms only have one minority director on their board of directors, followed by 27.2% of sample firms who have two minority directors on their board of directors. Around 22.4% of the firms have three minority directors, with only 12 firms (9.6%) having four or more. Firms in the manufacturing industry are most likely to have minorities on their board, as fifteen firms have three or more minorities on their boards.
Furthermore, as reported in Panel B, the lowest number of outside directors on the board in this sample is five. Most of the sample firms (35.2%) have at least nine or ten outside directors on their board, which shows a substantial higher amount in contrast to minority directors. Moreover, as with minorities, manufacturing firms in this sample are most likely to have a larger number of outside directors on their board.
Regression results for the hypotheses are presented in Table 3 - 7. Table 3 provides the results of the regression analyses of the effects of both CSP constructs on financial reputation, and Table 4 on public reputation. Table 5 provides the results of the regression analyses of the effect of the boardroom diversity variables on each of the two mediator variables (institutional and technical CSP). Table 6 and 7 provide the regression results for the mediation effects, all after controlling for the industry dummies, CEO duality, RoA, and firm size. Further analysis will investigate the support for each hypothesis.
Table 2. Breakdown of number of minority and outside directors for sample firms by industry. Panel A: The number and % of minority directors on boards by industry
Number of minority directors
0 1 2 3 4+ Industry Num. of firms % of firms Num. of firms % of firms Num. of firms % of firms Num. of firms % of firms Num. of firms % of firms 1. Manufacturing 5 10.6% 15 31.9% 12 35.5% 10 21.3% 5 10.6% 2. Transportation 1 6.7% 3 20.0% 5 33.3% 4 26.7% 2 13.3%
3. Wholesale & retail trade 3 14.3% 5 23.8% 4 19.0% 7 33.3% 2 9.5%
4. Financial 0 0.0% 6 35.3% 6 35.3% 3 17.6% 2 11.8%
5. Services 6 24.0% 7 28.0% 7 28.0% 4 16.0% 1 4.0%
Total 15 12.0% 36 28.8% 34 27.2% 28 22.4% 12 9.6%
Panel B: The number and % of outside directors on boards by industry
Number of outside directors
5-6 7-8 9-10 11-12 13+ Industry Num. of firms % of firms Num. of firms % of firms Num. of firms % of firms Num. of firms % of firms Num. of firms % of firms 1. Manufacturing 1 2.1% 9 19.2% 19 40.3% 15 31.9% 3 6.3% 2. Transportation 0 0.0% 2 13.4% 5 33.4% 7 46.7% 1 6.7%
3. Wholesale & retail trade 3 14.4% 6 28.7% 5 23.8% 7 33.4% 0 0.0%
4. Financial 2 11.8% 1 5.9% 8 47.1% 3 17.7% 3 17.7%
5. Services 4 16.0% 5 20.0% 7 28.0% 8 32.0% 1 4.0%
4.2 HYPOTHESES TESTING
Hypothesis 1a predicts that institutional CSP strength ratings are positively related to public reputation. After controlling for all industry dummies, CEO duality, RoA, and firm size, regression results show that institutional CSP strength ratings are positively and significantly related to public reputation (b = 0.395; p < 0.01) (Table 3, Model 2). Therefore hypothesis 1a is supported.
Hypothesis 1b predicts that institutional CSP strength ratings are negatively related to financial reputation. Regression results show that institutional CSP strength ratings are not related to financial reputation (b = 0.037; p = 0.696) (Table 4, Model 2). Therefore hypothesis 1b is not supported.
Hypothesis 1c predicts that technical CSP strength ratings are positively related to public reputation. The regression results show that technical CSP strength ratings are positively and significantly related to public reputation (b = 0.307; p < 0.01) (Table 3, Model 3). Hypothesis 1c is therefore supported.
Furthermore, in line with hypothesis 1d, technical CSP strength ratings are also positively and significantly related to financial reputation (b = 0.324; p < 0.01) (Table 4, Model 3). Hypothesis 1d is therefore supported.
Together with the control variables, 18.5% of firms’ public reputation is explained by their institutional CSP strength ratings and 14.0% is explained by their technical CSP strength ratings. For firms’ financial reputation, 12.8% is explained by their institutional CSP strength ratings and 22.7% by their technical CSP strength ratings. As such, firms’ CSP ratings explain a slightly bigger amount of variance (35.5%) in their financial reputation compared to their public reputation (32.5%).
Table 3. Regression results of relationships of CSP with public reputation𝒂𝒂. Variables𝒃𝒃 Model 1 Control Model 2 Main effect Model 3 Main effect CEO duality 0.056 (0.592) 0.007 (0.082) 0.071 (0.790) RoA -0.022 (-0.226) 0.056 (0.609) -0.013 (-0.140) Firm size 0.092 (1.002) 0.057 (0.662) 0.102 (1.162) Institutional CSP 0.395 (4.4381)** Technical CSP 0.307 (3.488)** R ² 0.050 0.185 0.140 F 0.874 3.282** 2.168*
𝑎𝑎Standardized coefficients are reported, with t-values in parentheses; n=125. 𝒃𝒃Industry dummies are included in the models.
* p < 0.05, ** p < 0.01.
Table 4. Regression results of relationships of CSP with financial reputation𝒂𝒂.
Variables𝒃𝒃 Model 1 Control Model 2 Main effect Model 3 Main effect CEO duality 0.037 (0.408) 0.032 (0.354) 0.053 (0.621) RoA 0.192 (2.072)* 0.199 (2.101)* 0.202 (2.229)* Firm size 0.007 (0.077) 0.003 (0.039) 0.017 (0.208) Institutional CSP 0.037 (0.392) Technical CSP 0.324 (3.883)** R ² 0.127 0.128 0.227 F 2.426* 2.127 4.263**
𝑎𝑎Standardized coefficients are reported, with t-values in parentheses; n=125. 𝒃𝒃Industry dummies are included in the models.
* p < 0.05, ** p < 0.01.
Table 5 provides the results of the regression analyses of the effects of the independent variables on institutional CSP (Model 2) and technical CSP (Model 4). Hypothesis 2a predicts that ethnic diversity is positively related to CSP strength ratings. The coefficient for ethnic diversity is positive and statistically significant for institutional CSP strength ratings (b = 0.392; p < 0.01), but not for technical CSP strength ratings (b = 0.113; p = 0.227). Accordingly, this hypothesis is not supported.
Hypothesis 2b predicts that the proportion of outside directors is positively related to CSP strength ratings. The coefficient for outside directors is positive and statistically
significant for institutional CSP strength ratings (b = 0.395; p < 0.01) and for technical CSP strength ratings (b = 0.413; p < 0.01). Accordingly, hypothesis 2b is supported.
Furthermore, the models in Table 5 show that together with the control variables, 35.5% of firms’ institutional CSP strength ratings is explained by the proportion of outside and minority directors on their board, while only 19.0% of firm’s technical CSP strength ratings is explained by the proportion of outside and minority directors on their board.
Table 5. Regression results of relationships with CSP𝑎𝑎.
Variables𝒃𝒃 Institutional CSP Technical CSP Model 1 Control Model 2 Main effects Model 3 Control Model 4 Main effects CEO duality 0.123 (1.366) 0.051 (0.636) -0.050 (-0.531) -0.144 (-1.589) RoA -0.197 (-2.132) -0.148 (-1.815) -0.029 (-0.298) 0.039 (0.427) Firm size 0.089 (1.018) 0.044 (0.564) -0.033 (-0.353) -0.021 (-0.244) Ethnic diversity 0.392 (4.797)** 0.113 (1.214) Outside directors 0.395 (4.610)** 0.413 (4.583)** R ² 0.134 0.355 0.044 0.190 F 2.581* 7.030** 0.762 3.005**
𝑎𝑎Standardized coefficients are reported, with t-values in parentheses; n=125. 𝒃𝒃Industry dummies are included in the models.
* p < 0.05, ** p < 0.01.
Hypothesis 3a predicts that institutional CSP strength ratings mediate the relationship between ethnic diversity and financial reputation. The first two steps of the analysis show that ethnic diversity is positively and significantly related to institutional CSP strength ratings (b = 0.392; p < 0.01) (Table 5, Model 2) but not significantly related to financial reputation (b = 0.042; p = 0.642) (Table 6, Model 3). Consequently, hypothesis 3a is not supported. The relationship between ethnic diversity and financial reputation is not mediated by institutional CSP strength ratings.
Hypothesis 3b predicts that institutional CSP strength ratings mediate the relationship between ethnic diversity and public reputation. The first step of the analysis, the regression of the mediator (institutional CSP) on the independent variable (ethnic diversity), shows that ethnic diversity is positively and significantly related to institutional CSP strength ratings
(b = 0.392; p < 0.01) (Table 5, Model 2). The second step of the analysis regresses the dependent variable (public reputation) on the independent variable (ethnic diversity), and this regression shows that ethnic diversity is positively and significantly related to public reputation (b = 0.327; p < 0.01) (Table 7, Model 3). The third step of the analysis includes regressing the dependent variable (public reputation) on both the independent variable (ethnic diversity) and the mediator (institutional CSP). This step requires that the relationship between the mediator and the dependent variable is significant, and regression shows that institutional CSP strength ratings are positively and significantly related to public reputation (b = 0.395; p < 0.01) (Table 7, Model 2a). Finally, the last step of the analysis shows that ethnic diversity is still a significant predictor of public reputation after controlling for the mediator institutional CSP (b = 0.204; p < 0.05). However, the effect of the independent variable on the dependent variable is less in the third step than in the second step. As can be seen in Table 7, Model 3 and 4, the standardized coefficient declined from 0.327 in step 2 to 0.204 in step 3. Thus, from the analysis can be concluded that the relationship between ethnic diversity and public reputation is partially mediated by institutional CSP (b = 0.298, p < 0.05). Hypothesis 3b is therefore partially supported.
Hypothesis 3c predicts that technical CSP strength ratings mediate the relationship between ethnic diversity and financial reputation. However, the first step of the analysis shows that ethnic diversity is not related to technical CSP strength ratings (b = 0.113; p = 0.227) (Table 5, Model 4). Subsequently, hypothesis 3c is not supported. The relationship between ethnic diversity and financial reputation is not mediated by technical CSP strength ratings.
Hypothesis 3d predicts that technical CSP strength ratings mediate the relationship between ethnic diversity and public reputation. However, since this hypothesis also requires ethnic diversity to be related to technical CSP strength ratings (as in hypothesis 3c),
hypothesis 3d is not supported. Technical CSP strength ratings do not mediate the relationship between ethnic diversity and public reputation.
Hypothesis 4a predicts that institutional CSP strength ratings mediate the relationship between outside directors and financial reputation. Outside directors are positively and significantly related to institutional CSP strength ratings (b = 0.395; p < 0.01). However, institutional CSP strength ratings are not significantly related to financial reputation (b = 0.037; p = 0.696) (Table 6, Model 2a). Therefore this hypothesis is not supported. The relationship between outside directors and financial reputation is not mediated by institutional CSP strength ratings.
Hypothesis 4b predicts that institutional CSP strength ratings mediate the relationship between outside directors and public reputation. The first two steps of the analysis show that outside directors are positively and significantly related to institutional CSP strength ratings (b = 0.395; p < 0.01) (Table 5, Model 2), and positively and significantly related to public reputation (b = 0.199; p < 0.01) (Table 7, Model 3). Furthermore, regression shows that institutional CSP strength ratings are positively and significantly related to public reputation (b = 0.395; p < 0.01) (Table 7, Model 2a). Finally, the last step of the analysis shows that outside directors are no longer a significant predictor of public reputation after controlling for the mediator institutional CSP (b = 0.028; p = 0.774), as can be seen Table 7, Model 4. Thus, from the analysis can be concluded that the relationship between outside directors and public reputation is mediated by institutional CSP strength ratings (b = 0.298, p < 0.01). Hypothesis 4b is therefore supported.
Hypothesis 4c predicts that technical CSP strength ratings mediate the relationship between outside directors and financial reputation. The first step of the analysis shows that outside directors are positively and significantly related to technical CSP strength ratings (b = 0.413; p < 0.01) (Table 5, Model 4). However, outside directors are not related to financial
