DO INVESTORS VALUE BOARD ETHNIC DIVERSITY? A CANADIAN STUDY

How to cite this paper: Talbot, C., Coulmont, M., & Berthelot, S. (2023). Do investors value board ethnic diversity? A Canadian study. Corporate Board: Role, Duties and Composition, 19 (1), 20–28. https://doi.org/10.22495/cbv19i1art2


INTRODUCTION
Over the last 10 years, a number of European countries have legislated to promote greater diversity in management positions and on boards of directors of listed corporations. While some have adopted the quota model to increase diversity, others rely on a purely voluntary initiative accompanied by a "comply or explain" approach. Most of these countries define diversity as gender diversity and issue recommendations to increase the proportion of women on their boards of directors and management teams (Organisation for Economic Co-operation and Development [OECD], 2021). In 2015, the Canadian Securities Administrators (CSA) amended the regulation on disclosure of governance practices to encourage issuers to adopt a policy on the representation of women and to publish it in their proxy circulars or to explain the reasons for its non-adoption.
Corporations have also realised that enhanced diversity in strategic corporate positions is strategically important to help align business strategies more effectively with a growing demographic diversity among stakeholders in the markets in which they operate (Vairavan & Zhang, 2020). Since then, a new broader definition of diversity, targeting adequate representation of the general population, has been promoted in financial markets. Responding to the evolution of the general public's conception of diversity, the government of Canada amended the Canada Business Corporation Act (CBCA) to encourage the nomination of a broader diversity of individuals on boards of directors as well as in senior management positions in public corporations (Dauphin & Allaire, 2021). Canadian federally incorporated corporations are required to provide shareholders with information on their diversity policy and practices respecting their boardrooms and strategic management positions. With this new regulation, Canada became the first jurisdiction worldwide to formally expand diversity beyond gender (Jeffrey et al., 2019). The objectives of these amendments, effective since January 2020, are to increase the nomination not only of women, but also of ethnic minorities, Aboriginal communities, and persons with disabilities.
Although these changes particularly affect federally incorporated corporations in Canada, 101 firms included in the S&P/Toronto Stock Exchange (TSX) Composite Index are subject to these requirements. While the other corporations in the index, incorporated at the provincial level, are not required to follow CBCA recommendations, they are subject to the same pressures for inclusion, equality, justice, and representativeness of society. Pressures from society and new requirements from CBCA to make boards more inclusive seem to have borne results. In 2021, 35% of the new director appointments in Canada's 100 largest publicly traded companies were filled by historically underrepresented groups 1 (Spencer Stuart, 2022).
To demonstrate the merits and thereby promote diversity on boards, supporters of the values of inclusion and ethnic diversity in strategic corporate positions hope to demonstrate that gender, race, or any other discriminating factor improves the performance of publicly traded corporations. Proponents refer to the agency and resource dependence theories to defend their position. The agency theory posits that since the board's role is to monitor and discipline managers (Jensen & Meckling, 1976), decisions will be made in the shareholders' best interests and should improve firm performance. In addition, the resource dependence theory proposes that boards operate on a strategic level to advise and counsel management (Pfeffer & Salancik, 1978). Here again, it may be proposed that a more diverse board will have access to more diversified resources and will therefore better advise management (Ben-Amar et al., 2013; Aggarwal et al., 2019), in turn improving corporate performance. While several works in the literature have studied the link between diversity and performance, empirical evidence has not reached a consensus to support this theory (Aggarwal et al., 2019). Opponents of board diversity instead base their position on social identity theory to illustrate that heterogeneity could reduce group cohesion and alter the efficiency of the board's monitoring and advisory function (Ntim, 2015). They may also refer to the theory of tokenism to justify the marginalisation of minorities and their lack of real effect on board efficiency.
We have identified several different forms of board diversity in previous studies. While the majority of studies define diversity as gender diversity ( Board diversity is defined as the percentage of visible minority directors on the board, while firm market value is measured using Ohlson's (1995) model. The empirical analyses were carried out using a sample of 563 observations of Canadian companies listed on the S&P/TSX Composite Index over the period 2019-2021 for which data on governance, gender and ethnic diversity on boards were collected. We control for possible endogeneity between firm market value and ethnic diversity by using a two-stage least square analysis. The results suggest that, in accordance with a portion of the literature, the financial market positively considers the share of visible minority members on boards of directors. This result holds after controlling for gender diversity, board independence and industry.
Our paper makes the following contributions to the literature. First, our results support the regulatory initiative that requires federally incorporated Canadian corporations to disclose information about board diversity, namely gender and visible minority, in their official documentation. Also, we extend the literature that addresses board diversity. While many studies have examined the link between diversity and performance, we have not seen any studies investigating whether investors take board diversity into account. In addition, we improve our understanding of ethnic diversity and its effects, an area that is still not actively researched despite its importance to firms and policymakers (Guest, 2019). The results of this study also suggest that although regulatory bodies are promoting diversity and inclusion values on boards, investors currently see more value in building a board that is gender rather than ethnically diverse.
The remainder of the paper is organised as follows. Section 2 reviews the relevant literature associated with board diversity. Section 3 describes the empirical model and the sample. Section 4 presents the study results and discussion, and lastly, Section 5 reports its main conclusions, limitations, and potential avenues for future research. . Moreover, the demand for a more diversified board is now seen as a priority in the marketplace due to increased diversity in the workforce in terms of age, gender and ethnicity (Darmadi, 2011). In North America, the concepts of diversity and inclusion have gained importance in the last decade, during which time representativity on boards and executive positions has extended well beyond gender (Dauphin, 2022). Canada is the first jurisdiction to require federally incorporated corporations to disclose the participation of women, visible minorities, Aboriginal communities, and people with disabilities in their official documents. Other jurisdictions are also sensitive to expanded inclusion values on boards but have so far refrained from imposing specific requirements. For example, the UK Corporate Governance Code mentions how important it is to promote diversity on boards in order to foster constructive debates. According to this Code, a diversified group includes but is not limited to, diversity of gender and race (Financial Reporting Council [FRC], 2016).

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
In the literature, board diversity has been studied through the perspectives of agency, resource dependence, and social identity theories, as well as tokenism. The agency theory illustrates the agency problems that can arise when managers' interests differ from those of shareholders (Jensen & Meckling, 1976). To mitigate these conflicts of interest, the corporate board of directors has a dual mandate -to monitor management's decisions and advise management (Jensen, 1993). From this perspective, proponents of board diversity suggest that a diverse board promotes board independence, improves monitoring of management (van der Walt & Ingley, 2003; Carter et al., 2007Carter et al., , 2010, provides enhanced disclosure that reduces agency costs and information asymmetry, and protects the reputation of board members (Lim et al., 2007), thereby enhancing performance.
According to resource dependence theory, an organisation's performance is influenced by its external environment (Pfeffer & Salancik, 1978). For a corporation to be successful, the board of directors and management must develop links with other corporations to reduce the challenges and uncertainties arising from their dependency on other corporations' resources (Ali et al., 2014). Consequently, resource dependence theory suggests that more diversity on boards will increase the legitimacy and resources, such as expertise, skills, and access to a diversified network (Hillman et al., 2000), provided by board members and ultimately affect a firm's short-and long-term performance. Many explanations have been put forward to support this idea since diversified resources enable a better match with the demographic representation of the population and enhance understanding of employees, customers, and suppliers (Robinson & Dechant, 1997). Cognitive diversity in a group is linked to better creativity, innovation and decision-making (Baranchuck & Dybvig, 2009). According to Adams et al. (2010), a more diverse group brings more diverse viewpoints, which in turn leads the board to make better decisions and play a better advisory role since it represents the population, minority customers and employees.
However, opponents of board diversity point out that since heterogeneity interacts with board processes, monitoring efficiency and decisionmaking, it could alter the board's performance (Ntim, 2015). According to the social identity theory, success in fulfilling a mandate's tasks is linked to the ability of individuals to function efficiently as a group. Since individuals prefer to build relationships with other people who belong to similar social categories (Williams & O'Reilly, 1998), age, gender and ethnic heterogeneity could also act as drivers of board conflicts, inhibit boardroom cohesion, and ultimately lead to negative performance (Ely & Thomas, 2001).
Ethnic appointments on boards could also have a neutral effect on monitoring or advisory efficiency. While diversity in the boardroom is desirable, nominating individuals with specific attributes solely to comply with diversity is not. The theory of tokenism (Kanter, 1977) illustrates that on a board with a skewed proportion of visibly different directors, minority members are seen as tokens because of their rarity. This token status could encourage majority members to have certain perceptions, such as greater visibility, polarization (exaggeration of differences within social groups), and assimilation of minority attributes to fit stereotypes (Rixom et al., 2023). When minorities are barely represented on boards, they could be marginalised and disregarded until they have attained a certain level of representation. Consequently, when investors or other market actors believe that these nominations do not enhance firm performance, tokenism results and the gain expected from a diverse board are not achieved (Rixom et al., 2023).
In the literature and with policymakers worldwide, more attention has been focused on board gender than on ethnic diversity (Guest, 2019). Particularly over the past two decades, initiatives have been introduced to make boards and executive positions more gender inclusive. Efforts made by investors groups, such as the Institute for Governance of Private and Public Organization (IGOPP) in Canada, and changes to the legislation adopted by the Canadian government improved the representation of women on boards from 15% to nearly 30% in 2020 (Dauphin & Allaire, 2021). Findings have emerged from previous studies that examined gender diversity in the boardroom. For example, the literature indicates that female directors behave differently than their male counterparts, particularly in terms of risk-aversion (Levi et al., 2014), women being seen as more riskaverse than men and as adopting more conservative dividend pay-out policies (Chen et al., 2017). Women are also associated with better performance in firms with agency problems (Adams & Ferreira, 2009). The representation of women is greater on key monitoring committees and in corporations offering lower compensation (Adams & Ferreira, 2009). As well, women are associated with better earnings quality (Srinidhi et al., 2011), a decrease in discretionary revenue recognition and a lower earnings management practice (Garcia Lara et al., 2017). Accordingly, previous work in accounting literature shows differences in behaviour across gender in director and executive positions and concludes that female directors are better at monitoring responsibilities than male directors.
In addition, with the different theories explaining the lack of diversity on boards in the academic literature and given arguments put forward by the general population to enhance gender representativity in strategic business positions, researchers have attempted to test empirically whether greater gender diversity on boards results in better or worse performance. Contrary to gender diversity, which has been addressed in numerous studies that see some tangible benefits in board gender diversity, ethnic diversity has received much less attention, despite the fact that better ethnic representation in key management positions should be a priority for the market and legislators (Guest, 2019). One strand of the empirical literature shows that diversity in the boardroom is associated with higher market valuation. For example, Erhardt et al. (2003) study the association between demographic diversity and the board of directors. The demographic diversity variable considers both ethnic and gender representation on boards. The authors find a positive association between financial indicators of firm performance and demographic diversity on US boards of directors. In a study of 169 organisations listed on the Johannesburg Stock Exchange (JSE) for the years 2002 to 2006, Ntim (2015) founds that board diversity is positively associated with market valuation, measured respectively by Tobin's Q, return on assets (ROA) and total share return. He also clearly demonstrated that the stock market places more value on ethnic diversity than on gender diversity.
Similarly, when Ameer et al. (2010) evaluated ethnic diversity according to the representation of outside and foreign directors, they found that diversity is associated with better performance. They argue that a more racially diverse board would have a better understanding of the different stakeholders in today's corporate environment. They rely on the resource dependency theory to justify the notion that more policies and strategies will be developed by a diverse board to accommodate and please the variety of stakeholders, which would also benefit the firm financially and please investors. Other studies measure the moderating effect of employee satisfaction and productivity on firm performance. A more diverse board is more likely to adopt corporate practices that enhance employee satisfaction and productivity (Creek et al., 2019), which in turn contributes to the firm's financial performance (Vairavan & Zhang, 2020).
Other empirical studies on ethnic diversity fail to demonstrate the beneficial effects of racial diversity in the boardroom on corporate performance. In a study based on a sample of 1,500 S&P firms, Vairavan and Zhang (2020) found that increasing board racial diversity has no significant direct effect on a firm's financial performance when measured by return on assets and Tobin's Q. In a similar study of 11,916 firm-years for the years 1996 to 2011, Guest (2019) found no relationship between board ethnic diversity and firm performance, nor any links to chief executive officer's (CEOs) compensation, accounting misstatements, CEO turnover, performance sensitivity or acquisition performance. One explanation he proposes is that a board member is cautiously selected and likely to be very similar to the rest of the Caucasian group. He also suggests that a visible minority member is pressured to conform to the rest of the group, positing here that minorities could be assimilated into the group majority and therefore their behaviour would not differ from that of the majority.
With respect to Canada and its two successive legal requirements, pertaining to gender in 2015 and visible minorities in 2020, to make corporate boards more inclusive and representative of the general population, and considering mixed empirical evidence from recent studies examining the association between gender and ethnic diversity and firm performance, we predict a statistically significant association between board diversity and market valuation, without specifying the direction of the sign of the coefficient. Therefore, our main hypothesis to be tested in this study is as follows: H1: There is a significant positive or negative relationship between a firm's market value and board diversity based on both ethnicity and gender.

Research design
where,  -represents the market value for firm i 4 months after fiscal year-end t;  -represents the book value (equity) for firm i at the end of the year t;  -represents the net earnings for firm i at the end of the year t;


-is a dummy variable equal to 1 when earnings are negative to control for the potential differential impact of negative earnings on market value for firm i in year t;

Sample and data collection
The sample used in this study is composed of all Canadian corporations included in The Globe and Mail's Board Games Director and Company Diversity Reports for the years 2019 to 2021 inclusively, representing 655 firms-years where information about the percentage of ethnic diversity, female representation and independence was collected. These reports pertain to major Canadian companies listed on TSX, primarily those included in the S&P/TSX Composite Index. Accounting and financial data were derived from the Capital IQ database. We eliminated income trusts (64 firmsyears) since their business strategy and tax incentives are not comparable with other corporations. We also eliminated 22 firms-years from the sample for firms whose financial structure has changed (public to private, merger) and 6 firms-years with negative book value, leaving 563 firms-years in our final sample, where 195 apply to 2019, 177 to 2020 and 191 to 2021. Table 1 presents the distribution of firms by sector. Four sectors appear to be more represented: 24% in the materials sector, 14% in financial, 13% in energy and 13% in the industrial sector.   Table 3 presents the Pearson correlation coefficients for the variables included in our equation. As expected, the largest correlation coefficients are between market value ( ), book value for common equity ( ), and earnings ( ). Also, the correlation coefficients representing a board's ethnic diversity (% ), representation of women (% ), and independence (% ) are weaker but still significantly correlated to the financial variables ( , and ). The coefficient of the control variables women representativity (% ), independence (% ) and ethnic diversity (% ) are not significantly correlated with each other, reducing the potential of multicollinearity in the regression analyses.  For both models, we ran least squares regressions. Neither of our regression models (M1 or M2) presents a variance inflation factor higher than the maximal prescribed threshold of 10 proposed by Hair et al. (2009), indicating serious multicollinearity problems. In addition, we ran the Dublin-Watson statistic test for autocorrelation problems. With both models, the value obtained was close to 2, indicating that autocorrelation does not seem to be problematic. Furthermore, considering that we noted a weak, but positive and significant, correlation between each of our interest variables, that is, ethnic diversity (% ), women's representativity (% ) and board independence (% ), and our independent variable's market value ( ) and book value ( ), there is a possibility that some of the effects of these three interest variables are already incorporated in the two accounting variables, making it difficult to measure their real effect on market value. To address this possibility, we first regressed each of the two accounting variables and on ethnic diversity, women's representativity and independence (% , % , and %

Regression analyses
). The residuals from these pre-regressions _ and _ replace the raw accounting variables in our models M1 and M2. These minor statistical replacements eliminate the correlation between the adjusted accounting variables and the market value.  less significant than in model M1. One possible explanation is the fact that the new legal requirement encouraging better representation of visible minorities on boards has only been in effect since 2020, in contrast to the requirement encouraging the nomination of women on boards, which was introduced in 2015. Board members are not necessarily replaced every year and new qualified board members proposed for election do not necessarily belong to a visible minority. Sectors continue to significantly impact the relationship between our independent and dependent variables, as was the case in M1. In addition, there is a slightly significant increase in the explanatory power of the independent variables over the dependent variable of Model 1 (M1), now at 75.2%, with the addition of the new board diversity measure (% ). The difference between the adjusted R 2 statistic is significant (F-test improved fit of 14.765).

Discussion
Although board ethnic diversity has received more attention from the financial market, academics, and legislators in some countries in recent years, the evidence supporting the benefits of greater board ethnic diversity is inconclusive (Carter et Morrone et al., 2022). The objective of this study is rather to evaluate how the market perceives board diversity, which brings a different perspective of analysis through a more sophisticated and complete empirical model. Another important distinctive feature of our study is that it is conducted in a jurisdiction where new requirements have been introduced to encourage the nomination of visible minority directors. In 2020, Canada established regulations requiring federally incorporated corporations to disclose the participation of women, visible minorities, Aboriginal communities, and people with disabilities in their official documents. This new requirement is a tangible response from a governing body to the general population for whom the values of gender and racial inclusion and equality now occupy an important place in societal debates. The objective of our study is to determine whether investors also attach importance to these same values. Our results are similar to those of EmadEldeen et al.

CONCLUSION
This study aims to examine whether investors value the ethnic diversity of boards of directors. Based on a sample of 563 firm-years from the S&P/TSX Composite Index, our results suggest that the market takes gender and ethnic diversity into account, thereby empirically supporting the participation of visible minorities on boards of directors. The results also show that the impact of ethnicity on value for investors is nearly 50% less than that of gender diversity. However, while the impact of board ethnic diversity is not substantial, it remains significant. Although the relationship between ethnic diversity and shareholder value is significant, including the ethnic diversity variable in the second regression model slightly improves its strength. One reason could be that the cause-and-effect relationship between ethnicity and perceived shareholder value is difficult to validate empirically. Another could be that investors view the nomination of visible minority candidates as a requirement for being considered a good corporate citizen, rather than as an initiative to bring more technical expertise to the board. Investors could therefore perceive these nominations as an indication that the corporation follows the rules and manages the business as a good corporate citizen, promoting sustainable values that eventually tie into corporate performance.
This study has certain limitations. While the number of firm-years constituting our sample is less than in certain previous studies that covered more than 10,000 firm-years (Guest, 2019), it is comparable to others (Rixom et al., 2023). It should be noted however that the composite index of the S&P/TSX on which we based our sample contains 51% of firms required to comply with the new disclosure requirement on board diversity. In addition, our sample also represents approximately 95% coverage of the Canadian equities market (TSX, 2023). Therefore, since the data used in the paper are based on large public firms, the results may not be generalized to small or private firms. Furthermore, the data used in this study was limited to three years because the new requirement is recent, thus, in turn, limiting the number of firm-years. A second limitation lies in the challenge of determining whether a director belongs to a visible minority. Although the new Canadian regulation requires federally incorporated corporations to disclose information on directors belonging to visible minority groups, a number of corporations in our sample are not subject to this regulation and those that are can still adopt the compliance or explain the approach and choose not to comply for the years under study. It is therefore possible that the information collected in The Globe and Mail database underestimates the directorships of visible minority groups or that the data collected contains errors. However, to mitigate the possibility of errors in the ethnic diversity variable, we compared the representation of visible minority directors in our sample with another study for the same period (MacDougall et al., 2022).
A future project could investigate whether women and members of visible minorities have few responsibilities as directors or become members of audit, compensation, and governance committees or even chair of the board. It could also be worthwhile examining whether the impact on the future benefits that shareholders appear to perceive translates into increased earnings thanks to better strategic decisions or more effective control of agency costs. In other words, it might be interesting to examine whether the contributions of women and visible minorities stem from their role as advisors and resource providers in terms of strategic decisionmaking or from their role of monitoring management.