* Candidato
a Doctor en Ciencias Administrativas, EGADE Business School Tecnológico de Monterrey. Profesor a Nivel Profesional, E-mail: A00809455@itesm.mx
Artículo recibido:
4 de septiembre de 2016 Artículo
aceptado: 29 de noviembre de 2016
Ciencias Administrativas. Teoría y Praxis Num. 1 Año
13, Enero-Junio 2017, pp. 11-27
El efecto de
la diversidad en el desempeño social corporativo: Un análisis empírico
The effect of diversity on corporate social performance: An
empirical analysis
Este artículo investiga el efecto de la diversidad de los empleados
(operativos y ejecutivos) en el nivel de Desempeño Social Corporativo de la
empresa. Basándose en la teoría de los stakeholders, el autor propone que la
diversidad ayuda a una organización a
mejorar su capacidad de identificar y satisfacer las diversas necesidades de
los stakeholders, por lo que se espera un mayor desempeño social corporativo
entre empresas con mayores niveles de diversidad tanto ejecutiva como
operativa. Utilizando datos secundarios y un análisis de regresión para probar
las hipótesis propuestas, este estudio contribuye a la literatura del Desempeño
Social Corporativo identificando a la diversidad como una causante del DSC,
presentando algunas diferencias contextuales que pudieran existir en aplicaciones prácticas y mostrando la
relación con estudios previos de la teoría de los Stakeholders. Además, se
desarrolla y presenta una nueva medida simple de DSC; Esta medida innovadora se
compara con las anteriores ya aceptadas en la literatura y se demuestra su
validez estadística. Finalmente, el autor discute las implicaciones para
investigaciones futuras.
Palabras clave: Desempeño social corporativo; Diversidad; Teoría de los stakeholders
This study
investigates the effect of employee diversity (operational and executive) on
the level of Corporate Social Performance
of the firm.
Building on Stakeholder
theory, the author argues that Diversity helps an organization improve
its ability to identify and meet diverse stakeholders’ needs, thus higher
Corporate Social Performance is expected among companies with higher levels of
both executive and operative employee diversity.
Utilizing secondary data and a regression analysis to test the proposed
hypotheses, this study contributes to the literature of Corporate Social
Performance by identifying this new driver,
presenting some contextual differences that might exist in practical
applications, and showing the relationship with previous studies of Stakeholder
theory and CSP. In addition, a new
simple measure of CSP is developed and presented; this innovative measure
is compared with previous
ones already accepted in the literature and its statistical validity is proven.
Finally, the author discusses implications for future research.
Keywords: Corporate
social performance;
Diversity; Stakeholder theory
* Candidato
a Doctor en Ciencias Administrativas, EGADE Business School Tecnológico de Monterrey. Profesor a Nivel Profesional, E-mail: A00809455@itesm.mx
Artículo recibido:
4 de septiembre de 2016 Artículo
aceptado: 29 de noviembre de 2016
Corporate social responsibility is an important and common practice among companies worldwide. Unilever recently implemented the Project Shakti (“empowerment”) in India where, instead of using its customary wholesaler- to-retailer distribution model to reach remote villages, the company recruits village women, provides them with access to microfinance loans, and trains them in selling soaps, detergents, and other products door-to-door. More than 65,000 women entrepreneurs now participate, nearly doubling their household incomes, on average, while increasing rural access to hygiene products and thus contributing to public health (Rangan, Chase, & Karim, 2015). The family-owned Mexican baking company Grupo Bimbo also offers a great example. Bimbo is the largest bakery in Mexico, with a workforce of nearly 100,000 and a similar number of small retailers in its network. Its comprehensive CSR programs focus on social welfare: It provides free educational services to help employees complete high school and offers supplementary medical care and financial assistance for dependents’ care to close the gaps in government health coverage (Grupo Bimbo, 2013).
These examples enlighten us about the impact and global reach of these corporate social responsibility practices. Some 93% of the world’s largest 250 companies now publish annual corporate responsibility reports, almost 60% of which are independently audited (KPMG, 2013). That means companies from sectors as diverse as financial services, information technology and consumer goods to oil, gas and mining making billions of dollars of public commitments to help solve societal challenges. This highlights the importance that this main topic plays in the business place nowadays.
Bad corporate reputation due to poor CSR performance can lead to several crises. Take the case of Nestlé as example; the company
was accused for causing infant illness and death in poor communities by promoting bottle- feeding and discouraging breast-feeding. Nestlé boycotts spread from Switzerland and Britain to the USA (Muller, 2013).
CSR is becoming more and more important, in part, because there has been growing attention on this issue by consumers. According to a global study (Nielsen, 2014) 55% of global online consumers across 60 countries say they are willing to pay more for products and services provided by companies that are committed to positive social and environmental impact. These tendencies no longer are true for consumers in developed countries; the propensity to buy socially responsible brands is strongest in Asia-Pacific (64%), Latin America (63%) and Middle East/Africa (63%). The numbers for North America and Europe are 42 and 40 percent, respectively.
Despite the numerous definitions that try to capture the essence of CSP, in this article I will employ the following definition: Corporate social performance refers to “actions that appear to further some social good, beyond the interest of the firm and that which is required by law” (McWilliams & Siegel, 2001). This definition will be employed due to its fit with the research purpose of this project; the CSP measure that I developed mainly takes into account business policies regarding CSR program initiatives which perfectly reflect the actions that McWilliams & Siegel mention in their definition. On the other hand, there are other authors that affirm that CSP concerns not only about these actions, but also about the outcomes of this actions (Wood, 1991). Nevertheless, according to theoretical evidence in the economics discipline, business policy and positive outcomes for those policies are highly correlated (Allsopp & Vines, 2015). Therefore, I can argue that the outcome component of the definition of CSP provided by other authors (Wood, 1991) is somehow captured by the business policy component as well. However, the CSP definition that will
guide the rest of the paper is the one provided
by McWilliams & Siegel (2001).
Talking about CSR judgments, there are two broad, but different (and somehow contradictory) perspectives in evaluating CSR: value maximization and stakeholder theory. Value maximization rests on the proposition that in implementing organizational change, managers must have a criterion for deciding what is better, and better should be measured by the increase in long-term market value of the firm, this is, the long-term market value of the firm is the only function that managers should be concerned about maximizing, the other objectives should be subordinated or restricted to this long-term value function (Jensen, 2002).
Stakeholder theory, on the other hand, says that managers should make decisions that take account of the interests of all the stakeholders in a firm. Stakeholders include all individuals or groups who can substantially affect the welfare of the firm, not only the financial claimants, but also employees, customers, communities, and governmental officials, and under some interpretations, the environment, terrorists, blackmailers, and thieves. This approach does not rest on a single dimension that must be maximized, but instead claims that multiple functions should be satisfied (Jensen, 2002).
This paper will use as guiding perspective Stakeholder theory, which currently is the dominant perspective taken in evaluating a firm’s corporate social performance. As mentioned earlier according to stakeholder theory, firms act in a socially responsible manner when they take the interests of multiple stakeholders (e.g., customers, employees, government, etc.) into account (McGuire, Dow, & Argheyd, 2003).
Because the executives make decisions regarding strategic directions, including corporate social responsibility programs. There are several studies that have investigated the impact of the CEO and the Top management on the corporate social performance of their
firm (Manner, 2010). This present research contributes to extend the literature by determining the impact of the diversity of the executive employees on the corporate social performance of the firm, it also explores the role and influence that the diversity of the operative employees may have in the social performance of the firm whether direct or through an influence on the executive team. This paper adds to knowledge of corporate social responsibility as well as to knowledge of stakeholder theory by identifying new drivers of this phenomenon and exploring the way these variables act and interact in the real world.
I argue that since it has been found that the degree of diversity within a group impacts the decision making (Jackson, May, & Whitney, 1995), the level of diversity within an organization will affect the corporate social performance of the firm as well, this is possible because the strategy of CSP is decided within an organization the same as any other investment decision (Campbell, 2007).
In most western nations, the corporate workforce has historically been dominated by white males. In other parts of the world, the story is similar: One or a few groups of people tend to be the majority in the workplace. A narrow definition of diversity is one that concentrates on race and gender, but diversity in the workplace can be broadly defined as differences, similarities and related tensions among people in the workplace based on visible dimensions, secondary influences, and work diversities. These differences affect the way people function within an organization. Visible dimensions include age, race/ ethnic heritage, gender, physical ability and qualities (including obesity), mental ability and qualities, and sexual or affectional orientation. Secondary influences include religious beliefs, socioeconomic class, background, and education. Work diversities include differences like management versus union; functional level; classification; and proximity or distance to the corporate headquarters. Other
differences include personality and work style (Tetteh, 2015). Worldwide organizations are hiring more women and minorities to create a diverse workforce that captures different points of view, backgrounds, opinions, perspectives, decision making strategies, etc. Due to their partnerships with external stakeholders such as minority communities and suppliers, organizations with diverse workforces gain reputations as considering all these different perspectives into their decision making (Tetteh, 2015).
According to the stakeholder theory, a company has a better corporate social performance when it better meets all the different stakeholder needs. So, the basic idea behind this work is that by having different perspectives and backgrounds within an organization, a better understanding of the different stakeholders of the firm will also occur, so higher corporate social performance is expected among those organizations with higher diversity, both employee diversity and executive diversity.
In this article, I aim to understand the relationship between a human resources strategy (diversity management in the workforce) and the level of corporate social performance of the organization. Specifically, I examined how the employee and executive diversity affect the level of social performance. This paper contributes to the extant literature by determining the effect of diversity in the workplace in corporate social performance of the firm. Using an econometric approach the present research adds to knowledge of corporate social responsibility literature by identifying a new driver, as well as to knowledge of diversity as a human resource strategy and stakeholder theory.
As previously mentioned, according to
stakeholder theory, a firm achieves higher
corporate social performance when it meets more stakeholder needs, whereas it attains lower corporate social performance when it meets fewer stakeholder needs. Further, any one stakeholder’s interests should not be satisfied at the expense of others, but rather, multiple stakeholders should be managed within a “mutually supportive framework” (Donaldson & Preston 1995). Since managers and executives are the ones responsible for taking decisions that must satisfy these different stakeholders, it is important to study the impact of this group and its characteristics in the corporate social performance of their organizations.
There have been several studies that have aimed to clarify the effect of the top management on the level of corporate social performance of the firm. These studies include the ones that have found links between the type of the strategic leadership of the top management and the level of social responsibility of the organization (Larrieta-Rubin de Celis, Velasco- Balmaseda, De Bobadilla, Alonso-Almeida, & Intxaurburu-Clemente, 2015). These bunch of literature include projects about personality, cognitive style, values, experiences, tenure, and functional background.
On the other hand, there is an interest in studying the impact of diversity characteristics within a top management group and the level of corporate social performance of the firm. Previous studies have found that board diversity is positively associated to financial performance (Erhardt, Werbel, & Shrader, 2003). Harjoto, Laksmana y Lee (2015), recently published an article that found a significant relationship between gender, race, age, outside directorship, tenure, power, and expertise of the board members and the level of corporate social performance of the firm. The study concludes by showing that these characteristics of the board members are positively associated with corporate social performance. This work helps extend this literature by examining the influence of the
executive team diversity of a firm and the corporate social performance not only isolated but also interacting with the level of diversity of the operating employees.
On the other hand, in order to meet stakeholder needs it is first necessary to identify those stakeholder groups relevant to the organization. In their seminal work, Mitchell, Agle, and Wood (1997), contribute to a theory of stakeholder identification and salience based on stakeholders possessing one or more of three relationship attributes: power, legitimacy, and urgency. By combining these attributes, authors generate a typology of stakeholder’s propositions concerning their salience to managers of the firm.
Later in their article, they pointed out the important role that executives and managers have in this theory: “Whatever the magnitude of their stake, each stakeholder is a part of the nexus of implicit and explicit contracts that con- stitute the firm. However, as a group, manag- ers are unique in this respect because of their position at the center of the nexus of contracts. Managers are the only group of stakeholders who enter a contractual relationship with all other stakeholders. Managers are also the only group of stakeholders with direct control over the decision-making apparatus of the firm”.
Authors bounded their stakeholder identifi- cation theory with managerial characteristics. They treated managerial characteristics as a variable and suggested that it will be an im- portant moderator of the stakeholder-manager relationship. This is, although groups can be identified reliably as stakeholders based on their possession of power, legitimacy, and ur- gency in relationship to the firm, it is the firm’s managers who determine which stakeholders are salient and therefore will receive manage- ment attention. In short, the specific charac- teristics of managers are important to identify stakeholders.
Agle, Mitchell, and Sonnenfeld (1999) quantitatively tested this proposition in a subsequent article. They found that indeed,
CEO characteristics (defined as CEO values) moderated the relationship between power, legitimacy, and urgency to stakeholder salience. More importantly, they also found that these CEO characteristics have a direct effect on the level of corporate social performance (defined as employee relations, products, environment, and community).
In summary, previous literature has demonstrated that the distinct characteristics of the executives, managers and CEO’s have a major impact on decision making within their organizations. They are the ones actually making decisions concerning CSR programs. The composition of the executive team will influence not only the degree of satisfaction of stakeholders’ needs, but also the identification and salience of the relevant stakeholder groups of a firm. Given this view, I expect that differences in the level of corporate social responsibility among firms arise as a function of the level of diversity within a top management group of a firm (among other variables already found in the literature). Developing this point further, I expect that diversity within a top management team evoke a major concern for all the stakeholder groups of the firm, thus a higher level of social responsibility is expected in the firm with higher level of executive diversity; on the other hand, low levels of executive diversity will lead to a homogeneous top management team, so less different point of views will be represented and few stakeholder needs will be met and identified, thus a lower level of corporate social responsibility is expected among firms with low levels of executive diversity. Formally stated:
H1: The level
of executive diversity
is positively related to corporate social performance
Diversity refers to the differences, similarities,
and related tensions and complexities that
can characterize a collective mixture like the workforce. These similarities and differences can be demographic in nature (for example, race, gender, ethnicity, sexual orientation and age), or they can represent behavioral variations (for example, thought, problem solving approaches, or behavioral traits associated with personality) (Thomas 2011).
Corporate social responsibility has been found as a precursor of a management diverse talent approach (Thomas 2011; Albinger & Freeman, 2000). This is, corporate social responsibility has made that companies hire more diverse employees, so link CSR causes employee diversity has been established. One of the aims of this study is to test the relationship the other way around: employee diversity causes corporate social performance, as well.
Previous research has shown that companies that employ more operative women and have a stronger European presence tend to exhibit a higher concern for climate change (Boulouta, 2013). On the other hand, it has also been investigated the role that affective diversity on the operative workforce has on the performance of the firm, finding that both positive and negative affective diversity are positively associated with managerial decision performance, although only the relationship with negative affective diversity is significant (Kouamé, Oliver, & Poisson-de-Haro, 2015). These results have been possible because of direct actions undertaken by the operative employees, not because of an influence of the top management, but because of actions and programs developed by the operative employees. Therefore, in synthesis, previous literature seems to suggest that operative employees have a direct effect on the level of corporate social performance of their firm. Taken together, these arguments suggest that operative employees with low diversity will cause their organizations to meet the needs of fewer stakeholders and, therefore, their firms will exhibit lower overall corporate
social performance. In contrast, companies characterized by high levels of diversity within their operative employees are likely to seek a variety of information and viewpoints in making strategic decisions, thus a higher level of corporate social performance is expected to occur in these organizations. Formally stated:
H2: The level of operative employee diversity is positively related
to corporate social performance.
Besides all mentioned earlier, in addition to trying to learn as much as possible about the issues facing their various stakeholders, diverse teams will also seek to identify integrative solutions, or ways to simultaneously satisfy multiple stakeholder needs within hierarchical levels of the company. Diverse teams should be more open to discuss and debate about different stakeholder needs. Besides, diverse members bring into play different perspectives and even can identify new stakeholder groups previously unidentified.
So far, I have hypothesized that executive and operative employee diversity have main effects on corporate social performance. However, on the other hand, I see some arguments that lead me to suggest that those two constructs have an interactive effect as well on the corporate social performance level of the firm. As discussed above, previous research suggests that the effect of operative employee diversity is not only direct (H2), but that it can also have an indirect effect through an influence in the top management team (H3).
Literature in the human resources discipline have found that not only do the higher levels of the organizational hierarchy have an influence on the lower levels, but that also the lover employees can influence the decision-making processes of the top management members (Franklin, 1975; Chun, Yammarino, Dionne, Sosik, & Moon, 2009; Oshagbemi, & Gill,
corporate social responsibility strategy. This should not be confused with an additive effect; I am not stating that by having both kinds of diversity the level of CSP will be bigger (that is obvious according to H1 and H2). What I am saying is that by having a higher level of operative diversity the relationship discussed in H1 (executive diversity affects positively CSP) will increase. Companies with higher level of operative diversity will have a stronger relationship between executive diversity and CSP than companies with lower levels of operative diversity. Formally stated:
H3:
Operative employee diversity moderates the relationship between executive
diversity and corporate social performance. This is, the greater the level of
operative employee diversity,
the greater the
effect of executive diversity on
corporate social performance; on the other hand, the lower the level of
operative employee diversity, the
lower the effect of executive diversity on corporate social performance.
Figure 1 integrates in a visual diagram the relationships proposed in this research. Executive diversity (H1) directly affects the level of CSP; operative employee diversity (H2) also has a direct effect on CSP. Finally, employee diversity similarly moderates the relationship between executive diversity and CSP.
Figure
1. Proposed Model
Sample. The sample consisted of
174 organizations listed on Bloomberg environmental, social and governance database that had available data concerning the variables involved in the proposed model. The sample includes firms from many different countries and all industries, the data was from year 2015. Table 1 summarizes the sample profile regarding country and industry. Country was coded as developed or developing nation, industry was coded as service or production/ manufacturing goods industry.
Countries and Industries in the sample
Dependent variable:
Corporate social performance.
Diverse items measuring CSR initiatives reported by Bloomberg ESG database were employed to construct a CSP Index. 16 dummy items were chosen to construct this CSP index: Energy efficiency policy, Emission reduction policy, Green building policy, Waste reduction policy, Sustainable packaging policy, Biodiversity policy, Water treatment policy, Sustainability commitment, Health/safety policy, Equal opportunity policy, Human rights policy, Training policy, Business ethics policy, Fair remuneration policy, Employee CSR training, and Anti bribe policy. In this part of the paper it is relevant to mention again that this measure of CSP is consistent with the definition of CSP provided by McWilliams & Siegel (2001).
Then, for each firm the total of the items that applied were counted and divided by the maximum number of possible items (16). For
example, an organization with 10 policies has a CSP index of 62.50% (10/16), an organization with 16 policies has a CSP index of 100%, and an organization with 0 policies has a CSP index of 0%, and so on.
Previous literature has argued that diversity is a component of CSP (Turban, & Greening, 1997). Nevertheless, the CSP index that I developed and used in this project does not include any aspects of diversity, so the econometric model will not have any endogeneity concerns since the dependent variable will not explain any of the independent variables involved in the model (Kim, 2010).
In order to provide statistical support for the existence of such a CSP index from these dummy variables, a polychoric factor analysis was performed; this statistical technique is similar to the common factor analysis with principal components, but with the difference that this technique allows analyze dichotomous variables. The variables involved in this research are dichotomous since they can only have 2 possible values (1 or 0). The polychoric correlation coefficient introduced by Pearson, is an alternative to Pearson’s r specifically for situations in which the variables of interest are noncontinuous and the measurement instruments yield data that may only be ordinal (Holgado–Tello, Chacón–Moscoso, Barbero–García, & Vila–Abad, 2010). Several studies have investigated its effects and have concluded with positive statistical accuracy from this technique (Flora & Curran, 2004). In (table 2) and table 3 the results of this polychoric factor analysis are shown.
Variance explained
Factor Eigenvalue Proportion
As can be seen in table 2, factor 1 explains basically all the variance involved in the 16 variables (96% of total variance). On the other hand, in table 3 all variables have a significant factor loading in factor 1, previous literature suggests that a minimum factor loading of 0.30 is sufficient to prove that a variable is grouped in a specific factor (Rigdon & Ferguson, 1991). All 16 variables surpass this threshold of 0.30 in their factor loading in factor 1.
The fact that factor 1 explains almost all the variance involved, and that all variables have a loading with factor 1 of at least 0.30 presents statistical support for the existence of such a CSP dimension in the data analyzed.
To assess the degree of executive diversity within a firm, I used the arithmetical average of the following items: percentage of women in management, percentage of minorities in management, and percentage of female’s executives.
These three items were used as proxies for executive diversity because they were the only diversity indicators measured at the executive level available on the Bloomberg ESG database. The average was computed in order to have a single measure, the arithmetical average is a common practice for this purpose (Chin, Hambrick, & Treviño, 2013).
To measure
the level of employee
diversity in an organization, I used the arithmetical average of the
following items: percentage of women in workforce, percentage of minorities in
workforce, percentage of disabled in workforce. As in the case of the executive
diversity measure, these
four indicators were
|
Factor 1 |
2.71 |
0.96 |
employed as
proxies for operative
employee diversity because they were the only
diversity |
|
Factor 2 |
0.056 |
0.02 |
indicators
measured at the operative level |
|
Factor 3 |
0.023 |
0.01 |
available on the Bloomberg ESG database. (table 3). |
Table 4 summarizes these metric variables
involved in
the
study
and
offers
basic
descriptive
statistics.
According
to previous studies that have dealt with CSP variables. I used firm size,
industry sector, country and financial performance as control variables
(Godfrey, Merrill, & Hansen, 2009). Firm size was measured using the
number of employees
of the firm, the industry
was measured using
the global
industry classification standard
(GICS) developed by MSCI and Standard & Poor’s, this was coded as service (1) or goods industry (0). Financial performance was measured with the return on assets (ROA) indicator, and country was coded as developed (1) or developing nation (0).
The industry control variable was coded as service or goods industry and the country variable as developed or developing in order to keep the model as simple as possible. Also, this is a common practice in the discipline (Godfrey, Merrill, & Hansen, 2009).
Table 5 presents the correlations among the study variables. I also present the variance inflation factor (VIF) in order to rule out any multicollinearity problem in the proposed model. These VIF indicators are shown in table 6. The highest observed variance inflation factor value in the study variables was 1.64, which suggests that multicollinearity was not a concern since this value is way below the conventional cutoff of 10.00 (Neter, Wasserman, & Kutner, 1989).
Besides multicollinearity, two very common problems with econometric models are heteroscedasticity and endogeneity. In order to discard heteroscedasticity issues the Breusch Pagan test, which is designed to detect any linear form of heteroscedasticity (Mizon, 1995), was computed. Since the chisquare of the test was small, I can conclude that heteroscedasticity was not a problem (χ² = 11.89, p > 10%).
In order to test for possible endogeneity problems, Davidson and MacKinnon (1993) suggest an augmented regression test, which can easily be formed by including the residuals of each endogenous variable, as a function of all exogenous variables, in a regression of the original model. To this purpose the Hausman test was performed to test for the endogeneity of both, executive and operative employee diversity. The F values of these tests indicate that endogeneity was not a significant concern (p < 5%). In synthesis, I can affirm that the following regression models are free of any multicollinearity, heteroscedasticity and endogeneity problems.
To test the first two hypotheses, I regressed corporate social performance on the control variables, executive diversity, and employee diversity. In order to test the hypothesis regarding moderating effects (hypothesis 3), the interaction between those variables was computed: executive.diversity * employee. diversity. The complete model would be as follows:
CSP = Controls + ExecutiveDiversity + EmployeeDiversity + ExecutiveDiversity* EmployeeDiversity + Residuals
Table 7 shows the results of these regression analyses. In model 1, the base model, I regressed corporate social performance on only the control variables in order to analyze just the effect of these controls on the dependent variable. These results show that return on assets, country, and industry are significant (p < 1%), number of employees
approximate significance (p < 10%). This model explains 37% of the total variance.
In model 2, I included executive diversity (H1) and executive diversity (H2) in order to test these hypotheses concurrently. This is, this model explores and tests the significance of executive and employee diversity on corporate social performance at the same time. This model gives strong statistical support for hypothesis 1; hypothesis 2 is not supported under these results. The R² significantly increases to 71%, adjusted R² increases as well in comparison to model 1. R² increments more than 34 percentage points in comparison of the previous regression.
Finally, model 3 tests all the hypotheses involved in the proposed model. This model regressed corporate social performance on executive diversity, employee diversity, interaction of executive and employee diversity, and controls. This regression shows the complete model proposed in this research. All the variables are analyzed simultaneously. Executive diversity is still significant at the 99% of confidence. Employee diversity and the interaction are not statistically significant at any conventional levels of statistical significance (1%, 5%, or 10%). Return on assets and industry are significant at the 95% and 99% of confidence, respectively.
Country and number of employees are not significant in this model. The overall R² is 71.18% and the adjusted R² is 69.96%, this represents 32% more percentage points of explained variance in comparison of model 1 that just regressed controls on the dependent variable CSP. In synthesis, these regression analyses offer strong statistical evidence for hypothesis 1, executive diversity is statistical significant at the 99% of confidence in models 2 and 3. Executive diversity coefficient is also positive in all models, thus hypothesis 1 is supported in the present research. Executive diversity increases the level of corporate social performance. On the other hand, employee
diversity and the interaction are not statistically different from 0 at any conventional level of significance, thus hypotheses 2 and 3 are not supported by these data.
Employee diversity does not influence the level of corporate social performance and employee diversity does not moderate the relationship between executive diversity and
corporate social performance. Finally, a slope main effect graph is presented in order to show in a simple manner the effect of executive diversity on corporate social performance. This slop graph is presented in figure 2, where is shown that high executive diversity firms have, on average, 30 more percentage points of CSP than those firms with low executive diversity.
OLS Regression results for overall
CSP regressed on executive and employee diversity
Figure 2. Effects of executive diversity on corporate social performance
Although strong statistical evidence was shown in order to prove the existence and validity of the CSP dimension used in this research. There exists more CSP proxies variables used in the literature that have been trusted as good measure for CSP for years. Thus, I conducted additional robustness checks in order to see if changing the measure of the dependent variable provided the same relevant conclusions. Previous literature has employed proxies of CSP using Kinder, Lydenberg and Domini scores (Waddock & Graves, 1997). Others have used Bloomberg, Thomson Reuters ASSET4 or the Sustainalytics Index (Ioannou, & Serafeim, 2012).
In this additional analysis, CSP was mea- sured using the index provided by Sustainalyt- ics. Sustainalytics supports investors with the development and implementation of respon- sible investment strategies. Sustainalytics’ re- search is used by investors to integrate envi- ronmental, social and governance factors into their investment processes (Sustainalytics). This proxy has been used in previous literature as a reliable variable measure for CSP (Orij, 2010; Torres, Bijmolt, Tribó, & Verhoef, 2010).
The other variables involved in the research were exactly the same as were explained earlier. The only difference was the dependent variable. Because fewer companies have data concerning this sustainalytics measure in the Bloomberg ESG database, the sample size was considerably lower, 94 companies were used in this additional regression analysis.
Table 8 presents the results for this robustness analysis. Model 1 only presents the effects of the control variables on the Sustainalytics CSP index. Return on assets is the only control significant variable; number of employees, country and industry are not significant at any conventional levels of statistical significance (1%, 5%, or 10%). The adjusted R² is quite low, less than 6% of total explained variance.
Model 2 indicates that executive diversity affects the degree of CSP, this coefficient is significant at the 99% of confidence. Employee diversity’s coefficient is not significant. Number of employees is the only significant control variable (p < 5%). The overall R² is 28% and the adjusted R² is almost 23%, this represents 18% more percentage points of explained variance in comparison of model 1 that just regressed controls on the dependent variable CSP.
Finally, model 3 also presents strong statistical evidence of hypothesis 1 (p < 1%). Hypotheses 2 and 3 are not supported by the analyzed data; these coefficients are not significant at any conventional levels of statistical significance (1%, 5%, or 10%). This analysis yielded the same main conclusions as ordinary least squares regressions using the proposed measure of CSP in this research. Thus, we can conclude that the same findings are found independently of the CSP measure employed.
Besides, it is important to note that the R² is significantly bigger using the proposed CSP measure than using the sustainalytics index. On the other hand, using the sustainalytics measure, gives no statistical significance for the control variables, but number of employees. (Table 9) compares this two CSP measures in several statistical indicators.
As can be seen, the mean is very similar between both CSP measures. The proposed one is slightly lower probably due to its minimum values of 0%, while the minimum value of the sustainalytics one is 9%; both have a maximum of 100%. This means that the proposed measure uses the entire range (0%
-
100%), while the
sustainalytics index does not uses very low measures (9% - 100%). On the other
hand, the proposed
measure has a standard deviation
almost five times lower than de sustainalytics index. The proposed measure’s
kurtosis of 2.06 and its skewness of 0.14 indicate
that the variable
is normally
distributed, whereas the sustainalytics kurtosis of -0.04 and its skewness of 1.02 indicate that this variable does not follow a normal distribution. Finally, the correlation between both measures is very high and positive, indicating its significant association (0.8051).
In synthesis, I can affirm that the proposed CSP measure is better in the following aspects: it uses the entire range, which allows more discrimination among observed units; it follows a normal distribution, which allows the use of classical statistical methods with normal distribution assumptions; and it has a lower standard deviation, which permits develop predictive models with less total variance to explain. Finally, I can also argue that the proposed CSP scale in fact measures a CSP existing dimension; the polychoric factor analysis and the high correlation with the sustainalytics index allow me to conclude this. I encourage researchers to adopt and use this new measure in future research projects concerning corporate social performance.
Understanding the impact of executive diversity on the level of corporate social performance of a firm requires considering the focus and the level of strategic decisions that this top management team has within an organization. The data used in this research has shown that not all the workplace diversity really counts in corporate social performance concerns. Previous literature suggests that executive and employee diversity play a different role in the decision making of a firm. This research found that the only type of diversity that really counts in CSP is the diversity within the top management team, it seems that the level of diversity of operative employees do not have a strong linear effect on the level of CSP of the firm.
This previous investigation demonstrated that the level of executive is positively related to corporate social performance. This means that the more diverse the team leaders are, the
higher the level of corporate social performance of their firm. I also investigated the direct and moderating role of employee diversity in the relationship between executive diversity and corporate social performance. It was found that these variables did not play any relevant role concerning the social performance of an organization.
This research has some limitations that might explain why employee diversity hypotheses (H2 and H3) were not supported. I encourage future research in these topics in order to address these limitations.
OnelimitationisthatIdidnottakeintoaccount the degree of centrality in the decision making of the firm. Strategic leadership researchers have noted that decision making occurs in broader organizational and environmental contexts (Finkelstein & Hambrick, 1996; Reger, 1997). One organizational variable posited to influence decision making (through its effect on information flow) is decentralization (Miller, 1987). Centralization occurs when decision- making power resides in the hands of a select few at the upper levels of an organization, whereas decentralization occurs when decision-making power involves individuals at various organizational levels (Steers, 1977). In other words, the more centralized the firm, the less top managers have delegated decision making. Decentralization then occurs when decision-making power is shared among all employees in all levels of an organizational hierarchy. Researchers have argued that decentralization is beneficial in integrating diverse information obtained from various organizational levels, which can improve decision making (Malone, 1997). Previous literature suggests that centralization plays an important role concerning corporate social performance strategic decisions in an organization. I argue that the direct and moderating effect of operative employee diversity could be uncovered if future research replicates this study adding a centralization measure for each firm. Centralization in
decision-making could be a possible bounding condition, this is, and operative diversity might have a direct effect on the level of corporate social performance as long as the organization has a decision-making decentralized.
Another limitation is that the age of the executive team was not part of the model. Previous literature suggests that the age of the executives influence the degree of the influence they have by lower employees. This research stream has found that the age of the top management team influences the type of decision making within their firm (Richard & Shelor, 2002; Murray, 1989). Taken these arguments together, I argue that by adding this demographic variable into the proposed model, the relationship between operative employee and corporate social performance will be uncovered. In other words, it may be possible that this relationship is significant only in young or old executives.
Finally, the effect of employee diversity might be different from a linear one; this construct can have a non-linear effect on CSP. So, a more sophisticated statistical analysis might help explore its effect on CSP, some examples of this type of models are: nonlinear least squares, exponential regression, Poisson regression, and Polytomous regression. More research is needed using non-linear regression models in order to explore this kind of relationship across these constructs.
Since the top executives are the ones actually making decisions concerning CSR initiatives, another explanation of these results is that basically operative employees do not have any influence concerning this dependent variable. These initiatives are made only by the top management team and they are not influenced neither direct nor indirectly by any low hierarchical level employee.
In summary, this investigation provides strong evidence about the main effect that executive diversity within the top management team plays in the level of corporate social
performance of the firm (H1). On the other hand, this work also shows that the operative employee diversity does not affect the social performance (H2) despite previous literature have shown that low hierarchical level employees can affect the decision making of the top management; this effect is not present regarding CSR decisions. Finally, I also presented evidence about operative employee diversity not having a moderating effect on the main executive diversity and CSP relationship (H3). Operative employee diversity does not have neither a direct nor an indirect effect on CSP.
This paper contributes to the extant literature by determi ning the effect of executive diversity in the workplace in corporate social performance of the firm. Using an econometric approach the present research adds to knowledge of corporate social responsibility literature by identifying a new driver, as well as to knowledge of diversity as a human resource strategy and stakeholder theory by finding that a higher level of executive diversity within a firm will cause the organization to identify their stakeholder groups and meet their needs, thus a higher corporate social performance is present.
This investigation also developed and validated a new measure of corporate social performance using dummy variables from Bloomberg ESG database and employing a polychoric factor analysis to subtract the CSP dimension from the data. This new measure was compared with a previously used one (Sustainalytics) and the same conclusions were reached independently the measure of CSP used. More importantly, the R² is significantly bigger using the new proposed measure and the significance of the control variables is not supported employing the Sustainalytics index. So, initial support for the validation of this new CSP measure is provided as well as initial findings that show that this new measure is better than previous ones in terms of total explained variance and significance of control
variables already tested as significant in the literature. I conclude this paper by reaffirming the importance of addressing these limitations and keep developing knowledge regarding this
constructs. I strongly encourage researchers interested in these topics to move this study forward.
Agle, B. R., Mitchell, R. K., & Sonnenfeld,
J.
A. (1999). Who
matters to Ceos? An investigation
of stakeholder attributes and salience, corpate performance,
and Ceo values. Academy of management
journal, 42(5), 507-525.
Albinger, H. S., & Freeman, S. J. (2000). Corporate social performance and attractiveness as an employer to different job seeking populations. Journal of Business Ethics, 28(3), 243-253.
Allsopp, C., & Vines, D. (2015).
Monetary and Fiscal Policy in the Great Moderation and the Great Recession. Oxford Review Of Economic Policy, 31(2), 134-167.
Boulouta, I. (2013). Hidden connections: The link between board gender diversity and corporate social performance. Journal of Business Ethics, 113(2), 185-197.
Campbell, J. L. (2007). Why Would Corporations Behave in Socially Responsible Ways? An Institutional Theory of Corporate Social Responsibility. The Academy of Management Review, (3), 946.
Chin, M. K., Hambrick, D. C., & Treviño, L.
K.
(2013). Political
Ideologies of CEOs The Influence of
Executives’ Values on Corporate Social Responsibility. Administrative Science Quarterly, 58(2),
197-232.
Chun, J. U., Yammarino, F. J., Dionne, S. D., Sosik, J. J., & Moon, H. K. (2009).
Leadership across hierarchical levels: Multiple levels of management and multiple levels of analysis. The Leadership Quarterly, 20(5), 689-707.
Davidson, R., & MacKinnon, J. G.
(1993). Estimation and inference in econometrics. Journal of Applied Econometrics, 10(3), 339–341.
Davidson, R., & MacKinnon, J. G. (1993).
Estimation and inference in econometrics.
Donaldson, T. & Preston, L. E. (1995). The stakeholder theory of the corporation: Concepts, evidence, and implications. Academy of management Review, 20(1), 65-91.
Erhardt, N. L., Werbel, J. D., & Shrader, C.
B. (2003). Board of director diversity and firm
financial performance. Corporate
governance: An international review, 11(2),
102-111.
Finkelstein, S.,
& Hambrick, D.
C. (1996). Strategic
leadership: Top executives
and their effects on organizations. West Publishing Company.
Flora, D. B., & Curran, P. J. (2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological methods, 9(4), 466.
Franklin, J. L. (1975). Down the
organization: Influence processes across levels of hierarchy. Administrative Science Quarterly,
20(2) 153-164.
Godfrey, P. C., Merrill, C. B., & Hansen, J. M. (2009). The
relationship between corporate
social responsibility and shareholder value:
An empirical test of the risk management hypothesis. Strategic Management Journal, 30(4), 425-445.
Grupo Bimbo (2013). Informe De Responsabilidad Social. Desempeño
En Responsabilidad Social. Recuperado de http://www.grupobimbo.com/informe/ Bimbo-Informe-Anual-2013/dm.html
Harjoto, M., Laksmana, I., & Lee, R. (2015). Board diversity and corporate social responsibility. Journal of Business Ethics, 132(4), 641-660Holgado–Tello, F. P., Chacón–Moscoso, S., Barbero–García, I., & Vila–Abad, E. (2010). Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables. Quality & Quantity, 44(1), 153-
166.
Ioannou, I., & Serafeim,
G. (2012). What drives
corporate social performance? The role of nation-level institutions. Journal of International Business Studies,
43(9), 834-
864.
Jackson, S. E., May, K. E., &
Whitney, K. (1995). Understanding the dynamics of diversity in decision-making
teams. Team effectiveness and decision
making in organizations, 204-261.
Jensen, M. C. (2002). Value Maximization, Stakeholder Theory, and the Corporate Objective Function. Business Ethics Quarterly, 12 (2). 235-256.
Kim, C. J. (2010). Dealing with endogeneity in regression models with dynamic coefficients. Now Publishers Inc, 3(3).
Kouamé, S., Oliver, D., & Poisson-de-Haro,
S. (2015). Can emotional differences be a strength? Affective diversity and managerial decision performance. Management Decision, 53(8), 1662-1676.
KPMG. (2013). Survey of Corporate
Responsibility Reporting. KPMG International Cooperative
Larrieta-Rubin de Celis, I., Velasco-Balmaseda, E., de Bobadilla, S., Alonso-Almeida, M., & Intxaurburu-Clemente, G. (2015). Does
having women managers lead to increased gender
equality practices in corporate social responsibility?. Business Ethics-A European Review, 24(1), 91-110.
Malone, T. W. (1997). Is empowerment
just a fad? Control, decision making, and IT. MIT Sloan Management Review, 38(2), 23.
Manner, M. H. (2010). The impact of CEO
characteristics on corporate social performance. Journal of
business ethics, 93(1), 53-72.
McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting interactions and moderator effects. Psychological bulletin, 114(2), 376.
McGuire, J., Dow, S., & Argheyd, K.
(2003). CEO Incentives and Corporate Social Performance. Journal of Business Ethics, 45(4).
341-359.
Mc Williams, A., & Siegel, D. (2001). Corporate social responsibility: A theory of the firm perspective. Academy of management review, 26(1), 117-127.
Miller,
D. (1987). Strategy
making and structure: Analysis and implications for
performance. Academy of management
journal, 30(1), 7-32.
Mitchell, R. K., Agle, B. R., & Wood, D. J. (1997). Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts. Academy of management review, 22(4), 853-886.
Mizon, G. E. (1995). A simple message
for autocorrelation correctors: Don’t. Journal of
Econometrics, 69(1), 267-288.
Muller, M. (2013). Nestlé baby milk scandal has grown up but not gone away. Recuperado de https://www.theguardian. com/sustainable-business/nestle-baby- milk-scandal-food-industry-standards
Murray, A. I. (1989). Top management group heterogeneity and
firm performance. Strategic Management
Journal, 10(1), 125-
141.
Nestlé. (2013). Nestlé baby milk scandal has grown up but not gone away. Retrieved from http://www.theguardian.com/sustainable-
business/nestle-baby-milk-scandal-food-
industry-standards
Neter, J., Wasserman, W., & Kutner, M.
H.
(1989). Applied regression models.
Homewood, IL: Irwin.
Nielsen. (2014). Global consumers are willing to put their money where their heart is when it comes to goods and services from
com- panies committed to social responsibility. Retrieved from http://www.nielsen.com/us/ en/press-room/2014/global-consumers- are-willing-to-put-their-money-where-their-
heart-is.html
Orij, R. (2010). Corporate social
disclosures in the context of national
cultures and stakeholder theory. Accounting,
Auditing & Accountability Journal, 23(7), 868-889.
Oshagbemi, T., & Gill, R. (2004). Differences
in leadership styles and behaviour across hi- erarchical levels in UK organisations. Lead-
ership & Organization Development Jour- nal, 25(1), 93-106.
Rangan, K., Chase, L., & Karim, S. (2015). The truth about CSR Most of these program aren’t strategic-and that’s OK. Harvard Business Review, 93(1-2), 41-49.
Reger, R. K. (1997).
Strategic leadership: top executives and their effects on
organizations. Academy of management
review, 22(3), 802-805.
Richard, O., & Shelor, R. (2002). Linking top management team age heterogeneity to
firm performance: juxtaposing two mid- range
theories. International Journal Of Human
Resource Management, 13(6),
958-974.
Rigdon, E. E. & Ferguson Jr, C. E. (1991). The performance of the polychoric correlation coefficient and selected fitting functions in confirmatory factor analysis with ordinal data. Journal of Marketing Research, 28(4), 491-497.
Steers, R. M. (1977). Organizational effective- ness: A behavioral
view. Michigan: ilustra- da, Goodyear Pub. Co.
Tetteh,
V. A. (2015). Diversity in the Workplace (Business). Research Starters: Business (Online Edition)
Thomas, R. R. (2011). The management of
workforce diversity: A continuing evolution. Employment Relations Today (Wiley), 38(3), pp.1-9.
Doi:10.1002/ert.20349
Torres, A., Bijmolt, T. H., Tribó, J. A., & Verhoef,
P. (2012). Generating global brand equity
through corporate social responsibility to key stakeholders. International Journal of Research in
Marketing, 29(1), 13-24.
Turban, D. B. & Greening, D. W. (1997). Cor- porate social performance and organiza- tional attractiveness to prospective em- ployees. Academy of management jour- nal, 40(3), 658-672.
Waddock, S. A., & Graves, S. B. (1997). The corporate social performance-financial performance link. Strategic management journal, 18(4), 303-319.
Wood, D. J. (1991). Corporate Social
Performance Revisited. The Academy of
Management Review, 16(4),
691-718.