Amsterdam Business School
Culturally diverse boards and the influence on sustainability performance
Name: Charlotte Pauline Kraaijenbrink Student number: 10701796
Thesis supervisor: ir. drs. A.C.M. de Bakker Date: June 8th 2017
Word count: 12.008
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
Statement of Originality
This document is written by student Lotte Kraaijenbrink 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
The objective of this study is to assess whether culturally diversity in a board of directors improves sustainability performance. In order to answer the research question the data of the American S&P 500 companies during the period 2012-2014 was used. The researched literature showed results that board diversity improves firm performance and financial reporting and indicates that board diversity would increase sustainability reporting and performance. Furthermore researched was whether collective culture representatives on the board would lead to improved sustainability performance. As the previous mentioned literature showed, results showed that such boards more invested more in sustainability.
Hofstede’s theory of collectivist vs. individualistic cultures implies that collective cultures tend to care more about the group, while individualistic cultures are more self-centered (Hofstede, 1980). This would result in collective cultures caring more about environment and sustainability, as this would be viewed as for the whole community. This has been put together with the theory that board diversity leads to better decision-making and finding solutions to dynamic situations (Amason, 1996). This leads to the question whether more culturally diverse boards improve sustainability performance.
We used the Newsweek Green rankings for the sustainability score given to firms.
The results of this study are not significant. The limitation lies in the low board diversity in America and the small sample size. Therefore, future research should be performed with a larger sample size and representation where board diversity has increased to an acceptable level.
Contents 1 Introduction ... 6 2 Theoretical Framework ... 8 2.1 Sustainability ... 8 2.2 Diversity ... 10 2.3 Collectivism vs. Individualism ... 11 2.4 Hypotheses Development ... 13 3 Methodology ... 15 3.1 Board diversity ... 15
3.2 Changes in board composition ... 18
3.3 Collective vs. Individualistic ... 18 3.4 Used Variables ... 19 3.5 Modified-Jones Model ... 20 4 Data ... 23 4.1 Sources ... 23 4.2 Descriptive Statistics ... 24 4.2.1 Correlation Matrix ... 27 4.2.2 Multicollinearity ... 29 5 Results ... 30 5.1 Results Hypothesis 1 ... 30 5.2 Results Hypothesis 2 ... 32 5.3 Results Hypothesis 3 ... 34 5.4 Summary Results ... 37 6. Conclusion ... 38 6.1 Conclusion of Research ... 38
6.3 Recommendations future research ... 39 7. References ... 40 Appendix ... 47
1 Introduction
With this paper, we want to examine if the cultural composition of the board of directors influences sustainability performance in the annual financial statements. We examine specifically whether culturally diverse boards are more inclined to improve sustainability performance than boards of directors with predominantly the same cultural background. Finally, this paper will relate the different composition of boards to the background of their culture, and whether this is collective or individual. Cultural background is relevant as a collectivist culture intends to give more attention to group responsibilities than individualist cultures. Coming from Hofstede’s work (1980), individualist countries differ from collectivist countries in basic beliefs, individualist countries are predominantly focused on self, and ego. While collectivist countries are engaged in the community and are group-oriented.
The board of directors has a significant role as they exercise control over top management (Kose and Senbet, 1998). Research has proven that board diversity has many advantages. According to Amason (1996), diverse groups, in a research on group-decision making, have more ideas, different perspectives and are more creative in finding solutions in dynamic situations.
This research follows up on the study of Feijoo et al (2013), who discovers that when there are 3 or more women on a board, the level of CSR (corporate social responsibility) in reporting increases, and the study of Marimthu (2008), who finds that culturally diverse boards improve financial performance. Contradicting research has been done by Pletzer et al (2015), who finds that gender diversity is not related to financial performance. First, to define diversity, this paper will focus on demographic diversity of culture. We will use Marimuthu’s definition: that of representation of different countries and ethnic differences in the board of directors (Chinese, Latin, Indian, African and others). This research will try to show that culturally diverse boards are expected to strive for better sustainability performance, especially when collective cultures are on the board. This research will therefore consider the history of sustainability performance in firms to assess whether when the board changed composition to more culturally diverse, sustainability performance increased..
To further understand the relation between sustainability performance and diverse boards, this research will relate the different ethnic backgrounds and whether they come from collective cultures, or individual cultures. The individualistic or collectivistic orientation of a country has been found to influence social behavior. More precise, person-level traits of individualism or
and Shrum (1994, 2001). Therefore, supporting the idea that collectivism countries prefer sustainability to individualism countries.
With this research, we try to answer the following question: Are culturally diverse boards positively associated with affecting sustainability performance? For the purpose of this study, we use the Hofstede’s theory. First, we assume that firms having a more culturally diverse board will perform better on sustainability than boards without cultural diversity. Second, we expect that culturally diverse boards with races from collective cultures will perform better on sustainability than culturally diverse boards from individualistic cultures. Finally, we expect firms that change directors and hire a culturally diverse member will change their sustainability performance in the following year. This research contributes to prior literature in several ways. First, it contributes how sustainability-performance changes as culturally diverse members are added to the board. This paper fills the gap by looking at culturally diverseness as variable. Furthermore, it will relate cultural background to collectivism and individualistic cultures and the cultures perspective on the environment. This research will collect data from the U.S. companies.
The remainder of this paper is organized as follows. In the second section, we discuss why sustainability and diversity, and why reporting and elaborating on these issues is beneficial for the firm and their reputation and how they are related. What influence the board of directors has and how different cultures affect this. In the third section, we discuss the research methodology. Subsequently we discuss the methodology chosen and furthermore this section will include the data, descriptive statistics and the results. Finally, in the fifth section we will present a conclusion, present the limitations of the paper and suggest further research.
2 Theoretical Framework
This section consists of a literature review concerning three research areas that are linked in this paper, namely diversity, sustainability and the collective or individualistic nature of cultures. The first part of the literature review consists of an examination of past and current research on sustainability, why sustainability has gained popularity and why it is important also and how does sustainability add value to a company? Last, what influence the board of directors has currently on sustainability. In order to develop a meaningful hypothesis, it is crucial to have a basic understanding what sustainability is and what it entails. The second part deals with literature concerning diversity and previous research done in this area in relation with sustainability. How diversity adds value and what do we understand when we say a board is culturally diverse. Last, we will relate cultural background more to Hofstede’s theory of collectivism and individualism and the relation to sustainability.
2.1 Sustainability
“Today humanity uses the equivalent of 1.5 planets to provide the resources we use and absorb our waste. This means it now takes the Earth one year and six months to regenerate what we use in a year. Moderate UN scenarios suggest that if current population and consumption trends continue, by the 2030s, we will need the equivalent of two Earths to support us. And of course, we only have one” (the Global Footprint Network, 2015).
The Sustainability Challenge is becoming more and more popular within organizations, and in everyday life (United Nations Climate Change Conference, 2009). Stakeholders increasingly demand that an organization releases a sustainability report. Corporate social reporting is a mechanism through which the environmental performance and social effects are communicated to stakeholders. The public closely follows companies, especially when they operate in countries where there are significant social inequalities in labor, ethics and environmental issues that could be caused by their business. This has increased the number of reports and information disclosed in the financial statements. Subsequently, “Corporate sustainability is a business approach that creates long-term shareholder value by embracing opportunities and managing risks derived from economic, environmental and social developments” (Dow Jones Sustainability Indexes, 2015). Reporting on sustainability might not immediately change the way businesses cares about the environment and their social responsibility but it will shed light on the way they evaluate the social
becoming important to stakeholders, firms will feel the pressure to change their behavior. Kates, Parris & Leiserowitz (2005) defines sustainability, as “humanity having the ability to make development sustainable to ensure that it meets the needs of the present without compromising the ability of future generations to meet their own needs.” This definition is the most popular definition and is also used by the United Nations World Commission on environment.
In these present times with our environment becoming of more importance every day and increasing awareness regarding the implications of misusing our environment, environment and sustainability is no longer an issue for the future. People often care more about the economy and in politics the environment is regarded, as something not interrelated to humans. While they are interrelated, nowadays more than ever our economy depends on the environment and society (Giddings 2002; White 2016). The KPMG survey of Corporate Responsibility Reporting in 2013 found that most of the 250 world’s largest companies report on Corporate Responsibility and that 62% of companies report on sustainability. Corporate sustainability performance has become such an issue that all the big four companies (KPMG, EY, PwC, and Deloitte) have started special consultancy branches just for sustainability.
The main drivers of sustainability performance are also gaining popularity; the level of investor support has doubled over the last decade, the growing importance of stakeholder relations, the rise of the information technology and the pursuit of alternative business strategies and furthermore the legislative and regulatory pressure are all signs that the topic is becoming more relevant than ever (Tonello, 2010).
Organizations realize the additional value to their transparency by reporting on issues like sustainability and furthermore realize the importance of focusing on not only financial performance but also their non-financial performance. In addition, Feijoo et al., (2013) have argued that reporting on sustainability is a vital part of strategic management. “The benefits of sustainability reporting go beyond relating firm financial risk and opportunity to performance and establishing license to operate. Sustainability disclosure on the relaxed performance can serve as a differentiator in competitive industries and foster investor confidence, trust and employee loyalty. Analysts often consider a company’s sustainability disclosures in their assessment of management quality and efficiency, and reporting may provide firms better access to capital” (EY, value of sustainability reporting, 2016). Sustainability reporting can entail many issues besides financial reporting in a company. Sustainability includes but is not limited to: workplace safety and operational integrity, environmental issues, community engagement, corporate citizenship, human rights, privacy & data security, political contributions and lobbying (Frias-Aceituno, 2013).
The SEC picked up on the growing importance and the legislative pressure added, in 2010, was that the SEC added guidance to public firms, to help with disclosure requirements in the changing climate matters (Federal register, vol. 75), and more recently more regulations are added, minor, but in the U.S. companies must disclose in 10-K fillings risk of damage due to physical environment such as changes in weather and rises of the sea level (IRRC, 2013).
Further specific requirements remain a ‘guide’.
2.2 Diversity
Currently the culturally diverse board members in the largest 200 companies in the U.S. are about 15% (Rhode, 2015). “Diversity of age, gender, ethnicity and cultural background is proving to favor new exchanges of ideas and ensures that multiple perspectives are represented as the board engages in strategic discussions and makes long-term business decisions. Recent studies haven proven a positive association between board diversity and both innovation and corporate reputation (Alliance for board diversity, 2005 & Miller and Triana, 2009).
Research has found that demographic diversity in the boardroom positively relates to firm performance (Carter et al., 2003; Erhardt et al., 2003). The trend among diversity is huge, not just because several countries have implemented a minimum or require a statement when you don’t meet the requirements. In December 2009, the SEC added a requirement for firms to report on board diversity. It requires firms to disclose in the annual statements whether diversity is taken into account when selecting the board of directors, if there are any policies in place and how they consider diversity and whether it is effective (IRRC, 2013).
Diversity is essential in today’s economy as the issues faced today are dynamic and complex. Research proves that a diverse board will handle these complex issues much better and will lead to effective corporate governance. Support for the idea that board cultures and gender diversity should be related to firm innovation and effective governance may also be found in the behavioral theory of the firm (Cyert and March, 1963). This behavioral theory of the firm entails that the extensiveness of the search and decision-making processes can influence innovation in organizations. Gender and cultural diversity will evaluate choices more meticulously as they propose more perspectives, a greater variety of ideas and therefore have more information available. Research has supported this statement by empirical research done on group-decision making. Research shows that heterogeneous groups make higher quality decisions especially when it comes down to complex tasks (Amason, 1996).
Previous studies have been done in this field, when it comes to board diversity in the sense of gender (Feijoo et al, 2013). Board diversity is important as “diversity is seen as providing boards with access to a wider pool of competencies, experiences and perspectives, which should be beneficial to board effectiveness” (Firoozi, 2016). Guthrie (1990), finds that when foreigners are present in the majority stakeholders it increases the reporting on CSR. However, Frias-Aceituno & his colleagues (2013) find no significant positive effect when foreigners are present in boards. According to Zhang (2012), strategic decision-making will improve when the board becomes more demographically diverse. Previous research done on demographic diversity in the management team shows that this leads to more innovative solutions (Bantel, 1994; Bantel and Jackson, 1989; Wiersema and Bantel, 1992). Furthermore, does Kosnik (1990) show that the different demographic backgrounds lead to different views and norms and values and therefore different solutions to strategic decisions. Based on that research, we can conclude that demographic diversity will allow for more cognitive abilities and more innovative solutions. Ibarra (1993) finds that demographically diverse boards can increase stakeholder’s relations and social performance as minorities and women have a broader network. This broader network is beneficial for the firm as it allows for understanding, and more help from their connections (Beckman and Haunschild, 2002).
Marimuthu (2008) finds in his research that ethnic diversity in top-level management will increase financial performance and Simons & Pelled (2008) find that diversity in the workforce increases financial performance. Additionally diversity has proven to have a positive effect on board performance. Multiple researches find and as previously mentioned, diversity adds a unique perspective (Helms et al., 2004; Van der Walt, Ingley, Shergill, & Townsend, 2006). A more diverse board will possess more information, as private information is present, especially the private information coming from the board member’s culture. This information can be shared with other board members. Therefore, all together a more diverse board will make better decisions, as more information is present. Subsequently, Adams & Feirrera (2009) find that a board will gain independence when they increase in diversity and therefore will increase their monitoring potential.
2.3 Collectivism vs. Individualism
When looking at diversity in boards we will look at different cultures and ethnic backgrounds, as culture is not the same as identity. An Afro-American female might identify herself with America after living there her entire life, but the culture will, in most cases, be that of Afro-Americans.
Identity is the question of where do we belong, while culture has to do with values and beliefs (Hofstede 2001). Your geographic origin will influence your value system and the consequence is that your family pattern and social emphasis will have a certain structure and functioning. Hofstede’s theory of collectivist & individualist cultures originates from 1980 and later updated in 2001, and is still relevant today. As mentioned earlier some countries are collectivists meaning they are more group oriented, while individualistic countries are more ego-oriented. “Collectivists are closely linked individuals who view themselves primarily as parts of a whole, be it a family, a network of co-workers, a tribe, or a nation. Such people are mainly motivated by the norms and duties imposed by the collective entity. Individualists are motivated by their own preferences, needs, and rights, giving priority to personal rather than to group goals” (Triandis, 1995). Drawn from Hofstede’s (2001) list examples of individualistic countries are: United States, Australia, Canada, The Netherlands, Germany, South Africa and Sweden. Examples of collective countries are: Indonesia, India, Brasil, Mexico, China, Russia and Egypt. The line was drawn at 61-70% individualism rating for a complete overview see appendix A.
This research does not attempt to proof whether collectivist cultures are more environmentally oriented than individualist cultures. Based on previous research we will assume that collective cultures tend to care more for sustainability than individualistic cultures. McCarty & Shrum (2001) found that collectivist cultures are more likely to recycle and have positive attitude towards recycling and that they are more willing to help others. This was in contrast with individualist culture countries. The same goes for resource conservation (Dunlap and Van Liere, 1984) and green purchase behaviors (Yeonshin and Sejong 2005). Drawing from these researches we find that collectivist cultures tend to care more for resource conservation and green purchase behavior than individualist cultures. Furthermore, Schultz (2002) finds in his study that people from the United States, an individualistic country, are less concerned with environmental issues than people from collectivistic cultures. Björn et al. (2015), find a significant result that collective countries care more for brand image. Additionally, Shen et al. (2015), find in their study that China is more willing to manage relationships with retailers, while the U.S. cares more about their own interests and are even willing to go as far as proactively make a bargain. Emphasizing that collectivist cultures tend not to just care more about the environment but tend to care more too when it comes down to other topics concerning sustainability like: operational integrity and the community.
Therefore, concluding that when sitting in a board of directors the board members whom possess a collective culture from their demographic background will more often choose for more sustainable options and reporting. While the board members with an individualistic background
will tend to care less about issues concerned with sustainability. Based on the previously mentioned investigations.
2.4 Hypotheses Development
Considering the above discussion, this research will use demographic diversity of the board of directors and their influence on sustainability performance. Whether sustainability performance improves when more culturally diverse members are added to the board. Relating culturally diverseness to demographic background, and when collective or individual cultures are added to the board, the sustainability performance improves.
First, we will look at sustainability reporting and cultural board diversity in general, predicting that when more culturally diverse board members are present in the board of directors the quality of the sustainability performance will increase. Boards play an important role as they decide on strategy and CSR policy (Adams, 2010). Leaning on the research that more diverse groups are more innovative, have more perspectives and have more information available. These groups will see the urgency of sustainability performance. According to Nidumolu (2009), sustainability is innovation, when firms become more eco-friendly firms, they become more competitive. Taking into consideration that the previous research mentioned that sustainability performance increases the stock price and overall performance (The Conference Board, 2012).
Consistent with these findings, we predict that there will be a positive association between boards with more culturally diverse members and the quality of sustainability performance. Therefore, we developed the following hypothesis about culturally diverse members and sustainability performance.
H1: There is a positive association between culturally diverse boards and sustainability performance
To further assess whether sustainability performance improves when board diversity increases, we will compare the sustainability reports of the previous years when there were less cultural diverse members present on the board with sustainability reports of the years when there were more culturally diverse members present. To see whether changes in sustainability performance happened when culturally diverse members were added. As Blau already stated in 1997, diversity was the number of different statuses in which a population is distributed, and using Hillman et al., research from 2002 that finds that race is considered a proxy for different perspectives individuals will deliver to the firm, and finds that demographic background influences strategic choices in an organization. Which is caused by the diversity in background and culture. Therefore, as mentioned
earlier, should diversity in the board lead to different and distinctive idea’s that originate from their different backgrounds.
Subsequently, we predict that sustainability performance improves when culturally diverse members are added to the board, and that the sustainability performance will change compared to previous years. Therefore, we developed the following hypothesis about sustainability performance and the changes in the board composition.
H2: The change in composition of the board towards more cultural diversity has a positive association with sustainability performance improvement
Considering prior research that shows evidence that collectivist cultures tend to care more about the environment than individualistic cultures (Yeonshin and Sejong 2005; Dunlap and Van Liere, 1984). We argue that when culturally diverse members are from collective cultures they value sustainability more than members that are from individualistic cultures. Therefore, we predict that boards with more collective cultures present will perform better on sustainability. This resulted in the following hypothesis:
3 Methodology
In this section, the research method that has been used to answer the research question will be introduced. We will discuss the variables of interest along with the control variables. Finally, the regression models that will be used to regress sustainability performance on the variables of interest culturally diverseness and the form of culture, collective or individual, and the control variables will be presented and explained. This will be done on a hypothesis-by-hypothesis basis.
3.1 Board diversity
H1: There is a positive association between culturally diverse boards and sustainability performance
To test the first hypothesis, we will start with describing the dependent, following the independent variables and finally describe our control variables. Using a cultural diversity measure as the independent variable and sustainability performance as the dependent variable.
Dependent Variable
As this is about whether board cultural diversity influences performance on sustainability, this makes sustainability performance the dependent variable.
Firms sustainability performance is measured using the green rankings score from Newsweek. Newsweek scores the S&P 500 firms on eight indicators. The indicators are: Combined Energy Productivity, Combined Greenhouse Gas Productivity, Combined Water Productivity, Combined Waste productivity, Green revenue score, Sustainability Pay Link, Sustainability Board Committee and Audited Environmental Metrics. The total of these scores are added and come to a score of 0-1. Therefore the dependent variable is measured on a scale from 0-1, equal to the green percentage score. With the best firm being with the score closest to 1, and the worst firm with lowest score close to 0 or 0.
Independent Variable
The independent variable is the cultural diversity of the board. Miller et al. (2009) measure of board diversity will be introduced. To measure board cultural diversity we used the IRRC cultural grouping, which has mapped the following 4 cultural groups in the board of directors: Caucasian, Hispanic, African-American, and Asian.
The measure used is Blau’s (1977) index of diversity. BLAU= Blau index of diversity [" = 1 − &'(] , where H is the board demographic diversity measure, !
members in each category and ! is the different demographic categories in the company. Using the same grouping of cultures as previously stated by the IRRC. The same method is used as mentioned by Miller et al. (2009) to calculate board diversity. Where board diversity ranges from zero to 0.75. The score of 0.75 means that the board is very diverse, while a score of 0 means that the board is not diverse at all. The Blau index is commonly used to measure for diversity; it is the more ideal measurement for diversity and the variations within the board of directors (Harrison and Klein, 2007). Blau’s index of diversity meets the four criteria for good measurement: it includes complete homogeneity (zero), and it includes larger numbers for more diversity and ignores negative values, and lastly it is not unbounded (Harrison and Sin, 2006). In this study we will use Blau’s index of diversity measurement.
Control variables
We must control for factors that previous literature has indicated to affect sustainability reporting and performance. Further using a model designed by Al-Shaer and Zaman (2016), who have done a similar study between gender diversity and sustainability reporting, we control for board size, number of meetings and number of independent directors on the board of the directors. Previous studies have showed that larger boards tend to report more sustainability (De Villiers et al., 2011 and Denies et al, 2015), they find that more members will see it as a priority. Therefore including board size to control for the influence on sustainability reporting.
Subsequently, transparency and quality of the reporting is influenced by the amount of board meetings held, and impacts the effectiveness of decision making in the board (Laksmana, 2008 and Kanagaretnam et al., 2007). Therefore including board meetings, as a control variable.
Arora & Dharwadkar indicate that independent directors on the board will increase time spent on sustainability reporting and reduces agency costs (Arora and Dharwadkar, 2011). Subsequently evidence has been found for the positive relationship between the number of independent directors and integrated reporting (Frias-Aceituno, 2013). Moreover previous research has shown that outside directors improves corporate governance (Zhang, 2012). Therefore included is independence of the board, and measured as the proportion of independent directors present in the board.
Other control variables used in the model are underlying reporting incentives and reporting behavior, those are based on characteristics of the firm. These reporting incentives and behaviors are based on the study of Daske et al. (2013). Using the three proxies made by Al-Shaer & Zaman (2016), that most accurately describe the factors that influence reporting incentives and reporting
behavior. The first proxy is reporting incentive, which is based on profitability, financial leverage, firm size, growth opportunities, ownership concentration, and internationalization. These firm characteristics are measures through: log of total assets, return on assets, debt to total assets ratio, book-to-market ratio, percentage of closely-held shares and foreign sales over total sales. Al-Shear & Zaman used factor analysis to cluster these reporting incentives effects, to get one single factor. This study will use the logarithm of total assets, return on assets, debt to total assets ratio and book-to-market ratio. As we consider these as the most important and use these as separate control variables.
The second proxy variable is reporting behavior (RB) this control variable is based on earnings management. Daske et al. (2013), shows in their study that firms with less earnings management have better reporting behavior. Earnings management is measured using the Modified Jones model from Jones (1991) and Dechow (1995). Lower values mean that earnings management is lower and that the reporting is more transparent. We have chosen to alter from Arora & Dharwadkar model here, as based on research conducted by Dechow et al. (1995) the Modified Jones model is the best model for detecting earnings management and produces the least amount of error. For a description of this variable we refer to 3.5
The third control proxy is external reporting environment (RE). The external reporting environment is measured through counting the number of analysts that are keeping track of the firm. The more analysts that are keeping track of the firm, the higher the value of external environment (Daske et al., 2013).
Lastly we have added year dummies to differentiate for different years per company. We have included the years 2012-2014, therefore we have 2 dummy years: 2013 and 2014. 2012 is the default year.
To test for hypothesis 1 a linear regression analysis is used and we state that board diversity is positively related to sustainability performance. We will regress sustainability performance on the board diversity and the control variables. To examine whether board diversity is positively related to sustainability performance, the following equation will be used:
!"!#$% = '( + '+,-." + '/ ,01!23$+ '4,01251+ '6,017$$8 + '9log 8. + '=#0. +'>18. +'?,87 +'@#, +'+(#$+ ABCD1EFFGBH + I
For a description of each variable see section 3.4
3.2 Changes in board composition
H2: The change in composition of the board towards more cultural diversity has a positive association with sustainability performance improvement
To test for hypothesis two, we will have to measure multiple years of sustainability performance and measure the board composition over these years. Excluding the financial crisis and adjusting for financial reports available we will the Newsweek reports from 2014-2016, these are based on the financials from 2012-2014. Therefore using financial information from 2012-2014. As previously mentioned, to be included in the sample firms had to be listed in the S&P 500. We state in our previous hypothesis that sustainability performance should change in relation to the board composition changes. For this reason, we compare board composition change to sustainability performance change for year to year. The regression equation we will use for testing this hypothesis is similar to the one used for testing hypothesis 1. However the dependent variable SUSREP from the first regression equation is replaced by the change in value ∆ SUSREP. Analogous the independent variable BLAU is replaced by the ∆ BLAU. The following equation is therefore formulated:
∆"#"$%& = ()+ ∆(,-./# + (0 -12"34%+ (5-12362+ (7-128%%9 + (:log 9/ + (>$1/ +(?29/ +(@-98 +(A$- +(,)$%+ BCDE2FGGHCI + J
For a description of each variable see section 3.4
3.3 Collective vs. Individualistic
H3: There is a positive association between collective cultures in the boards and sustainability performance To test for hypothesis three, we will use an independent variable, based on Hofstede’s research on collective vs. individualistic cultures and assign a score to each ethnic group. As we take Blau’s index of diversity as a starting point we will have 4 ethnic groups. Every ethnic group will receive a score from most collective to least collective culture. Using Appendix A and the research done by Freeberg & Stein (1996), the Asian ethnicity group is seen as most collective and will receive a score of 3, followed by Hispanics who receive a score of 2, Afro-Americans are hard to determine from the appendix, therefore drawing from Acevedo (2003) study we find that Afro-Americans have a combination of both, and therefore receive score 1, but are more collective than Caucasians. Caucasians are dominantly an individualistic in their cultural aspects and will therefore receive a score of zero. We expect that when the score moves towards 3, the sustainability performance
improves. All control variables remain the same. !"#$ =&'(*) . !"# , is the collectivism score of board member ! , ! is the number of board members.
!"!#$% = '( + '+,-.! + '/ 0-1!23$+ '40-1251+ '60-17$$8+ '9log 8= + '>#-= +'?18= +'@087 +'A#0 +'+(#$+ BCDE1FGGHCI + J
For description of the variables see section 3.4
3.4 Used Variables
The following variables are used to test our hypotheses. Table 1- Variable Description
Dependent Variable
SUSREP Sustainability performance using the Newsweek Green ranking score of 0-100% ∆ SUSREP Change in sustainability performance from the Newsweek Green ranking score
of 0-100% Independent Variables
BLAU Board cultural diversity is measured using
BLAU= Blau index of diversity [Η = 1 − Σ&'2
] where !" is the percentage of board members of culture 1
∆ BLAU Change in board cultural diversity measured using BLAU= Blau index of diversity [Η = 1 − Σ&'2 ] where !" is the percentage of board members of culture 1
COLS Collectivism score using a 0-3 scale [3=Asian, 2=Hispanic, 1=Afro-American, 0=Caucasian]. Cols= !"#%$ : !"# , is the collectivism score of board member ! , ! is the number of board members
Control Variables
BODSIZE Total number of directors on the board
BODMEET Number of board meetings per year LOG (TA) Logarithm of total assets
ROA Return on assets
DTA Debt to total assets ratio BTM Book-to-market ratio
RB Reporting Behavior, the absolute value of discretionary accruals RE Reporting Environment, total analysts following the company. Year
Dummy1
By dummy=1, company year 2013
Year Dummy2
By dummy=1, company year 2014
Error Term
! Error Term
3.5 Modified-Jones Model
For control variable reporting behavior (RB) we will be using the ‘Modified-Jones’ model (Dechow et al., 1995). This model is often used for determining the absolute discretionary accruals (Healy, 1985). The original Jones model has been adjusted as the shortcomings were too prevalent, and earnings management was still possible. Therefore the model was modified in 1995, and was called the Modified-Jones model, to test for earnings management and determine the reporting behavior. In order to determine the discretionary accruals we will have to determine the total accruals. Total accruals is made up to the non-discretionary accruals (NDA) and discretionary accruals (DA). In order to calculate total accruals we will use the formula mentioned in Healy (1985) and Dechow (1995).
TA# = (∆CA#- ∆CL#- ∆Cash#+ ∆STD#- DEP#) /A#45
The variables are explained in the table 2 below. After calculating the total accruals, the year specific parameters have to be calculated. To determine the year specific parameters (α$, α&, α') an OLS regression has to be performed. This is determined by dividing the total accruals by the
TA#/A%&' = α'(1/A%&') + α/( ∆REV#/A%&') + α4(PPE#/A%&') + ℰ# The variables are explained in the table 2 below. The regression is performed per year;
subsequently the calculated coefficients are used to estimate the non-discretionary accruals. The following formula is used:
NDA$ = α'( 1
A$*') + α.( ∆REV$- ∆REC$) + α5(PPE$)
After performing the regression the total accruals and the non-discretionary accruals are known therefore we can calculate the discretionary accrual. This is done by the previously mentioned method of:
DA = TA − NDA
We need the absolute value of the discretionary accruals DA , in our regression model as the control variable reporting behavior (RB).
Table 2- Variable Description Modified Jones Model TA# Total accruals in year t
∆CA$ Mutation in current assets of the firm in year t ∆CL$ Mutation in current liabilities of the firm in year t ∆Cash& Mutations in cash of the firm in year t
∆STD% Mutation in short term debts of the firm in year t DEP$ Depreciation and amortization of the firm in year t A"#$ Total assets of the firm in year t-1
!", !$, !% Year specific regression coefficients estimates
PPE# Gross fixed assets of year t divided by the total assets of the previous year
DA Absolute discretionary accruals
∆REV% Mutation in revenue of year t to the previous year ∆REC% Mutation in receivables in year t to the previous year !" Error term
4 Data
In this section the data used as sample selected for this research will be discussed, together with the sources. Finally the descriptive statistics will be presented.
4.1 Sources
Data was collected from several sources, to collect the data for sustainability performance we used the Newsweek Green Rankings. This ranks the largest 500 companies in the United States on sustainability performance. For our dependent variable data we used firms in the S&P 500 in the US. The S&P 500 was used as these firms are the best in the industry and other studies researching races of board members prefer using the S&P 500 (Miller et al. 2009; Zhang, 2012). To be included in the final sample the firms should have been listed in the S&P 500 and for the 3 consecutive years in the period 2012-2014 and in the largest 500 companies as selected by the Newsweek. It is important to note that the Newsweek Green Rankings are based on the performance data of two years before, this means that the rankings from 2014 are based on the performance of the firms in 2012. The Newsweek Green Ranking was chosen because it is a credited source, acknowledged by multiple institutes like Governance & Accountability institute and the Global Reporting Initiative and is often seen as a reputational list (G&A institute, 2012). A second reason the Newsweek green ranking was chosen is because it supplies the most extended data. While the G&A institute and the Global Reporting Initiative have databases of their own, their most recent database is from 2013. The sample period extends from 2012-2014, as the data available only goes back until 2012. As mentioned earlier the firms in the Newsweek Green rankings are ranked on the basis of eight indicators.
From the Newsweek Green rankings 100 firms were selected for the sample, to avoid selecting just the 100 best ranked firms; we have randomly selected 25 firms from every ranking quartile in starting year 2012. Resulting in 100 randomly selected firms, we had to merge this with the presence of these firms in the S&P 500. This is where we lost 6 firms.
Our initial sample was limited to a 100 firms as the data needed for our sample is hand collected and analyzed. Due to our large amount of control variables and independent variable measure COLS we had to use several databases and were data was missing we had to hand collect by going through the financial statements of the specific company. This is another reason our sample was limited to a 100 firms, as having to collect more data for more firms would not have been viable in this time frame. We believe that the 100 firms will be sufficient to give significant results and is enough to validate our research.
For the dependent variable, cultural diversity of the board of directors, we retrieved data from Wharton Research Data Services (WRDS). With our selected sample of 100 firms we retrieved the data from the ISS (Institutional Shareholder Services) on director data. The data showed for all years 2012-2014 the ethnicity of the board members. The ISS has 6 ethnicities: Caucasian, Asian, Hispanic, Afro-American, Middle-Eastern and Indian. As Indian and Middle-eastern are not prevalent and are currently not yet official separate ethnicities, Indian was considered Asian and Middle-Eastern was considered Caucasian (USA TODAY, 2016).
For our control variables the WRDS database was used as well, except for data on board meetings per year. For this control variable we had retrieved data from the Stuart & Spencer Data on Boards. For some of our selected firms in the sample data were not available, we filled the missing data by looking up the financial statements. Some firms of our selected sample were lost due to the fact that they were not in the S&P 500 for 3 consecutive years. Leading to a final sample of 79 firms.
4.2 Descriptive Statistics
Before the regression analysis was performed, the control variable RB (reporting behavior) had to be determined by estimating year specific parameters to find our discretionary accruals. This is based on the described literature in 3.5. By using an OLS regression we found our specific parameters. The estimated specific year parameters are presented in Table 4:
Table 3 – Observations included
Description Amount
Selected data of Newseek Green Rankings on sustainability reporting of the years 2014-2016 (2012-2014)
300
-/- Not in the S&P 500 18
-/- Not in the S&P 500 for 3 consecutieve years -/- Missing Data
27 18
Table 4 – specific year regression coefficients
2012 2013 2014
!" 0,539 1,134 0,876
!" 0,082 0,057 -0,040
!" 0,337 -0,228 -0,106
After the discretionary accruals were determined using the Modified Jones Model, these were adjusted to absolute discretionary accruals. Finally, with the absolute discretionary accruals determined, we had all data available that was required for this study. Table 5 presents the descriptive statistics on the dependent and control variables.
Table 5 shows that there are 237 observations. The mean of the sustainability score is around 34,75%. The Blau score per firm has a mean of 0,232, which is considerably low out of 0,75 (30,9%). Furthermore the standard deviation of the scores shows that there is a lot of variance to explain. The collectivity score mean is 37,2%.
We have tested for outliers on the dependent variable, as this is the most important one. We have found no outliers. The variable is skewed to the right, but further checking the z scores, no score was larger than 3. This is why we found no reason to winsorize our data.
Table 5- Descriptive Statistics
Variable Mean Median Std. dev. Max Min N
Dependent Variable SUSREP 34,75 32,7 20,033 84,4 0 237 ∆ SUSREP -0,443 0,5 14,032 50,80 -43,20 156 Independent Variables BLAU 0,232 0,219 0,164 0,580 0,0 237 ∆ BLAU -0,014 0 0,076 0,25 -0,28 156 COLS 0,272 0,231 0,234 1,167 0 237 Control Variables BODSIZE 10,911 11 1,899 6 17 237 BODIND 0,853 0,889 0,093 1 0,4 237 BODMEET 7,865 7 3,377 22 0,0 237 LOG (TA) 4,373 4,201 0,660 6,41 3,11 237 ROA 0,167 0,163 0,10 0,425 -0,165 237 DTA 0,616 0,575 0,202 1,243 0,115 237 BTM 0,450 0,338 0,351 1,943 0,004 237 RB 111,049 27,550 439,537 5330,84 0 237 RE 15,45 14 6,552 18 1 237 Year Dummy1 0,333 0 0,472 1 0 237 Year Dummy2 0,333 0 0,472 1 0 237
4.2.1 Correlation Matrix
In this section we will discuss the correlation matrix, we do this with the Pearson correlation coefficient. This will look at the correlation of the different dependent and independent variables. The correlation results from Pearson are between +1 and -1. +1 displays a positive linear relationship between the variables while -1 displays a negative linear relation. A high correlation is considered when the results are higher than 0.7 or lower than -0.7 (Field, 2013).
The results are presented in table 6 below.
The correlation matrix shows that there are no variables that strongly correlate with each other this indicates that there is no multicollinearity. There are several variables that are significantly related to one another. The correlation matrix indicates that for example, the BLAU score significantly relates to the sustainability score. Which would indicate that a higher BLAU score would lead to a higher sustainability performance score. Cautiously we can say that board diversity leads to better sustainability performance. There is no strong correlation between collectivism and sustainability performance, but when focusing on the significance score, we find that there is a significance of 0,05, and that a high collectivity score could lead to an improved sustainability performance. This would also be the case with our BLAU score that shows no high correlation with the collectivism score but a significance of (0,05) with the collectivism score. As mentioned in our previous literature we find that log of total assets and debt to asset ratio has a positive relation with sustainability performance (Al-Shear & Zaman). A significant relationship is also found with sustainability. The highest correlation score is that between Board Meetings and Log of total assets 0,485 this indicates the amount of board meetings is bigger in big companies. Furthermore we find that the BLAU score has a significant correlation with return on assets, book to market ratio, board independence and board size. Indicating that board diversity positively relates with the amount of board members on the board and the independence of these board members. All together there are no absolute high correlations between the dependent variables, therefore we can conclude there is no indication for multicollinearity. In the next section we will further analyze multicollinearity
Table 6 – Pearson Correlation Matrix
SUSREP ∆ SUSREP BLAU ∆ BLAU BOD SIZE BOD IND BOD MEET LOG (TA)
ROA DTA BTM RB RE COLS Year
Dummy1 Year Dummy2 SUSREP 1 -0,32*** -0,17** 0,15** -0,07 -0,2 0,15** 0,14** 0,03 -0,15** 0,03 0,05 0,05 -0,13** 0,04 -0,02 ∆ SUSREP -0,32*** 1 -0,07 -0,06 0,01 0,01 0,06 0,03 -0,07 0,04 0,23 0,01 -0,05 0,04 0,09 -0,01 BLAU -0,17** -0,07 1 0,21*** 0,03*** 0,13** 0,02 0,11* 0,12** 0,06 -0,18** -0,01 -0,59 0,19*** 0,01 -0,04 ∆ BLAU 0,15** -0,06 0,20*** 1 0,09 0,16** -0,07 0,03 -0,09 0,10* 0,04 0,04 -0,21*** -0,03 0,13** 0,04 BODSIZE -0,07 0,01 0,03*** 0,09 1 0,13** -0,01 0,19*** 0,07 -0,05 -0,18*** -0,05 0,07** 0,09* -0,02 0,033*** BODIND -0,2 0,01 0,13* 0,16** 0,13** 1 0,09* 0,25*** -0,09* 0,05 0,07 0,07 -0,07 0,13 0,04 0,035 BODMEET 0,15** 0,06 0,04 -0,07 -0,01 0,09* 1 0,49*** 0,08 -0,14** -0,01 -0,07 0,02 -0,15 0,03 0,044 LOG (TA) 0,14** 0,03 0,19* 0,04 0,19*** 0,25*** 0,49*** 1 0,13** -0,02 -0,20*** 0,09 0,13** -0,05 0,0 0,021 ROA -0,03 -0,07 0,12* -0,09 0,07 -0,09* 0,08 0,13* 1 -0,38* -0,65** -0,01 0,15* 0,07 -0,02 -0,038 DTA -0,15** 0,04 0,06 0,10* -0,05 0,05 -0,14** -0,02** -0,39*** 1 0,27*** 0,04 -0,06 -0,01 0,01 0,026 BTM 0,03 0,23 0,18* 0,04 -0,18*** 0,07 -0,01 -0,20*** -0,65*** 0,27** 1 0,07 0,02 0,12** -0,04 -0,131** RB 0,05 0,01 -0,01 0,04 -0,05 0,07 0,07 0,09 -0,04 0,04 0,07 1 0,08 0,06 -0,06 -0,075 RE 0,05 -0,05 0,59 -0,21*** 0,07 -0,07** 0,02 0,13 0,15** -0,06 0,02 0,08 1 0,09 -0,02 -0,153*** COLS -0,13** 0,04 0,19*** -0,03 0,09* 0,13 -0,15** -0,05** 0,07 -0,01 0,12 0,06 0,09 1 -0,01 -0,075 Year Dummy1 0,04 0,09 0,01 0,13** -0,02 0,04 0,03 0,02 -0,02 0,01 -0,04 -0,06 -0,02 -0,01 1 -0,5*** Year Dummy2 -0,02 -0,01 -0,04 0,04 0,03*** 0,04 0,04 0,02 -0,04 0,03 -0,13** -0,08 -0,15*** -0,01 -0,50*** 1
4.2.2 Multicollinearity
Multicollinearity is an analysis that can be done to discover which of the independent variables that are included in the regression are linearly related with all other independent variables together, and how much they relate. It is important to test for multicollinearity because including independent variables that relate strongly with one another can influence the regression results, and would impair our results (Wonsuk et al., 2014). To test for the relation and the amount of relation between the independent variables we use the Variance Inflation Factor (VIF). When the VIF score is higher than 5, we can expect that this variable will relate strong to other variables (O’brien, 2007). We can solve this by eliminating a variable, as the two variables will explain the same difference. When performing the test for multicollinearity we find that no independent variable is strongly related to other variables. The highest VIF score is 4,633 but bears no significance to doubt our results.
Therefore we can conclude that multicollinearity is not a problem, and our variables won’t bias the results. The results of our VIF test are presented in table 7.
Table 7 - Multicollinearity Variable VIF BLAU 1,181 BODSIZE 4,288 BODIND 4,633 BODMEET 1,506 LOG (TA) 1,742 ROA 1,947 DTA 1,218 BTM 2,082 RB 1,166 RE 1,266 Year Dummy1 1,791 Year Dummy2 1,846 ∆ BLAU 1,170 COLS 1,164
5 Results
In this section the regression results will be presented and explained. Three regressions were run to test whether the hypotheses are valid. In the previous section the descriptive statistics were presented, this section will focus on the results of the in section 2.4 formulated hypotheses.
5.1 Results Hypothesis 1
In this section we will test hypothesis 1 to see whether if there is a more culturally diverse board, sustainability performance will improve. We used sustainability performance as our dependent variable and the BLAU score as our independent variable for cultural diversity. In the mentioned theory we predicted that diversity has a positive effect in the board of directors, and therefore a positive effect on sustainability performance. Sustainability has continuous proved to be an emerging topic. Therefore we predicted that culturally diverseness would have a positive effect on sustainability performance and the following hypothesis was formulated:
There is a positive association between culturally diverse boards and sustainability performance
To be able to use the results from this hypothesis and the regression analysis, the research model must be valid. In order for the research model to be valid we analyzed the significance of the model and the F-score. Furthermore, we analyzed the Adjusted R-Square to see if the variables explain the difference by the means of the sustainability performance, often this test is referred to as “the goodness of fit”. The regression analysis shows us that the adjusted R-Square is 0,034, and that the dependent variable explains 3.4% of the differences in sustainability performance. The model itself shows a significance of 0,072 > 0,05, this would make our model not significant. 0,072 < 0,1 means that there is some significance at the 0,1 level. We should take cautious with interpreting the results.
Further analysis of the regression results shows that when culturally diverseness increases the sustainability performance decreases, and that this result is significant. This is contradicting with our theory, but this could be caused by the BLAU scores being low and the culturally diverseness on the board of directors is still low in the US. It shows that sustainability performance scores lowers with 25,39% when the BLAU score increased with 1%. Even though the model is significant, this result is not in line with the predicting theory based on diverseness in boards and increased performance on sustainability. Therefore we have to reject hypothesis one, there is no positive relation between culturally diverse boards and sustainability performance. This could also be due to different results coming from diverseness on boards, if we look at Pletzer et al. (2015), they find
that female presentation on Corporate boards do not improve financial performance. This is another study that finds diverseness of the boards not always improve the performance of the firm.
No other variables are significant in the model, the log of total assets does show to have positive relation with sustainability performance but this relation is not significant. The coefficient is 2,587 but the significance is 0,303 which is > 0,05. The standard deviation shows us that there is a lot of variance to be explained, but the control variables used don’t seem to be significant for the model. Furthermore noteworthy is that the Return on Asset ratio, the Book to Market ratio and the size of the board do seem to negatively relate to sustainability performance, none of these control variables are, as mentioned earlier, significant. This is contradicting based on the previous theory by Zaman & Al-Shaer that these variables do relate to sustainability performance. Using different measures for sustainability performance could cause differences. They focused their study on the United Kingdom, while we used the S&P 500 of America. These results, while not significant, seem to indicate firms that perform badly find incentives to focus more on sustainability reporting. Furthermore, we can conclude when testing for normality of the residues with a Kolmogorov-Smirnov test, that our residues are normally distributed.
Table 8 – Regression analysis hypothesis 1
Variable Expected relation ! p-score
Constant 23,094 BLAU + -24,817 0,004** BODSIZE + -0,554 0,443 BODIND + 22,387 0,129 BODMEET + 0,203 0,659 LOG (TA) + 2,675 0,815 ROA + -10,533 0,555 DTA - -11,153 0,091 BTM + -4,316 0,406 RB - -0,002 0,953 RE + 0,149 0,900 Year dummy 1 +/- 0,973 0,768 Year dummy 2 +/- -0,821 0,809 Adjusted R-Sqaure 0,034 Model significance 0,072 * Significance of 0,10 ** Significance of 0,05 *** Significance of 0,01 5.2 Results Hypothesis 2
In this section we will test hypothesis 2 to see if in case of a change to a more culturally diverse board, sustainability performance will improve over the years. We used the change in sustainability performance as our dependent variable and the change in BLAU score as our independent variable for cultural diversity. In the mentioned theory we predicted that the change to more diversity has
a positive effect in the board of directors, and therefore a positive effect on sustainability. The following hypothesis was formulated:
The change in composition of the board towards more cultural diversity has a positive association with sustainability performance improvement
This regression was ran like the regression described in 5.1, except due to only being able to compare 2 years our sample shrunk to 156 firms. Where we are comparing the change from 2013 on 2012 and 2014 on 2013. For the BLAU score and sustainability performance we created a variable called ∆ BLAU and ∆ SUSREP. The regression results show us an adjusted R-Square of
0,051. This means there is an explanation of the variance of 5,1%, the variables. The model is not significant with a score of 0,971 >0,1 this is what is in line with the adjusted R-square. This could be due to the low number of samples included. Due to the model not being significant we have to reject the hypothesis. We can conclude when testing for normality of the residues with a Kolmogorov-Smirnov test, that our residues are normally distributed.
Table 9 – Regression analysis hypothesis 2
Variable Expected relation ! p-score
Constant 32,395 BLAUDELTA + -17,77 0,271 BODSIZE + -1,982 0,178 BODIND + -33,302 0,280 BODMEET + 0,134 0,861 LOG (TA) + 4,513 0,250 ROA + -9,787 0,721 DTA - 15,937 0,194 BTM + -14,639 0,089 RB - 0,005 0,446 RE + -0,331 0,357 Year Dummy 1 +/- 3,907 0,214 Year Dummy 2 +/- 1,610 0,627 Adjusted R-Square 0,051 Model significance 0,971 * Significance of 0,10 ** Significance of 0,05 *** Significance of 0,01 5.3 Results Hypothesis 3
In this section we will test hypothesis 3 to see if sustainability performance will improve when there is a more collective board. We used sustainability performance as our dependent variable and the collectivity score as our independent variable for the collectivism of the board. In the mentioned theory we predicted that a board with more cultures that are collective has a positive effect in the board of directors, and therefore a positive effect on sustainability performance. The
There is a positive association between collective cultures in the boards and sustainability performance
The regression was ran exactly like the regression analysis performed for hypothesis 1, but instead of the BLAU score for diversity, we used the COLS score for collectivism on the board. The regression results show us an adjusted R-Square of 0,015. This means there is an explanation of the variance of 1,5%. The model is not significant with a score of 0,224 >0,10 this is in line with the adjusted R square. This result could be because, as mentioned earlier, the diversity on boards in the US is very low, this would also reflect in the collectivity score, as the score would be higher if there was more diversity on boards. We find that the collective score is significant, but correlates negatively just as the BLAU score in hypothesis 1. As the model is not significant we can disregard this result. Due to the fact that the model is not being significant we have to reject hypothesis 3, and find no relation between the collectivist culture of the board and sustainability performance. We can conclude when testing for normality of the residues with a Kolmogorov-Smirnov test, that our residues are normally distributed.
Table 10 – Regression analysis hypothesis 3
Variable Expected relation ! p-score
Constant 27,373 COLS + -12,636 0,034** BODSIZE + -0,802 0,266 BODIND + 21,232 0,154 BODMEET + 0,053 0,909 LOG (TA) + 2,352 0,352 ROA + -13,771 0,444 DTA - -13,882 0,065 BTM + -3,867 0,461 RB - -0,002 0,352 RE + 0,224 0,536 Year Dummy1 +/- 1,074 0,747 Year Dummy2 +/- -0,589 0,865 Adjusted R-Square 0,015 Model significance 0,224 * Significance of 0,10 ** Significance of 0,05 *** Significance of 0,01
5.4 Summary Results
Table 11 – Overview of the research results
Hypotheses Result
H1: There is a positive relation between culturally diverse boards and sustainability performance
Rejected
H2: The change in composition of the board towards more cultural diversity has a positive association with sustainability performance improvement
Rejected
H3: There is a positive association between collective cultures in the boards and sustainability performance
6. Conclusion
In this section, a conclusion will be drawn based upon the results presented in section 5, whether culturally diverseness and collectivism positively influences sustainability performance. Conclusion will be drawn per hypotheses and an overall conclusion will be made on the basis of the research question. Subsequently we will discuss the shortcomings of this research and the recommendations for further research.
6.1 Conclusion of Research
This research paper answered the research question whether culturally diverseness positively affects sustainability performance with S&P 500 firms in the US. We examined 237 firms and by means of hand collected data from different databases from 2012-2014 to conclude whether culturally diverseness improves sustainability performance.
In the regression in the first hypothesis whether board culturally diverseness positively relates to sustainability performance, we found that there was a significant relationship between board cultural diversity and sustainability performance. We found a negative relation between board diversity and sustainability performance, this caused a rejection of our hypothesis 1. This negative relationship was significant, and not in line with all found literature and predicted results. It is not the first time a negative result is found when relating firm performance to board diverseness. Pletzet et al. (2015), finds that firm performance is negatively related to the number of women on a board of directors, while contradicting research is done by Al Shaer & Zaman (2015), who find that there is a positive relationship between gender board diversity and sustainability reporting. Carter et al. (2013), finds a positive relation between board diversity (minorities & gender) and firm value. While research done by Shrader et al. (1997), concluded a negative relationship between firm performance and board diversity. Sustainability reporting and culturally diversity remains an issue in the boardroom. The percentage of minorities on the board is still significantly low and can influence the results. Only 8 of the 79 companies included a BLAU score of 0.5, with maximum being 0.543.
The second hypothesis was created to find evidence whether over the years as board diversity increased, sustainability performance improved. We performed a regression analysis and found no significant results; our model did not test to be significant. This makes it difficult to draw certain conclusions on our data. A reason for our model not being significant is the lacking sample, our database used for sustainability performance in the US only included the years 2012-2014; no data
hypothesis and conclude that there is no improvement of sustainability performance when culturally diverseness is increased on the board.
The third hypothesis was designed to add additional support that diversity in the boards in the US will lead to better sustainability performance. As predominantly Caucasian, and being an individually based country, adding diversity to the board would lead to a more collectivist culture on the board. Based on the mentioned literature a more collective cultured board would improve sustainability performance. The regression analysis performed, again proved not to be significant as the model was not significant. We therefore could not conclude on the result and the third hypothesis was rejected. There is no relation between a collective culture and sustainability performance. A reason for our model not being significant could be our COLS score that is a new designed score or the fact that board diversity is low in the US and that reaching a convincing COLS score would be difficult.
6.2 Shortcomings
A shortcoming of our research is the previously mentioned low cultural diversity in the S&P 500 in the US. This can definitely influence our results. Subsequently our sample was on the small side, a lot of firms were lost when aligning the S&P 500 with the Newsweek green rankings and finding all the control variables for the firms over the year. Furthermore our discretionary accruals were measured using the Modified Jones Model, and estimates were used, this could pose some limits on our research.
6.3 Recommendations future research
This research did not indicate that there is a significant positive relationship between culturally diverseness and sustainability performance. We concluded this on the basis of the Newsweek Green rankings; future research could use a different measure of sustainability performance. Additionally not all our control variables did produce the results we expected and some were negatively related where a positive relationship was predicted. Therefore, we recommend investigation by using other control variables. Likewise with board diversity still being low in the US, waiting a few years for board diversity to increase and repeat the research could affect the results and the sample could be enlarged to reach a valid conclusion. Most importantly the negative relationship resulted from hypothesis 1 between sustainability performance and culturally diverseness should be investigated for negative impacts of board diversity.
7. References
Acevedo, Ignacio David, "Understanding Ethnicity: The relation among ethnic identity, collectivism, and individualism in African Americans and European Americans" (2003). University of Kentucky Master Adams, B. E. (2010). The Role of Boards of Directors in Corporate Governance: A Conceptual
Framework and Survey. Journal of Economic Literature .
Adams, J. d. (2015). Board Diversity: Moving the Field Forward. Corporate Governance: An International Review, .
Adams, and Ferreira. 2009. Women in the Boardroom and Their Impact on Governance and Performance. Journal of Financial Economics, 94(2): 291–309
Ahmad-Zaluki, Nurwati A. "Board Ethnic Diversity in Newly Listed Malaysian Companies." Indian Journal of Corporate Governance 5.1 (2012): 24-32. Business Source Alumni Edition. Web. 24 Mar. 2017.
Alliance for Board diveristy. (2005). 3 8, 2017, from catalyst: http://www.catalyst.org/press-release/117/alliance-for-board-diversity-fact-sheet
Al-Shaer, H., & Zaman, M. (2016). Board gender diversity and sustainability reporting quality. Journal of Contemporary Accounting & Economics, 12(3), 210-222. doi:10.1016/j.jcae.2016.09.001
Al-Tuwaijri, S. A., Christensen, T. E., & Hughes, K. (n.d.). The Relations Among Environmental Disclosure, Environmental Performance, and Economic Performance: A Simultaneous Equations Approach. SSRN Electronic Journal. doi:10.2139/ssrn.405643
Amason, A.C. (1996). Distinguishing the effects of functional and dysfunctional conflict on strategic decision-making: Resolving a paradox for top management teams. Academy of Management Journal, 39(1), 123-148.
Arora,. Dharwadkar, Corporate governance and corporate social responsibility (CSR): the moderating roles of attainment discrepancy and organization slack, Corp. Gov, 19 (2) (2011), pp. 136–152 Bantel, K. A. (1994). Strategic planning openness: The role of top team demography. Group & Organization
Management, 19, 406–424.
