Internationality of the Board: Corporate Social Responsibility and Corporate
Social Performance
Author: Mark van ‘t Klooster (s2778637)1
Supervisor: Dr. N. Selmane
MSc Finance 12/06/2019
Abstract: This study examines the impact of board internationality on firms’ corporate social responsibility (CSR). With a dataset of 292 listed firms in the U.S. and Europe containing 100812 director observations over the period 2002-2016 robust evidence is found for a correlation between the degree of internationality and CSR score. Moreover, evidence is found for the environmental, social and corporate governance pillar explicitly. Whether a CEO is foreign or domestic is found to be of significant influence on CSR performance. No robust evidence is found based on the diversity of legal origins in the board.
JEL classifications: G34, G39, J10, M14, M51
Keywords: Corporate Governance, Board Structure, Origin, Nationality, CSR, CSP, Diversity, CEO
1 Correspondence information: University of Groningen, Faculty of Economics and Business,
1
1. Introduction
Diversity and Corporate Social Responsibility (CSR) have become a more pressingly matter recently. Both go beyond profit optimization for firms, rationale behind this is found in various directions. Certain is that a lot of research is performed on CSR, much of this research is done on the board and its effect on CSR and some of this is aimed at the effects of the board’s composition on CSR (e.g., Harjoto, Laksmana and Lee, 2015; Gulzar et al., 2019). Various effects are found in these studies, which are discussed later on in this paper. Motivations for such studies vary, but it is certain that in the European Union, United States and some other mostly developed countries it is required to strive for increased board diversity and to be transparent in these practices2. This is expected to put pressure on firms to increase their diversity by having to inform the market of these corporate governance practices. Therefore, it becomes highly relevant whether board diversity influences the firms’ performance.
Kitzmueller and Shimshack (2012) define CSR economically as a manifestation of oneself in some observable and measurable behaviour or output where it exceeds levels set by obligatory regulations or the standards set and enforced by the law. CSR is a trending topic nowadays and considered as increasingly important by all stakeholders (Kitzmueller and Shimshack, 2012). The objective of this research is to find if there is a relation between the degree of internationality in the board and CSR score. Not only the effect of internationality on CSR will be focused on, but also reverse causality is kept in mind. As earlier research finds a mostly positive relation between diversity and CSR, it is possible that the degree of CSR influences the degree of internationality. The difference is critical in the applications of the results of this study. Therefore, the main research question will be; till which extent does the internationality of a board influence CSR score? Literature has left room for elaboration on this relation and allows to contribute to filling this gap.
This research question is relevant for multiple reasons of which one is diversity and its commonly acceptation as a subject of high priority3. CSR is often measured through environmental, social and governmental (ESG) dimensions. These are considered as a measure of standards in order to assess how socially conscious a company operates. Assembling a board
2See, European Banking Authority. (2017). On the assessment of the suitability of members of the
management body and key function holders under Directive 2013/36/EU and Directive 2014/65/EU. URL;https://eba.europa.eu/documents/10180/1972984/Joint+ESMA+and+EBA+Guidelines+on+the+a ssessment+of+suitability+of+members+of+the+management+body+and+key+fu.nction+holders+%28 EBA-GL-2017-12%29.pdf (accessed on 17/05/19)
3 See, Jacobs. E. (2019). Europe, data and diversity and HR’s own changing role. Financial Times.
2
could be given an additional factor for consideration. Liang and Renneboog (2017) provide us with a number of reasons why firms would engage CSR in their business; as a method in order to manage risks, to reduce harm, stress and waste, reduce the cost of capital, improve relations with regulators and stakeholders and finally for their reputation. Investors are interested in CSR, in what CSR stands for and in the potential of obtaining abnormal returns that could come along when constructing a portfolio (Kempf and Osthoff, 2007). More relevancy is found by Chatterji et al. (2009) whom describe four motivations for investors to include the CSR of firms in their decisions. The first possible motivation is financial reasoning, as most research link better CSR results to better financial performance. Secondly, deontological reasoning is possible; to avoid investing in irresponsible firms. Thirdly, there is the motivation of a consequentionalist; whom reward good firms and punish bad firms for their behaviour. Lastly, they argue that investors include CSR for expressive reasoning, as a tool to express their social and self-esteem. Nowadays we see a shift in globalization, in which large economies such as the U.S. under president Trump and the U.K. with their Brexit prioritize their own citizens over other foreigners. The results found by this study could add arguments and evidence in the debate on such topics as I find evidence for a positive association between internationality and CSR score. This paper continues as follows; Section 2 discusses previous literature on the topic which lead to hypotheses being formed. This is followed by research and data in Section 3 wherein the methodology and data are motivated and discussed. Thereafter, Section 4 reports and discusses the results. Finally, in section 5 there is a conclusion which summarizes the main findings and its implications.
2. Literature Review
3
Zhu, and Ding (2013) already find evidence for outsider directors and women on the board being directly linked to an improvement in CSR score. The relationship between the origin of the board members and corporate social performance is still a topic for which its influences are rather undiscovered and elaboration on the specifics of nationalities in the board is still missing to the best of my knowledge. Studies such as those by García-Meca et al. (2015) give reason to believe this could be similar for their performance in CSR. Oxelheim and Randøy (2003) suggest that the inclusion of at least one foreign board member further aligns the interest of shareholders and managers, partly due to their active presence but also due to signalling. Moreover, Cai and He (2014) find evidence for value creation due to firms’ environmental strengths whom can be measured by the environmental pillar of CSR.
The focus of this paper lies on listed companies worldwide. This allows the results of this study to be most conclusive and includes mostly larger companies. Larger companies tend to be more pressured by the public to engage in CSR and therefore are expected to be more transparent about it. Research by Estélyi and Nisar (2016) on why a firm would appoint a director of foreign nationality has found that there are positive effects involved for an international board, though they did not focus on CSR related matters.
More arguments in the CSR debate can be found in the effects of CSR on firm performance. The composition of the board has proven to be of significant importance with regard to a multitude of performances (see, e.g., Arora and Sharma, 2016; Bhagat and Black, 1999; Dahya and McConnell, 2007; Abidin et al. 2009), while others are less conclusive (see, e.g., Chapple and Humprey, 2014; Rao and Tilt, 2016) Characteristics such as age, gender, race and educational background have been investigated. Examples of evidence are given by Dam and Scholtens (2015) whom find a strong foundation for conditional evidence of a positive correlation between CSR and financial performance. Johl, Kaur, and Cooper (2015) researched certain characteristics of the board and their relation to firm performance and found that the size of the board and different aspects of expertise are positively correlated with the firm’s performance in Malaysia. Cao et al. (2018) find internationality in the board to be influencing firm decision making.
4
exists a significant correlation. Furthermore, these investors will be interested in anything impacting the CSR score, as earlier research found these to have significant impact on market returns (Friede, Busch, and Bassen, 2015). When a company’s priority of obtaining CSR goals is high enough on the priority list, the different nationalities could help in mitigating towards this target, depending on the results of this study. Therefore, I test if there is a relation between the degree of internationality of a board and it’s CSR score and come up with the following hypothesis:
Hypothesis 1: the degree of internationality in the board influences the CSR score positively
And subsequently, this hypothesis follows logically:
Hypothesis 2: the degree of internationality in the board influences the
Environmental/Social/Governance pillar positively
As one could expect that different people with different backgrounds will have varying effects on the management and performance of a company. A variety of people, or a diverse board, is expected to outperform a board with only likeminded on it (see, e.g., Estélyi and Nisar, 2016). Different perspectives and insights lead to debate and conversation and are expected to increase the CSR score as was already found by other research on other dimensions of diversity (Harjoto, Laksmana, and Lee, 2015).
5
different perspective if it comes from a nation where CSR practices are more prevalent and their ability to translate this into the firm. However, this goes both ways. Therefore, the following hypothesis is set up in order to test whether this holds for internationality of a CEO;
Hypothesis 3: there is a significant relation between whether a CEO is domestic or foreign and CSR score and its respective pillars.
Earlier research finds that the degree of CSR score is related to the market in which a company is operating (Liang and Renneboog, 2017). They differentiated between type of common and civil law systems in the operating country of the firm. Whether this translates into foreign board membership influences on CSR as well could provide useful insights in the relevance of board diversity. Relevant here is the sign of this relation; is it positively influencing the CSR or negatively? They found that common law countries score lower on CSR than civil law countries, which gives justification to suspect this translates into a positive association between legal origin diversity and CSR scoring. Therefore, a fourth hypothesis is tested;
Hypothesis 4: a diversity of legal origins in the board has a positive influence on the CSR score and its respective components
3. Research method and data
3.1 Data
6
From the more than 7000 firms included in the dataset nowadays, only a few hundred have consistently reported during the span of its existence. In the final dataset, a total of 292 firms are included based in 11 countries, being: Belgium, Denmark, France, Germany, Greece, the Netherlands, the Republic of Ireland, Spain, Switzerland, the United Kingdom and the United States. More specifically, the dataset contains 100812 director observations, with a focus on firms based in the United States. It covers the period between 2002 up and included until 2016 and is always denominated in dollars whenever applicable.
As stated before, ESG scores are retrieved from the Reuters ASSET4 dataset through Datastream. It contains data on multiple ESG components from 2002 and onwards of which the
Environmental, Social and Governance are the obtained ESG pillars. With the latest year
consisting of over 7000 firms it is one of the most complete datasets regarding ESG scores. However, as most firms only recently started to report their ESG data and a large number of those wo did report the numbers did so inconsistently, only a fraction of that database was fitting for use. As a firm whom does not report consistently could have a reason to do so, this could either be most likely negative or potentially positive. Either way, it is impossible to obtain the correct data for these firms. Thomson Reuters combines all ESG scores in order to arrive at one score covering all components; TotalESG rating. This study includes only firms whom have constantly reported over the 2002-2016 period in order to get the most consistent results. Ratings are rewarded relative to industry peers both domestic and international, which allows the data to be unbiased on country specific variables.
For the remaining selection of companies, firm and director level data are retrieved from BoardEx. The director level data regarding nationalities of the specific directors was transformed into firm level variables. Firm level control variables such as Firm Size, Return on Assets (ROA) and Sales growth (Salesgrwth) are obtained from Datastream. As a country level control variable the natural logarithm of Gross Domestic Product per Capita (GDPC) are taken from Datastream as well.
3.2 Descriptive statistics
7
The average TotalESG score which measures the overall ESG score has a mean of 65.89, being higher than the average of 504. This could be related to the fact that these firms consistently reported their ratings since Datastream started to implement them. TotalESG is calculated as an overall score for ESG5. More explicitly, it is calculated as follows: [(34% 𝐸𝑛𝑣𝑖𝑟𝑜𝑛𝑚𝑒𝑛𝑡𝑎𝑙 + 35.5% 𝑆𝑜𝑐𝑖𝑎𝑙 + 30.5% 𝐺𝑜𝑣𝑒𝑟𝑛𝑎𝑛𝑐𝑒) + 𝐸𝑆𝐺 𝐶𝑜𝑛𝑐𝑒𝑟𝑛𝑠] ⁄ 2 (1) and is provided by Datastream. The degree of internationality as measured by %Foreign is 0.13 on average, implying that a firm in the sample on average has 13% foreign representation in both the executive and supervisory board. The average firm has an Environmental score of 54.76, a Social score of 58.40 and a significantly higher mean score of 76.14 on Governance. The means of the number of directors, standard deviation of age and standard deviation of the time on board of the boards are 11.19, 6.92 and 6.01 respectively. Regarding firm size, the mean of the natural logarithm of the market value of each firm (FirmSize) is 9.39.
Table A.2 in the Appendix shows director descriptive statistics by distribution. Here, it can be observed that there is a bias toward American directors in this sample. Additionally, European directors are also relatively well represented. This is in line with the origins of the firms whom are in these regions. Table A.3 shows the country qualification per legal origin group and the count of each group. The qualification is based on English common law, French civil law, German civil law, Scandinavian civil law and Socialistic law of which the latter is not represented in the sample. As the United States is well represented in the sample, this translates into a large proportion of English common law in the sample. This qualification is shown per nationality in Table A.4 of the Appendix. In Table A.5 it can be observed how TotalESG is distributed over the different industries. It shows that the distribution over firms is well diversified and all industries are covered. Interestingly, in some industries such as oil & gas there is a wide distribution of CSR scores ranging from almost zero till hundred. This can be explained by the relative nature of these CSR scores, as they are awarded relative to industry peers internationally. This makes it more reasonable to compare the transport sector with banks regarding environmental scores for example. Table A.6 displays descriptive statistics regarding their TotalESG scoring and the distribution of each legal origin on all firms. In contrast to findings of Liang and Renneboog (2017) where the mean CSR score in civil law countries is
4 The average is 50 due to the fact that ratings are relative to industry and country peers.
5 see; Thomson Reuters. (2019). Thomson Reuters ESG Scores. URL;
8
found to be higher on average than in common law countries, in this sample it is on average the other way around.
Table 1. Descriptive Statistics CSR data
Notes: ESG scores obtained from Reuters ASSET4. Total ESG is a weighted average score of all individual components of ESG included in the dataset of Reuters ASSET4. Total numbers of firms in the sample equals 292 whom all include date in the whole period covering 2002-2016. %Foreign depicts the number of domestic directors over total directors in a firm. LagESG is a lagged variable of TotalESG. Details of variables are elaborated in the Appendix in Table A.1
Variables N Mean S.D. Min Max
TotalESG 2,313 65.90 27.55 2.55 98.67 Environmental 2,313 54.76 31.80 8.13 97.35 Social 2,313 58.40 29.04 3.58 99.34 Governance 2,313 76.14 18.13 2.09 98.10 LagESG 2,313 64.34 27.32 3.43 98.71 LagEnv 2,313 51.47 31.79 8.13 97.35 LagSoc 2,313 55.64 29.57 3.58 99.34 LagGov 2,313 74.67 19.52 1.49 98.10 %Foreign 2,313 0.13 0.17 0.00 0.90 ForCEO 2,313 0.18 0.37 0.00 1.00 LegalBlau 2,313 0.11 0.13 0.00 0.72 %Male 2,313 0.85 0.09 0.33 1.00 #Directors 2,313 11.19 2.68 4.00 28.00 SDtimebrd 2,313 6.01 2.81 0.00 15.90 SDage 2,313 6.92 2.00 0.80 15.30 ROA 2,313 6.73 7.13 -46.80 47.48 Salesgrwth 2,313 0.08 0.21 -1.90 2.81 TobinsQ 2,313 0.63 0.21 0.07 1.70 FirmSize 2,313 9.39 1.27 4.39 13.12 Leverage 2,313 3.53 21.00 -271.30 866.90 GDPC 2,313 10.74 0.17 9.67 11.39 # Firms 292 # Director observations 100,812 # CEOs 2,594 3.3 The model
The main regression will be ran as shown in formula 2 below:
9
profitability to be positively related to firm performance. The sales growth over the year (SalesGrwth) and the natural logarithm of the Gross Domestic Product per capita (GDPC) as a country level control variable are included. Furthermore, 𝜔𝑡 are time fixed effects, 𝛾𝑖 are time fixed effects and the error term is represented by 𝜀𝑖,𝑡. The abovementioned control variables are used in accordance with earlier mentioned literature regarding diversity and CSR score (see, e.g., Kaplan, 2012; and Ferrell et al. 2016)
The main independent variable in this formula will be the number of foreign directors over the total number of directors in the board (%Foreigners). The dependent variable will be a weighted rating taken over all ESG components as provided by Datastream. This score consists for 34% of Environmental score, 35.5% Social score and 30.5% Corporate Governance score which is then equally weighted with a score for ESG controversies as is given by Datastream. Furthermore, there is always a lagged variable of the dependent variable included to decrease the chance of endogeneity. Here that is TotalESGi,t-1 which is the TotalESG score of t-1
(LagESG). This is included in order to control for potential serial correlation. In addition, each ESG pillar is used as a dependent variable. This is used for testing the hypothesis of each ESG pillar being effected by the internationality of the board. Furthermore, all regressions are included with firm and year fixed effects.
In order to account for potential endogeneity issues as a method of robustness, also a two-stage least square (2SLS) regression is performed. Earlier research already finds evidence that board diversity is endogenous (Gul et al., 2011). Instrumental variables used here are similar to those of earlier research by e.g. Gul et al. (2011), and Harjoto, Laksmana, and Lee (2015). I use
ForCEO as it influences the %Foreign directly, SDtimebrd to control for the recent trend in
demand for diversity in the board, TobinsQ and FirmSize as larger, more visible and more profitable firms are more inclined by stakeholders to perform well on diversity related practices. The factors impacting diversity is examined in the first stage of this regression. The second stages is performed thereafter in order to inspect the impact of internationality on CSR scores. 3.4.1 Legal Origin
10
outperform common law countries with respect to CSR scores. After all restrictions have been imposed on the dataset used in this study, the sample allows to investigate the influences of legal origin in boards of firms with mostly the United States as their home country. As the United States imposes English common law, the influence of board members based on legal origin is investigated. Table A. 6 in the Appendix shows descriptive statistics based on the legal origin of all countries. Tables A.3 and A.4 respectively show the legal origin classification of each country and nationality in the dataset.
In order to analyse the model, Blau’s Index of heterogeneity is used to test for an association of a diverse board based on legal origin and CSR scores (BlauLegal). It is calculated as 1 − ∑ 𝑃𝑖2, in which P is the proportion of directors in each category of origin; English, French, German,
Scandinavian and Socialist and i the number of categories (Blau, 1997). It allows for improved
comparability and is more commonly used in literature. 3.4.2 CEO specification
Until this point all testing discussed was focused on the complete boards, both supervisory and executive. To add value to this paper, an additional approach is added in the form of a CEO focus. A CEO is at the top of the firm and therefore one of the, if not the most, influential positions in a firm. Accordingly, the CEO specifically is expected to have a larger influence on firm performance and thus on the CSR score than an average board member. Studies by the likes of Haleblian and Finkelstein (1993), Brick et al. (2006), and Peni (2014) already found a relationship in which CEO characteristics, amongst other factors, significantly influence the performance of the firm. Thus, in line with hypothesis 3, it is expected that the origin of the CEO influences the CSR score of a firm significantly. In order to examine this, there is no dummy used as some firms have multiple CEOs, but a ratio which is almost always zero or one. 3.5 Testing
11 Table 2. Pearson correlation matrix
12
4. Results
In Table 3 the results of the main OLS regression are displayed and these are discussed below in this section. The earlier stated hypothesis 1 expected that internationality in the board has a positive correlation with CSR score of the firm. The table shows that this is indeed the case as the beta of %Foreign is positive and statistically significant at a 5% level. This implies that a larger number of nationalities in the boardroom of a firm leads to an increase of the firm’s CSR score. One reason for this could be that an internationally diverse board would be better to comply with all stakeholders requirements due to their different backgrounds. This diversity in backgrounds allows for decision-making in which more angles are discussed. Additionally, from the 2SLS regression which was performed in order to correct for endogeneity it follows that these results are robust and there is statistically significant correlation between the two variables at a 5% level. These results are shown in Table 4. Therefore, the regressions confirm hypothesis 1 by using two different models. These results are in line with some of the earlier research whom found there to be a statistically significant association between board diversity and firm performance. Accordingly, it could be useful for a firm whom is in the selection process of appointing a new director to take this effect of an increased CSR score into account to arrive at their decision.
13
significantly correlated with internationality, while in regressions performed later on it is also influenced the least of all CSR related scores.
Table 4. shows the results of a two stage least square regression. It was conducted to control for endogeneity in the model. The Durbin-Wu-Hausman test is performed in order to test whether %Foreign is endogenous, the results, Table A.11 in the Appendix, confirm this. Therefore, the results of the ordinary least square regression are inferior to those of a 2SLS regression. In the first stage it is examined which variables influence %Foreign and make it to be endogenous. Earlier research found that it is predicted that several factors could influence the diversity of the board or in this case the internationality (see, e.g., Gul et al. 2011; Harjoto, Laksmana and Lee. 2015). In the second stage it is tested whether the same variables as in the OLS regression influence the TotalESG, Environmental, Social and Governance score. It turns out that concurring with earlier findings in the OLS regression there is a significant association between the internationality and ESG score. The coefficients are significant at a 5% statistical level for TotalESG and Social. At a 10% level, there is also a significant correlation between the degree of internationality and the environmental pillar of ESG. All of these found associations are of positive nature. Therefore, it is possible to say that the effect of the amount of foreigner directors in the board is positive. Hypothesis 1 stated that the degree of internationality in the board influences the CSR score positively, the evidence found concurs with this and therefore I fail to reject this hypothesis. The second hypothesis stated that the degree of internationality in the board influences the Environmental/Social/Governance pillar positively. As there is evidence found for the Environmental and Social pillar I can confirm the hypothesis for these elements. The degree of internationality in the board is not found to be statistically significantly associated to Governance on the other hand, therefore the hypothesis is rejected for the Governance pillar.
CEO specification
14
Nevertheless, the level of statistical significance is lowered to a 10% level for both. However, for Environmental this does not hold in the 2SLS and thus I cannot find robust evidence for an association for this pillar. Nonetheless, as evidence is found in this regression, the null hypothesis of a statistically significant association between the degree of the CEO(s) being foreign and the TotalESG score of the firm in hypothesis 3 is confirmed.
Earlier research sometimes found evidence for a correlation between specifications of a CEO and certain performance measurements of a firm (see, e.g., Peni, 2014). There is evidence at a 10% level to believe this to be the case here as well. Still, the evidence does not hold for all pillars. One explanation for this is that having a foreign nationality does not necessarily imply that one has grown up there and carries its culture, norms and values. In this globalised world it is not unlikely that one lived most, if not all, of its life in a different country from its nationality. Moreover, having a foreign nationality does not necessarily imply that a director conducts a different type of stakeholder management than a domestic CEO would have conducted in the same seat. Yet, the impact is significant and positive at an economical level according to these results, implying that appointing a foreign CEO is related to improved CSR scoring.
Table 3. Ordinary least square regression %Foreign
Notes: All regressions include firm and year fixed effects. In parenthesis the standard errors are included. Variable descriptions can be found in Table A.1 of the Appendix. ***, ** and * indicate statistical significance at 1, 5 and 10 % respectively.
Variables Total ESG Environmental Social Governance
15 Table 4. Two stage least square regression %Foreigners
Notes: This table displays the results of a two stage least square regression. The instrumental variables here are ForCEO, SDtimebrd, FirmSize and TobinsQ. All regressions include firm and year fixed effects. In parenthesis the robust standard errors are included. Variable descriptions can be found in Table A.1 of the Appendix. ***, ** and * indicate statistical significance at 1, 5 and 10 % respectively.
1st stage 2nd stage 1st stage 2nd stage 1st stage 2nd stage 1st stage 2nd stage
Variables %Foreign TotalESG %Foreign Environment %Foreign Social %Foreign Governance
16 Legal origin
Thirdly, the results are separated based on the legal origin differences between the board members with different nationalities for which the results are shown in Table 6 and A.9 of the appendix. In line with Liang and Renneboog (2017) a research is conducted on the basis of the legal origin of the home country of a firm and the influence of directors whom are foreign to this law system. English common law, French civil law, German civil law, Scandinavian civil law and socialist law are separated and regressed in a similar manner as before of which the results are shown in Table 6 for the OLS regression and A.9 for the 2SLS. In the initial OLS regression evidence is found at a 10% level for a correlation between diversity of legal origin and Total ESG. The difference in beta’s are not statistically significant at any of the commonly accepted levels in the 2SLS regression. We therefore cannot accept hypothesis 4 that a diverse board based on the legal origin would positively influence the CSR performance of the management. Thus, the findings of Liang and Renneboog (2017) that firms in certain legal
Table 6. Legal origin, Ordinary least square regression
Notes: LegalBlau is Blau’s index of heterogeneity. All regressions include firm and year fixed effects. In parenthesis the standard errors are included. Variable descriptions can be found in Table A.1 of the Appendix. ***, ** and * indicate statistical significance at 1, 5 and 10 % respectively.
Variables TotalESG Environmental Social Governance
17
origin regions outperform others on CSR score do not translate in directors from those regions transforming those scores in CSR performance of their own firm. This implicates a limitation to the previously results discussion on %Foreign, as an internationally diverse board was found to be positively correlated with CSR score, whereas now it is found to be insignificant if we rank this international diversity based on legal origin. A possible reason for this could be that CSR is regulated or subsidiarised in certain countries, but that this is not the case for a director originated from such a country which operates out of that jurisdiction for a foreign firm. Moreover, implying that there would be no subsidiaries or regulation accompanying the director. Furthermore, this gives reason to argue that earlier results cannot be traced back to a variance of legal origin in a board, but other variables arriving at an internationally diverse board are the cause of this association. In line with earlier research I also find evidence for a positive correlation between female representation in a board and CSR pillar scores at a 5% level for Environmental and Governance, For Social this is found at a 10% level implying again that a more diverse board influences CSP.
5. Conclusion and recommendations
Nowadays, diversity and CSR are increasingly important for publicly listed firms. Regulators, stakeholders and the public are increasingly pressing for more diversity in firms and boardrooms. Those same regulators, stakeholders and public additionally demand more CSR. This study examined the relation between one aspect of diversity, namely internationality, and socially responsible activities and initiatives or CSR. It is argued that a greater diversity of nationalities in the boardroom in this study relates to CSR, which is of interest and mostly in favour of all stakeholders. The nationalities of both executive and supervisory board members and ESG measures from ASSET4 as CSR score are used.
18
The association between the origin of the CEO of a firm and CSR scoring is also examined. Evidence is found that employing a foreign CEO is positively correlated with an increase in overall CSR score. When explicitly investigating the distinct pillars this only holds for the Social pillar.
It is argued that the legal origin influences the CSR score of a firm, however in this study I do not find robust evidence for a relation between the legal origin and CSR. Explicitly, the degree of diversity with respect to legal origin is neither found to be positively nor negatively be of influence on the overall CSR or individual pillars.
This study contributes to the existing research by increasing our knowledge of the impact of board diversity on firm performance. The effectiveness of an internationally diverse board on managers’ influence on CSR score is examined more specifically. This study is in line with increasing demand for more board diversity by regulators stakeholders and the public and accordingly not in favour of limiting diversity. Limitations of this study include the high degree of U.S. firms in the sample used and the limitations of the data available. Suggestions for further research include a similar study on this impact in different markets when enough data is available or on specific firms, sectors, nations or other criteria. It should also be mentioned that this does not imply that there is no negative effect of internationality on other factors involved in firm performance. These results do not necessarily support initiatives for quotas on international diversity in firms, yet they do support the rationale of internationality in a corporate board.
19
6. References
Abidin, Z.Z., Kamal, N.M. and Jusoff, K., 2009. Board structure and corporate performance in Malaysia. International Journal of Economics and Finance, 1(1), pp.150-164.
Berger, A.N., Kick, T. and Schaeck, K., 2014. Executive board composition and bank risk taking. Journal of Corporate Finance, 28, pp.48-65.
Bhagat, S. and Black, B., 1999. The uncertain relationship between board composition and firm performance. The Business Lawyer, pp.921-963.
Blau, P.M., 1977. Inequality and heterogeneity: A primitive theory of social structure (Vol. 7). New York: Free Press.
Brick, I.E., Palmon, O. and Wald, J.K., 2006. CEO compensation, director compensation, and firm performance: Evidence of cronyism?. Journal of Corporate Finance, 12(3), pp.403-423.
Cai, L. and He, C., 2014. Corporate environmental responsibility and equity prices. Journal of Business Ethics, 125(4), pp.617-635.
Campbell, J.L., 2007. Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility. Academy of management Review, 32(3), pp.946-967.
Cao, J., Ellis, K.M. and Li, M., 2018. Inside the board room: the influence of nationality and cultural diversity on cross-border merger and acquisition outcomes. Review of Quantitative Finance and Accounting, pp.1-38.
Chatterji, A.K., Levine, D.I. and Toffel, M.W., 2009. How well do social ratings actually measure corporate social responsibility?. Journal of Economics & Management Strategy, 18(1), pp.125-169.
Dahya, J. and McConnell, J.J., 2007. Board composition, corporate performance, and the Cadbury committee recommendation. Journal of financial and quantitative analysis, 42(3), pp.535-564.
20
Djankov, S., La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., 2008. The law and economics of self-dealing. Journal of financial economics, 88(3), pp.430-465.
Estélyi, K.S. and Nisar, T.M., 2016. Diverse boards: Why do firms get foreign nationals on their boards?. Journal of Corporate Finance, 39, pp.174-192.
Ferrell, A., Liang, H. and Renneboog, L., 2016. Socially responsible firms. Journal of Financial Economics, 122(3), pp.585-606.
Friede, G., Busch, T., & Bassen, A., 2015. ESG and financial performance: aggregated evidence from more than 2000 empirical studies. Journal of Sustainable Finance & Investment, 5(4), pp.210-233.
Galbreath, J., 2018. Is board gender diversity linked to financial performance? The mediating mechanism of CSR. Business & Society, 57(5), pp.863-889.
García-Meca, E., Garcia-Sanchez, I.M. and Martínez-Ferrero, J., 2015. Board diversity and its effects on bank performance: An international analysis. Journal of Banking & Finance, 53, pp.202-214.
Green, C.P. and Homroy, S., 2018. Female directors, board committees and firm performance. European Economic Review, 102, pp.19-38.
Gul, F.A., Srinidhi, B. and Ng, A.C., 2011. Does board gender diversity improve the informativeness of stock prices?. Journal of Accounting and Economics, 51(3), pp.314-338. Gulzar, M.A., Cherian, J., Hwang, J., Jiang, Y. and Sial, M.S., 2019. The Impact of Board Gender Diversity and Foreign Institutional Investors on the Corporate Social Responsibility (CSR) Engagement of Chinese Listed Companies. Sustainability, 11(2), pp.307-325. Haleblian, J. and Finkelstein, S., 1993. Top management team size, CEO dominance, and firm performance: The moderating roles of environmental turbulence and discretion. Academy of management journal, 36(4), pp.844-863.
Harjoto, M., Laksmana, I. and Lee, R., 2015. Board diversity and corporate social responsibility. Journal of Business Ethics, 132(4), pp.641-660.
21
Kempf, A. and Osthoff, P., 2007. The effect of socially responsible investing on portfolio performance. European Financial Management, 13(5), pp.908-922.
Khan, W.A. and Vieito, J.P., 2013. CEO gender and firm performance. Journal of Economics and Business, 67, pp.55-66.
Kitzmueller, M. and Shimshack, J., 2012. Economic perspectives on corporate social responsibility. Journal of Economic Literature, 50(1), pp.51-84.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A., 2008. The economic consequences of legal origins. Journal of economic literature, 46(2), pp.285-332.
Liang, H. and Renneboog, L., 2017. On the foundations of corporate social responsibility. The Journal of Finance, 72(2), pp.853-910.
Oxelheim, L. and Randøy, T., 2003. The impact of foreign board membership on firm value. Journal of Banking & Finance, 27(12), pp.2369-2392.
Peni, E., 2014. CEO and Chairperson characteristics and firm performance. Journal of Management & Governance, 18(1), pp.185-205.
Porta, R.L., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R.W., 1998. Law and finance. Journal of political economy, 106(6), pp.1113-1155.
Rao, K., and Tilt, C., 2016. Board composition and corporate social responsibility: The role of diversity, gender, strategy and decision making. Journal of Business Ethics, 138(2), pp.327-347.
Wolfers, J., 2006. Diagnosing discrimination: Stock returns and CEO gender. Journal of the European Economic Association, 4(2-3), pp.531-541.
22
Appendix
Table A.1 Variable descriptions
Variable Name Description Measurement
Total ESG A weighted rating taken over all aspects of CSR score as provided by Datastream
34% 𝐸𝑛𝑣𝑖𝑟𝑜𝑛𝑚𝑒𝑛𝑡𝑎𝑙 + 35.5% 𝑆𝑜𝑐𝑖𝑎𝑙 + 30.5% 𝐺𝑜𝑣𝑒𝑟𝑛𝑎𝑛𝑐𝑒
LagESG Lagged Total ESG Total ESGi,t-1
Governance A score rewarded on all performances that fall under the governance pillar of Reuter’s ESG score.
[(62.3% 𝑀𝑎𝑛𝑎𝑔𝑒𝑚𝑒𝑛𝑡
+ 23.0% 𝑆ℎ𝑎𝑟𝑒ℎ𝑜𝑙𝑑𝑒𝑟𝑠 + 14.8% 𝐶𝑆𝑅 𝑆𝑡𝑟𝑎𝑡𝑒𝑔𝑦) + 𝐶𝑆𝑅 𝐶𝑜𝑛𝑐𝑒𝑟𝑛𝑠]/2
LagGov Lagged Governance
Social A score rewarded on all performances that fall under the social pillar of Reuter’s ESG score.
45.1% 𝑊𝑜𝑟𝑘𝑓𝑜𝑟𝑐𝑒 + 12.7% 𝐻𝑢𝑚𝑎𝑛 𝑅𝑖𝑔ℎ𝑡𝑠 + 22.5% 𝐶𝑜𝑚𝑚𝑢𝑛𝑖𝑡𝑦
+ 19.7% 𝑃𝑟𝑜𝑑𝑢𝑐𝑡 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑖𝑏𝑖𝑙𝑖𝑡𝑦
LagSoc Lagged Social
Environmental A score rewarded on all performances that fall under the social pillar of Reuter’s ESG score.
32.4% 𝑅𝑒𝑠𝑜𝑢𝑟𝑐𝑒 𝑈𝑠𝑒
+ 35.3% 𝐸𝑚𝑖𝑠𝑠𝑖𝑜𝑛𝑠 + 32.4% 𝐼𝑛𝑛𝑜𝑣𝑎𝑡𝑖𝑜𝑛
LagEnv Lagged Environmental
% Foreign The ratio of foreign board members over the total number of board members
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑜𝑟𝑒𝑖𝑔𝑛 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠 𝑖𝑛 𝑓𝑖𝑟𝑚𝑖,𝑡
𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠 𝑖𝑛 𝑓𝑖𝑟𝑚𝑖,𝑡
ForCeo Ratio of foreign CEOs in the firm over all CEOs. Most of the time this either 0 or 1, however some firms have multiple CEOs. In such cases I decided to include both CEOs.
𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑜𝑟𝑒𝑖𝑔𝑛 𝐶𝐸𝑂𝑠 𝑖𝑛 𝑓𝑖𝑟𝑚𝑖,𝑡
𝑇𝑜𝑡𝑎𝑙 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐶𝐸𝑂𝑠 in 𝑓𝑖𝑟𝑚𝑖,𝑡
BlauLegal A diversity measurement as calculated by Blau’s index of heterogeneity. Here P is the ratio of either English,
French, German,
Scandinavian or socialistic legal origin.
1 − ∑ 𝑃𝑖2,
#directors Number of directors on the board of a firm
SDtimebrd Standard deviation of time on board of all directors in a firm SDage Standard deviation of age of all directors on board in a firm
ROA Return on Assets 𝑁𝑒𝑡 𝐼𝑛𝑐𝑜𝑚𝑒 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠⁄
TobinsQ Tobin’s Q ratio is a measurement of profitability.
𝐵𝑉 𝐴𝑠𝑠𝑒𝑡𝑠 − 𝐵𝑉 𝐸𝑞𝑢𝑖𝑡𝑦 + 𝑀𝑉 𝐸𝑞𝑢𝑖𝑡𝑦 𝐵𝑉 𝐴𝑠𝑠𝑒𝑡𝑠
Leverage Natural logarithm of total debt divided by total assets
23
Salesgrwth Growth of sales 𝑆𝑎𝑙𝑒𝑠𝑖,𝑡− 𝑆𝑎𝑙𝑒𝑠𝑖,𝑡−1
𝑆𝑎𝑙𝑒𝑠𝑖,𝑡−1
GDPC Natural logarithm of Gross
Domestic Product per capita 𝑙𝑛
𝐺𝐷𝑃𝑖,𝑡
𝑐𝑎𝑝𝑖𝑡𝑎𝑖,𝑡
%Male Ratio of male representation on the board
𝑀𝑎𝑙𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠𝑖,𝑡
𝑀𝑎𝑙𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠𝑖,𝑡+ 𝐹𝑒𝑚𝑎𝑙𝑒 𝑑𝑖𝑟𝑒𝑐𝑡𝑜𝑟𝑠𝑖,𝑡
FirmSize Firm size measured as the natural logarithm of the market value of a firm
𝑆ℎ𝑎𝑟𝑒 𝑝𝑟𝑖𝑐𝑒𝑖,𝑡
× 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑠ℎ𝑎𝑟𝑒𝑠 𝑜𝑢𝑡𝑠𝑡𝑎𝑛𝑑𝑖𝑛𝑔𝑖,𝑡
24 Table A.2 Director descriptive statistics
Notes: Here nationality depicts the nationality of a director and N the number of observations included in the dataset. The average percentage of each nationality on a board is shown under mean, the same holds for the median, standard deviation (SD), minimum percentage and maximum percentage of each nationality on all boards included. The table contains information over the period of 2002-2016 of all included firms obtain from WRDS. The director level data was transformed to yearly firm level data and thus these statistics quantify all Nationality data used. Here both supervisory and executive directors are included. Nationality is a dummy variable, either 0 or 1.
Nationality N Mean Median SD Min Max
26 Table A.3 Country Legal Origin qualification
Notes: Here each country is provided with the legal origin of each respective nation and their number of observations. Table A.1 of the Appendix shows detailed variable descriptions.
Country Legal Origin N
Belgium French civil 36
Denmark Scandinavian 19
France French civil 17
Germany German civil 9
Greece French civil 5
Netherlands French civil 20
Republic Of Ireland English common 49
Spain French civil 23
Switzerland German civil 54
United Kingdom English common 53
27 Table A.4 Legal Origin per nationality
Notes: Here each nationality is provided with the legal origin of each respective nation and their number of observations. Table A.1 of the Appendix shows detailed variable descriptions.
Nationality Legal origin N Nationality Legal origin N
American English 24289 Irish English 135
Argentine French 5 Israeli English 40
Australian English 164 Italian French 142
Austrian German 33 Ivorian French 4
Bangladeshi English 1 Jamaican English 9
Barbadian English 3 Japanese German 78
Belgian French 299 Lebanese French 3
Bermudian English 27 Luxembourger French 5
Brazilian French 31 Malaysian English 24
British English 1928 Mexican French 84
Cameroonian French 1 New Zealander English 32
Canadian English 407 North Korean Socialist 1
Chilean French 4 Norwegian Scandinavian 87
Chinese Socialist 30 Pakistani English 2
Croatian German 11 Panamanian French 13
Cuban Socialist 13 Polish Socialist 7
Cypriot English 5 Puerto Rican French 25
Czech Socialist 1 Russian Socialist 33
Danish Scandinavian 209 Saudi English 6
Dutch French 219 Singaporean English 18
Ecuadorian French 3 South African English 47
Egyptian French 12 South Korean German 7
Emirian English 8 Spanish French 403
Filipino French 4 Sri Lankan English 8
Finnish Scandinavian 44 Swedish Scandinavian 109
French French 511 Swiss German 247
Georgian German 9 Syrian French 2
German German 490 Taiwanese German 7
Ghanaian English 2 Turkish French 6
Greek French 81 Ugandan English 3
Guatemalan French 1 Uruguayan French 3
Icelander Scandinavian 1 Venezuelan French 12
Indian English 121 Vietnamese Socialist 4
Indonesian French 2 Zimbabwean English 6
28 Table A.5 ESG scores per industry
This table shows the descriptive statistics per industry of all firms included in the dataset. The industries are as defined by BoardEx. TotalESG is an equally weighted score of all ESG components provided by Thomson Reuters ASSET4 in order to quantify the average CSR score of a firm.
TotalESG Score N Mean SD Min Max
Aerospace & Defence 56 76.35571 25.95915 5.42 98.2
Automobiles & Parts 55 54.30491 25.11471 17.25 94.65
Banks 93 61.68516 21.8197 20.58 96.03
Beverages 28 63.95393 36.63323 7.28 98
Business Services 55 50.55436 20.85683 6.23 91.79
Chemicals 107 76.18308 24.03309 5.73 96.53
Clothing & Personal Products 60 82.822 24.20622 7.74 98.28 Construction & Building Materials 88 53.8708 29.58496 4.64 96.55
Consumer Services 7 30.09714 13.04068 14.39 46.85
Containers & Packaging 16 66.94438 32.45214 16.27 96.04 Diversified Industrials 37 93.32081 11.60393 25.87 97.98
Electricity 7 86.67 12.72707 61.34 96.4
Electronic & Electrical Equipment 191 73.56874 25.12774 13.64 97.86 Engineering & Machinery 82 68.65659 26.63733 8.23 97.39 Food & Drug Retailers 27 72.72926 15.76113 39.58 95.79 Food Producers & Processors 72 72.13361 28.77839 6.84 97.47
Forestry & Paper 14 31.67143 27.6313 12 96.5
General Retailers 135 63.64015 26.78202 2.55 97.69
Health 71 62.8807 24.11552 11.42 98.16
Household Products 25 59.3576 30.16187 11.57 95.86
IT Hardware 62 83.91419 22.72899 9.87 98.41
Insurance 168 55.55077 24.72446 5.74 95.69
Leisure & Hotels 75 71.78987 23.79306 16.79 98.37
Leisure Goods 12 80.77667 21.14318 20.49 95.47
Life Assurance 6 78.71833 18.17362 51.16 92.71
Media & Entertainment 23 41.76087 28.5968 3.78 90.54
Mining 7 88.57714 8.212528 79.21 96.02
Oil & Gas 128 71.89586 26.43635 4.3 98.67
Pharmaceuticals and Biotechnology 75 58.11133 31.1996 7.86 97.53
Publishing 36 51.03778 24.50563 7.87 93.95
Real Estate 50 50.3826 18.30541 16.08 87.28
Renewable Energy 18 92.07278 2.570049 87.51 96.68
Software & Computer Services 119 62.39479 28.66813 11.16 98.23 Speciality & Other Finance 74 51.50905 28.46578 5.04 94.51
Steel & Other Metals 15 53.06 21.74452 15.96 82.81
Telecommunication Services 57 64.02404 27.12495 6.1 96.88
Tobacco 14 81.06929 14.69477 47.56 95.79
Transport 44 58.88364 30.2436 4.57 96.46
Utilities – Other 104 75.96183 21.28478 13.2 98.2
29 Table A.6 Descriptive Statistics Legal Origin – Country
Notes: This table depicts the descriptive statistics by groups of countries, here each country is placed in its group based on its respective legal system. TotalESG is an equally weighted score of all ESG components provided by Thomson Reuters ASSET4 in order to quantify the average CSR score of a firm.
Total ESG
Legal Origin N Mean SD Min Max
English 2131 66.30992 27.23416 4.3 98.67
French 76 69.51618 23.24201 3.78 96.03
German 87 60.86471 33.46322 2.55 96.7
Scandinavian 19 28.15316 19.74543 4.74 70.32 Grand 2313 65.89702 27.54584 2.55 98.67 RatEng RatFre RatGer RatSca RatSoc
Legal Origin Mean
30 Table A.7 Home country Total ESG score summary
Notes: Total ESG is an equally weighted score of all ESG components provided by Thomson Reuters ASSET4 in order to quantify the average CSR score of a firm. This table shows the descriptive statistics by the Home Country of each firm.
Home Country N Mean SD Min Max
31 Table A.8 CEO origin, Two stage least square regression
Notes: This table present the second stage of the 2SLS regression. Instrumental variables include SDtimebrd, TobinsQ and FirmSize. ForCEO is the ratio of foreign CEOs in a firm, mostly this is either 1 or 0. All regressions include firm and year fixed effects. In parenthesis the robust standard errors are included. Variable descriptions can be found in Table A.1 of the Appendix. ***, ** and * indicate statistical significance at 1, 5 and 10 % respectively.
Variables Total ESG Environmental Social Governance
32 Table A.9 Legal origin, Two stage least square regression
Notes: This table present the second stage of the 2SLS regression. Instrumental variables include
ForCeo, SDtimebrd, TobinsQ and FirmSize. LegalBlau is Blau’s index of heterogeneity. All
regressions include firm and year fixed effects. In parenthesis the robust standard errors are included. Variable descriptions can be found in Table A.1 of the Appendix. ***, ** and * indicate statistical significance at 1, 5 and 10 % respectively.
Variables Total ESG Environmental Social Governance
33 Table A.10. Results Hausman Test
Notes: This table presents the results for the Hausman test. It tests whether fixed or random effects are appropriate to use in regression. A small p-value indicates fixed effects are preferred.
Variable p-value
%Foreign 0.00
ForCEO 0.00
LegalBlau 0.00
Table A.11 Results Durbin-Wu-Hausman Test
This table presents the results for the Durbin-Wu-Hausman test for endogeneity. A small p-value indicates that an OLS regression is not consistent and a two stage least square regression should be performed.
Variable p-value
%Foreign 0.00
ForCEO 0.00
