Political Connections, CEO Duality
and, the firm’s ESG Disclosure:
Evidence from the Financial Sector
Master Thesis Accountancy
Personal Details
Name Rick Schaper
Student number S2900181
Email r.j.schaper@student.rug.nl
University University of Groningen
Faculty Economics and Business
Master Accountancy
Supervisor Dr. S. Mukherjee
Date June 24, 2019
1
Abstract
This study investigates the relationship between politically connected board members and ESG disclosure within the financial sector. Further, the interacting effect of CEO duality is explored in this study. Financial organizations are still heavily discussed due to their disclosure levels. Based on previous studies and the resource dependency theory, I expect that politically connected directors and conservative-leaning directors negatively influence the level of ESG disclosure in the financial sector. While from an agency perspective, I expect that CEO duality enhances both negative relationships. Based on a sample of 3183 firm-observation of 739 firms over a time-span 2003-2015, I performed a fixed-effect (within) regression. The results show that politically connected directors and conservative-leaning directors are negatively related to ESG disclosures in the financial sector while CEO duality enhances both negative relationships. These results can be explained because,
connected financial organizations: have fewer incentives, are afraid of exposure of political favors, or have reduced corporate governance quality. Besides, the reluctance towards ESG activities by conservative-leaning boards may also explain the lower levels of ESG
disclosure. This study contributes to the existing disclosure and corporate governance
literature by expanding the literature on ESG disclosure within the financial sector and filling the gap left by other scholars who researched the relationship between politically connected board members and disclosures.
Keywords: Resource Dependency Theory, ESG Disclosure, Political Connections, Board of Directors, Financial Sector, Political Leaning, Conservative, CEO Duality
2
Inhoud
1. INTRODUCTION ... 3
2. THEORETIC FRAMEWORK & HYPOTHESES ... 7
3. RESEARCH DESIGN ... 11
3.1 Sample and Data ... 11
3.2 Variables ... 12
3.2.1 Dependent Variable: ESG Disclosure ... 12
3.2.2 Explanatory Variables: Politically Connected Board & CEO Duality ... 12
3.2.3 Control Variables ... 13
3.3 Empirical Model ... 15
4. RESULTS ... 17
4.1 Descriptive Statistics ... 17
4.2 Correlations ... 17
4.3 Regression and Findings ... 18
4.4 Robustness tests ... 19
5. DISCUSSION AND LIMITATIONS ... 22
REFERENCES ... 24
TABLES ... 33
APPENDIXES ... 41
3
1. Introduction
The financial sector is globally seen as highly important in encouraging social and environmental disclosures of other industries (Chaudhury, et al., 2011; Scholtens, 2008). However, compared to other industries, their disclosure level on CSR remains low (Day & Woodward, 2009). This has led to increasing pressure from the government and the society on financial organizations to be more transparent about CSR(Chaklader & Gulat, 2015; Dissanayake, et al., 2016; Weber, et al., 2014). In fact, the government has taken an activist role in promoting social and environmental reporting of organizations (Steurer, 2010). A way for organizations to shield themselves from political interference and possible unfavorable regulation is by appointing former politicians to their board since it provides a link with the government (Mellahi, et al., 2016). This brings advantages to the firm, but lowers the overall monitoring quality of the board of directors which may be beneficial for the CEO in pursuing their course of action and gaining personal benefits (Agrawal & Knoeber, 2001; Faccio, et al., 2006; Faccio, 2006; Kang & Zhang, 2018; Ye & Li, 2017). Hence, the purpose of this thesis is to examine the influence of politically connected directors on Environmental, Social & Governance (ESG) disclosure and the moderating role of CEO duality. ESG disclosures slightly differ from CSR disclosures since it is more focused on sustainable, ethical, and corporate governance issues (Chadwick, 2013). However, in previous literature, ESG disclosure is often used as a proxy for CSR due to the similarities between the two disclosures (Giannarakis, 2014). It is found that social and human capital significantly influences disclosure levels although, the political background of the directors is often ignored(Johnson, et al., 2013; Niu & Chen, 2017; Said, et al., 2013; Stevenson & Radin, 2009). Moreover, the political leaning of those directors are expected to have an impact on corporate decisions and outcomes, such as ESG activities, since liberals and conservatives have different views and values which influences those decisions and outcomes (Chin, et al., 2013; Hambrick & Mason, 1984; McCright & Dunlap, 2010). Therefore, I will also examine this in this thesis. In addition, when the CEO is also chair of the board, it will influence the overall corporate governance quality of organizations and they will have the power to appoint such directors, which may be in their favor to achieve wanted actions and outcomes (Baliga, et al., 1996; Haniffa & Cooke, 2002; Rechner, 1989).
Since ESG disclosures are mainly voluntary by nature, firms have different reasons to disclose them (Gamerschlag, et al., 2011). Due to external pressures, for example, but also due to the expectations of society, legal requirements, economic incentives, and reputational
4
concerns (Aldrugi & Abdo, 2014). Especially the governmental pressure on financial
organizations cause that they are more likely to engage in ESG related activities (Weihena, et al., 2017). Based on the resource dependency theory, one way to deal with the governmental pressure is to appoint politically connected directors since they provide external linkages between the government and the organizations (Hillman, 2005; Kim & Cannella, 2008; Lester, et al., 2008). The politically connected board members may (1) provide information about the public policy process (Hillman, et al., 1999); (2) provide a communication channel or access to the government with whom the board member is aligned (Hillman, 2005); (3) have influence over political decisions (Pfeffer, 1972); and (4) legitimacy, in which a firm can find approval for their activities (Galaskiewicz & Wasserman, 1989). All this may result in advantages for the firm like lighter taxation, preferable treatment in competition for government contracts, easier access to debt financing, a higher probability of corporate bailouts and relaxed regulatory oversight of the company (Agrawal & Knoeber, 2001; Faccio, et al., 2006; Faccio, 2006). These advantages are not only beneficial for the firm as a whole, but also for the CEO (Awasthi, et al., 2016; Goldman, et al., 2009). This gives CEOs, who are also chair of the board, next to the protection of unfavorable regulation and sanctions, more incentives to appoint such directors. However, both CEO duality and politicians on the board results in bad corporate governance quality which gives the CEO better chances to gain personal benefits and less incentives for voluntary disclosures, while politically connected board members, and mainly the conservative board members, are reluctant to ESG related activities and disclosures due to their political values, and exposure risks (Bona‐Sánchez, et al., 2014; Kang & Zhang, 2018; Leuz & Oberholzer-Gee, 2006; McCright & Dunlap, 2010; Rechner, 1989; Ye & Li, 2017).
Studies of Owolabi &Iyoha (2017), Sobhani et al., (2011) and Viganò & Nicolai (2009) all show that financial organizations have improved their CSR reporting in the past years. However, there are mixed results concerning the influence of the political background of directors on corporate disclosures. Some scholars found a positive association between political connections and the overall disclosure levels of the firms (Harymawan & Nowland, 2016; Watts & Zimmerman, 1978), while others found a negative association between political connections and the overall disclosure levels (Banerji, et al., 2018; Chaney, et al., 2011; Habib, et al., 2018; Hung, et al., 2018; Leuz & Oberholzer-Gee, 2006). With regard to CSR related disclosures, some scholars found a positive relationship (Fernández‐Gago, et al., 2018; Marquis & Qian, 2014; Rahman & Ismail, 2016), while others found a negative
5
relationship between political connections and CSR related disclosures (Bona‐Sánchez, et al., 2014; Muttakin, et al., 2018a; Ramón-Llorens, et al., 2018). Overall the results show a
negative relationship between political connections and corporate disclosures. In addition, most studies found a negative association between the conservative political leaning and CSR related disclosures (Chin, et al., 2013; Di Giuli & Kostovetsky, 2014; Gupta, et al., 2017; Hoover & Fafatas, 2014). The negative results above can be enhanced by chairman CEOs, since most studies found a negative association between CEO duality and CSR related disclosures (Giannarakis, 2014; Husted & Sousa-Filho, 2018; Lagasio & Cucari, 2019; Lattemann, et al., 2009; Li, et al., 2010; Ong & Djajadikerta, 2017; Samaha, et al., 2015). The sample of this research consists of 3183 firm observations from 739 organizations out of 33 different countries over the period 2003-2015. A fixed-effect (within) regression was performed in this study to test the hypotheses. This study found that a politically connected board is significant and negative related to ESG disclosure of financial organizations, which means that financial organizations with a politically connected board member are less likely to disclose on ESG. Next, the study performed a test regarding the political leaning of the board, since the political leaning of board members may influence the actions of the
organization (Hutton, et al., 2014; Layman, 1997). The results show that a conservative board is significant and negative related to ESG disclosure in the financial sector. Furthermore, the results show that CEO duality significantly enhances both negative relationships, meaning that the power of the CEO in a politically connected board result in even less ESG disclosures in the financial sector.
Additional robustness tests were performed to control for reverse causality. I test whether the last time a former minister was appointed in the government influences the level of ESG disclosure in the financial sector. The results show that when a former minister’s last
governmental appointment was more in the past, those ministers have a significant and more negative effect on ESG disclosure, then when a minister was more recently in a political position. The same can be said about the conservative ministers, and the results are similar for the moderating effect of CEO duality however, this effect is not significant for
conservative ministers. Thus, the robustness test shows that former politicians on the board influence ESG disclosure in the financial sector. Furthermore, using the propensity score matching (PSM) method, we also find a significant and negative result between politicians on the board and ESG disclosure, and between the conservative leaning of the board and ESG
6
disclosure of financial organizations. Thus, the reverse causality issues are reduced based on the robustness tests performed in this thesis.
This research will contribute to the ESG disclosure literature in two ways. First, to my knowledge, this is the first study to explore the influence of political connections and the political leaning on ESG disclosures together with the interacting variable CEO duality. Therefore, this research tries to fill the gap left by previous research which focused on the effects of political connections and financial reporting quality (Banerji, et al., 2018; Chaney, et al., 2011; Habib, et al., 2018; Harymawan & Nowland, 2016; Leuz & Oberholzer-Gee, 2006) general voluntary disclosure (Hung, et al., 2018) or CSR reporting (Bona‐Sánchez, et al., 2014; Fernández‐Gago, et al., 2018; Marquis & Qian, 2014; Muttakin, et al., 2018a; Rahman & Ismail, 2016; Ramón-Llorens, et al., 2018), but not on the relationship between political connections and ESG disclosures, while there is also limited research of interacting effect of CEO duality and political connections. In addition, studies on the effects of political connections and studies on CSR disclosures were mostly performed in specific countries (Amran & Haniffa, 2011; Benlemlih, et al., 2016; Ezhilarasi & Kabra, 2017; Fernández‐ Gago, et al., 2018; Goldman, et al., 2009; Iatridis, 2013; Osazuwa, et al., 2015), while this study is conducted globally, making the results more generalizable. Second, the literature on ESG/CSR disclosures of financial organizations is still limited. Regarding CSR disclosures, the role of financial organizations is highly ignored in the existing literature and mostly not seen as important by academics (Haniffa & Hudaib, 2004). This, while they received
increasing pressure to be more transparent on CSR, and play a crucial role in changing other industries behavior regarding CSR reporting since their reports are mostly seen as good examples (Carnevale, et al., 2012; Douglas, et al., 2004; Lock & Seele, 2015). This research, therefore, contributes to the existing literature on CSR related disclosures in the financial sector.
The remainder of the thesis is build-up as follows: in the next section, I provide the theoretical framework and present the hypothesizes. Next, I will explain the data and data sources used in this research, together with the explanation of all the variables in this thesis. After that, the statistical results are shown and discussed. To conclude, the final remarks, limitations, and suggestions for further research are presented.
7
2. Theoretic Framework & Hypotheses
The financial sector is a heavily regulated industry with great dependence on the government (Carretta, et al., 2012). This creates risks and uncertainty for financial institutions (Pfeffer & Salancik, 1978). Following the resource dependency theory, politically connected directors are more common in regulated industries to reduce these risks and uncertainty, since they provide an external linkage between the government and the firms (Hillman, 2005; Lang & Lockhart, 1990; Mahon & Murray, 1981; Pfeffer & Salancik, 1978). In contrary to other directors, former ministers are mainly appointed for their networking (Kim & Cannella, 2008; Lester, et al., 2008; Ramón-Llorens, et al., 2018). Therefore, Braiotta & Sommer (1987) argue that former politicians on the board lack the required expertise and background. This means that they influence the overall corporate governance quality of the firm since they do not bring the knowledge, expertise, and skills that are normally required of board members (Muttakin, et al., 2018b; Reeb & Zhao, 2013). However, they can be used as a strategic asset to gain advantages over non-connected firms (Hillman, 2005; Salamon & Seigfried, 1977). For example, Firth et al., (2011) found that politically connected firms tend to have more favorable regulatory treatment, lower litigation risk, and lower propriety cost. Hence, Bona-Sanchez et al., (2014) argue that politically connected firms will be less likely to disclose CSR information in order to prevent that competitors uphold their competitive advantages. This can be explained by the research of Leuz & Oberholzer-Gee (2006) who argue that high levels of transparency expose political favors of questionable legality, and state that firms therefore disclose less information.
Previous studies found mixed results regarding the disclosure levels of politically connected firms. Watts and Zimmerman (1978) and Marquis & Qian (2014) argue that politically connected firms have more general disclosures and more CSR disclosures due to greater scrutiny and pressures. However, Habib et al., (2018) found that politically connected firms have more inferior financial reporting quality than non-connected firms. In addition, Hung et al., (2018) found that firms disclose less voluntary information because they have fewer incentives for voluntarily disclosures than non-connected firms. Ramón-Llorens et al., (2018) found evidence that board members with a political background in Spain negatively affect CSR reporting. Muttakin et al., (2018a) found the same result in Bangladesh and suggests that corporate political connections are negatively related to CSR disclosures.
8
Following the resource dependency theory, the external linkages between the government and the firms are mainly based on the network of the politically connected directors (Pfeffer & Salancik, 1978). Therefore, the political party/parties for which the former politicians served the government, are the core of their network. These parties have their own political leanings, which reflect the values for which the parties stand for (Hutton, et al., 2014; Layman, 1997). These values influence the decisions of individuals, and thereby influence outcomes
(Hambrick & Mason, 1984). For example, Detomasi (2008; p.815) argues in his study that “conservative parties tend to value individualism, free market mechanisms, and a limited role for the government.” In organizations, conservatives care more about profits and thus prefer the shareholders’ wealth over stakeholders’ needs (Tetlock, 2000). This may explain why conservatives care less about CSR since they feel that CSR is not profit-maximizing for organizations (Walters, 1977). In fact, McCright & Dunlap (2010) argue in their study that conservatives are very hostile towards environmental and climate change activities. Hence, conservatives are less likely to engage in CSR activities or increase their performances (Gupta, et al., 2017). Therefore, organizations are less likely to disclose on CSR since firms with low CSR performances are less likely to disclose it (Clarkson, et al., 2008).
Previous studies found some significant results between political ideology and CSR. Di Giuli & Kostovetsky (2014) found that the political leaning of organizations and the political leaning of their founders, influences the CSR activities and that companies in liberal states and with liberal founders engage more in CSR than conservative ones. Hoover & Fatatas (2014) found similar results, and state that firms headquartered in Republican-leaning states have lower environmental disclosure scores. Chin et al., (2013) found that conservative-leaning CEOs engage less in CSR activities than liberal-conservative-leaning CEOs, and state that they only engage in CSR engagement when the financial performances allow it, showing that conservatives are more profit-oriented. Gupta et al., (2017) found that conservative-leaning firms engage less in CSR activities and that these results are even stronger when the
performances of CSR are low within the industry. As stated in the introduction, CSR and ESG are quite similar to each other. Hence, this leads to the following hypotheses:
H1: Organizations within the financial sector with politically connected board members have a lower level of ESG disclosure
H1a: Organizations within the financial sector with conservative board members have a lower level of ESG disclosure
9
The power of a Chief Executive Officer (CEO) has a significant effect on organizational outcomes and disclosure decisions (Muttakin, et al., 2018b). When a CEO also holds the position of chair of the board, it will give the CEO more control over the board of directors and the organization (Jensen, 1993). The CEO then has the power to set agenda’s, determine directors’ tenures, appoint directors who are in favor of them and control organizational outcomes (Baliga, et al., 1996; Haniffa & Cooke, 2002; Imhoff, 2003). Furthermore, CEOs have control over the information provided to other board members, and can therefore withhold important information (Krause, et al., 2013). This leads to increased information asymmetry between the board and the CEO (Kim, et al., 2009). Rechner (1989) argued that organizations have weak corporate governance when the CEO is also chair of the board, due to monitoring ineffectiveness. This can be a result of impaired independence of the board of directors since there is no separation of powers and the CEO undermining the functions of the board. (Barako, et al., 2006; Ugwoke, et al., 2013). In that case, the CEO can constrain the board to control the organizations effectively (Millstein, 1992; Tuggle, et al., 2010). From the agency perspective, this would lead to conflicts of interests in which the CEO engages in their own interests (Jensen & Meckling, 1976). This has led to higher compensations of chairman CEOs in the financial sector (Ting & Huang, 2018). Thus, organizations with CEO duality give CEOs great power to make decisions that could increase their individual wealth at the costs of the shareholders’ wealth. (Said, et al., 2009). Hence, the dual role of the CEO allows them to ignore the interest of the stakeholders when they make decisions, resulting in less interest in the society and the environment, and consequentially the related disclosures (Muttakin, et al., 2018b). The latter will most likely be the case since CEOs have fewer incentives to engage in social and environmental activities since it is seen as costly and can reduce the individual wealth of the CEO (Ahmad, et al., 2017).
While some studies show no significant relationship between CEO duality and CSR related disclosures (Barako, et al., 2006; Buniamin, et al., 2008; Ezhilarasi & Kabra, 2017; Hu & Loh, 2018), others did found CEO duality to have a significant impact. Gul & Leung (2004) and Forker (1992) found that CEO duality is negatively associated with voluntary corporate disclosures, because CEO duality weakens the oversight role of the board. This is in line with the research of Haniffa & Cooke (2002), who found that role duality negatively affects the extent of voluntary disclosures in Malaysia. Allegrini & Greco (2013) also found a significant and negative relation between CEO duality and voluntary disclosures Giannarakis (2014) and Li et al., (2010) found that organizations with CEO duality disclose less information on CSR,
10
while Latteman et al., (2009) found that organizations with CEO duality have a low CSR communication intensity compared to other companies. Husted & Sousa-Filho (2018) studied the relationship between CEO duality and ESG disclosure in Latin America, where he also found negative results due to the importance of the separation of the roles of board chairman and CEO in Latin America. Ong & Djajadikerta (2017) studied the impact of CEO duality on the separate economic, social, and environmental disclosures and found that firms with CEO duality have significant less disclosures than those without CEO duality.
Chairman CEOs may have various incentives to appoint or hold politically connected directors to its board of directors. First, the political connections of board members can extract additional rents and lower tax payment, which will lead to improved profits in the future along with the connected compensation (Awasthi, et al., 2016). In addition, the announcement of appointing a former politician to the board leads to a positive abnormal stock return (Goldman, et al., 2009). Moreover, Gupta and Wowak (2016) show that the political ideology of the board influence CEO compensation. They found that a conservative-leaning board results in a higher compensation of the CEO than a liberal-conservative-leaning board, giving chairman CEOs more incentives to appoint or hold former conservative ministers to the board. In addition, the political connections of directors give the firms lower litigation risks (Firth, et al., 2011). In addition, politically connected directors may also use their connections to receive personal benefits (Bertrand, et al., 2018). As stated above, both CEO duality and politically connected directors lower the corporate governance quality of the firms, but this may be even more the case when those directors are directly appointed by the CEO, since it may results in higher loyalty and social obligation towards to CEO, reducing its independence (Chen, 2014; Muttakin, et al., 2018b). Since ESG disclosures are more focused on ethical and governance-related issues, higher disclosures may expose these benefits and governance inefficiencies, which may be harmful for the firms, the CEO and the directors while the decrease of corporate governance quality is also expected to have an influence (Chadwick, 2013; Husted & Sousa-Filho, 2018; Leuz & Oberholzer-Gee, 2006). This leads to the following hypotheses:
H2: CEO duality enhances the negative relationship between politically connected board members and ESG disclosure within the financial sector
H2a: CEO Duality enhances the negative relationship between conservative board members and ESG disclosure within the financial sector
11
3. Research Design
3.1 Sample and Data
The data used in this thesis is collected from different sources. First, the data regarding ESG disclosure is collected from the Thomson Reuters ASSET4 database. Second, the financial and accounting data is collected from the Thomson Reuters Worldscope Database. Third, Corporate Governance data is retrieved from the BoardEx Database. Last, the data regarding the political background of the directors are hand-collected by two students of the University of Groningen. The data on politically connected individual directors was not available in online databases, and therefore, the hand-collected data is unique and improves the reliability of the research. The data from the various databases and the hand-collected data ware later matched with the individual Director IDs and the individual Firm IDs.
I used some criteria to come to the number of observations used in this research. To start, all observations which had no data on ESG disclosure available were excluded. Next, I included only the observations within the financial sector following the classification of Fama and French (1997). In addition, when countries had no eight year-span data available, those observations were excluded. I also excluded observations with a negative book-to-market ratio and observations that were missing control variables used in this research. To conclude, the year 2002 is not included in this research, because the quality of board data is not
sufficient for this year, so only the years 2003-2015 were left after. After excluding the firm observations, there were 3183 firm observations left from 739 organizations out of 33
different countries from 2003-2015. Table 1 shows the distribution of the observations among the countries and years used in this research. This table shows that the United States has far more observations than any other country (1389 observations, covering 43,6% of the total dataset). Further, it can be stated that the number of observations increases over the years. This can be the result of the growing external pressures or awareness to disclose on CSR related issues (Benlemlih, et al., 2016; Weihena, et al., 2017).
12 3.2 Variables
3.2.1 Dependent Variable: ESG Disclosure
Based on previous literature, I used the ESG_SCORE to measure ESG disclosure (Fatemi, et al., 2018; Giannarakis, et al., 2014). However, compared to these researches, I used the Thomson Reuters ASSET4 database to collect the ESG_SCORE. The ESG_SCORE is measured on 178 indicators and a total of three pillars, namely Environmental (based on Resource Use, Emissions, and Innovation), Social (based on Workforce, Human Rights, Community, and Product Responsibility), and Governance (based on Management, Shareholders, and CSR Strategy). The category scores are calculated as follows:
Score = 𝑛.𝑜𝑓 𝑐𝑜𝑚𝑝𝑎𝑛𝑖𝑒𝑠 𝑤𝑖𝑡ℎ 𝑎 𝑤𝑜𝑟𝑠𝑡 𝑣𝑎𝑙𝑢𝑒+
𝑛.𝑜𝑓 𝑐𝑜𝑚𝑝𝑎𝑛𝑖𝑒𝑠 𝑤𝑖𝑡ℎ 𝑡ℎ𝑒 𝑠𝑎𝑚𝑒 𝑣𝑎𝑙𝑢𝑒 𝑖𝑛𝑐𝑙𝑢𝑑𝑒𝑑 𝑖𝑛 𝑡ℎ𝑒 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑜𝑛𝑒 2
𝑛.𝑜𝑓 𝑐𝑜𝑚𝑝𝑎𝑛𝑖𝑒𝑠 𝑤𝑖𝑡ℎ 𝑎 𝑣𝑎𝑙𝑢𝑒
Each category is given a weight which will lead to their total scores. The database uses 150 trained content research analysts to collect the ESG data from different sources like
companies Annual Reports, company websites, NGO websites, Stock Exchange Filings, and CSR Reports. The total ESG_SCORE is discounted for ESG controversies, like lawsuits and fines, which results in a more reliable score. To ensure the quality of the data, which results in the ESG_SCORE, the database uses various methods to ensure this goal, like independent audits and management reviews. The score ranges from zero to 1, in which zero is the lowest possible score, and 1 is the highest possible score.
3.2.2 Explanatory Variables: Politically Connected Board & CEO Duality
A firm is politically connected following the definition of Boubakri et al., (2009; p.369) as “when at least one member of its board of directors or its supervisory board is or was a politician, a member of parliament, a minister, or any other top appointed bureaucrat.” In this research, I focus solely on the board of directors, and thereby classify a board as politically connected when a director previously held a governmental position. POLITIC_BRD is thereby measured in line with the research of Chen et al., (2014b) with a dummy variable ‘1’ if a firm’s board member has ever served the government and ‘0’ otherwise. Next, the
political leaning can be defined as “an interrelated set of attitudes and values about the proper goals of society and how they should be achieved” (Tedin, 1987; p.65). The liberal(left)-conservatism(right) spectrum is identified to determine the political leaning (CONS_BRD) of the board (Schwarz, 1996). I used a dummy variable ‘1’ if a firm’s board is conservative and
13
‘0’ otherwise. The board is listed as conservative if the board has a member who previously held a governmental position for a conservative party. The data on the political directors is mainly hand-collected. BoardEx identified the directors who had previously held
governmental positions. To determine the political leaning of the directors, and indirectly of the board of directors, A fellow student and I searched online for which political party the individual director served the government (See Appendix A). Based on that political party, the political leaning could be determined. If a director served for more than one political party with various leanings, we listed the political party and the political leaning of the director for which he/she served the government the latest or the political party for which he/she served the longest, indicating that this was his/her preferred political party. Last, the interacting variable CEO_DUALITY, defined by Tuggle et al., (2010; p.954) as “CEO’s also being the chairperson of the board” is measured as a dummy variable ‘1’ if the CEO of a firm is also chair of the board, and ‘0’ otherwise. This is consistent with the research of Ahmad et al., (2017). This data is collected from the BoardEx database.
3.2.3 Control Variables
The financial performance of firms has a significant influence on the level of CSR disclosure (Fatemi, et al., 2018; Sial, et al., 2018). First, Return on Assets (PROFIT) and Tobin’s Q (FIRM_VALUE) are both positively related to the level of voluntarily disclosure, due to the fact that these firms are more visible and have better resources and channels to communicate more information (Brammer & Pavelin, 2008; Chen, et al., 2014a; Sial, et al., 2018). The studies of Andrikopoulos & Kriklani (2013) and Setyorini & Ishak (2012) also show that profitability influences the level of CSR disclosure, where the Operating Return of Assets (OP_INCOME), and negative Return of Assets (LOSS_DUMMY) can be seen as other proxies for profitability. The studies suggest that profitability influences the level of CSR disclosure positively due to better resources, where losses negatively affect disclosures. There are mixed results regarding the relationship between leverage ratios (LEVERAGE,
FOREIGN_ASSETS, DEBT_INTEREST_RATE, CASH) and voluntarily disclosure. While some scholars argue that high leverage ratios have a positive effect on disclosures (Clarkson, et al., 2008; Leftwich, et al., 1981), others disagree (Gao & Connors, 2011; Zarzeski, 1996) Next, Aboody & Lev (2000) found that investment intensity leads to information asymmetry, and thereby CAP_EXPENDITURE and R&D are negatively associated with disclosures. Moreover, firms with a high amount of closely held shares (CLOSELY_HELD_SHARES)
14
have fewer incentives to disclosure more voluntarily information (Ghazali, 2007; Lin & Vos, 2006).
Corporate Governance Control Variables
Corporate Governance mechanisms play an important role in monitoring the firms and influence the level of information disclosed to the public. Therefore, the structure of the board of directors influences the level of CSR disclosure of organizations. First, regarding board composition, Liao et al., (2015) found that board independence
(BRD_INDEPENDENCE) and female directors (FEMALE_BRD) have a positive influence on the level of greenhouse gas disclosure due to their better monitoring role within the board while Khan (2010) found the same results for the proportion of foreign directors
(FOREIGN_BRD). Furthermore, Abduh and AlAgeely (2015) found that board size
(BRD_SIZE) is negatively associated with CSR reporting in the banking industry, which may be a result of inefficient governance. In addition, studies on the relationship between board tenure (BRD_TENURE) and CSR reporting show mixed results (Hafsi & Turgut, 2013; Krüger, 2009). While the results of Khan et al., (2013) and Kiliç (2016) suggest that directors with experience in quoted firms (BRD_QOUTED_AFFIL) have a positive influence on CSR disclosure.
Country Control Variables
Country-specific characteristics cause that there are differences among organizations in the way they operate, due to different legislation and economic environments. The study of Jorgensen and Soderstrom (2006) show that social and environmental disclosure differ among developed and less developed countries, and that countries with a higher gross domestic product per capita (GDP_PER_CAPITA) (a proxy for economic development (Adhikari & Tondkar, 1992)) have better disclosures due to greater attention and awareness. In more developed countries, shareholder protection (GUILLENCAPRON_SHR) tends to be significantly higher than in less developed countries (Siems, 2007). As a result, information asymmetry is reduced in the form of better disclosures (Koch, et al., 2014).
Next, to the control variables mentioned above, I further control for business segments (BUSINESS_SEG), geographic segments (GEOGRAPIC_SEG), and reporting regulation (IFRS). Table 2 shows the measurements of the control variables.
15 3.3 Empirical Model
This research is conducted over the years 2003-2015 and makes use of panel data. The panel data allows us to control for variables that we cannot observe/measure/or change over time and will add accuracy to the predicted variables, and in general result in more accurate results (Hsiao, 2007). I used a fixed-effect (within) regression to tests the hypotheses in this
research. The first model examines the relationship between the dependent variable
(ESG_SCORE) and the independent variables (POLITIC_BRD& CONS_BRD) with all the control variables.
Model 1 (Hypotheses 1)
ESG_SCOREit = β0 + β1 POLITIC_BRD β2 CONTROLS it + αi + μit
Where ESG_SCOREit is the ESG disclosure score based on the Thomas Reuters ASSET4
database for firm i in year t. βi represents the coefficients. The score ranges from zero to 1, in which zero is the lowest possible score, and 1 is the highest possible score. POLITIC_BRD
represents the dummy variable, whether a firm has a political board member. CONTROLS it
represents all the control variables listed in Table 2 for firm i in year t. αiis the unobserved time-invariant individual effect for firm i. μitis the error term for firm i in year t. If coefficient β1 is negative and statistically significant, it indicates that our main first hypothesis is
accepted. This means that firms with politicians on the board have a negative influence on the ESG disclosures of the firm. When it is not significant, it means that politicians in the board do not affect the firm’s ESG disclosure.
The second model examines the relationship between the dependent variable (ESG_SCORE) and the interacting variables POLITIC_BRD * CEO_DUALITY & CONS_BRD *
CEO_DUALITY. This is shown below.
Model 2 (Hypotheses 2)
ESG_SCOREit = β0 + β1 POLITIC_BRD + β2 CEO_DUALITY + β3 (POLITIC_BRD * CEO_DUALITY) + β2 CONTROLS it + αi + μit
Where CEO_DUALITY is the dummy variable whether the CEO is also chairman of the board. POLITIC_BRD*CEO_DUALITY is the interacting variable where the CEO is chair of the board in a firm with politically connected directors. This is measured with a dummy variable ‘1’ if the CEO of a firm is also chair of the board, and ‘0’ otherwise. The
16
coefficient β3 is negative and statistically significant, it indicates that our second hypothesis is accepted. This means that the influence of the politicians on the board on ESG disclosure of the firms is negatively affected by the CEO, who is chair of the board of directors. When it is not significant, it means that the chair CEO doesn’t influence the relationship between the politicians on the board and the firm’s ESG disclosure. Table 2 shows the measurements of all the variables, including all the control variables. Some variables are winsorized at 1% level at the top and bottom to make sure that outliers do not influence the results.
17
4. Results
In this section, the results of the tests are presented. In the first paragraph the descriptive statistics will be discussed, followed by the Pearson correlation matrix in the second section. Third, the main results of the fixed-effect (within) regression are presented and discussed. To conclude, the final paragraph shows the results of the robustness tests of this study using other independent variables to measure the influence of the politically connected board members and a test using the propensity score matching (PSM) method.
4.1 Descriptive Statistics
Table 3 shows the descriptive statistics of the dependent variable, the independent and interacting variables, and the control variables used in this study. Most of the variables were winsorized at 1% at the top and the bottom to eliminate the outliers in this research (See Table 2). The descriptive statistics show that the average ESG_SCORE is 0.506 with a standard deviation of 0.175. This means that the financial companies score sufficient (B-) on the Thomson Reuters Score Scale on their ESG disclosures. The mean of the POLITIC_BRD is 0.077, with a standard deviation of 0.267 and the average CONS_BRD 0.054, with a standard deviation of 0.054. This means that 7.7% of our sample has a former politician on their board, indicating that they are politically connected and most of those former politicians were conservative, with 5.4% of our sample has a conservative-leaning board. Furthermore, we see that the average CEO_DUALITY is 0.273 with a standard deviation of 0.446. Thus 27,3% of our sample has a CEO who is also chair of the board.
[TABLE 3]
4.2 Correlations
Table 4 shows the results of the Pearson correlation tests. This test is performed to determine the correlations between the variables and to determine whether multicollinearity exists. Multicollinearity is the case when the correlation between two or more variables is high, which results in large sampling errors (Blalock, 1963). When a coefficient equals or exceeds 0.7, a multicollinearity problem may occur however, it does not have to be the case since it just serves as a warning (Yu, et al., 2015). We can see that the correlation between
FIRM_VALUE and OP_INCOME is above the 0.7 threshold (0.728) and that the correlation between PROFIT and LEVERAGE is near the 0.7 limit (0.683). To make sure that
multicollinearity is not a problem, Bollen (1989) states that the variance inflation factor (VIF) can predict whether multicollinearity will be a problem. Kutner et al., (2004) proposed that
18
when a VIF is exceeding the value of 10, multicollinearity will be a problem where Yu et al., (2015) use a value of 5 as a limit. When estimating the VIFs of the variables, no value goes above the limit stated by Yu et al., (2015), which increases the credibility of the results. Hence, it can be concluded that multicollinearity will not be a problem during the regressions. Furthermore, the correlation matrix can be used to determine relations between variables. The results show that all variables, except for CASH, are significantly correlated to
ESG_SCORE, with only GDP_PER_CAPITA correlated at a 10% level. In addition, the results show a significant positive relationship between ESG_SCORE and POLITIC_BRD (r = 0.054; p < 0.01), which is in contrast to our first hypotheses. A significant negative relation is found between CEO_DUALITY and ESG_SCORE (r = 0.61; p < 0.01) which is in line with the second hypotheses.
[TABLE 4]
4.3 Regression and Findings
This study examines the impact of a politically connected board and their political leaning on ESG disclosures and the interacting effect of CEO duality. A fixed-effect (within) regression is performed to test the hypotheses. The fixed-effect regression model is a frequently used model with panel data. Table 5 shows the results of the regression.
Model 1 shows the base model. This model shows some significant results. It indicates that
higher ESG disclosures are statistically associated with firms with higher firm size (β = 0.024; p < 0.01), with a higher operating return of assets (β = 0.175; p <0.05), with a higher proportion of woman on the board of directors (β = 0.063; p < 0.1), in countries with higher Gross Domestic Product per Capita (β = 0.087; p < 0.01), and firms with a lower number of business segments (β = 0.014; p < 0.05),.
Model 2 shows the result for the relation between the independent variable (POLITIC_BRD)
and the dependent variable (ESG_SCORE). The model shows a significant negative relation between a politically connected board and ESG disclosure (β = 0.037; p <0.05), supporting the first hypotheses (H1). This means that former politicians on the board are associated with less disclosure of ESG related information of organizations. Model 2a tests the
sub-hypotheses regarding the political leaning of the board. The model shows that a conservative-leaning board of directors (CONS_BRD) is significant and negatively related to ESG
19
in line with the reasoning that conservatives are not supportive of the social and environmental activities of organizations (McCright & Dunlap, 2010).
Model 3 shows the results of the interacting relationship (POLITIC_BRD *
CEO_DUALITY) and the dependent variable. This model shows a significant negative relationship between the interacting variable and the dependent variable (β = 0.065; p < 0.05), supporting our second hypotheses. This result implies that when a CEO is also chair of the board with a former politician on it, the ESG disclosures will be even less for firms with only former politicians on the board. The model further shows that CEO_DUALITY alone, is significantly negative related to ESG_SCORE (β = 0.013; p < 0.1). The model also shows a significant positive relationship between FIRM_VALUE and ESG_SCORE (β = 0.009; p < 0.1). Model 3a shows the results for the interacting relationship when a CEO is also chair of the board with a conservative board member. These results are consistent with model 3, but at a lower significance (β = 0.061; p <0.1), supporting the sub-hypotheses (H2a).
All the models testing the hypotheses are showing the same results as the base model, except for the relationship between FEMALE_BRD and ESG_SCORE, which is no longer
significant in the models. PROFIT, OP_INCOME, GDP_PER_CAPITA, BUSINESS_SEG, and GEOGRAPHIC_SEG are all significant positive related to ESG_SCORE.
[TABLE 5]
4.4 Robustness tests
Robustness tests were performed to control for reverse causality. Therefore, additional regressions were performed. The values of the political parties change over time regarding social and environmental issues since there is a greater concern for those issues, which demands more accountability of organizations (Mchavi & Ngwakwe, 2017). Hence political parties change their policies and respond to the desire of the voters (Adams, et al., 2004; Haque & Islam, 2012; Mchavi & Ngwakwe, 2017). Therefore, politicians who were more recently in governmental positions have more experience with social, and environmental issues and are more willing to social, and environmental activities (Hambrick & Mason, 1984; Khatri & Ng, 2000). Since views of politicians and political parties change over time, I performed regressions which tests whether the time when a politician last served the
government matters regarding ESG disclosures. Therefore, I made two variables to test this relationship. First, LST_GOV_APP measures the years that a former politician in the boardroom was last appointed in the government. The higher the years, the more years back
20
the former politician was last appointed in the government. Second,
LST_CONS_GOVN_APP measures the years a conservative politician in the boardroom was last appointed in the government, and thus for a conservative party. The higher the years, the more years back it was that a conservative politician was last appointed in the government. These results are shown in table 6. Model 4 shows that a former politician whose last appointment is more back in time (LST_GOV_APP) is significant and negative related to ESG_SCORE (β = 0.011; p < 0.1), supporting our main results which state that politicians on the board matter regarding ESG disclosure. These results are even more significant regarding the last appointment of conservative politicians, (LST_CONS_GOVN_APP) (β = 0.017; p < 0.01), shown in model 5. This also implies that the views of politicians change over time and that the conservative views regarding social and environmental activities were even severe back in time. Model 6a shows the interacting relationship between LST_GOVN_APP and CEO_DUALITY, and ESG_SCORE, which is also significant and negative (β = 0.033; p < 0.5) while model 6b also shows a negative relation, but no significant relation, for the interacting relationship between LST_CONS_GOVN_APP and CEO_DUALITY, and the dependent variable.
[TABLE 6]
Next, to these robustness tests, I performed additional robustness tests using the propensity score matching (PSM) method to test whether the influence of the former politicians and former conservatives on the board has a negative and significant influence on ESG
disclosures of firms. Based on the research Lin (2019), a logit estimation is used to estimate a propensity score for the used variables to find a matching index. Table 7 shows the
coefficient estimates of the logit model. Model 7 shows the coefficient estimations of the variables regarding the politicians (treatment group) on the board while model 7a shows the coefficient estimations of the variables regarding the conservatives on the board (treatment group). The models, for example, show that politicians and conservatives are more likely to sit on the board when firms have lesser values and profits, have lower leverage, have bigger boards, and are more likely to sit on the board in firms that are active in more geographic segments.
21
Furthermore, the quality of the balancing of the control and the treatment samples is checked using following the research of Lin (2019) by performing a pstest. The quality of the
balancing can be assured when all the variables have insignificant t-values after matching (Largoza, et al., 2015). Table 8 shows the results of the pstest for the treatment group
politicians on the board (model 8) and the treatment group conservatives on the board (model
8a). Both models show that there all the variables have insignificant t-values after matching,
so the quality of the balancing is assured to tests the treatment effects on firms’ ESG disclosures.
Table 9 shows the results of robustness tests using the PSM method. These results indicate that politicians on the board (β =0.060; p<0.05), and conservatives on the board (β =0.029; p<0.1) are both significant and negative related to the firm’s ESG disclosure in the financial sector. This is consistent with our previous results from the regressions to tests our
hypotheses. Based on the robustness tests using the last time a former politician was appointed in the boardroom as an independent variable and the results of the PSM method, the results of this thesis are robust which reduces the reverse causality issues.
22
5. Discussion and limitations
There are several reasons why organizations would be more transparent about CSR, like governmental pressure, economic incentives, or reputational concerns (Aldrugi & Abdo, 2014). Currently, the financial sector has improved its social and environmental disclosure level. However, their disclosure level is compared to other industries still relatively low, while the financial sector is seen as highly important to encourage CSR related disclosures around the globe. This study examines the influence of politically connected board members on ESG disclosure within the financial sector. As if today, there is not much known about the role of the political connections of board members and their influence on disclosures,
especially on ESG disclosure. This research tries to contribute to the literature by filling the gap left by previous research. This research used data from 3183 firm observations from 33 countries over the period 2003-2015. A fixed-effect (within) regression was used to tests the hypotheses in this research.
The results show that politically connected board members have a significant negative
influence on ESG disclosures, meaning that financial organizations with politically connected board members have significantly less ESG disclosures than firms that are not connected. This is in line with the research of Ramón-Llorens et al., (2018), who found that board members with a political background negatively affect CSR reporting. Further, the results show the same results regarding conservative board members, indicating that the political leaning of board members influences ESG disclosures in the financial sector. This is in line with the research of Chin et al., (2013), who found that conservative executives are less likely to engage in CSR reporting. The study also examined the influence of CEO duality on both the political connection and the political leaning of the board members. The analysis shows that CEO duality significantly enhances both negative relationships. This can be explained due to the lack of independence of the appointed board members, the power the CEO has to set agendas in the board, and the control the CEO has over the information flow among the board members.
This research has some limitations. First, the research is only focused on the financial sector, which means that it may not be generalizable to other industries. Second, a total of 43.6% of our dataset comes from the United States. The research includes 33 different countries, but it is likely that the results are heavily influenced by the United States, putting a limitation to the generalizability of the results. Finally, within the sample of the former politicians on the
23
board, there is a relatively high population of conservative-leaning board members, which indicates that the conservatives mainly influence the results of this thesis. Therefore, the overall results of the influence of the general politically connected board members may be different with a higher population of liberal-leaning board members.
For future research, it may be interesting to research what drives the politically connected board members to influence the ESG disclosures of their firms negatively. Since there can be multiple explanations why firms with politically connected board members disclose less information (fewer incentives, reputation, afraid of exposure, etc.), it is still not clear why they cause this negative impact. Next, it may be worthy to research the difference between developed and less developed countries, since the GDP per capita has a positive and significant influence on the ESG Score, meaning that there are differences in ESG Scores between countries. It might be an interesting research, since the effects of the political connections also differ between developed and less developed countries (Faccio, 2010).
24
References
Abduh, M. & AlAgeely, H. A. M., 2015. The Impact of corporate governance on CSR disclosure in Islamic banks: empirical evidence from GCC countries. Middle East Journal of Management, Volume 2, 283-295.
Adams, J., Clark, M., Ezrow, L. & Glasgow, G., 2004. Understanding Change and Stability in Party Ideologies: Do Parties Respond to Public Opinion or to Past Election Results. Britisch Journal of Political Science, Volume 34, 589-610.
Adhikari, A. & Tondkar, R. H., 1992. Environmental Factors Influencing Accounting Disclosure Requirements of Global Stock Exchanges. Journal of International Financial Management and Accounting, 4(2), 75-105.
Agrawal, A. & Knoeber, C., 2001. Do some outside directors play a political role?. Journal of Law and Economics, Volume 44, 180-198.
Ahmad, N. B. J., Rashid, A. & Gow, J., 2017. CEO Duality and Corporate Social Responsibility Reporting: Evidence from Malaysia. Corporate Ownership & Control, 14(2), 69-81.
Aldrugi, A. & Abdo, H., 2014. Determining the Motives or Reasons that Make Companies Disclose Environmental Information. Journal of Economics, Business and Management,, 2(2), 117-121. Allegrini, M. & Greco, G., 2013. Corporate boards, audit committees and voluntary. Journal of Management and Governance, Volume 17, 187-216.
Amran, A. & Haniffa, R., 2011. Evidence in development of sustainability reporting: a case of a developing country. Business Strategy and the Environment, 20(3), 141-156.
Andrikopoulos, A., Samitas, A. & Bekiaris, M., 2014. Corporate social responsibility reporting in financial institutions: evidence from Euronext. Research in International Business and Finance, Volume 32, 27-35.
Awasthi, K., Yayavaram, S. K., George, R. & Sastry, T., 2016. Effect of Political Connections on Profit Persistence: Evidence from India. Academy of Management Proceedings, Volume 1.
Baliga, B. R., Moyer, R. C. & Rao, R. S., 1996. CEO duality and firm performance: what's the fuss. Strategic Management Journa, 17(1), 41-53.
Banerji, S., Duygun, M. & Shaban, M., 2018. Political connections, bailout in financial markets and firm value. Journal of Corporate Finance, Volume 50, 388-401.
Barako, D. G., Hancock, P. & H.Y. Izan, H., 2006. Relationship between Corporate Governance Attributable and Voluntary Disclosures in Annual Reports: The Kenyan Experience. Financial Reporting, Regulation and Governance, 5(1), 1-25.
Benlemlih, M., Qiu, Y. & Shaukat, A., 2016. Environmental and Social Disclosure and Firm Financial Risk: Evidence from UK. Journal of Business Ethics, 152(3), 613-626.
Bertrand, M., Kramarz, F., Schoar, A. & Thesmar, D., 2018. The Cost of Political Connections. Review of Finance, 22(3), 849-876.
Blalock, H. M., 1963. Correlated Independent Variables: The Problem of Multicollinearity. Social Forces, 42(2), 233-237.
25
Bona‐Sánchez, C., Pérez‐Alemán, J. & Santana‐Martín, D. J., 2014. Politically Connected Firms and Earnings Informativeness in the Controlling versus Minority Shareholders Context: European Evidence. Corporate Governance: An International Review, 22(4), 330-346.
Boubakri, N., Cosset, J.-C. & Guedhami, O., 2009. From state to private ownership: Issues from strategic industries. Journal of Banking & Finance, 33(2), 367-379.
Braiotta, L. J. & Sommer, A. A., 1987. The essential guide to effective corporate board. NJ ed. Englewood Cliffs: Prentice-Hall .
Brammer, S. & Pavelin, S., 2008. Factors Influencing the Quality of Corporate Environmental Disclosure. Business, Strategy and the Environment, Volume 17, 120-136.
Buniamin, S., Alrazi, B., Johari, N. H. & Rahman, N. R. A., 2008. An Investigation of the Association between Corporate Governance and Environmental Reporting in Malaysia. Asian Journal of Business and Accounting, 1(2), 65-88.
Carnevale, C., Mazzuca, M. & Venturini, S., 2012. Corporate Social Reporting in European Banks: The Effects on a Firm’s Market Value.. Corporate Social Responsibility and Environmental Management, Volume 19, 159-177.
Carretta, A., Farina, V., Gon, A. & Parisi, A., 2012. Politicians ‘on Board’: Do Political Connections Affect Banking Activities in Italy. European Management Review, Volume 9, 75-83.
Chadwick, P., 2013. IEDP. [Online]
Available at: https://www.iedp.com/articles/understanding-the-esgcsr-value-chain/ [Accessed 22 March 2013].
Chaklader, B. & Gulat, P. A., 2015. A Study of Corporate Environmental Disclosure Practices of Companies Doing Business in India. Global Business Review, 16(2), 321-335.
Chaney, P., Faccio, M. & Parsley, D., 2011. The quality of accounting information in politically connected firms. Journal of Accounting Economics, 51(1-2), 58-76.
Chaudhury, S. K., Das, S. K. & Sahoo, P. K., 2011. Practices Of Corporate Social Responsibility (Csr) In Banking Sector In India: An Assessment. Research Journal of Economics, Business and ICT, Volume 4, 76-81.
Chen, H., 2014. Board Capital, CEO Power and R&D Investment in Electronics Firms. Corporate Governance: An International Review, 22(5), 422-436.
Chen, J. J., Cheng, X., Gong, S. X. & Tan, Y., 2014a. Do higher value firms voluntarily disclose more information? Evidence from China. The British Accounting Review, 46(1), 18-32.
Chen, Y.-S., Shen, C.-H. & Lin, C.-Y., 2014b. The Benefits of Political Connection: Evidence From Individual Bank-Loan Contracts. Journal of Financial Services Research, 45(3), 287-305.
Chin, M. K., Hambrick, D. C. & Treviño, L. K., 2013. Political Ideologies of CEOs: The Influence of Executives' Values on Corporate Social Responsability. Administrative Science Quarterly, 58(2), 197-232.
Clarkson, P. M., Li, Y., Richardson, G. D. & Vasvari, F. P., 2008. Revisiting the relation between environmental performance and environmental disclosure: An empirical analysis. Accounting, Organizations and Society, 33(4-5), 303-327.
26
Day, R. & Woodward, T., 2009. CSR reporting and the UK financial services sector. Journal of Applied Accounting Research, 10(3), 159-175.
Detomasi, D., 2008. The political roots of corporate social responsibility. Journal of Business Ethics, Volume 82, 807-819.
Di Giuli, A. & Kostovetsky, L., 2014. Are red or blue companies more likely to go green? Politics and corporate social responsibility. Journal of Financial Economics, 111(1), 158-180.
Dissanayake, D., Tilt, C. & Xydias-Lobob, M., 2016. Sustainability reporting by publicly listed companies in Sri Lanka. Journal of Cleaner Production, Volume 126, 169-182.
Douglas, A., Doris, J. & Johnson, B., 2004. Corporate Social Reporting inIrish Financial Institutions. TQM Magazine, 16(6), 387-395.
Ezhilarasi, G. & Kabra, K. C., 2017. The Impact of Corporate Governance Attributes on Environmental Disclosures: Evidence from India. Indian Journal of Corporate Governance, 10(1), 24-43.
Faccio, M., 2006. Politically–connected firms: Can they squeeze the State?. The American Economic Review, 96(1), 369-386.
Faccio, M., 2010. Differences between Politically Connected and Nonconnected Firms: A Cross‐ Country Analysis. Financial Management, 39(3), 905-928.
Faccio, M., Masulis, R. & McConnell, J., 2006. Political contributions and corporate bailouts. Journal of Finance, Volume 61, 2597-2635.
Fama, E. & French, K., 1997. Industry costs of equity. Journal of Financial Economics, 43(2), 153-193. Fatemi, A., Glaum, M. & Kaiser, S., 2018. ESG performance and firms value: The moderating role of disclosure. Global Finance Journal, Volume 38, 45-64.
Fernández‐Gago, R., Cabeza‐García, L. & Nieto, M., 2018. Independent directors' background and CSR disclosure. Corporate Social Responsability and Environmental Management, 25(5), 991-1001. Firth, M., Rui, O. M. & Wu, W., 2011. The Effects of Political Connections and State Ownership on Corporate Litigation in China. The Journal of Law & Economics, 54(3), 573-607.
Forker, J. R., 1992. Corporate governance and disclosure quality. Accounting and Business Research , 22(86), 111-124.
Galaskiewicz, J. & Wasserman, S., 1989. Mimetic processes within an interorganizational field: An empirical test. Administrative Science Quarterly, Volume 34, 454-479.
Gamerschlag, R., Möller, K. & Verbeeten, F., 2011. Determinants of voluntary CSR disclosure: empirical evidence from Germany. Review of Managerial Science, 5(2-3), 233-262.
Gao, L. S. & Connors, E., 2011. Corporate Environmental Performance, Disclosure and Leverage: An Integrated Approach. International Review of Accounting, Banking and Finance, 3(2), 1-32.
Ghazali, N. A. M., 2007. Ownership structure and corporate social responsibility disclosure: some Malaysian evidence. Corporate Governance: The international journal of business in society, 7(3), 251-266.
Giannarakis, G., 2014. The determinants influencing the extent of CSR disclosure. International Journal of Law and Management, 56(5), 393-416.
27
Giannarakis, G., Konteos, G. & Sariannidis, N., 2014. Financial, governance and environmental determinants of corporate social responsible disclosure. Management Decision, 52(10), 1928-1951. Goldman, E., Rocholl, J. & So, J., 2009. Do Politically Connected Boards Affect Firms Value?. The Review of Financial Studies , 22(6), 2331-2360.
Guillén, M. F. & Capron, L., 2016. State Capacity, Minority Shareholder Protections, and Stock Market Development. Administrative Science Quarterly, 61(1), 125-160.
Gul, F. A. & Leung, S., 2004. Board leadership, outside directors' expertise and voluntary corporate disclosures. Journal of Accounting and Public Policy, Volume 23, 351–379.
Gupta, A., Briscoe, F. & Hambrick, D. C., 2017. Red, blue, and purple firms: Organizational political ideology and corporate social responsability. Strategic Management Journal, 38(5), 1018-1040. Gupta, A. & Wowak, A. J., 2016. The Elephant (or Donkey) in the Boardroom: How Board Political Ideology Affects CEO. Administrative Science Quarterly, 62(1), 1-30.
Habib, A., Ranasinghe, D., Muhammadi, A. H. & Islam, A., 2018. Political connections, financial reporting and auditing: Survey of the empirical literature. Journal of International Accounting, Auditing and Taxation, Volume 31, 37-51.
Hafsi, T. & Turgut, G., 2013. Boardroom diversity and its effect on social performance: Conceptualization and empirical evidence. Journal of Business Ethics, Volume 112, 463-479. Hambrick, D. C. & Mason, P., 1984. Upper Echelons: the organization as a reflection of its top managers. Academy of Management Review, 9(2), 193-206.
Haniffa, R. & Hudaib, M., 2004. Disclosure practices of Islamic financial institutions: An exploratory study. Brisbane, Australia, The International Conference on Accounting, Commerce and Finance: The Islamic Perspective,.
Haniffa, R. M. & Cooke, T. E., 2002. Culture, Corporate Governance and Disclosure in Malaysian Corporations. Abacus, 38(3), 317-349.
Haque, S. & Islam, A., 2012. Stakeholder Pressues and Climate Change Disclosures: Australian Evidence. Accounting & Finance Association of Australia and New Zealand (AFAANZ) Conference. Harymawan, I. & Nowland, J., 2016. Political connections and earnings quality: How do connected firms respond. International Journal of Accounting and Information Management, 24(4), 339-356. Hillman, A., 2005. Politicians on the board of directors: Do connections affect the bottem line?. Journal of Management, 31(3), 464-481.
Hillman, A., Zardkoohi & Bierman, L., 1999. Corporate political strategies and firm performance: Indications offirm specific benefits from personal service in the US government. Strategic Management Journal, Volume 20, 67-81.
Hoover, S. & Fafatas, S., 2014. Political environment and voluntary disclosure in the U.S.: Evidence from the Carbon Disclosure Project. Journal of Public Affairs, 18(4), 1-12.
Hsiao, C., 2007. Panel data analysis—advantages and challenges. TEST, Volume 16, 1-22.
Hu, M. & Loh, L., 2018. Board Governance and Sustainability Disclosure: A Cross-Sectional Study of Singapore-Listed Companies. Sustainability, 10(7), 1-14.
28
Hung, M., Kim, Y. & Li, S., 2018. Political Connections and Voluntary Disclosure: Evidence from Around the World. Journal of International Business Studies, 49(2), 272-302.
Husted, B. & Sousa-Filho, J. M. d., 2018. Board structure and environmental, social, and governance disclosure in Latin America. Journal of Business Research.
Hutton, I., Jiang & Kumar, A., 2014. Corporate policies of Republican managers. Journal of Financial Quantitative Analysis, Volume 49, 1279-1310.
Iatridis, G. E., 2013. Environmental disclosure quality: Evidence on environmental performance, corporate governance and value relevance. Emerging Markets Review, Volume 14, 55-75.
Imhoff, E. A., 2003. Accounting quality, auditing, and corporate governance. Accounting Horizons, Volume 17, 117-128.
Jensen, M. C., 1993. The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance, 48(3), 831-880.
Jensen, M. C. & Meckling, W. H., 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305-360.
Johnson, S. G., Schnatterly, K. & Hill, A. D., 2013. Board Composition Beyond Independence: Social Capital, Human Capital, and Demographics. Journal of Management, 39(1), 232-262.
Jorgensen, B. & Soderstrom, N., 2006. Environmental Disclosure Within Legal and Accounting Context: An International Perspective. Columbia University: s.n.
Kang, J. & Zhang, L., 2018. Do outside directors with government experience create value?. Financial Management, Volume 47, 209-251.
Khan, A., Muttakin, M. B. & Siddiqui, J., 2013. Corporate Governance and corporate social
responsibility disclosures: evidence from an emerging economy. Journal of Business Ethics, 114(2), 207-233.
Khan, H. U. Z., 2010. The effect of corporate governance elements on corporate social responsibility (CSR) reporting. International Journal of Law and Management, Volume 52, 82-109.
Khatri, N. & Ng, A., 2000. The role of inntuition in strategic decision making. Human Relations, 53(1), 57-86.
Kiliç, M., 2016. Online corporate social responsibility (CSR) disclosure in the banking industry: Evidence from Turkey. International Journal of Bank Marketing, 34(4), 550-569.
Kim, K.-H., Al-Shammari, H. A., Kim, B. & Lee, S.-H., 2009. CEO duality leadership and corporate diversification behavior. Journal of Business Research, 62(11), 1173-1180.
Kim, Y. & Cannella, A. A., 2008. Toward a Social Capital Theory of Director Selection. Corporate Governance: An International Review, 16(4), 282-293.
Koch, C. v., Nilsson, O. & Eriksson, K., 2014. Does shareholder protection affect the performance of analysts as a gatekeeper?. Journal of Management & Governance, 18(2), 315-345.
Krause, R., Semadeni, M. & Cannella, A. A., 2013. CEO Duality: A Review and Research Agenda. Journal of Management, 40(1), 256-286.
29
Krüger, P., 2009. Corporate Social Responsibility and the board of directors. Toulouse School of Economics, France, Working Paper (Job Market Paper).
Kutner, M. H., Nachtsheim, C. J. & Neter, J., 2004. Applied Linear Regression Models. fourth ed. Homewood: McGraw- Hill Irwin.
Lagasio, V. & Cucari, N., 2019. Corporate governance and environmental social governance disclosure: A meta-analytical review. Corporate Social Responsibility and Environmental Management, 1-11.
Lang, J. & Lockhart, D., 1990. Increased environmental uncertainty and changes in board linkage patterns”.. Academy of Management Journal, Volume 33, 106-128.
Largoza, G. L., Favorada, I. F. C., Tan, C. V. G. R. S. G. T. & Thai, J. S., 2015. Evidence for causal interpretations of the Easterlin paradox: Propensity score matching and dose-response functions on Philippine data from the 2004 and 2014 World Values Surveys. De La Salle University, Manila, Philippines, DLSU Research Congress.
Lattemann, C. et al., 2009. CSR Communication Intensity in Chinese and Indian Multinational Companies. Corporate Governance: An International Review, 17(4), 426-442.
Layman, G. C., 1997. Religion and political behavior in the United States: The impact of beliefs, affiliations, and commitment from 1980 to 1994. Public Opinion Quarterly, Volume 61, 288-316. Leftwich, R., Watts, R. & Zimmerman, J., 1981. Voluntary Corporate Disclosure: The case of interim reporting. Journal of Accounting Research, Volume 18, 50-77.
Lester, R. H., Hillman, A., Zardkoohi, A. & Cannella, A. A., 2008. Former government officials as outside directors: The role of human and social capital. Academy of Management Journal, Volume 51, 999-1013.
Leuz, C. & Oberholzer-Gee, F., 2006. Political relationships, global financing, and corporate transparency: Evidence from Indonesia. Journal of Financial Economics, 81(2), 411-439.
Liao, L., Luo, L. & Tang, Q., 2015. Gender diversity, board independence, environmental committee and greenhouse gas disclosure. The British Accounting Review, 47(4), 409-424.
Lin, H., 2019. Government–Business Partnerships for Radical Eco-Innovation. Business & Society, 58(3), 533-573.
Lin, J. & Vos, E., 2006. Closely-Held Corporation Participate Less In Corporate Social Reporting. Corporate Ownership & Control, 4(1), 160-176.
Li, S. et al., 2010. Corporate social responsibility in emerging markets. The importance of the governance environment. Management International Review, 50(5), 635-654.
Lock, I. & Seele, P., 2015. Analyzing Sector‐Specific CSR Reporting: Social and Environmental Disclosure to Investors in the Chemicals and Banking and Insurance Industry. Corporate Social Responsibility and Environment Management, 22(2), 113-128.
Mahon, J. & Murray, E., 1981. Strategic planning for regulated companies. Strategic Management Journal, Volume 2, 251-262.
Marquis, C. & Qian, C., 2014. Corporate Social Responsibility Reporting in China: Symbol or Substance?. Organization Science, 25(1), 127-148.
