Corporate governance and corporate social responsibility: The case of small, medium, and large firms

Gul, Sajid; Muhammad, Faqir; Rashid, Abdul Article Corporate governance and corporate social responsibility: The case of small, medium, and large firms Pakistan Journal of Commerce and Social Sciences (PJCSS) Provided in Cooperation with: Johar Education Society, Pakistan (JESPK) Suggested Citation: Gul, Sajid; Muhammad, Faqir; Rashid, Abdul (2017) : Corporate governance and corporate social responsibility: The case of small, medium, and large firms, Pakistan Journal of Commerce and Social Sciences (PJCSS), ISSN 2309-8619, Johar Education Society, Pakistan (JESPK), Lahore, Vol. 11, Iss. 1, pp. 1-34 This Version is available at: https://hdl.handle.net/10419/188279 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/ Pakistan Journal of Commerce and Social Sciences 2017, Vol. 11 (1), 01-34 Pak J Commer Soc Sci Corporate Governance and Corporate Social Responsibility: The Case of Small, Medium, and Large Firms Sajid Gul (Corresponding author) Air University School of Management, Islamabad, Pakistan Email: [email protected] Faqir Muhammad Air University School of Management, Islamabad, Pakistan Email: [email protected]u.edu.pk Abdul Rashid International Islamic University, Islamabad, Pakistan Email: abdulras[email protected]u.pk Abstract This paper aims to explore whether and how firm-level governance mechanisms affects Corporate Social Responsibility (CSR) using a large sample of firms listed at Pakistan Stock Exchange. Further, the paper investigates the differential effects of corporate governance (CG) on CSR across small, medium, and large firms. The findings strongly support the hypothesis that CG alone is not sufficient to induce firms to provide more CSR information. Rather, we show that both CG and ownership structure matters and have a vital role to play in firms’ choice of CSR engagement. The results reveal that better governed firms have higher CSR disclosure when compared with lower CG firms controlling for the level of insider ownership. Specifically, the results suggest that firms are more likely to be involved in CSR when insiders’ ownership is at medium level (25% to 50%) as compared to low (0 to 25%) or high level (>50%). Nonetheless, the estimates suggest that CSR involvement decreases when the insider ownership goes beyond the 50% level. Finally, the results reveal that there are significant differences in the effects of CG and other underlying empirical determinants of CSR across different sized firms. Keywords: corporate social responsibility; corporate governance; insiders’ ownership; system GMM 1. Introduction The effectiveness of CG and accountability has been the subjects of heated debate after the corporate accounting scandals around the world. The term CG, although used extensively in the literature, generally lacks a well-accepted definition (Ekanayake, 2011). The existing definitions of CG, at best, can be classified into two categories based on accounting and finance related CG literature, narrow and broad, depending on the extent to which CG issues are addressed therein. The narrow perspective is orientated towards only corporate accountability to shareholders. However, CG within the broader Corporate Governance and Corporate Social Responsibility 2 perspective requires corporate accountability to shareholders and other stakeholders (e.g., investors, lenders, employees, customers, suppliers, government auditors). ‘Securities and Exchange Commission of Pakistan’ [SECP] (2005) defined CG as: “The mechanism by which the agency problems of corporation stakeholders, including the shareholders, creditors, management, employees, consumers, and the public at large are framed and sought to be resolved.” Along with the acceleration of CG mechanisms, one of the most important corporate trends of the previous decade is the growth of CSR (Khan et al., 2012). As in Ghazali (2007), in these days, the main objective of the organizations is not only to make maximum profit but it also includes an element of CSR and accountability. This is because the business could not prosper in segregation. Rather, it also significantly depends on the society to operate efficiently. There are several studies that have separately explored CSR and CG in different aspects. However, there is relatively less research on the link between CG and CSR. Ghazali (2007) argues that CSR reporting means information disclosure on community development, environmental reporting, products and services, and human resource aspects. Corporate governance needs to be considered in determining CSR disclosure especially board composition and ownership structure because CSR reporting is affected by the motives, values, and choice of those who are involved in organizations decision making process (Lau et al., 2016). Since, both ownership and board structure significantly differ not only across firms but also across industries and nations, it is predicted that CSR will also vary subsequently. In a firm with a high level of ownership concentration, the incentive should be higher to support the controlling owner entrenchment. Nevertheless, it will be difficult for management to use CSR to mask their opportunistic behaviors, if an effective system exists to monitor their decisions. However, Choi et al. (2013) suggest that in weak governed firms CSR involvement induced by opportunistic managers would be more prominent in order to fulfill personal objectives by overinvesting in CSR. Hence, CSR debate continues to flourish without a clear consensus on its meaning or value. 1.1 Research Problem In principle, CG, ownership structure, CSR may have different strengths of relationship, depending on the market’s level of legal protection and the extent of concentrated ownership. However, empirical evidence on this relationship is very limited. Yet, it is important to know how CG is related to CSR. It is also important to understand how the level of ownership structure and degrees of ownership concentration affect the association between CG and CSR. 1.2 Literature Gaps and Contributions of the Study The role of CG in stakeholders’ value creation has become the subject of great interest recently. Nonetheless, most of the previous studies follow a fragmented approach, investigating the link between isolated aspects of governance and CSR. However, it could be argued that firms choose sets of governance arrangements, which in turn, are analysed as a whole by investors. Furthermore, it is assumed that CG mechanisms are Gul et al. 3 complements and can act as substitutes for each other (Cheung et al., 2011). Therefore, rather focusing on just individual aspects of CG, the current study also employs a composite measure of CG. In addition, the review of the literature on CG has shown that the previous existing studies have included either only relatively large firms in their sample or pooled all listed firms. However, different sized firms have different characteristics including ownership structure and information asymmetries (Ronnie Lo, 2009). Hence, this study contributes to the literature by classifying the sample firms into large, medium, and small firms. By doing this, the study provides a novel contribution to the literature as shown by results that firms of different sizes differ in terms of their CG. Moreover, the literature review also suggests that there is lack of longitudinal studies on CG. Thus, we do not have robust evidence on the role of CG in firms’ CSR activities. Nevertheless, the panel nature of our data enables us to examine the CG-CSR link over a period of twelve years. Definitely, combining cross sectional variations across time in CG will yield more reliable and robust estimates. Furthermore, CG measures used in previous studies are not devised to rank firms on CG quality. However, in this study, firms are ranked into high CG and low CG on the basis of their median score of CG. This study also contributes further evidence on the issue whether CSR is a joint function of the CG and the ownership structure of firms. The previous existing studies on the association between CG and CSR do not emphasize on the issue on these lines. Hence, this study presents a novel contribution to the literature by showing that firm’s social performance varies with the level of its insiders’ ownership, and the pattern differs relying jointly on CG and insider ownership. 1.3 Significance of the Study The findings of this study have important implications. For the managers, the empirical results presented in the paper clarify that the costly exercise of disclosure of CG information can help in enhancing CSR. Similarly, the regulators may take the results into consideration when they are going to determine the appropriate levels of disclosure and compliances of regulations in the future. The results show that the implementation of CG codes that came into effect from 2002 are effective and have enhanced CG level. The CG role of mitigating agency conflicts to maximize wealth of shareholder has evolved to now creating value for not only shareholders but also protecting all other stakeholders’ interests. In Pakistan, the ownership of firms is highly concentrated in families and business is conducted under a week legal investor-protection regime (Javid & Iqbal, 2010). The separation of ownership and control in Pakistani firms is not as clear as in the Western countries. Therefore, the empirical findings on the role of CG in CSR activities based on a sample of Pakistani firms can have significant meanings and implications for those firms operating in markets in which high concentration in family ownership is prevalent. 1.4 Theoretical Orientation For developing and testing the research hypotheses, it is important to establish a theoretical base. The theoretical foundation of this study is premised primarily on agency, Corporate Governance and Corporate Social Responsibility 4 stakeholder, and legitimacy theories. Furthermore, most of the results of this study are supported by these theories. The most prominent is the agency theory as majority of the CG research evolves from this theory. According to Mees (2015), in the CG literature, agency theory is commonly used to describe firm managers’ CG decisions. However, due to the differences in environmental characteristics, the applicability of agency theory in Pakistan may be different than developed countries. Hence, the risk of expropriation by the dominant or controlling shareholder is the main agency problem and not the managershareholder conflict (Javid & Iqbal, 2010). As this study investigates the impact of CG on CSR, CG is the outcome of agency problem (ownership structure). Therefore, the theory of agency is used in this study to understand the managers’ behaviour. The agency theory is pertinent as it explains the conditions under which a company is likely to adopt CG practices. Contrary to the agency theory, the theory of stakeholder is a broader concept that considers the interests of diverse constituents comprising interest groups related to environmental, social, and ethical considerations (Mainardes et al., 2011). Pease and Macmillan (1993) argue that in the modern concept of CG, a set of legislative, regulatory and legal market mechanisms are put in place inviting firms to consider all other interest groups. Under a stakeholder approach, the support of all of their stakeholders would be required by a company in order to be successful and survive in the long run. Hence, social pressure is likely to influence the future course of CG, where firms have to focus their attention on larger stakeholders’ community. The stewardship theory unlike the agency theory is an alternative theory for researchers, which states that the agents are trustworthy and good stewards of firms’ resources. This makes monitoring unnecessary (Hu & Alon, 2014). The legitimacy theory asserts that the survival of companies depends on the extent to which they operate according to the bounds and norms of the society. CG can be related to the legitimacy concept because management has to legitimise its actions in order to be accepted in the society (Frynas & Stephens, 2015). Through CG information, organisations signal to all their stakeholders that they are abiding with the terms of the social contract and hence attain the legitimacy imperative for their continued survival. Furthermore, both legitimacy and stakeholder theories can be regarded as complementing each other. In particular, the stakeholder theory provides the basis for legitimacy theory. Based on the above discussion, we can conclude that there is some degree of consistency, complementarity, and compatibility of these theories with each other. 2. Literature Review Jo and Harjoto (2011) find that after controlling for various firm characteristics several governance mechanisms including board leadership and independence, analyst following, institutional ownership, and antitakeover provisions appears to have a direct positive influence on CSR engagement. However, one should note that Jo and Harjoto use only institutional ownership to study its impact on CSR, whereas Khan et al. (2012) also used foreign and public ownership and percentage of ownership held by management. The results provide evidence that foreign and public ownership, presence of audit committee, Gul et al. 5 and board independence have significant positive effect on CSR disclosures, however negative in the case of management ownership. Nevertheless, they fail to find any significant evidence between CEO duality and CSR disclosure. They also document that the dominance of family ownership on CSR disclosure is alleviated to some extent by sound CG mechanisms. In contrast to Jo and Harjoto (2011) and Khan et al. (2012), Ghazali (2012) finds that boards with independent directors are less involved in CSR activities. The result may imply that for independent directors, CSR engagement is not the primary concern. However, the result was statistically significant at margin (the 10% level). Ghazali (2007) finds evidence that CSR disclosure is significantly higher in firms that are larger in size and have higher government ownership. However, similar to Khan et al. (2012), Ghazali finds that high proportion of management ownership and CSR disclosure is negatively correlated. On the other hand, Said et al. (2009) show that only the presence of audit committee, ownership concentration, and government ownership significantly positively impact the extent of CSR disclosures. Their results contradict the findings of previous studies discussed above that ownership concentration and CSR are negatively correlated. The results presented by Giannarakis (2014) indicate that profitability, the board commitment to CSR, and firm size positively affect the extent of CSR disclosure, whereas financial leverage and CSR disclosure is inversely related. Rees and Rodionova (2015) examine whether CG promotes CSR disclosure through its effect on managers’ intention and whether CG mediates the influence of family ownership on social and environmental improvements. Their results reveal that ESG performance is negatively affected by closely held equity and family ownership. Nevertheless, after controlling for CG, closely held equity and ESG score is no longer associated but family ownership retains a significant inverse relationship. Branco and Rodrigues (2008) find that company size is a crucial determinant for both types of disclosure. The factor of media exposure is an important characteristic only for annual reports while financial leverage has a negative effect on the extent of disclosure based on the web sites. Finally, profitability is positively significant only to products and consumers information of annual reports. Reverte (2009) finds evidence that industry type, firm size, and media exposure are the most crucial determinants of the extent of CSR disclosure. In addition, the results also reveal that firms with higher CSR ratings have a less concentrated ownership. Nevertheless, the variable financial leverage does not seem to explain differences in CSR disclosure, hence contradicting with the findings of Giannarakis (2014). Mallin and Michelon (2011) investigate whether board reputation (board diversity, leadership, composition, competence, and structure) affect social performance of firms. The results suggest that corporate social performance is positively related to the proportions of independent, community influential and female directors. Nevertheless, community influential directors with multiple directorships and CEO duality significantly negatively affect corporate social performance. Corporate Governance and Corporate Social Responsibility 6 Jain and Jamali (2016) critically review the previous literature on the impact of CG mechanisms at the institutional, firm, group, and individual levels on CSR. They argue that theoretically there is a strong case for CG as an antecedent of CSR and promising patterns are beginning to emerge in the literature. Yet, the empirical evidence remains mixed and inconclusive in some areas. They recommend that greater scholarly attention needs to be accorded to disaggregating variables and comprehending how multiple configurations of CG mechanisms interact and combine to impact firms' CSR behavior. Lau et al. (2016) provide evidence that state ownership positively affects CSR performance, whereas concentrated ownership negatively affects CSR. The variable board composition and composition of TMT is not statistically significant. However, board size as well as board members having foreign experiences have significant effects on CSR. Jizi et al. (2014) find that even after controlling for profitability, audit committee characteristics, size and risk, and board meeting frequency, both size and independence of the board positively affect disclosure of CSR. However, in contrast to Khan et al. (2012) who find insignificant link between CEO duality and CSR, Jizi et al. (2014) find a significant positive association between CEO duality and CSR. Liu and Zhang (2017) examine the link between CG, CSR, and enterprise value. The results reveal that supervisory board meetings, state ownership, number of directors, and managerial shareholding positively affect CSR while ownership by largest shareholders negatively affects CSR. Moreover, they find that CSR information is not beneficial for the short-term profit of an enterprise but can increase its long-term value. Manasakis et al. (2014) show that hiring ‘individually’ socially responsible CEO acts as a commitment device for the firm's owners and signal to consumers that the missioned CSR activities will be undertaken. They further show that for consumers as well as firms corporate social responsibility activities are welfare enhancing. To some extent our study is closer in spirit to Ghazali (2007) and Rees & Rodionova (2015). However, they examine the impact of CG on managers’ intention to promote CSR. They have simply studied the impact of management and/or family ownership on CSR. We substantially depart from these papers as we investigate how CSR is related to firms’ state of CG, subject to the various levels of ownership structure and degrees of ownership concentration, controlling for the impact of external CG and other firm characteristics. Theoretical framework constructed based the review of the existing literature is presented in Figure 1. The figure shows that CG index is developed based on board composition, transparency and auditing, and disclosure factors. The figure also shows that the constructed index is then related to SCR activities as per proposed by the agency theory, legitimacy theory, and the stakeholder theory. The summary of the literature review is presented in Table 1. Gul et al. 7 Table 1: Summary of Literature Review on Corporate Governance and Corporate Social Responsibility Author (s) Sample Time Period CSR Measure Model Main Finding Jo and Harjoto (2011) 2,952 U.S. firms 1993– 2004 Kinder, Lydenberg, and Domini’s Index Probit, Tobit, and 2SLS Positive Khan et al. (2012) 116 Dhaka Stock Exchange listed firms 20052009 Selfconstructed index Multiple regression model Management ownership (-) Outside directors (+) Governance committees (+) Ghazali (2012) 27 Bursa Malaysia listed firms 2005 and 2007 Selfconstructed index Paired-sample t-test Positive Ghazali (2007) 87 companies listed on Bursa Malaysia 2001 Selfconstructed index Multiple regressions Firm size (+) Government ownership (+) Management ownership (-) Said et al. (2009) 150 Malaysian public listed companies 2006 Selfconstructed index Hierarchical regression analysis Insignificant Giannar akis (2014) 100 U.S. firms 2011 ESG score Multiple linear regression Insignificant Rees and Rodion ova (2015) 3,893 firms from 46 countries 20022012 ESG score OLS Closely held equity (-) Family ownership (-) Branco and Rodrig ues (2008) 49 companies listed on Portuguese Stock Exchange 2003 Selfconstructed index Multiple linear regression Company size (+) Financial leverage (-) Profitability (+) Reverte (2009) 35 largest Spanish firms 2005 and 2006 The Observatory on CSR Linear regression model Industry type (+) Firm size (+) Media exposure (+) Concentrated ownership (-) Corporate Governance and Corporate Social Responsibility 8 Mallin and Michel on (2011) 100 firms listed in the Business Ethics 100 best corporate citizens 20052007 KLD database OLS Positive Jain and Jamali (2016) N/A 20002015 N/A N/A Theoretically there is a strong case for CG as an antecedent of CSR. The empirical evidence remains mixed and inconclusive in some areas Lau et al. (2016) 471 firms in China 2010 and 2011 KLD and Global Reporting initiative OLS State ownership (+) Concentrated ownership (-) Board size (+) Board with foreign experiences (+) Jizi et al. (2014) 193 US commercial banks 2009– 2011 Selfconstructed index Tobit and Linear Panal regression Board independence (+) Board size (+) Liu and Zhang (2017) 968 firms in China 2008 to 2014 CSMAR measure OLS Positive Manasa kis et al. (2014) Two large publicly traded firms in oligopolistic markets 2011 Selfconstructed index Hackner (2000) and Manasakis et al. (2013) Socially responsible CEO acts as a commitment device for the firm's owners Gul et al. 15 4.3 GMM Estimation Results: Corporate Governance and Corporate Social Responsibility The Hansen J statistic test indicates that the instruments used in the study are valid. In addition, the autocorrelation test of Arellano–Bond presents no evidence of model misspecification. The results document that CG enhancement is linked with a significant increase in CSR in pool, large, medium, and small Cap firms. Specifically, the results suggest that an increase of one unit in CG−score is associated with an increase of 0.10 in CSRS in pooled regression, 0.09 in large sample firms, 0.12 in medium sample firms, and 0.06 in small sample firms. Hence, the empirical evidence suggests that CG is an important determinant of enhancing CSR disclosures in annual reports. Khan et al. (2012) argue that closely held small companies may be less active in investing in social activities because there will be relatively low level of public interest and because the costs of such investment may far outweigh its potential benefits. As previously mentioned, prior literatures have used isolated aspects of CG rather using a composite measure. Hence, this paper provides a novel contribution to the literature about the link between CG and CSR by using a broad measure of 18 CG provisions. Further, the variable board score used in this study is comprised of 10 CG practices. No single study has used such a broad measure they either use board independence or board size (e.g., Jo & Harjoto, 2011; Khan et al., 2012; Ghazali, 2012) as a proxy for board compositions. The results reveal that board of directors sub-score is significantly positively associated with CSR in all samples. From the perspective of agency theory, better governed boards are more interested in the long term sustainability of a company i.e. CSR engagement rather than focusing on just short-term financial performance targets (Jizi et al., 2014). In addition, from the Legitimacy theory perspective, better governed boards with higher independence are helpful in constituting and preserving corporate legitimacy for their credibility and reputation (Liu and Zhang (2017). Nonetheless, for a company in an emerging economy like Pakistan in order to preserve its competitiveness a significant consideration is the financial performance. In order to sustain its financial performance a firm with better governance would be in a better position which will ultimately influence its CSR performance. Hence, we conclude that social performance is the result of a properly designed governance structure. Moreover, audit and disclosure sub-scores are positive and statistically significantly related to CSR. The agency theory suggests that audit effectiveness mitigates asymmetric information between firm insiders and outside investors and stakeholders and in turn enhancing the reliability of corporate reporting (Jizi et al., 2014). The audit committee works in collaboration with the board and is recognized as indispensable to the effectiveness of the board. Nevertheless, it is unlikely that members of the audit committees understanding about CSR activities is associated to their financial expertise, however generally they might have a more positive attitude to disclosure. Said et al. (2009) and Khan et al. (2012) also find similar result however they used a narrow measure. From the coefficients on BRD−score, AUD−score, and DSC−score in the Corporate Governance and Corporate Social Responsibility 16 three samples, it can be seen that audit sub-score plays a major role in influencing CSR as compared to other categories of CG. The results also suggest that there is a significant negative relation between the debt ratio and CSR in all sized firms. Hence, the negative association suggests that creditor stakeholders will apply less pressure to inhibit managers’ prudence over activities of CSR. This finding is consistent with the findings presented in Branco and Rodrigues (2008) and Giannarakis (2014). However, Reverte (2009) and Sariannidis (2014) do not find any statistically significant association. On the other hand, Esa and Ghazali (2012) find that firms with high leverage ratios enclose more information on CSR. However, they use web pages as the source of CSR information rather than annual reports. The results show that firm size (LnAssets) is positive and highly significant in explaining the likelihood of choosing CSR engagement. Stakeholder groups provide more attention to larger firms and therefore to exhibit social responsibility such firms would be under greater public pressure (Ghazali, 2007). The empirical results of prior studies document that size and the extent of social disclosure have positive correlation (e.g., Ghazali, 2007; Said et al., 2009; Giannarakis, 2014; Jizi et al., 2014). On the other hand, Rees and Rodionova (2015) do not find any significant association. The reason may be that they have used a different proxy to measure firm size and a small sample. The literature has provided different evidence on the association of profitability with CSR in the form of a negative, a positive, or an uncertain link. As shown in case of pooled sample and small Cap firms, ROE and CSR have a significant positive association confirming the legitimacy theory hypothesize, suggesting that firms with higher financial performance demonstrate their contribution to society’s well-being and will act in more socially responsible ways than unprofitable companies because in general it is a costly decision. Similar result was found by Said et al. (2009) and Jizi et al. (2014). Nevertheless, the study fails to find any significant evidence between ROE and CSR in large and medium Cap firms. Said et al. (2009) also find no association between profitability and CSR disclosure. The results further document that firms’ involvement in social activities is negatively affected by high ownership concentration measured by insiders’ ownership. At higher level of managerial ownership, public accountability will be less concerned for directors, resulting in low disclosure of CSR (Ghazali, 2007). The result further implies that high share ownership by executive and non-independent director’s results in less CSR disclosure. In contrast, Khan et al. (2012) provide evidence that higher management ownership is associated with increased CSR disclosure in export-oriented industries. The reason behind this result is the fact that management ownership in export-oriented industries is very small hence pressures exerted by powerful stakeholder groups as a principal driver of CSR reporting. Further, the results reported show that family firms as compared to non-family firms avoid engagement in CSR. Rees and Rodionova (2015) argue that families oppose CSR investments as being value destroying. Family firms do not have the reputational pressure for social and environmental responsibility from the beneficiaries due to their large and Gul et al. 17 long term ownership stakes and will therefore oppose to excessive investment in CSR because it may not bring personal benefits. Rees and Rodionova (2015) find that ESG performance is negatively affected by closely held equity and family ownership. As compared to younger firms, there will be more CSR involvement and reputation for older firms due to their deep societal presence and comparatively more legitimacy. Nevertheless, opposed to this expectation, firm age is found to have insignificant association with CSR, suggesting that age of the firm does not influence the decision of CSR disclosure. The reason behind this association is the fact that company age is measured in this study by the actual listing status i.e. the time span between when a firm was listed on PSX and the study period. Nonetheless, there are recently listed firms in Pakistan that were present for a long time even before their listing. Corporate Governance and Corporate Social Responsibility 18 Table 8: Regression Results for Corporate Governance and Corporate Social Responsibility (Pooled Sample) Variables (1) (2) (3) (4) CSRS (-1) 0.9382*** (0.000) 0.9432*** (0.000) 0.9921*** (0.000) 0.9972*** (0.000) CG-Score 0.0993*** (0.000) BRD-Score 0.04915*** (0.002) AUD-Score 0.1179*** (0.000) DSC-Score 0.0264*** (0.000) Debt/Assets -0.0016 (0.314) 0.0001 (0.940) -0.0007 (0.713) -0.0014 (0.256) FRAGE 0.0002 (0.717) 0.0007 (0.175) 0.0004 (0.411) -0.0002 (0.621) GR -0.0044*** (0.000) -0.0049*** (0.000) -0.0048*** (0.000) -0.0021** (0.042) LnAssets 0.0136*** (0.000) 0.0123*** (0.000) 0.0112*** (0.000) 0.0075*** (0.000) INSIDOWN -0.0004*** (0.000) -0.0004*** (0.000) -0.0004*** (0.000) -0.0003*** (0.000) Net Income/ComEq 0.0000* (0.067) 0.0000** (0.025) 0.0001*** (0.007) 0.0000** (0.019) Family -0.0664*** (0.000) -0.0705*** (0.000) -0.0687*** (0.000) -0.0455*** (0.000) Cons -0.1706*** (0.000) -0.1812*** (0.000) -0.2059*** (0.000) -0.0713*** (0.002) Obs 1668 1668 1668 1668 Instruments 107 107 107 107 Groups 200 200 200 200 AR (1) -7.67 -7.69 -7.70 -7.74 [P-Value] (0.000) (0.000) (0.000) (0.000) AR (2) -0.19 -0.01 -0.22 -0.18 [P-Value] (0.852) (0.992) (0.828) (0.859) Hansen test 107.33 107.47 104.88 108.70 [P-Value] (0.222) (0.220) (0.275) (0.196) F-Significance (0.000) (0.000) (0.000) (0.000) Note. *, **, ***= significance level at 0.10, 0.05 and 0.01 percent. Gul et al. 19 Table 9: Regression Results for Corporate Governance and Corporate Social Responsibility (Large Sample) Variables (1) (2) (3) (4) CSRS (-1) 0.08872*** (0.000) 0.9185*** (0.000) 0.8928*** (0.000) 0.8872*** (0.000) CG-Score 0.0923*** (0.001) BRD-Score 0.0449** (0.047) AUD-Score 0.0968*** (0.000) DSC-Score 0.0482*** (0.000) Debt/Assets -0.0196** (0.026) -0.0239*** (0.008) -0.0227*** (0.006) -0.0238*** (0.009) FRAGE 0.0004 (0.588) 0.0003 (0.653) -0.0003 (0.652) -0.0003 (0.647) GR -0.0391*** (0.000) -0.0446*** (0.000) -0.0378*** (0.000) -0.0282*** (0.000) LnAssets 0.0147*** (0.000) 0.01791*** (0.000) 0.0148*** (0.000) 0.0116*** (0.001) INSIDOWN -0.0018*** (0.000) -0.0020*** (0.000) -0.0020*** (0.000) -0.0019*** (0.000) Net Income/Common Equity -0.0000 (0.856) -0.0000 (0.947) -0.0001 (0.748) -0.0001 (0.558) Family -0.0964*** (0.000) -0.1096*** (0.000) -0.0959*** (0.000) -0.01050*** (0.000) Cons -0.1347*** (0.000) -0.1532*** (0.000) -0.1322*** (0.000) -0.0499 (0.101) Obs 434 434 434 434 Instruments 47 47 47 47 Groups 70 70 70 70 AR (1) -4.24 -4.26 -4.24 -4.15 [P-Value] (0.000) (0.000) (0.000) (0.000) AR (2) 0.66 0.95 0.80 0.41 [P-Value] (0.507) (0.340) (0.422) (0.680) Hansen test 40.88 41.10 40.10 40.35 [P-Value] (0.304) (0.296) (0.334) (0.325) F-Significance (0.000) (0.000) (0.000) (0.000) Corporate Governance and Corporate Social Responsibility 20 Table 10: Regression Results for Corporate Governance and Corporate Social Responsibility (Medium Sample) Variables (1) (2) (3) (4) CSRS (-1) 0.8935*** (0.000) 0.9954*** (0.000) 0.9785*** (0.000) 0.9173*** (0.000) CG-Score 0.1198*** (0.000) BRD-Score 0.0849*** (0.000) AUD-Score 0.0881*** (0.000) DSC-Score 0.0525*** (0.000) Debt/Assets -0.0079* (0.055) -0.0107*** (0.004) -0.0106*** (0.001) -0.0105* (0.059) FRAGE 0.0002** (0.018) 0.0005** (0.043) 0.0001 (0.499) 0.0001 (0.374) GR -0.0061* (0.061) -0.0200*** (0.000) -0.0129*** (0.000) -0.0039*** (0.008) LnAssets 0.0159*** (0.000) 0.0135*** (0.000) 0.0107*** (0.000) 0.0062*** (0.000) INSIDOWN -0.0002*** (0.000) -0.0001** (0.031) -0.0001*** (0.004) -0.0001* (0.064) Net Income/Common Equity 0.0005 (0.114) 0.0004 (0.225) 0.0005 (0.269) -0.0000 (0.949) Family -0.0393*** (0.000) -0.0232*** (0.000) -0.0333*** (0.000) -0.0116*** (0.007) Cons -0.1694*** (0.000) -0.1689*** (0.000) -0.1424*** (0.000) -0.0224* (0.052) Obs 835 835 835 835 Instruments 99 99 99 99 Groups 143 143 143 143 AR (1) -5.36 -5.33 -5.32 -5.53 [P-Value] (0.000) (0.000) (0.000) (0.000) AR (2) -0.30 -0.22 -0.15 -0.41 [P-Value] (0.767) (0.827) (0.879) (0.682) Hansen test 87.13 93.30 86.95 86.52 [P-Value] (0.536) (0.357) (0.542) (0.669) F-Significance (0.000) (0.000) (0.000) (0.000) Gul et al. 21 Table 11: Regression Results for Corporate Governance and Corporate Social Responsibility (Small Sample) Variables (1) (2) (3) (4) CSRS (-1) 0.9049*** (0.000) 0.9381*** (0.000) 0.9479*** (0.000) 0.9341*** (0.000) CG-Score 0.0584*** (0.000) BRD-Score 0.0297* (0.061) AUD-Score 0.0271*** (0.004) DSC-Score 0.0268** (0.016) Debt/Assets -0.0087** (0.049) -0.0084* (0.057) -0.0091** (0.036) -0.0085** (0.049) FRAGE 0.0002 (0.631) -0.0001 (0. 851) -0.0003 (0.415) -0.0003 (0.546) GR -0.0069 (0.112) -0.0048 (0.265) -0.0034 (0.434) -0.0018 (0.659) LnAssets 0.0109** (0.013) 0.00770** (0.038) 0.0056* (0.075) 0.0067* (0.069) INSIDOWN -0.0003*** (0.001) -0.0003*** (0.000) -0.0002** (0.031) -0.0003*** (0.001) Net Income/Common Equity 0.0023*** (0.000) 0.0022*** (0.000) 0.0017*** (0.000) 0.0023*** (0.000) Family -0.1131*** (0.000) -0.0892*** (0.000) -0.0588*** (0.001) -0.0666*** (0.000) Cons -0.1638*** (0.004) -0.1090** (0.031) -0.0648 (0.124) -0.0601 (0.176) Obs 398 398 398 398 Instruments 57 57 57 58 Groups 87 87 87 87 AR (1) -3.45 -3.41 -3.42 -3.49 [P-Value] (0.001) (0.001) (0.001) (0.000) AR (2) -0.77 -0.76 -0.75 -0.79 [P-Value] (0.440) (0.447) (0.454) (0.430) Hansen test 46.21 45.25 47.22 51.39 [P-Value] (0.505) (0.545) (0.463) (0.342) F-Significance (0.000) (0.000) (0.000) (0.000) Corporate Governance and Corporate Social Responsibility 22 4.3.1 Testing the Joint Effect of CG and Insider Ownership on CSR This section provides a novel contribution by investigating whether CSR can be jointly determined by CG and insider’s ownership. From the results given in Table 12, it can be seen that the coefficients on interaction terms DH×DP and DL×DP are negative and statistically significant in pooled regression. The result indicates that firms with a predominant shareholding are less involved in CSR engagement as compared to the base category of DH×DL . Nevertheless, the coefficient of high CG group is higher as compared to the low CG category implying that high CG firms are more involved in social disclosure for the same level of predominant ownership. The CSR involvement of DH×DP group is lower by 0.02 from the base category of DH×DL and the DL×DP is lower by about 0.05. Although the difference between the DH×DP and DL×DP is 0.0343, this difference is statistically significant. Figure 2 shows the coefficients of interaction terms on Y-axis and insider ownership levels on X-axis for pooled sample. The blue line is high CG line and the red line is low CG line. The blue line is above the red line implying that high CG firms are more involved in social activities than low CG firms. However, there is no statistical evidence to say that DH×DM , DL×DL and DL× DM are different from the base category of DH×DL . In case of large Cap firms, the results reveal that better governed firms have higher CSR disclosure when compared with lower CG firms controlling for the level of insiders ownership. As can be seen from Figure 3, the blue line (high CG line) is above the red line (low CG line). The coefficient of the interaction variables DH×DM , DH×DP , DL× DM , and DL×DP are negative and statistically significant. However, firms in the high CG-medium ownership category disclose more information on CSR. Thus, the result indicates that CG alone is not sufficient to induce firms to provide more CSR information in annual report, both CG and ownership structure matters in influencing firms’ choice of CSR engagement. Specifically, the result implies that high CG rank firms are more likely to be involved in CSR when insiders’ ownership is in the range of 25% to 50%. Nonetheless, when insider ownership goes beyond 50% then CSR involvement decreases by 0.1316 as compared to the base category. Further, low CG firms have less CSR information in annual reports as compared to high CG firms. However, they still are able to disclose more when lower agency and entrenchment problem exists as in medium ownership category (25%-50%). Further, for medium Cap firms, it can be seen that all low CG categories DL×DL , DL× DM , and DL×DP have lower coefficients as compare to the base category (DH×DL ). The results document that medium Cap low CG rank firms have lower CSR disclosure as compared to the base category. Furthermore, low CG firms with low or predominant ownership have lower CSR disclosure as compared to firms with medium ownership. However, DH×DM and DH×DP groups do not appear to be statistically different from the base category. For small Cap firms, the coefficient of interaction terms DH×DP , DL×DP , and DL×DL are negative and significant. Thus, the estimates suggest that high CG-predominant ownership group and low CG-predominant ownership group disclose Gul et al. 23 less CSR information as compared to the base category of high CG-low ownership category and the difference is 0.044 for DH×DP and 0.06 for DL×DP . The low CG-low ownership category also discloses less CSR information than the base category by about 0.09. Table 12: Regression Results for Joint CG-Insiders’ Ownership and Corporate Social Responsibility Variables Model 2_Pool Model 2_Large Model 2_Medium Model 2_Small CSRS (-1) 0.9722*** (0.000) 0.8921*** (0.000) 0.9435*** (0.000) 0.9921*** (0.000) BRD-Score 0.0567*** (0.000) 0.0309 (0.524) 0.0526*** (0.007) 0.0192 (0.681) AUD-Score 0.0987*** (0.000) 0.0923*** (0.005) -0.0018 (0.906) -0.0027 (0.870) DSC-Score 0.0250*** (0.000) 0.0297* (0.060) 0.0695*** (0.000) 0.0973*** (0.000) Debt/Assets 0.0003 (0.536) -0.0223*** (0.007) -0.0143*** (0.000) -0.0088** (0.017) FRAGE 0.0002 (0.675) -0.0005 (0.321) 0.0004 (0.323) 0.0005 (0.264) GR -0.0033*** (0.000) -0.0315*** (0.001) -0.0198*** (0.000) -0.0019 (0.632) LnAssets 0.0057*** (0.001) 0.0168*** (0.000) 0.0175*** (0.003) 0.0032** (0.034) Net Income/CommEq 0.0002** (0.015) -0.0001 (0.211) 0.0005 (0.108) 0.0023** (0.011) Family -0.0453*** (0.000) -0.0433*** (0.000) -0.0310*** (0.000) -0.0654*** (0.000) 𝐃𝐇×𝐃𝐌 -0.0199 (0.655) -0.0951*** (0.000) -0.0019 (0.323) -0.0345 (0.434) 𝐃𝐇×𝐃𝐏 -0.0154*** (0.000) -0.1316*** (0.000) -0.0134 (0.121) -0.0439** (0.043) 𝐃𝐋×𝐃𝐋 -0.0016 (0.115) -0.0116 (0.546) -0.1113*** (0.000) -0.0899*** (0.000) 𝐃𝐋×𝐃𝐌 -0.0231 (0.332) -0.1642*** (0.000) -0.1034** (0.012) -0.0329 (0.787) 𝐃𝐋×𝐃𝐏 -0.0497** (0.044) -0.2282*** (0.004) -0.1327** (0.032) -0.0619* (0.054) Cons 0.0532 (0.283) 0.1142** (0.013) 0.0976*** (0.000) 0.1213** (0.024) Corporate Governance and Corporate Social Responsibility 24 Obs 1668 398 835 399 Instruments 107 47 97 57 Groups 200 69 143 87 AR (1) -7.62 -4.15 -4.78 -3.28 [P-Value] (0.000) (0.000) (0.000) (0.001) AR (2) -0.25 0.69 -0.22 -0.68 [P-Value] (0. 803) (0.491) (0.645) (0.496) Hansen test 106.27 34.58 79.14 33.00 [P-Value] (0. 131) (0.440) (0.546) (0.838) F-Significance (0.000) (0.000) (0.000) (0.000) Figure 2: Coefficients of Joint CG and INSIDEOWN Interaction Variables on CSR (Pooled Sample) High Low High Med High Pred Low Low Low Med Low Pred -0.06 -0.05 -0.04 -0.03 -0.02 -0.01 0 Low Med Pred Cofficients Insider's Ownership High Low Gul et al. 31 Choi, B. B., Lee, D., & Park, Y. (2013). Corporate social responsibility, corporate governance and earnings quality: Evidence from Korea. Corporate Governance: An International Review, 21(5), 447-467. Ekanayake, A. (2011). The role of accounting in corporate governance of banks in a developing country: evidence from Sri Lanka (Doctoral thesis). Macquarie University, Australia. Esa, E., & Ghazali, M. A. N. (2012). Corporate social responsibility and corporate governance in Malaysian government-linked companies. Corporate Governance, 12(3), 292-305. Frynas, J. G., & Stephens, S. (2015). Political corporate social responsibility: Reviewing theories and setting new agendas. International Journal of Management Reviews, 17(4), 483-509. Ghazali, M. A. N. (2007). Ownership structure and corporate social responsibility disclosure: Some Malaysian evidence. Corporate Governance: The International Journal of Business in Society, 7(3), 251-266. Ghazali, M. A. N. (2012). Corporate social responsibility and corporate governance in Malaysian government-linked companies. Corporate Governance: The International Journal of Business in Society, 12(3), 292-305. Giannarakis, G. (2014). Corporate governance and financial characteristic effects on the extent of corporate social responsibility disclosure. Social Responsibility Journal, 10(4), 569-590. Gujarati, D. N. (2003). Basic econometrics, (4th Eds.). New York: McGraw-Hill. Habbash, M. (2010). The effectiveness of corporate governance and external audit on constraining earnings management practices in the UK (Doctoral thesis). Durham University, UK. Hu, W. H., & Alon, I. (2014). Are Chinese CEOs stewards or agents? Revisiting the agency-stewardship debate, In Chinmay Pattnaik, Vikas Kumar (Ed.), Emerging Market Firms in the Global Economy, International Finance Review (pp. 255-277). Emerald Group Publishing Limited. Jain, T., & Jamali, D. (2016). Looking inside the black box: the effect of corporate governance on corporate social responsibility. Corporate Governance: An International Review, 24(3), 253-273. Javid, A.Y., & Iqbal, R. (2010). Corporate governance in Pakistan: Corporate valuation, ownership and financing. PIDE Working Paper No. 57. [Online] Available: http://www.pide.org.pk/pdf/Working%20Paper/WorkingPaper-57.pdf (November 24th, 2016). Corporate Governance and Corporate Social Responsibility 32 Jizi, M. I., Salama, A., Dixon, R., & Stratling, R. (2014). Corporate governance and corporate social responsibility disclosure: Evidence from the US banking sector. Journal of Business Ethics, 125(4), 601-615. Jo, H., & Harjoto, M. (2011). Corporate governance and firm value: The impact of corporate social responsibility. Journal of Business Ethics, 103(3), 351-383. Khan, A., Muttakin, M. B., & Siddiqui, J. (2012). Corporate governance and corporate social responsibility disclosures: Evidence from an emerging economy. Journal of Business Ethics, 114(2), 207-223. Lau, C., Lu, Y., & Liang, Q. (2016). Corporate social responsibility in China: A corporate governance approach. Journal of Business Ethics, 136(1), 73-87. Liu, X., & Zhang, C. (2017). Corporate governance, social responsibility information disclosure, and enterprise value in China. Journal of Cleaner Production, 142(2), 10751084. Mainardes, E. W., Alves, H., & Raposo, M. (2011). Stakeholder theory: Issues to resolve. Management Decision, 49(2), 226-252. Mallin C. A., & Michelon, G. (2011). Board reputation attributes and corporate social performance: An empirical investigation of the US Best Corporate Citizens. Accounting and Business Research, 41(2), 119-144. Manasakis, C., Mitrokostas, E., & Petrakis, E. (2014). Strategic corporate social responsibility activities and corporate governance in imperfectly competitive markets. Managerial and Decision Economics, 35(7), 460-473. Mees, B. (2015). Corporate governance as a reform movement. Journal of Management History, 21(2), 194-209. Pease, G., & McMillan, K. (1993). The independent non-executive director. Longman Professional, Melbourne. Rees, W., & Rodionova, T. (2015). The influence of family ownership on corporate social responsibility: An international analysis of publicly listed companies. Corporate Governance: An International Review, 23(3), 184-202. Reverte, C. (2009). Determinants of corporate social responsibility disclosure ratings by Spanish listed firms. Journal of Business Ethics, 88(2), 351-366. Ronnie Lo, H. K. (2009). Voluntary corporate governance disclosure, firm valuation and dividend payout: Evidence from Hong Kong listed firms (Doctoral Thesis). Glasgow University, UK. Said, R., Omar, N. & Abdullah, W. N. (2013). Empirical investigations on boards, business characteristics, human capital and environmental reporting. Social Responsibility Journal, 9(4), 534-553. SECP (2005). Manual of corporate governance. [Online] Available: http://www.nzibo.com/IB2/manual-CG.pdf (October 13th, 2016). Gul et al. 33 Suphakasem, J. (2008). An evaluation of corporate governance disclosure: Evidence from Malaysia, Singapore and Thailand (Doctoral thesis). University of Strathclyde, Glasgow. Appendix 1: Corporate Governance Score/Index A. Sub Index-Board Composition Percentage of Independent Non-Executive Directors (INED’s) on board Presence of Independent Chairman Size of board Is the role of Chairman and CEO split? Number of board meetings held during the year Percentage of total director’s attendance at board meetings Percentage of board meetings attended by INED’s Minority shareholders representation on board Gender diversity on board Board and individual directors performance evaluation B. Sub IndexTransparency and Auditing Does the company have an audit committee? What percentage of audit committee constitutes INED’s? Independence of audit committee Chairman Whether a system is in place to protect whistle blowers C. Sub IndexDisclosure Does the company disclose board members biographies? does it list the other boards its directors sit on? Policy for handling conflict of interest Code of ethics for all directors and employees? Attendance record of each director at committee meetings Corporate Governance and Corporate Social Responsibility 34 Appendix 2: Corporate Social Responsibility Index A. Community Involvement Charitable donations Support for housing (infrastructure) Community program B. Environmental Policies related to environment Pollution prevention Recycling C. Employee Information Human resource Relations with employees Employees Welfare Training and development of employees Profit sharing with employees Health and safety of workers Child labour and related actions Strong retirement benefits D. Information Related to Product and Service Types of products disclosed Product research and development Quality and safety of product Customer service and satisfaction Customer Award/Rating Received Statement of value added