The Impact of Corporate Governance on Corporate Social Responsibility Disclosures: Evidence from Botswana Stock Exchange

The nexus between corporate governance (CG) and corporate social responsibility (CSR) is well researched and established in literature. However, the debate regarding the influence corporate governance exerts on corporate social responsibility engagements and reporting is still inconclusive. This study investigates the impact of CG mechanisms on CSR disclosure among the listed companies in Botswana. The study employs content analysis to extract relevant data from annual reports for period 2014 to 2017. Comprehensiveness of descriptions of entity engagements in community involvement, human resources, environment and product and consumers is used as dependent variable while board size, board meeting frequency, ratio of independent directors to board size, the size of audit committee and its meeting frequency are used as independent variables. Company size and Return on assets are controlling variables. Least square regression method is used to analyse the effect of CG attributes on CSR disclosures. Comprehensiveness of disclosure is measured by sentences of reporting, availability of comparative figures, presence of photos and quantified CSR descriptions. The results suggest that the board size, board independence and audit committee meetings frequency have a statistically significant positive influence on corporate social responsibility reports. Company size also reflected a significant positive association with CSR disclosure. These results have implications on composition and structure of the board of directors which should be accorded due attention as the firms work to improve responsiveness to their stakeholders’ demands through CSR investments and disclosures.

. In additions, comprehensive CSR disclosure is credited with lessening information asymmetry between managers and stakeholders particularly equity holders and thereby assisting in the supervision and control of managers. Woodward, Edwards and Birkin, (1996) point out that CSR disclosure can enhance corporate reputation through gaining trust and support by various stakeholders. Given the array of benefits accruing to companies due to CSR disclosure Jamali, Safieddine and Rabbath (2008) contend that the board of directors are expected to promote CSR disclosure.
The concept of CSR has expanded the basis of corporate performance whereby social, environmental and ethical aspects have been popularised and being evaluated along with financial results. Jamali and Rabbath (2007) argue that as the multiple bottom-line performance attributes are gaining popularity companies are pressured to reconsider their strategies and objectives and to integrate new societal, environmental and ethical concerns in their operations. The role of big corporations is no longer limited to contributing to global economy but rather to reconcile and skilfully balance bottom line and manage the interests of multiple stakeholders (Lantos, 2001). Jamali and Rabbath (2007) add that this is where the concepts of corporate governance (CG) and CSR converge. Cheffins (2015) attributes the rise of the corporate governance concept to growing division between share ownership and managerial control, inadequacies in managerial accountability and shareholders indifferences. Thus corporate governance emerged to protect and advance shareholders' interests. The concept emerged in 1970s in the US and became popular in public companies in the early 1990s with the publication of 1992 Cadbury Report in the UK and the 1994 King I Report in South Africa. Both reports provided recommendations for improving company board structures in order to alleviate corporate governance risks and failures. Despite its longevity the corporate governance concept has not received a commonly acceptable definition (Claessens, 2003). Cadbury defines corporate governance as the systems by which companies are directed and controlled (Cadbury, 2000, p.8). Wright (2014) as cited by Cheffins (2015) views corporate governance as the process that encompasses the checks and balances affecting those who run companies. From the perspective of shareholders supremacy, Page (2005) views corporate governance through the lenses of power and responsibility of decision making in the organisation and broadly describes it as consisting of the legal, contractual, and implicit frameworks that define the exercise of power within a company, that influence decision making, that allow the stakeholders to assume their responsibilities, and that ensure that their rights and privileges are respected (Page, 2005, p.2). The World Bank (2005) stresses that whether a broad or narrow perspective is taken in defining corporate governance it is important that the fundamental values of transparency, accountability, fairness, and responsibility be respected in order for firms to build and sustain the confidence of investors, stakeholders, and society as a whole.

Corporate Governance as a concept
Corporate governance is viewed as a critical condition for equitable and effective allocation of resources in both private and public sectors of the economy. Cadbury in Claessens (2003) opines that it is only when principles of fairness, transparency, accountability, and responsibility, which are key pillars of corporate governance, have been widely applied equitable and efficient allocation of resources can be achieved. Earlier on Cadbury in the foreword in Iskander and Chamlou (2000) had argued that adherence to corporate governance standards benefits both private and public sector entities. According to Iskander and Chamlou (2000) compliance with standards helps the private sector entities to achieve their corporate aims and attract investments; and assists the states to strengthen the economy and discourage fraud and mismanagement. The authors further note that good corporate governance is a source of competitive advantage and essential for economic and social progress. It attracts both foreign and local investors as decisions to invest are not only based on company's outlook but also on its reputation and its governance. Oman and Blume (2005) assert that good governance is now accepted as vital to achieving the United Nations Millennium Development Goals and as a pre-requisite for sustainable economic growth.

Relationship between Corporate Governance and Corporate Social Responsibility
The relationship between CG and CSR has always been reflected in various definitions of corporate governance as they underscore the need for the boards of directors to strive to balance the interests of shareholders and stakeholders in their decision making. Claessens (2003) points out that corporate governance should extend to the issue of corporate social responsibility, including such aspects as the dealings of the firm with respect to culture and the environment. Cadbury in the foreword of the work of Iskander and Chamlou (2000) notes that corporate governance is about holding the balance between economic and social goals and between individual and communal goals. Gill (2008) opine that in modern businesses corporate governance is being associated more and more with business practices that are stakeholder-friendly.
Millstein as cited by World Bank (2005, p.5) defines corporate governance as "that blend of law, regulation and appropriate voluntary private sector practices which enables the corporation to attract financial and human capital, perform efficiently and thereby perpetuate itself by generating long term economic value for its shareholders, while respecting the interests of stakeholders and society as a whole." Iskander and Chamlou (2000) emphasize that CG is about maximising value subject to meeting the corporation's financial and other legal and Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697(Paper) ISSN 2222-2847(Online) Vol.10, No.20, 2019 52 contractual obligations. This accentuated the need for board of directors to balance the interests of shareholders with those of other stakeholders such as employees, customers, suppliers, investors, communities in order to achieve long-term sustained value for the corporations. King II Report (2009) when underlining the importance of sustainability principle of corporate governance stated that nature, society, and business are interconnected in complex ways that should be understood by decision-makers. Jamali, Safieddine and Rabbath (2008) highlight that both CG and CSR require companies to fulfil their fiduciary and moral responsibilities towards their stakeholders. And are both perceived to command important long-lasting benefits and to assure the sustainability of the business. Jamali, Safieddine and Rabbath (2008) perceive the relationship between CG and CSR from three dimensions. The authors recognise that CG is a pillar for CSR, CSR is an attribute of CG and CG and CSR are co-existing components of the same continuum.

Corporate Governance and Corporate Social Responsibility in Botswana.
In an effort to enhance corporate governance practices in the country, Botswana introduced Botswana Corporate Governance Code (BCGC) in 2013 (Josiah, Themba & Matenge, 2016;Magang & Kube, 2018). The code was preceded by Botswana Stock Exchange Code of Best Practice whose main principles were retained in the new code. In pursuance of its mandate to promote high standards of corporate governance and compliance with professional ethics, Botswana Accountancy Oversight Authority (BAOA), in 2016 directed public companies in Botswana to adopt South African King Report as the best corporate governance model. BAOA asserts that after conducting a comprehensive research it found that King Reports which are adopted on the "apply or explain" basis are more suitable to Botswana since they are more flexible and all-encompassing and are akin with Botswana Code of CG. Strong economic ties existing between Botswana and South Africa whereby 70% of the former's imports come from the latter and a large number of South African companies participating in various sectors of Botswana's economy (Seitshiro, 2016;Magang and Kube, 2018) could have been a catalyst for BAOA directive. Consequently a good number of public companies in Botswana make reference of their compliance with King III Report in their annual reports. Also at the time of conducting this study a mention of King IV Report adoption was starting to emerge in Botswana public companies' corporate governance reports. However, BAOA cautions that Botswana public entities are only expected to adopt corporate governance principles that meet the minimum requirements of King Reports without necessarily adopting the Codes themselves (Botswana Accountancy Oversight Authority, 2016) Firms from both private and public sectors in Botswana seem to have a strong belief in the importance of giving back to the community through CSR initiatives especially those focusing on community involvement. And prior studies in the country have provided evidence of the apparent existence of CSR activities and their disclosures among the Botswana business entities. See for example Mbekomize and Dima (2013) about CSR and CSRD in State owned entities and listed companies and Botshabelo, Mbekomize and Phatshwane (2018) on CSR and CSRD in Commercial banks. Despite the recognisable spread of CSR initiatives and reporting about the same in Botswana the practice is still mainly voluntary since currently there is no specific legislation compelling firms to engage in and report about CSR activities. However, there are sentiments among the public that the unregulated undertaking of the philanthropic actions may be abused. Consequently, there have been calls for CSR policy in Botswana aimed at standardising the CSR initiatives and curbing the tendencies of providing the same with intention of scoring personal gains.

5.
To determine the effect of frequency of audit committee meetings on CSRD

Theoretical framework
From the corporate governance perspective this study is informed by agency theory; and from corporate social responsibility view point the study is hinged on legitimacy theory, stakeholders' theory and institutional theory. While agency theory focuses on explaining the relationship between the owners and managers of firms the other three concentrate on explaining the firm's motive for CSR disclosure practices (see for example, Kiyanga, Wingard and Cronje, 2016;Ruangviset, Jiraporn and Kim, 2014;Khan, Muttakin and Siddiqui, 2013;Ali, Sandhu, Iqbal and Tufail, 2016). 4.1.1 Agency theory Agency theory has been a dominant theory of explaining organisational relationships in corporations characterised by separation of ownership and control (Roberts, 2004;Elhabib, Rasid, Basiruddin, 2014). Agency theory suggests that the agent should work to serve the interests of the principal. In contemporary large corporations the shareholders (owners) are the principals and the managers serve as agents who are expected to work for the interests of principals while disregarding their self-interests. However, this has not been always the case as agents tend to heighten their interests at the expense of the principals'. According to Berle and Means (1932) separation of ownership and control generates goals incongruences between the interests of managers and those of owners creating agency problem. Roberts (2004) adds that the combination of managers' assumed autonomy and selfinterested motivation creates the problem within agency relationships. Therefore, practitioners as well as academicians have been preoccupied with developing different governance mechanisms which may ensure effective corporate control to cause managers to always act in the best interests of shareholders and other stakeholders (Bonazzi and Islam, 2007). The authors assert that among the many mechanisms that have been advocated to direct corporate managers at enhancing shareholders value, designing an effective board of directors has remained vital and feasible option. This view is supported by Elhabib, Rasid and Basiruddin, (2015) who highlight that various definitions of corporate governance are centred on managing and controlling the relationship between corporate managers and equity owners in order to mitigate the negative impact of the agency conflict. Roberts (2004) has a view that if board of directors have to succeed to align managers' desires with those of shareholders the boards must be dominated by independent non-executive directors who have clear specific roles; the role of the non-executive chairman must be separated from the role of chief executive and board subcommittees should have members who come from independent non-executive cohort. In addition to the aforementioned internal controls, external controls in the form of financial performance and social and environmental disclosures should be enhanced in order to improve monitoring of executives' performance. This study investigates the impact of board of directors' attributes on corporate social responsibility disclosure which is one of the strategies corporations use to enhance stakeholders value (Akinleye & Adedayo, 2018). 4.1.2 Legitimacy theory For the most part of the period of its existence since 1970s corporate social and environmental reporting has been voluntary (Jupe, 2005;Tilt, 2009, van der Laan, 2009) although some efforts to make the disclosure standardised and mandatory have been made (Yongvanich & Guthrie, 2006;KPMG, 2005). The voluntary nature of CSR initiatives and disclosure spurred researchers to develop keen interests in investigating the reasons behind. Legitimacy theory, stakeholders' theory and institutional theory are among the frequently cited theories to explain voluntary disclosure of non-financial performance (Hahn & Kühnen, 2013).
Legitimacy theory holds that no organisation has an inherent right to exist but that any business operation owes its existence to a greater acceptance approved by society (Hahn & Kühnen, 2013). Tilling (2004) points out that according to legitimacy theory the society has the power to grant or deny the business entity the right to operate. Therefore, the legitimacy of the business entity is threatened whenever the society does not perceive the business operations as being congruent to society expectations. Deegan (2002) adds that a business entity needs to have legitimacy as a social licence to function. According to Suchman (1995, p574) "Legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions". The main purposes of pursuing legitimacy are to facilitate the attraction of economic resources and to gain social and political support (Suchman, 1995;Ogden & Clarke, 2005). Tilling (2004) argues that legitimacy is a resource that a business entity requires in order to operate successfully. It may be increased or decreased by certain actions. Therefore through some actions an organisation can go through stages of establishing, maintaining, extending or defending legitimacy. In the broader context of economic theory whereby the business entity takes certain actions to avoid some costs, legitimacy theory argues that business entities disclosure practices are a means of establishing or defending legitimacy to influence public opinion or policy (Tilt and Symes, 2000). The business need to appear acting within the bounds of what society deems to be socially acceptable behaviour since being legitimate attracts resources essential for survival (Jupe, 2005, Tilt, 2009).
Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.10, No.20, 2019 Voluntary CSRD has been frequently used as a legitimation strategy. Patten (1992) indicate that extent of corporate responsibility disclosure increases when the firm faces environmental or social difficulties such as environmental pollution or human rights violation. The company will use different media outlets and increase amount of information in order to defend its legitimacy. O'Donovan (2002) as cited by Siddique (2009) examined the environmental disclosures in the annual reports of environmental sensitive Australian Companies with a view of establishing the relationship between a potentially legitimacy threatening issue with managers' choice of legitimation tactics and thus determine the predictive power of legitimacy theory. O'Donovan concluded that environmental disclosure decisions were made on the basis of presenting the corporation positively to the society. Lupu and Sandu (2010) analysed legitimacy strategies in a recently privatised oil company in Romania and found out that as the social conflict about privatisation intensified there was a corresponding increase in moral legitimacy strategies especially around community involvement activities. Jupe (2005) analysed corporate environmental reports of 200 UK companies and found that legitimacy theory was an explanatory factor for the environmental reports. 4.1.3 Stakeholder theory Stakeholder theory suggests that organisations should be managed in the interests of all their constituents, not only in the interests of shareholders (Laplume, Sonpar, & Litz, 2008). Stakeholder theory is an extension of legitimacy theory in the sense that it proposes the focus on stakeholders with interest in the business instead of the society at large. It calls for integrating stakeholders concerns into business process. Stakeholder theory is a branch of social and political theories which suggests that business disclosures are a response to competing pressure from different stakeholders such as governments, employees, environmental groups, customers, creditors, suppliers and the public (Adler and Milne 1997). The stakeholder theory recognises the existence of various groups of stakeholders with different power to affect the organisation, different legitimacy relationship with the organisation and different urgency of claims from the organisation. The organisation will respond to the influence, legitimacy or claim of each cohort of stakeholders depending on how management assesses the importance of each attribute (Mitchell, Agle & Wood, 1997). Due to the importance of stakeholders in the survival of the company CSRD has been one of the strategies of maintaining legitimacy in the eyes of stakeholders. Nikolova and Arsić, (2017) assert that increased CSR makes business establishments more attractive to stakeholders, and as such enterprises should engage in CSR ventures. 4.1.4. Institutional theory Institutional theory recognizes that the institutional environment can strongly influence the development of formal structures in an organization. It underscores the concept that the forms, outcomes and dynamics of firms are influenced by other social institutions (Brammer, Jackson & Matten (2012). According to institutional theorists organisations tend to follow the formal structures and procedures that are valued in their social and cultural environment in order to achieve legitimacy and to secure the resources that are essential for their survival (DiMaggio & Powell, 1983;Katsikas, 2017).
From CSRD view point Dagiliene (2014) identifies four factors affecting the firm's practices namely mandatory factors, normative factors, economic factors and copying factors. Mandatory factors refer to formal and informal impact of public authorities and other social forces. Normative factors refer to the effect of professional organisations that set standards and principles of CSR reporting that firms have to comply with. Economic factors refer to forces emanating from the idea of economic changes which spark companies to disclose non-financial performances to existing and potential investors. Copying factors means tendency of companies to behave in similar manner as their flourishing counterparts (motivated to engage in CSR disclosure because your successful counterpart is doing it).
Dagiliene (2014) opine that mandatory and normative institutional factors mostly affect the development of CSRD in business companies due to consistent improvement of sustainability methodologies from organisations such as Global Reporting Initiative (GRI), OECD and UNHC. Brammer, Jackson & Matten (2012) add that corporate agency is largely shaped by the dominant institutions of the business systems or organisational field within which the firm operates. They further expound that in the context of Anglo-Saxon context, the influence of organisational field may result in mostly voluntary policies and programmes of CSR while in other countries like Japan and Germany CSR may be shaped by legal, customary or religious institutions. Drawing from extant literature on CSR, Brammer, Jackson & Matten (2012) argue that given the wide variations in CSR practices and dynamics in the concept and its applications, motives for CSR are broader than managers' voluntary responses to society or stakeholders expectations. Khan, Muttakin and Siddiqui (2013) analysed the relationship between corporate governance and corporate social responsibility in Bangladeshi companies. They analysed the contents of annual reports using legitimacy theory lens. The independent variables used included managerial ownership, public ownership, foreign ownership, board Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.10, No.20, 2019 independence, CEO duality, and existence of audit committee. The dependent variables were represented by the corporate governance characteristics. The control variables used were percentage of share owned by the directors, percentage of shares owned by the public and percentage of shares owned by foreign investors. Others included firm's size measured as natural logarithm of total assets, the number of years the company has been in existence, ratio of total debt to total assets and profitability measured as return on total assets. The dependent variable (CSRD) was measured using the CSR disclosure index. Khan, Muttakin and Siddiqui (2013) results indicate a significant positive association between CSRD with managerial ownership especially among export-oriented entities. The results also reveal that public ownership, foreign ownership, board independence and existence of audit committee have a positive significant effect on disclosure of corporate social responsibility. The authors concluded that corporate governance characteristics play an important role in ensuring organisational legitimacy through CSRD.

Empirical CG and CSR
Jizi, Salama, Dixon and Stratling (2014) examined the impact of corporate governance on corporate social responsibility disclosure in US banking sector. With respect to corporate governance they used board independence, board size, CEO duality as independent variables. They measured the quality and comprehensiveness of the corporate social responsibility disclosures using four themes of community involvement, environment, employees and social products and services quality. In addition, corporate governance variables that are expected to influence CSRD were brought in as control variables and they included firm size, financial expertise and meetings frequency of audit committee. Others included frequency of meetings of board of directors, company profitability measured as return on assets, company leverage, company's systematic risk and company size. Jizi, Salama, Dixon and Stratling (2014) found that board independence and board size were positively and significantly associated with CSR disclosure and CEO duality had a positive and significant effect on CSR disclosure.
Ali, Sandhu, Iqbal and Tufail (2016) investigated the effect of corporate governance characteristics on CSRD in Pakistan companies and found that only the multiple directorships of a chairperson had a significant positive relationship with CSRD. Existence of non-executive directors on the board and presence of non-executive directors in the audit committee did not reflect a significant relationship with CSR. The authors therefore concluded that in Pakistan companies corporate governance mechanisms carry no weight in driving the CSR reporting agenda.
Employing document analysis on the sample from Lebanon companies, Jamali, Safieddine and Rabbath (2008) investigated the nature of relationship between CG and CSR and found that in most cases CG was a critical component for sustainable CSR.
Adnan, Van Staden and Hay (2009) examined the influence of culture and corporate governance structure on quality of CSRD in companies operating in China, India, Malaysia and United Kingdom. The quality of CSR disclosure as a dependent variable was measured using the Global Reporting Initiatives (GRI) Index. The independent variables were corporate governance and culture. The former was measured using board composition and existence of CSR committee on the board while the later was represented by individualism, power distance, uncertainty avoidance and masculinity index. Control variables included CSR assurance statement, being audited by big-four auditors, ISO1400 certification and countries' environmental performance. The results of Adnan, Van Staden and Hay (2009) work suggested that the quality of information in CSR standalone reports and website was better than that disclosed in the annual reports. The CSRD quality was different across countries. The study also submit that social and environmental information was less likely to be disseminated through annual reports since they focus on the shareholders who are normally interested in economic performance rather than social and environmental performance. Ruangviset, Jiraporn and Kim ( 2014) used the USA data for the period 2001 to 2004 to examine the relationship between CG and CSRD and found a negative relationship between them.

5.Methodology
The target population for this study was 23 listed domestic companies in the Botswana Stock Exchange. However, due to various reasons such as delisting and new listings within the period of 4 years under review and lack of involvement in CSR and/or non-disclosure by certain firms, particularly those from property sector, we ended up analysing 11 firms. Sources of information for analysis were annual reports for the period 2014 to 2017. Although CSR engagements are nowadays reported using different media such as standalone statements, company websites and advertisements, annual reports have been a dominant channel due to their advantages over the other outlets. With annual reports corporate disclosures are easily comparable from year to year and the reporting company has more editorial control (Lavers, 1993). The annual reports are considered to be the main document used by various stakeholders to make a variety of decisions (Deegan & Rankin, 1997;Jizi, Salama, Dixon & Stratling, 2014). We however note that the work by Adnan, Van Staden and Hay (2009) indicated that CSR standalone and websites reports provide higher quality information than annual reports.
The wide usability and easy accessibility of annual reports of public companies in Botswana attracted us to confine ourselves to this medium. In line with prior studies (see for example Jizi, Salama, Dixon & Stratling, 2014;Adnan, Van Staden and Hay, 2009) content analysis was employed to extract the information regarding corporate social responsibility engagement and reporting which was divided in five categories namely community Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.10, No.20, 2019 involvement, environment, human resources, product and customers and general information for the easy identification of data of interest in the annual reports. The unit analysis for CSRD was the number of sentences reflecting activities in the above mentioned five themes. The information about corporate governance attributes and control variables (board size, board meeting frequency, board independence, audit committee size and its meeting frequency) were also obtained from annual reports.

Control Variables
Company size COZ Natural logarithm of total asset value Return on Assets ROA Ratio of profits before tax over total assets

Dependent Variable
The study aimed at determining the effect of the corporate governance mechanisms on comprehensiveness of CSR disclosure. Therefore the comprehensiveness of CSR disclosure was the dependent variable for this study. Prior studies interested in examining what influences corporate disclosures have used different methods including measuring the extent of reporting using words or sentences describing the activities undertaken by companies (See Abu-Baker, 2000, Thompson & Zakaria, 2004, Rahman & Muttakin, 2005, Haniffa and Coke, 2005 or determining the quality and comprehensiveness of disclosures (See for example Adnan, Van Staden and Hay, 2009 ;Jizi, Salama, Dixon & Stratling, 2014). Of the two methods, quality and comprehensive of disclosures approach is considered to be more rigorous and therefore superior to using words, sentences or paragraphs. However, the quality of disclosures approach would generally suffer from being arbitrary than the extent approach as measures of quality may be difficult to determine adequately. In prior studies the tool used to measure quality of information has been the assigning of scores on information disclosed. For example Adnan, Staden and Hay (2009) used scores from 0 to 5 to measure the quality of information disclosed based on 51 items according to GRI Index while Jizi, Salama, Dixon and Stratling (2014) used scores from 1 to 5 depending on existence and comprehensiveness of information reported including existence of comparative and quantified information. In this study we decided to use the number of sentences describing CSR engagements to measure comprehensiveness of disclosures. We are of the view that the more the company talks about a CSR theme the more the importance it attaches to that theme. The disclosure of photos, comparable figures and quantification of activities including the amount spent were considered to enhance the comprehensiveness of disclosure.
Therefore after capturing company's number of sentences per CSR category per year we totalled all the sentences from each category to make the sum of sentences for each year. We then sorted the totals of the 44 observations from 11 firms which had a minimum of 21 and maximum of 815 sentences in the ascending order (from smallest to largest). After that we grouped the 44 observations in five categories of nine observations in each group except the upper group which had eight observations. Then the five groups were assigned score as follows: Groups of number of sentences Scores 21 to 50 1 51 to 91 2 92 to 144 3 145 to 173 4 174 to 815 5 A comprehensive disclosure of photos, comparable figures and quantification of activities were each awarded one point while the absence of each was awarded a zero. What deserves one point or a zero was agreed upon by two principal investigators which were involved in extracting the information from annual reports. Therefore the Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697(Paper) ISSN 2222-2847(Online) Vol.10, No.20, 2019 57 maximum scores for each company per year were 8 points.

Independent variables
Five Corporate governance attributes formed the independent variables to be regressed against CSRD comprehensiveness. They included board size (BOZ), board meeting frequency (BMF), board independence (BOI), audit committee size (ACZ) and frequency of meetings of audit committee (ACF). In line with prior studies we included controls variables to reflect additional variables that can increase the likelihood of firms to engage in CSR activities (Ruangviset, Jiraporn, Kim, 2014). Control variables were company size (COZ) measured as total assets and return on total assets (ROA) representing performance of the company. Company size has been said to influence the extent of CSR activities and disclosure (Damak-Ayadi, 2010). Large companies have greater impact and society and environment and are therefore subject to public attention and scrutiny (Brammer and Pavelin, 2008). Although the association between CSR and corporate financial performance is still debatable some studies have concluded a positive and significant correlation between the two (Uadiale and Fagbemi, 2012;Adeneye and Ahmed, 2015). We formulated the linear regression model to explain the association between CSRD and CG attributes as follows: CSRD = α +β1 BOZ+ β2 BMF+ β3 BOI+ β4 ACZ+ β5 ACF + β6 COZ + β7 ROA+ ε Where CSRD = scores of comprehensives of corporate social responsibility disclosure. α represents the value that CSRD will have when all independent variables are equal to zero. β1…. β7 = regression coefficients to be estimated representing the quantified effect that each corporate governance variable has on the corporate social responsibility disclosure ε is an error term Table 2 shows that the mean scores for comprehensiveness of CSR disclosure among the sampled companies was around 4.5 points out of the maximum 8 points signifying that the quality of CSRD was around 56%. .18509118 The number of board members in the sampled companies ranged between 6 and 14 with an average of around 9 members. The board of directors in sampled companies had an average of about 4 meetings in year. The average ratio of independent members in the sampled boards of directors was around 56% signifying that the companies are properly managed. King III Principle 2.18 recommends that the majority of non-executive board members should be independent in order to achieve balance of power in the board (King Committee on Corporate Governance, & Institute of Directors (South Africa) (2009) The number of audit committee members ranged between 2 and 7 with an average of about 4 members and were meeting almost quarterly on average. Table 2 also reflects that the company size measured as natural logarithm of total assets value was between 11.558826 and 16.977238 with a mean of 14.47777614. For the sampled companies, the mean value of return on assets was 0.1504; the minimum was negative 0.01547 while the maximum was 1.14815. Table 3 reflects the correlation among the study variables. Among the five corporate governance mechanisms only size of board of directors has a significant and positive association with CSRD (r = 0.484; p = 0.001). Table  3 also shows that among the control variables only company size has a significant and positive association with CSRD (r = 0.444; p = 0.003).

Analysis and Discussion
Among the independent variables, board meeting frequency has a significant positive and strong relationship with audit committee meeting frequency (r = 0.691; p = 0.001). On average board of directors of companies in the sample have almost the same number of meetings in year like the audit committees reflecting the importance the firms attach to the audit committees.  .000 **. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Independence of the board of directors has a significant positive but weak correlation with audit committee size (r = 0.391; p = 0.009). Company size has a significant positive but moderate association with audit committee meeting frequency (r = 0.418, P =0.005); a significant positive but weak association with board meetings frequency (r = 0.364, p = 0.015) and audit committee size (r = 0.321, p = 0.034). Company size reflected a significant negative moderate association with return on assets (r = -0.530; p = 0.000). Table 4 reflects the results of multivariate regression analysis of comprehensiveness of CSRD. The model is good predictor of the comprehensiveness of CSRD with F = 7.947; p = 0.000 and adjusted R 2 of 53.% implying that 53% fluctuations in the comprehensiveness of CSRD can be attributed to the combination of CG characteristics imbedded in the model. Consistent with prior studies, for example, Jizi, Salama, Dixon and Stratling (2014) and Barakat, Pérez and Ariza (2015), the results suggest that CSR disclosure is mainly driven by board size with standardised beta of 60.7% and p-value of 0.000 Among the five corporate governance mechanisms tested only board size had a significant positive association with CSRD at 1% level of significance. Audit committee meetings frequency and board independence had a significant positive but weak relationship with comprehensiveness of CSR disclosure at 0.05 level of significance, β = 0.342, p = 0.038 and β = 0.267, p = 0.038 respectively. Board meetings frequency and audit committee size reflected non-significant relationship with CSR disclosure. The results of the current study also suggest that company size has a significant positive impact on CSRD (β = 0.451, p = 0.003). These results are also supporting the numerous prior studies which suggest that company size has a significant influence of CSRD (see for example Brammer and Pavelin, 2008;Sadou, Alom, and Laluddin, 2017;Ali, Sandhu, Iqbal and Tufail, 2016). Another control variable, return on assets, did not appear to have significant impact on CSRD.

Conclusion
This study used the data obtained from annual reports of companies listed on Botswana stock exchange for the period 2014 to 2017 to examine the influence of CG characteristics on CSR disclosure. Overall, the results of this research suggest a moderate association between CG and CSR disclosure. Consistent with prior studies the results of this study provide evidence of the three CG characteristics namely, board size, board independence and audit committee meeting frequency, having a significant positive impact on the CSRD. Of the three, the size of the board Research Journal of Finance and Accounting www.iiste.org ISSN 2222-1697 (Paper) ISSN 2222-2847 (Online) Vol.10, No.20, 2019 59 of directors has the most significant influence on CSRD. This implies that large boards which are normally associated with large firms bring in diverse ideas of how to respond to various stakeholders' demands through corporate social responsibility activities. In light with legitimacy theory large boards would pay more attention to stakeholders' claims and will respond to them through extra social and environmental activities and disclosures. The board consisting of majority of independent members is more likely to push for more CSR engagements as members may tend to have a better understanding of external stakeholders' needs since that's where they belong. In light with agency theory the larger the board size the more likely the board will act in congruence with stakeholders' interests. Honesty and transparency of the firm through the corporate disclosures would be enhanced if the board size is properly aligned with the size of the firm.
A worth noting observation has been made by this study in light of institutional theory which attributes the CSRD to practices found in organisational field in which the firm operates. This study noted a persistent lack of CSR investments and their disclosures in the property development and management sector firms. The reasons for this phenomenon are not clear but it may be assumed that firms in the property sector could be considering themselves operating in an industry which is not socially and environmental sensitive and therefore do not see the necessity of building their legitimacy through CSR investments. This could be an area where further empirical studies could be conducted to inquire from property company managers what could be the reasons behind lack of active CSR engagements and disclosure.