The Relationship between Corporate Governance and Voluntary Disclosure: The Role of Boards of Directors and Audit Committees

The objective of the present research is to examine the relationship between corporate governance and voluntary disclosure, and to determine how certain factors enhance governance practices and consequently increase voluntary disclosure. The study considers the content analysis of 22 Saudi listed companies from 2015 to 2019. A comprehensive index is developed, with a check-list covering 30 items to extract and measure corporate governance practices and levels of voluntary disclosure. The researchers use ordinary least squares (OLS) regression to examine whether corporate governance-specific mechanisms can explain any differences in voluntary disclosure levels among the listed companies. The results indicate a statistically significant relationship between the number of non-executive directors and board size and the level of voluntary disclosure. This study concluded that non-executive directors and board size are ranked the highest in terms of their positive effects on voluntary disclosure. The relationship between the independent directors and audit committees and voluntary disclosure is insignificant. The results suggest that the high number of non-executive directors and the increase in the number of directors on the boards lead to greater voluntary disclosure of information. This study helps regulators of corporate governance and company directors understand the factors affecting voluntary disclosure. Corporate governance regulators should require an increase in the minimum number of boards and non-executive directors for listed companies in order to gain the desired levels of voluntary disclosure and transparency. Saudi listed companies are advised to willingly increase their board members to the maximum number specified by regulation. This study has some limitations as participants represented a small sample; hence, the results cannot be generalised. Furthermore, the voluntary disclosure data were collected only from annual reports; sources such as websites, public announcements and press releases, were not taken into account, but would have provided many relevant details.


Introduction
Financial reporting and disclosure are important tools that management may use to communicate the performance and governance of a firm to outside investors. The quality of voluntary disclosure (VD) included in companies' annual reports is at the heart of modern financial reporting. Corporate VD is often a major source of information for capital markets (Frenkel et al., 2020), and it is important for their effective functioning because it increases potential shareholder and investor confidence by clearly communicating corporate governance and performance (Saha & Kabra, 2020). Thus, more disclosure transparency increases stakeholder's awareness and confidence (Matoussi & Chakroun, 2008). Investors and other key stakeholders have been dissatisfied by some practices of financial reporting, leading to demands that companies provide more comprehensive voluntary information about their strategies and long-term performance (Kumar, 2007). Additional VD helps investors reduce the likelihood of making uninformed investment decisions by reducing information asymmetries between shareholders and management (Schuster & O'Connell, 2006).
Recent financial scandals have prompted examinations of the potential relationship between a company's governance and its disclosure practices. Cormier et al. (2010) asserted that it would be very difficult to arrive at a conclusion about the expected failures of companies relying only on mandatory disclosure. Although financial scandals have been reported about well-known companies with many attributes of good governance (including boards of directors and audit committees), information emerging in the wake of their failures reveals that the actual governance processes of these companies were ineffectively dominated by top management, and that their disclosure practices were unsatisfactory. Even though all listed companies are required to meet minimum disclosure requirements, these companies differ in the amount of additional information they chose to disclose to capital markets. One way for a company to improve its credibility is through large-scale VD that enables investors to understand its business strategy and management (Gunawan & Lina, 2015).
The purpose of the present study is to examine the relationships between corporate governance (CG) and VD, and to determine how certain CG factors and practices lead to increasing such disclosure in nonfinancial listed companies in Saudi Arabia. The researchers developed a number of hypotheses concerning the relationship between CG and VD, focusing on board size, non-executive directors, board independence and audit committee as independent variables, and VD as a dependent variable.
Prior research on VD has analyzed the relationship between the level of VD and different variables, including CG mechanisms, ownership structure and firm-specific characteristics. However, most of these studies were conducted in developed countries in the global east and west, and only a few studies have centered on developing countries, including in Saudi Arabia. Research in Saudi Arabia has not considered the relationship between disclosure and some of the CG mechanisms. The present study fills this gap in the literature by examining a comprehensive set of variables across CG in Saudi listed companies, and by covering a period comparatively longer than those used in previous Saudi studies (Al-Janadi et al., 2013; Al-Moataz & Hussainey, 2013). A majority of the listed companies have board concentration; hence, it is common for boards to possess considerable control in the management (OECD, 2004). Al-Al-Ghamdi and Rhodes (2015) asserted that such concentration leads to conflicts of interest between the stakeholders and controlling board members. Therefore, it is very important to understand the CG and the effect of a board's role on disclosure in a given company (Gad, 2020;Hashed & Almaqtari, 2021).
This study makes several contributions to the general CG literature, as well as to Saudi Arabia's own body of research. Its findings contribute significantly to understanding the current state and issues of CG practices in Saudi Arabia, a vested interest for stakeholders such as regulators, policymakers, academics and the entire community. This type of research helps in identifying some of the determinants of VD in Saudi Arabia as well as in providing information about VD and its relationship to CG.

Corporate Governance (CG)
CG has become a constant element in academic and policy debates, and the primary business-related preoccupation of our time. Nowadays, CG is an element that shapes as business. Its importance to corporate success as well as to social well-being cannot be overemphasised. Recent publicity around corporate collapses that resulted from weak CG systems have highlighted an international need to improve and reform CG. Researchers have recognised that CG definitions vary widely, ranging from a narrow perspective focusing on companies and their shareholders (i.e. performance, efficiency, growth, financial structure) to broader definitions concerned with the accountability of companies to stakeholders (i.e. rules under which companies operate, derived from sources such as the legal and judicial systems and financial and labour markets (Onakoya et al., 2012). Solomon and Solomon (2004) defined CG as 'the system of checks and balances, both internal and external to companies, which ensures that companies discharge their accountability to all their stakeholders and act in a socially responsible way in all areas of their business activity' (p. 14). Accordingly, CG is the mechanism that protects the interests of shareholders from the inappropriate behaviour of managers in a company (Maskati & Hamdan, 2017 The results indicate that independence of the audit committee from the board showed a significant positive effect on the quality of financial reporting. Agency theory has been the primary basis for the development of CG standards, principles, and codes and has been used extensively by academics and practitioners and still dominates CG literature to date (Christopher, 2010).

Agency Theory
Agency theory states that because of the separation between ownership and control, problems may arise in the relationship between a principal (shareholder) and an agent (corporate manager). Zahra and Pearce (1989) argued that the premise of the agency model assumes or emphasizes a conflict between agent and principal, which becomes important for controlling and monitoring a board's function. Agency theory then attempts to reduce information asymmetries by incorporating monitors (Shapiro, 2005) and establishing mechanisms that can defend shareholders from conflicts of interest on the part of management (Fama & Jensen, 1983).
According to Mallin (2007), agency theory provides the best explanation for CG roles with respect to ownership and control through the use of internal CG mechanisms. From an agency perspective, boards of directors and audit committees are used as effective monitoring device for protecting shareholder interests (Bonazzi & Islam, 2007;Eisenhardt, 1989;Shapiro, 2005). The board of directors and the audit committee make up parts of the internal governance mechanisms, because their use depends on internal decision-makers and incorporates four internal variables: board size, structure, composition, and audit committee independence (Khanchel El Mehdi, 2007). In an effort to deal with agency problems, CG literature emphasises the significant role that the board of directors plays in monitoring top management, so as to ensure that managers act in the interests of shareholders. It also emphasises the role of audit committees in monitoring the decision control system (Fama & Jensen, 1983;Jensen & Meckling, 1976;Peasnell et al., 2005). Given assumptions of agency theory, enhancing CG mechanisms should result in the reduction of information asymmetry. It is expected that companies with stronger CG, reflected in larger board sizes and independence, are able to resolve the principal-agent conflicts and thus improve disclosure practices, with the ultimate objective of increasing shareholder value.

Corporate Disclosure (CD)
Corporate disclosure continues to be of great interest to analysts, investors and shareholders. It is an integral part of financial reporting and, technically, is the last step in the accountability process in terms of presenting information in the form of financial statements. Al-Razeen and Karbhari (2004) defined disclosure as any financial information, quantitative or qualitative, which is intentionally released by the company through official or unofficial channels. Disclosure of information in the annual report may be mandatory or voluntary. Mandatory disclosure is subject to the rules and principles laid down by the regulators; by contrast, VD is subject to the discretion of the company's management (Tsang et al., 2019). Mandatory disclosure refers to the information that a company is required to disclose according to the securities law, accounting principles, and regulatory agencies (Tian & Chen, 2009).
Stock market regulations worldwide require all listed companies to disclose financial information on an annual basis. Companies must meet minimum disclosure requirements, which vary according to different regulators (Alturki, 2014;Gunawan & Lina, 2015). Even with governmental regulations, full disclosure is not guaranteed, since corporate reporting regulation aims to provide outside investors with minimal information (Al-Razeen & Karbhari, 2004). A major issue in the literature, concerning disclosure, is how much information is enough and what the nature of this information is (Gunawan & Lina, 2015). A general consensus, however, is that disclosure helps investors better understand a company's affairs, thus reducing the knowledge gap between management and investors (Akhtaruddin et al., 2009). Therefore, there are calls and demands for more VD to provide investors with the information they need to satisfy their requirements (Alsaeed, 2006).

Voluntary Disclosure (VD)
There have been many attempts to define corporate VD. Although there is, to date, no one definition on which all researchers agree, most agree on the broader concept, which considers any non-mandatory disclosure as VD. Hence, VD discloses additional important information beyond the minimum requirements of applicable capital market regulations required by law and regulatory authorities  Maskati and Hamdan (2017) defined VD as the action that makes information related to a current situation, action, or decision available, accessible, and understood. The concept of corporate VD has developed over time to include all types of financial and nonfinancial information that are not mandatory and can be provided by a company's management in various forms and through different channels (Bhattacharyya, 2012).
By mastering the regulatory rules of the information disclosure market, VD aims to introduce and explain companies' capabilities to investors; drive liquidity in capital markets; ensure more effective capital allocation; reduce capital costs; and achieve more positive communication with investors (Tian & Chen, 2009). The Financial Accounting Standards Board (FASB) categorises corporate VD into six groups: company background; business data; business data analysis; forward-looking information; management and shareholders information; and unrecognised, intangible assets information. It is important to note that because the disclosure here is voluntary, companies may disclose information in different ways and through different channels, including annual reports, interim reports, analyst presentations, investor relations, press releases, conferences and websites. It is also noteworthy that structured annual reports remain the primary source for independently verifying comprehensive information about managers' performance at regular intervals (Saha & Kabra, 2020).
Studies have called for more expansive VD to enhance a company's performance (Wadesango et al., 2020), promote confidence in the financial reports (Owais, 2021), and eventually improve shareholders' value. Owais (2021) created a questionnaire to collect data from 374 external and internal Jordanian auditors to determine the role of VD in building confidence in financial reporting. The study results showed that VD promotes confidence in the financial reports of services companies listed in the Jordanian securities commission. Other studies highlight the link between the ownership structures and VD. Desta et al. (2019) investigate the determinants of VD in the annual reports of 107 Tanzanian manufacturing firms for the period 2011-2015. Their results show that concentration of ownership is negatively associated with VD in the sample companies and that foreign ownership is positively associated with VD.

Corporate Governance and Voluntary Disclosure
Research has shown that CG and VD are interrelated. Previous studies have found that CG mechanisms affect VD (e.g. Maskati  . These studies measured CD by developing an index or score to measure VD in annual reports. It was found that CG characteristics are related to VD levels. Gad (2020) examine selected CG mechanisms in order to determine their impact on the disclosure related to the control system concerned with financial reporting in Germany and Poland. The results indicate that selected CG mechanisms, including the number of board committees and the number of board meetings, have a significant positive influence on the level of disclosure. El-Deeb et al. (2021) examine the extent to which CG affects disclosure tone and their empirical result shows a significant positive impact of CG on disclosure. In Cheng and Courtenay (2006), it was found that companies with boards which held a majority of independent directors had significantly higher levels of VD. Ghazali and Weetman (2006) determined that director ownership was significantly correlated with the extent of VD. Furthermore, the largest shareholders have a powerful interest in firm performance and the capability to increase VD (Li & Qi, 2008). Sallehuddin (2016) asserted that companies where executive directors held higher proportions of shares contained more VD in their annual reports. Khanifah et al. (2020) examine the effect of CG disclosure on bank performance by building and using an index for 10 Islamic banks operating in Iran, Saudi Arabia and Malaysia. The most frequently reported and disclosed elements are board structure and audit committee. The results provide evidence that Islamic banks with a higher level of CG disclosure reported high operating performance.
Previous studies have adopted many theories to examine the relationship between CG and VD, including agency theory, stakeholder theory and signalling theory (see, e.g. Akhtaruddin  Agency theory posits that firms may voluntarily disclose more information to mitigate the information asymmetry that arises from conflicts between shareholders and managers (Lambert, 2001;Zayoud et al., 2011). Accounting researchers have also examined the possible determinants of disclosure using agency theory to explain the existing variation in VD practices (Elfeky & Nasiri, 2017).

CG and VD in Saudi Arabia
In 1965, the Ministry of Commerce issued the Companies Law, which stipulated CG provisions in order to protect shareholders' interests and regulated the CG practices of listed companies in Saudi Arabia (Haniffa & Hudaib, 2007). The law emphasises shareholders' rights by setting conditions for several aspects of business, such 682 The Relationship between Corporate Governance and Voluntary Disclosure: The Role of Boards of Directors and Audit Committees as legal frameworks through which companies can be established, operated and monitored. Likewise, it provides general guidelines concerning registration requirements; minimum capital to be maintained; number of partners and directors and ways of nominating them; monitoring of company accounts; annual audit of the accounts and so on. It consists of 12 parts and regulates all companies in Saudi Arabia. Part five relates to joint-stock companies and includes 98 articles that mostly cover the composition of boards of directors and their levels of independence. During the mid-1960s, the requirements for disclosure in Saudi Arabia were few and the Companies Law was the major source of CG.
In 1985, the Ministry of Commerce aggressively pursued CG reforms to significantly improve the framework and mechanisms for Saudi companies' CG practices by issuing the disclosure and transparency standard. Later, in 2003, the Capital Market Law was released, and the Capital Market Authority (CMA), i.e. a government agency empowered with complete legal, financial and administrative independence, was also instituted. The Capital Market Law brought the entire Saudi capital market under the umbrella of the CMA, a single regulator. The institution of the CMA marked a big step forward for Saudi Arabia, as more significant externally-focused corporate reform initiatives in the country's history were witnessed from that point onwards (Hussainey & Al-Nodel, 2008). One of the country's most recognised CG initiatives was the development and promulgation of the Corporate Governance Law (Al-Janadi et al., 2013). In 2006, the CMA released the first set of Corporate Governance Regulations (CGR), which mostly originated in the Companies Law of 1965. Since that time, there have been many updates to the CGR, to keep abreast of reforms and changes in the broader economy.
In 2007, the formation of the Saudi Stock Exchange (Tadawul) was approved by the Council of Ministers as a joint stock company and the official source for all market information. To date, Tadawul is the only entity in Saudi Arabia licensed to act as the securities exchange regulator (the Exchange) for listing and trading securities and depositing, transferring, clearing, settling and registering ownership of securities. The responsibilities of the stock exchange and the depository center are clearly defined in the Capital Market Law.
In 2017, the CMA approved the amended CG regulation in light of the amendments made to the Companies Law. The main objectives of Saudi's CGR were to provide a universal guide to the regulations, rules, and practices for listed firms and investors (Buallay et al., 2017). The CGR provide the standards and detailed rules that regulate the management of listed firms, in order to ensure compliance with best practices of CG and thereby protect the rights and interests of stakeholders and shareholders. The CGR are composed of eight main parts: rights of shareholders, board of directors, company committees, internal control, external auditors, shareholders, professional and ethical standards, and disclosure and transparency (CMA, 2017).
CGR recommend an ideal board size of 3-11 persons, and a non-executive majority among board members. The minimum number of non-executive members should be at least two, or one-third of the board, whichever is greater. Additionally, the role of the board chair must not be held by an individual who is also the CEO or who occupies another executive position. Directors may sit on the boards of multiple companies, up to a maximum of five at a time (CMA, 2017). All board members are expected to have adequate access to information about the firm in order to enable the board to effectively discharge its duties (Robertson et al., 2013). Furthermore, the CMA recommends that audit committees be composed of three to five members; they can be shareholders or others. At least one member should be an independent director, and no executive director is to be on the committee. Finally, without specifying any minimum number, the CGR encourage board chairs to hold regular meetings. As is the case in many other countries, current regulations on financial reporting in Saudi Arabia oblige companies to adhere to a minimum disclosure requirement, and companies can choose to disclose any additional information beyond this requirement. Various forces influence companies in carrying out CD, the main one being the effort to reduce information asymmetry among market participants and conflicts of interest in listed companies (

Hypotheses Development
VD in annual reports is currently a rapidly-growing area of research in the field of accounting (Alturki, 2014). Several studies examine how variables such as CG mechanisms, ownership structure and firm-specific characteristics affect VD (Ullah et  To conduct the study, the author constructed a disclosure index consisting of 60 items that represent the basic items of VD. The results indicated a wide variation in the nature and extent of VD among the companies studied. The study also found that companies tended to voluntarily disclose more than the required information. The following subsections further shed light on these studies.

Independent Directors and VD
A larger ratio of independent directors to the board's total size helps to assure board's independence, meaning that monitoring practices will have greater effectiveness,

Audit Committee Size and VD
Empirical evidence from previous studies has shown mixed results concerning the effect of audit committees on VD. In a study of listed companies in Kuwait in 2007, Al-Shammari and Al-Sultan (2010) indicated that a voluntary audit committee was likely to demand more VD. They showed that even though the Kuwaiti company law during that time did not oblige companies to establish audit committees, a number of listed companies created and formed them to encourage VD. Further, some empirical studies found that having a greater number of members on an audit committee had a significant positive influence on VD, i.e. the greater the number of members on the audit committee, the higher the level of disclosure (2021) examine the impact of CG mechanisms on the disclosure of 314 non-listed companies across the Gulf region, concluding that the audit committee makeup has no effect. Al-Nimer (2019) has gone further and reported a negative association between the size of the audit committee and level of VD among listed Jordanian firms. Following the majority of empirical findings, which indicate a positive association between audit committee size and VD, this study hypothesises the following: H4: There is a positive relationship between the size of the audit committee and the level of voluntary disclosure in the annual reports of Saudi companies.
In summary, previous research has examined the relationships between CG mechanisms and VD which may enhance transparency in annual reports. Some of these studies have analyzed the relationships among the level of VD and other variables, including CG mechanisms, ownership structure and firm-specific characteristics. However, most of these studies were conducted in developed countries in the global west and east, and only a few studies attended to developing countries, including in Saudi Arabia. Further, previous research concerning Saudi Arabia neglects some CG mechanisms, such as board size and number of independent, non-executive board members. The present study narrows this gap by examining the relationship between CG mechanisms and VD. It focuses on board size, non-executive directors, board independence and audit committees. Furthermore, the present study covers a longer period than those used in previous Saudi studies

Methodology
This study employs a correlational research design that attempts to explore the relationship between VD and CG in the annual reports of Saudi listed companies. Table 1 summarises the description and measurement of the dependent variable, independent variables and other control variables.

Dependent Variable
The dependent variable for this study is defined as voluntary disclosure. The unweighted approach is used to create a VD index scoring the VD of companies. This approach has been used by several studies (see, e.g. Akhtaruddin Patelli & Prencipe, 2007). This study developed an index after checking its items against the mandatory disclosure requirements of the Saudi regulations, in order to account for the levels of VD by Saudi firms in their annual reports. The index used in this study is composed of 30 items. Each company was investigated, and each index item received a score of 1 if the company met the disclosure requirements and a score of 0 if it did not. The VD index includes the level of strategic, nonfinancial and financial information that is voluntarily disclosed.

Independent Variables
CG was chosen as the main independent variable-in particular, board size, board independence, audit committee and non-executive directors. These independent variables were chosen following previous studies (Ke et Ghazali & Weetman, 2006). The measurement of these variables is presented in Table 1.

Control Variables
Previous studies have found that other firm characteristics may affect VD. Therefore, certain control variables were used in this study to control the relationship between the dependent and independent variables, including firm size To test the present study's hypotheses, the following model was used: Model:

Data Collection
Secondary data for CG was mainly collected manually, directly from annual reports. Reports were accessed through the Tadawul website or through the companies' websites. VD data were collected via a content analysis of annual reports, as in previous studies (Al-Shammari & Al-Sultan, 2010; Ghazali & Weetman, 2006;Habbash et al., 2016). The data collected from the companies' annual reports and financial statements were classified and arranged in an Excel file. The VD data were manually extracted from the contents of each annual report and compared to the 30-item VD checklist. Then, a score of 1 was attributed if the information was disclosed and score of 0 if not.

Sample
The sample for this study covers the materials sector from nonfinancial listed companies as classified by Tadawul. This sector consists of an industry group including chemicals, building materials, metals and mining, containers and packing, and paper and forest production, which represent 26.92% of the nonfinancial segment of the market, with 42 listed companies. Twenty companies were excluded due to the lack of annual reports for 2015 and 2016. After the exclusion, the sample consisted of 22 companies with 110 observations (n = 110). The study covered the 2015-2019 period. Table 2 presents the sample composition.

Statistical Tools for Data Analysis
Data were analysed using the Stata software for statistical analysis. The analysis of the results is based on the outcome of the statistical tool.

Descriptive Analysis
This study examines the relationship between VD and CG mechanisms. Table 3 shows the descriptive statistics for all the main variables. The VD ranges between 13% and 50%, and the mean is 29% lower than average and consistent with Alsaeed (2006). This result shows that, on average, the sample discloses only 29% of the checklist items. This average rate is considered very low compared to the maximum rate of VD by the firms, which is 50%.
The BS shows that the maximum number is equal to 11 and the minimum is equal to five. This result meets the CG regulation postulating that the board size should be more than three and less than 11 members (CMA, 2017). It also indicates that some firms have five members on their boards, while others have 11. Further, the results show that the average percentage of independent directors (IND) out of the total board size is 90%. This means that about 90% of the sample firms' directors are independent. Therefore, the sample indicates that Saudi listed companies tend mostly to appoint outside directors as their board members. This percentage is consistent with the Saudi CGR requirement stating that the board of directors must have at least two independent members or a number representing one-third of its total members (whichever is greater). The average percentage of non-executive directors (NEX) out of the total number of directors on the board is 38.6%. Such results agree with the CMA (2017) in that when forming the board, non-executive directors should be the majority of the board. This means that 38.6% of the directors for the sample firms are non-executive directors.
The results show that the audit committee size (AC) is between three and seven members, which meets the CGR requirement that the number of audit committee members shall not be less than three or greater than five. The maximum number did not meet the criteria for a year in one company due to the replacement of members during that year.

Pearson's Correlation Coefficients
Before discussing the regression analysis results, it is important first to review the results in terms of possible multicollinearity. The Pearson's correlation coefficient is viewed as an effective and powerful analytic tool for detecting any multicollinearity among the variables. According to Bryman and Cramer (2002) 'the Pearson's between each pair of independent variables should not exceed 0.80; otherwise the independent variables that show a relationship at or in excess of 0.80 may be suspected of exhibiting multicollinearity' (p. 244). This can affect the results of multiple regressions. In this study, no high correlation among the independent variables was found, as shown in Table 4. Table 4 presents the Pearson's correlation coefficient between VD and independent variables. It reveals that some variables have significant correlations with the disclosure index. The analysis shows that there is a significantly positive correlation relationship between VD, BS (0.365), IND (0.192), AC (0.207), and NEX (0.479).
Additionally, the results show that control variables have a positive correlation with the dependent variable. However, their correlations are not significant.
Notably, there is a significant positive correlation between the BS and IND. It is noteworthy that there is a significant negative correlation between FE (0.377) and LV, and a relatively less negative correlation between AS (0.039) and LV. By contrast, other independent variables show some positive correlations: AC and IND (0.301), and NEX and IND (0.330).

Regression Analysis and Hypothesis Testing
Multiple regression analysis was used to examine the relationship between VD and CG in nonfinancial Saudi firms' annual reports between 2015 and 2019. Table 5 shows the model's regression parameters. The F-value of Saudi companies between 2015 and 2019 is 7.08 (significant at 0.1), which means that the model as a whole is significant and all the independent variables significantly affect the dependent variable. The probability is equal to 0.00, which is less than 0.05; this means that the independent variables reliably predict the dependent variable. This lower value indicates that at least some of the regression coefficients are nonzero and that the regression equation has some validity in the fit of the data. The model's R^2 is (0.389), which denotes that it can explain approximately 38.9% of the variability in the data. This means that other variables that were not included in the model account for approximately 61.   Table 5 shows a positive relationship between NEX and the level of VD (coef. = 0.192), with (0.00) p-value; there is a statistically significant relationship between the NEX and the level of VD. The coefficient of NEX means that a one percent change in NEX brings a 19.2% change in the VD average. The result suggests that a greater number of non-executive directors on the board leads to more VD. If listed companies desire to increase VD and transparency, they must increase the number of NEX members of their boards. CG regulators striving to increase transparency and VD must design their regulations to oblige companies to increase the number of NEX members of their boards.  Table 5 shows a positive relationship between IND and the level of VD (coef. = 0.049), with (0.581) p-value. Statistically, the relationship between the IND and the VD is insignificant. This result is consistent with (Sallehuddin, 2016) and with many results in the literature, which demonstrated that when independent directors are the majority on the board of directors, the company will tend to release more information (e.g. Akhtaruddin Table 5

Robustness Test:
To test the homoscedasticity of the linear regression model, Breusch-Pagan test was used and the result is shown in Table 6. Constant variance is rejected because the Prob chi2 is greater than 0.05, which indicate that the data are free from homoscedasticity.

Discussion
From the analysis, it can be concluded that if listed companies wish to increase VD, the number of non-executive members on the board must be increased. CG regulators should oblige listed companies to increase the number of non-executives in order to increase transparency and VD. The result supports the CMA's composition of board number, whereby the majority of boards should be non-executive (CMA, 2017). Further, CG regulators should increase the minimum number of non-executives on the boards to gain the desired increase in the level of VD. Also, Saudi listed companies are advised to increase their numbers of board members to the maximum number specified by the CMA, i.e. 11 members.
Further, the findings show that the CG mechanisms used in this study (size of the board of directors, number of non-executive directors, independent directors, and audit committees) influence VD. This may be due to the fact that more members means more opinions and points of view. The increase in non-executive members may be a reason for increased VD, in terms of the companies' desire to form positive relationships with stakeholders by sharing and disclosing more information to them. However, the findings show no statistically significant association between IND, AC and the level of VD in Saudi listed companies. AC is insignificantly related to VD due to the focus on internal processes rather than external stakeholders. IND is also insignificant in relation to VD due to the lack of qualifications and expertise of these members resulting from weak selection mechanisms.
Agency theory is key to interpreting these results. The findings support the theory's presumption that external and internal processes of control and monitoring (characteristics of the boards) complement the mitigation of agency conflicts and information asymmetry. Both the boards and the audit committees are positively associated with VD. For the reduction of information unevenness, it can be observed that disclosure is a useful monitoring tools for aligning the different interests of managers and shareholders (Craswell & Taylor, 1992). However, for disclosure to achieve its objectives, the supervisory role of the board of directors is vital to ensure that disclosed information provides a sufficiently comprehensive view of the company (Donnelly & Mulcahy, 2008). In order to be effective as monitors, the board must have certain characteristics in terms of boards' size and composition. The results show that larger, more active boards with a majority of non-executive directors tend to favour greater VD. The results also show that the size of the board can contribute to the successful monitoring of a company, favouring VD, and as a result reducing potential agency costs. Larger boards also have a greater ease in performing their monitoring functions, as these involve the presence of many members and a diversity of skills and experiences. Thus, larger boards may better mitigate agency conflicts among shareholders (Di Pietra et al., 2008;Larmou & Vafeas, 2010).
Further, the results highlight how board independence and non-executive directors may positively influence VD. The presence of non-executive and independent members is an effective tool for supervising information collection, processes of representation, and the actions of management (Rosenstein & Wyatt, 1990). The increase of independent board members leads to voluntarily providing more information, which in turn reduces opportunistic behaviours. This suggests that higher numbers of independent members, and subsequent increases in monitoring may prompt managers to increase VD. Further, a greater percentage of independent members may help to ensure the effectiveness of the audit committee, again producing more VD. In fact, effective supervision by the audit committee can improve both the veracity of the disclosure and the information asymmetry between shareholders and managers. Overall, these results support arguments in favour of a complementary relationship between governance and disclosure.

Conclusions
This study examined the relationships between CG and VD. It determined how board size, independent directors, audit committees and non-executive directors enhance governance practices, thereby increasing VD in Saudi Arabia. This study investigated VD in Saudi Arabia from 2015 to 2019 by analysing the panel data of 110 nonfinancial firm-year observations, using both content analysis and OLS regression analysis. The study used a VD index comprised of 30 items. The results provide evidence of a significant positive relationship between NEX, BS and the level of VD, as well as a positive relationship between IND, AC and the level of VD. This indicates that these variables in CG are the main mechanisms driving VD in Saudi Arabia. The study shows that the independence of the board directors positively and significantly affects the extent of VD, which means that Saudi firms with a high number of independent non-executive directors are expected to voluntarily disclose more information. Furthermore, increasing board size will convince shareholders that the company has better governance.
The findings reported in this study have notable implications for regulators, policy-makers, listed companies and researchers who want to elevate corporate VD by improving CG mechanisms. This study helps regulators of CG and company directors understand the factors that have an effect on VD. The findings in this study could support calls to extend disclosure requirements in agency settings with a large presence of controlling shareholders. By extending VD, potential opportunistic behaviours by controlling managers and shareholders (which attempt to hide detailed information useful for all stakeholders) could be limited. CG regulators should oblige listed companies to increase the minimum number of board members and the number of non-executive members to gain the desired increase in VD and transparency and to protect shareholders. Saudi listed companies are advised to increase their board members to the maximum number specified by CMA to strengthen transparency and disclosure practices, and thereby attract foreign investment in the stock market and contribute to achieving the Saudi Vision objectives. It is expected that listed companies with stronger CG, reflected in a larger board with a majority of non-executive director members, are able to resolve crisis problems and thus improve VD.
The study has a number of limitations. First, the sample 690 The Relationship between Corporate Governance and Voluntary Disclosure: The Role of Boards of Directors and Audit Committees is too small for any generalisations to be applied to all listed companies. Second, the VD data were collected from annual reports only, and other sources like companies' websites and press releases were not taken into account. This represents an opportunity for future studies. Finally, this study covers only certain listed nonfinancial companies in the materials sector; other nonfinancial materials and other listed companies were not accounted for in the sample.
Other avenues for future studies can investigate the effect of the same independent variables on mandatory disclosure in Saudi Arabia, or investigate the effects of the other independent variables covered in this study on voluntary disclosure. Further, investigations of other nonfinancial and financial sectors can generate interesting results. Therefore, it is recommended that the sample be increased in a future study to include all nonfinancial sector companies. Also, when gathering information related to voluntary disclosure, it is recommended to consult companies' websites and all relevant press releases.