Does Corporate Governance Affect Financial Reporting Quality? Evidence from Listed Entities in Sri Lanka

The purpose of this paper is to investigate the relationship between selected CG characteristics and the level of FRQ in Sri Lanka. The study was carried out using secondary data obtained through published annual reports from 209 companies listed in the Colombo Stock Exchange (CSE) from 2017 to 2020 including 836 firm-year observations. Six selected CG characteristics (Board Size, Board Independence, CEO Duality, Audit Committee Size, Audit Committee Independence, and Audit Committee Accounting Expertise) and the level of FRQ has been evaluated by the absolute value of discretionary accruals (ADA) using Panel linear regression analysis. It was found that a significant positive relationship between the audit committee accounting expertise and FRQ, while a significant negative relationship was found between Board Size and FRQ. However, the other remaining CG characteristics did not significantly influence the level of FRQ. Overall, this analysis highlights the importance of having a comparatively smaller board size and composition of the members in the audit committee with financial and accounting backgrounds to enhance FRQ and transparency. The findings of this study expect to have a significant policy implication for policymakers and regulators in terms of formulating strategies and policies on CG best practices in Sri Lanka. Similarly, the entities should promote smaller board size and recruiting, especially the majority of independent non-executive directors with sufficient accounting skills and financial experience to curtail the adverse earnings management practices to improve FRQ. In relation to the variables, firm shows a significant negative correlation with a -0.086 value of coefficient (p=0.0370) represents a significant positive correlation with a 0.095 coefficient (p=0.0000). These suggest that the show a propensity on FRQ. of Silva et al. (2017) and have similar findings on this in the Sri Lankan context.


Introduction
Corporate Governance (CG) plays a vital role, especially in the process of assuring Financial Reporting Quality (FRQ) (Alzeban, 2019;Cohen, Krishnamoorthy & Wright, 2004). The ultimate purpose of preparing financial reports is to improve the quality of decisions made by all the current and potential investors. However, the users can make quality decisions, only with the availability of quality financial information. Thus, the corporations should certify the financial reporting quality in terms of reliability, relevance, timeliness, and biasness.
According to past studies, several factors could impact the level of FRQ. According to Alzeban (2019), an empirical relationship could be found between the audit committee, CEO, the quality of external audit, and FRQ. Further, Uwuigbe, Felix, Uwuigbe, Teddy, and Irene (2018) found that the board size, board independence, and director's tenure and its impact on the quality of financial reporting. However, the literature disclosed that weak CG leads to a financial crisis. Similarly, less attention on CG characteristics may be the reason for the recent corporate collapses that happened over the globe including Sri Lanka [1]. For example, the Freddie Mac scandal in December 2003, the American International Group scandal in 2005 can be seen as global failures of the companies in the recent past which provides evidence for CG weaknesses. Therefore, studying separately on these governance characteristics is more important to meet the quality of financial reports to satisfy the expectations of current and potential investors.
Although considerable empirical work has been published regarding the relationship between CG and FRQ (Alzoubi, 2014;Johl, Johl, Subramaniam & Cooper, 2013;Phuong & Hung, 2020), the scope of the above studies was different from the current study in terms of context, the tested CG characteristics and the tested time. In this paper, the researchers tried to collectively examine the relationship between six different CG characteristics namely, Board Size, Board Independence, CEO Duality, Audit Committee Size, Audit Committee Independence, Audit Committee Accounting Expertise, and the level of FRQ in Sri Lanka as significantly less recent studies were found in connection to the current scope in the Sri Lankan context. Further, it has been given mixed conclusions by previous empirical studies relating to this area where we are trying to examine the relationship in the local context. Hence the present study contributes to the literature by meeting the research objective of investigating the impact of selected CG characteristics on FRQ in Sri Lanka. Additionally, it will enhance the awareness of board governance and AC characteristics in CG being an important factor in the business arena followed by the corporate managers and Board of Directors (BOD), as they are the major party responsible to the owners of the company. In other words, a high FRQ might result only from the presence of an effective BOD and AC, determined by reference to its independence and expertise. Similarly, this will be useful for the regulatory institutions, policymakers, and professional accounting bodies to rethink the future CG provisions.

Literature Review
Understanding the concept of Corporate Governance (CG) There is no single definition of CG that can be applied to all situations and jurisdictions. Nevertheless, there are some commonly accepted definitions. The most widely used is "the system by which companies are directed and controlled" (Cadbury Committee, 1992). . Although corporate governance has been exhaustively defined as a mechanism for controlling, leading, and investigating the activities of the firm by promoting corporate fairness, transparency, and accountability to create shareholders' wealth, for this study it would be emphasized on the wider definitions that embrace a set of policies, structures, customs, laws, and procedures which define the controlling and administrating of owner's resources (Onuorah & Chi-Chi, 2016). Despite the different definitions of the concept of CG, there is a commonly accepted rationale behind the emergence of that notion; the 'Agency Problem' (Berle & Means, 1932;Jensen & Meckling, 1976). The agency problem aroused due to the separation of ownership and management in large corporations, where the managers (agents) try to maximize their benefits at the cost of owners (principals).
When considering the CG in the Sri Lankan context, it has become significant during the last two decades due to happening some isolated incidents of corporate failures, certain economic reforms, and a series of recent scandals (i.e Pramuka Bank, Vanik Corporation, Trading suspension of Entrust PLC, and Swarnamahal Finance PLC) in Sri Lanka. The first Sri Lanka code of best practices on CG was introduced in 1997 by the Institute of Chartered Accountants of Sri Lanka (CASL) to deal with Sri Lankan listed companies. It was based on the Anglo-Saxon model of corporate governance and was a blueprint of the Cadbury code (1992). The 1997 code was replaced by the CASL code of best practices which was introduced in March 2003 based on the Hampel report (1998). During the year 2008, the standards of CG were introduced into the CSE listing rules and made as mandatory compliance for listed companies in Sri Lanka by developing a joint initiative of CASL and SEC (Colombo stock exchange, 2008). Subsequently, certain revisions for the code of CG best practices have been made by CASL and SEC over a period to incorporate with recent global developments (Senaratne & Gunaratne, 2008;Weerasinghe et al., 2017). Accordingly, the Sri Lankan listed companies are now delt with a comprehensive model of corporate governance, which represents a mixture of both mandatory (companies act 2007, CSE listing rules, SEC directives and codes, CBSL directives for banks and financing companies) and voluntary (code of best practices of CG -CASL 2017) rules of corporate governance.

Understanding the concept of Financial Reporting Quality (FRQ)
Financial reporting in its ideal sense is the first source of independent communication about the management performance which helps investors, creditors, and other users to assess the basis for economic decision making on providing resources to the entity. The publicly held firms evaluate their financial position and performance through financial reporting information generated as a major outcome of the external reporting system. Although financial as well as non-financial disclosers might be communicated in different means, the traditional financial statements are always the most significant reports (Alzoubi, 2014;Paulinus et al., 2017;Trai, Kha, Trung & Dung, 2018).
Quality is a broad concept that can be measured comparatively. According to Cohen et al. (2004), the past researchers show a vague concept of financial reporting quality. In light of these, Cooray, Gunarathne, and Senarathne (2020) suggested that the concept of FRQ is an important aspect for its related parties to assure that financial reports, disclosers, and other non-financial reports minimize any asymmetry of information and improve the usefulness of decision making. Cohen et al. (2004) argued that the auditors, audit committee participants, and managers are now struggling with defining financial reporting quality. So, accounting literature has much focused on the factors such as earnings management, financial restatements, and frauds which demonstrate the attainment of high-quality financial reports, rather than defining FRQ.
Consequently, the latest research studies, are aware of the two general perspectives that are widely emphasized in the assessment of FRQ. The first perspective focuses on providing useful financial information to the users whereas, the second perspective relies on the sufficiency of the information provided to the users in order to guarantee the protection of investors (Paulinus et al., 2017). The stakeholders of the entity have different and possibly conflicting information needs. Thus, the perspectives on the evaluation of financial reporting quality are also varying accordingly. As a result, this study almost puts effort to concentrate on FRQ in terms of the accruals model that will be considered as a comprehensive measurement tool for financial reporting quality.

Theoretical background
The concept of CG originated with multiple theories. The most prominent two theories considered to explain the agency conflicts and the need for CG are Agency theory and Stakeholder theory. They can be viewed as two main theories amongst many other applicable theories such as Resource dependence theory and Stewardship theory. Agency theory is the predominant theory since it is considered to be given birth to the concept of CG. However, all of them can be treated as different lenses to see the concept of CG from different and interesting angles. The particular angle or the perspective of CG from which looking at will decide the appropriateness of a more suitable theory.
Agency Theory was defined in two aspects as, the economic perspective (Ross, 1973) and the institutional perspective (Mitnick, 1973). Costs associated with a lack of goal congruence between two parties were brought to the fore by Ross (1973) and were further explored by Jensen & Meckling (1976). The separation of ownership and management provides the opportunity for management (agents) to act in their self-interest by maximizing their wealth and power at the expense of the owners (principals) (Fama, 1980;Jensen & Meckling, 1976). Since this relationship is not harmonious; indeed, so-called agency conflicts or conflicts of interest between agents and principals arise. This is known as the "Agency Problem". Hence, the companies try to limit this agency problem through solid and effective CG policies such as including independent non-executive directors to the board, forming an audit committee to ensure fair management, and establishing two separate positions for CEO and chairman (Uwuigbe et al., 2018;Weerasinghe et al., 2017). Thus, such kind of adequate control mechanisms should be established along with good corporate governance to direct the behavior of the managers and to compel them to act in the best interests of the shareholders Stakeholder theory founded by Freeman (1984) emphasizes the existence of different stakeholder groups of a corporation and therefore, the necessity of serving the interests of all stakeholders. This theory suggests that the companies have a social responsibility to restructure the CG framework, apart from the owner-manager relationship, and identify each set of interest groups (Paulinus et al., 2017). In stakeholder theory, the principal-agent problem has been further widened, because of concerning the interests of multiple principals as central to the sustainability of the business firm. Compared with the agency theory, this theory demonstrates corporate governance in a holistic view, as a control mechanism created for efficient operations of a firm (Manawaduge, 2012). According to Weerasinghe et al. (2017), CG best practices, as the stakeholder theory point of view, should protect the interests of broad stakeholders and not only shareholders.
Resource dependence theory views a firm as an open system, dependent upon external organizations and environmental contingencies (Pfeffer & Salancik, 1978). According to the resource dependency theory, a larger number of directors on the board might be beneficial for high-quality financial reporting. Larger the size of the board directors in the board may give more opportunities than smaller boards. Similarly, agency theory also supports the Board Independence and CEO duality which will also have a positive impact on FRQ. Regarding audit committee characteristics and FRQ, the resource dependency theory presumes that the audit committee serves as a source of advice and counsel for the BOD to bring valuable resources to the organization (Zábojníková, 2016) and the have a positive relationship between audit committee characteristics and FRQ. When a small AC lacks with diversified skills and expertise provided by a large audit committee, it would make them ineffective. An audit committee with an appropriate amount of members allows them to make use of their knowledge for stakeholders' advantage (Goodstein, Gautam, & Boeker, 1994).
Stewardship Theory presents a contrasting view to Agency theory, which is 'managers are essentially trustworthy individuals and therefore good stewards of the resources entrusted to them' (Donaldson, 1990;Donaldson & Davis, 1991. Donaldson and Davis (1991) state that 'managers are principally motivated by achievement and responsibility needs'; and once they are given the responsibility, self-directed work, organizations may be better served to free-managers under non-executive director dominated boards. Accordingly, Stewardship theory suggests that managers are naturally trustworthy; thus, there will be no major agency costs (Donaldson & Davis, 1994). It assumes that in general, managers want to do a good job and will act as effective stewards of an organization's resources. As a result, senior management and shareholders are better seen as partners. Hence, the main function of the board is not to ensure managerial compliance with shareholders'/members' interests as per agency theory, but to improve organizational performance.

Empirical Evidence between Corporate Governance and FRQ
The impact of CG characteristics on FRQ has been comprehended through several prior studies (Abdulmalik & Ahmad, 2016;Amrah & Obaid, 2019;Sepasi, Deilami & Tavakoli, 2017). However, contradictions in the literature were found in various CG attributes and their subcomponents. When investigating the relationship between Board Size (BDSIZE) and FRQ variety of views was highlighted. In line with the agency theory, companies with large director boards boost the transparency and credibility of the accounting data by acting as a strong controlling mechanism (Phuong et al., 2020;Shah, Rashid & Shahzad, 2019). In contrast, Hassan (2011) and Paulinus et al. (2017) have expressed that large board size misplaced the strong governance as they have a lower level of coordination and a higher probability of issues related to free-riders. Further, in comparison to larger boards, smaller boards will foster a high degree of communication and be more active in timely decision-making. This is justified by the studies of Chalaki, Didar, and Riahinezhad (2012) and Chiedu (2014) which emphasize the negative association between board size and financial reporting quality measured through discretionary accruals.
Secondly, when analyzing the literature on Board Independence (BDIND) and FRQ, plenty of studies (Alzoubi 2014;Ibrahim & Jehu, 2018) have shown a significant positive relationship between the board independence and FRQ which suggest independent members of BODs voluntarily disclose more relevant information on corporate financial administration and thereby indirectly serving to maximize shareholders value. Inversely, Chalaki et al. (2012) and Onuorah et al. (2016) have proved that there is no such relationship between board independence and financial information quality in their studies, where the independent directors are less competent as they possess limited information on the company's operation and their presence on the board may have no impact on FRQ.
CEO duality emphasizes the situation where one person simultaneously carries the positions of both chairman of the board and CEO. Agency theory advocated the importance of separation between these two persons to ensure better management and satisfy the shareholders' interests (Shah et al., 2019). Alzoubi (2014) has stated that CEO duality has negatively impact on financial reporting quality measured through abnormal accruals, as it damages the firm value and sustainability of the firm in long run. Nevertheless, other studies have indicated CEO duality improves the firm value because of bringing an extra power to CEO to mitigate the ambiguity in decision making. Trai et al. (2018) has focused on the positive relationship between CEO duality and FRQ as it minimizes the extra cost which arises due to the separation of two roles.
Considering the Audit committee and FRQ, theoretically as per the resource dependency theory, larger audit committee size (ACSIZE) higher the resourcefulness of the entity which improves FRQ. This is due to having a wider knowledge base and varied expertise members in the committee. Accordingly, Eyenubo, Mohammed and Ali (2017), Shankaraiah and Amiri (2017) has shown a positive relationship between the variable of audit committee size and FRQ. On the other hand, some studies (Majiyebo, Okpanachi, Nyor, Yahaya, & Mohammed, 2018) have suggested larger audit committees are more likely to increase risk of material misstatements by depicting a negative relationship among audit committee size and FRQ. As a result, many regulatory bodies such as Blue-Ribbon committee in US (1999), ASX corporate governance council in Australia (2003), and combined code in UK (2008) have recommended, at least three members as the appropriate size for the audit committee.
Audit committee independence is a foremost characteristic influencing committee competence and effectiveness in managing the financial information process. The presence of an audit committee with a majority of independent directors ensures a strong controlling mechanism by reducing the agency cost, improving the internal controls, and thereby paving the way to high-quality disclosures. The extant literature provides evidence that firms with a higher proportion of independent directors report lower discretionary accruals, thus enhancing overall FRQ (Hasan, Kassim & Hamid, 2020;Kibiya, Che-Ahmad & Amran, 2016). However, Majiyebo et al. (2018) and Shah et al. (2019) in their studies, have documented that role of the independent audit committee is rather limited in controlling and monitoring the management behavior over reducing earnings manipulation and therefore present an adverse association between audit committee independence and FRQ.
It is very crucial to have audit committee members with in-depth knowledge and experience on accounting standards and auditing procedures to improve the credibility of financial statements, hence enhancing the FRQ more effectively. This characteristic of the audit committee has received considerable attention in the previous literature that indicates a positive relationship between audit committee financial, accounting expertise, and FRQ (Alzeban 2019;Hassan 2011;Nuraini 2015). In contrast, some studies have found an insignificant relationship which argues that having a financial expert in the audit committee is not enough component for improving the quality of financial information (Hasan et al., 2020;Kibiya et al., 2016). According to Shankaraiah et al. (2017), some accounting experts in the audit committee assess the subjective judgments of the management, meet more frequently, and more often disagree with the financial reporting decisions of the management, and finally result in inadequate monitoring over the reporting process.
Considering the empirical evidence in the local context, there is a dearth of studies on CG and FRQ in Sri Lanka due to the nature of the emerging economy. Kankanamge (2015) has studied the impact of board characteristics on earnings management measured using discretionary accruals model, in Sri Lanka during the period from 2012-2015 and found that there is a significant relationship between board size, board independence, board financial expertise, and earnings management of the firms and thereby enhancing the FRQ and transparency. Further, Kankanamge and Shantha (2015) in their study, have documented the impact of audit committee characteristics (audit committee size, independence, meetings, financial expertise) on FRQ which results in a strong negative relationship between these characteristics and discretionary accruals. According to De Silva, Manawaduge, and Ajward (2017) audit committee financial and accounting expertise has a negative relationship on the level of earnings management, while other selected corporate governance characteristics have not any influence on the level of absolute discretionary accruals. Moreover, Cooray et al. (2020) has emphasized the effect of governance mechanisms on the quality of integrated reporting that shows plenty of mixed evidence between corporate governance components and qualitative characteristics of integrated reporting. (See Appendix A for a summary of the empirical evidence.) When exploring the empirical evidence in the global and local context, contradictory mixed results were found on the relationship between CG characteristics and FRQ. Hence, the current study focuses on filling the empirical gap through further studying on the same area by introducing six independent variables (Board Size, Board Independence, CEO Duality, Audit Committee Size, Audit Committee Independence, and Audit Committee Accounting Expertise) and the level of FRQ in Sri Lanka.

Data collection and Sampling
Data were collected by the researchers to excel spreadsheet and analyzed using E-views software. Annual reports of companies downloaded from the CSE website were used excluding the banks, finance companies, and insurance companies as these industries are strictly regulated and apply different methods in their financial reporting including fundamentally different cash flows and accrual processes. The sample consisted of 209 companies listed in CSE during 2017 and 2020. The stratified random sampling technique was used to select the companies for the sample. 51 companies were excluded because of missing information relating to key variables. Table 1 shows the composition of the sample of the study.

Hypotheses Development
Companies with large director boards boost the transparency and credibility of the accounting data by acting as a strong controlling mechanism. Most of the literature (Phuong et al., 2020;Shah, Rashid & Shahzad, 2019) supports the argument that a higher board size would be preferred for a lesser board size to generate high-quality financial reporting. Similarly, it was evident that the independent members of BODs voluntarily disclose more relevant information on corporate financial administration and thereby indirectly serve to maximize shareholders' value (Alzoubi 2014;Ibrahim & Jehu, 2018) which ultimately supports a positive impact on FRQ.
Considering the CEO duality, agency theory also advocated the importance of separation between these two persons to ensure better management and satisfying the shareholders' interests (Shah et al., 2019). Hence, with the support of these arguments in the literature, the first three hypotheses were developed as follows.
H2: There is a significantly positive relationship between board independence and FRQ (Ibrahim et al., 2018).
Based on the prior literature, the audit committee is considered as an additional internal governance mechanism whose influence may improve the quality of financial reporting. In this respect, an audit committee has three main characteristics that should be taken into consideration, these are; audit committee expertise, audit committee size, and independence. Hence, in the study, we consider those three characteristics will have a positive impact on FRQ. However, as stated in the literature review section, some studies proved the contrary situations. Hence, we believe it is necessary to test these three hypotheses to discover the relationship between audit committee size, independence, expertise attributes, and FRQ in the Sri Lankan context. Therefore, with the support of the literature, AC characteristics (Cooray et al., 2020;De Silva et al., 2017;Kankanamge et al., 2015;Shankaraiah et al., 2017) following hypotheses were designed.
H5: There is a significant positive relationship between audit committee independence and FRQ (Hasan et al., 2020).
H6: There is a significant positive relationship between audit committee accounting expertise and FRQ (Alzeban 2019; Madhurangi and Abeygunasekera 2021).

Conceptualization
The study has adopted the following basis of measurement presented in Table 2, as proxies of quantifying each governance component, FRQ, and control variables.

Model Specification-Determining Financial Reporting Quality
Among accounting-related measures of FRQ, accruals earnings quality is very important in reflecting the performance of the organization and thereby produces more relevant financial information. The dependent variable of the study is FRQ, measured as the absolute value of the discretionary accruals. The absolute value of discretionary accruals depicts a negative relationship with FRQ. Therefore, when the absolute value increases, the magnitude of earnings management raises hence the FRQ decreases. Much prior research has applied the accrualbased earnings management technique when evaluating the association between corporate governance and FRQ (Alzeban 2019;Alzoubi 2014;De Silva et al., 2017). Different models can be used to estimate discretionary accruals, and Dechow et al. (1995) provide evidence that the modified Jones model is the most powerful compared to other existing models to observe FRQ. Accordingly, the absolute value of discretionary accruals is derived based on the study of De Silva et al. (2017) as follows: In the first stage, the total accruals (TA) are estimated, and then the non-discretionary accruals (NDA) are subtracted, thereby yielding the discretionary part of TA.
Step 1 -The total accruals are estimated by subtracting net cash flows from operating activities from net income as the cash flow approach.
TAi,t= NIi,t -CFOi,t Step 2 -The calculation of the coefficient parameters separately for each industry (industry average α values) by regressing equation 2 (modified Jones model) below.
DAi,t / Ai,t-1 = TAi,t / Ai,t-1 -NDAi,t / Ai,t-1 (04) Where: TAi,t = Total Accruals of the firm i for the year t NI i,t = Net Income before discontinued segments of the firm i for the year t

Regression model
The dependent variable of this study "financial reporting quality" (FRQ) is approximated by the Absolute discretionary accruals (ADA). The increase in ADA represents a lower level of FRQ and vice versa. To investigate the impact of CG characteristics on FRQ, a regression model has been established following De Silva et al. (2017). We have employed the below multiple regression models using E View statistical package.

Control variables
FSIZEi,t = Natural logarithm of sales LEVi,t = Ratio of total debt to total assets Before deriving the final output, we have tested the pre-diagnosis tests of the regression analysis. Thereby we tested for Stationary, Normality, Multicollinearity, and Autocorrelation. As per the normality testing, the data set should be well modeled by a normal distribution to use the regression estimates to test the hypotheses. Similarly, one or two extreme residuals cause a rejection of the normality assumption as well. Non-normality occurs due to the nature of the secondary data. Even previous literature has experienced the non-normality issue mostly when the study sample differs from its population. In this research study, a sample of 209 companies has been selected. Therefore, we used the option of adding 3 dummy variables analyzing the extreme values using E-views and rectified the normality assumption, and obtained a desired Jarque-Bera test statistic value of 1.0000.
Considering multicollinearity, the VIF results depict that the centered VIF values of all the variables are below 10, thus it was assumed no multicollinearity issues. Furthermore, when determining whether to employ a fixed or random model, the Hausman test was done, and based on the findings of the Hausman Test, the probability value is greater than 0.05. Hence, the random effect of the panel model is used to test the hypotheses.

Descriptive Statistics
Under the descriptive analysis, at first, a descriptive statistic of variables with mean, median, maximum, minimum, standard deviation have been presented in Table 3. Accordingly, there is an 836 number of observations, and the mean value represents the average value of observations. Considering the mean value, ADA shows 0.070 of average value, BDSIZE represents 8 directors, BDIND represents 3 directors, whereas ACSIZE demonstrates 3 members, ACIND represents 3 members, and ACAEXP represents 1 member. The maximum is the highest value among each of the stated variables and the minimum depicts the lowest value compared with related variables. Standard deviation measures the level of spread from the central value. Accordingly, ADA presents the lowest standard deviation of 0.075, which emphasizes that most of the values are concentrated around the mean value, while BDSIZE presents the highest standard deviation of 2.033, which describes the values are highly spreading over the mean value.
The normal distribution of the observations is measured through skewness and kurtosis of the descriptive analysis. In light of this study, the majority of the variables illustrate a right-skewed distribution indicating most of the companies record values of each variable less than its mean value. Moreover, this study demonstrates a leptokurtic distribution for all of its dependent and independent variables that represent most of the values are closely gathered around the average/mean value.  118 0 -----Note: **p<0.05 denotes the significance less than 5%, ***p<0.01 denotes the significance less than 1% Table 4 presents Pearson correlations of variables in this study. It can be observed that the board size shows a weak positive significant relationship with the absolute discretionary accruals (+0.058) and thereby represents a negative impact on FRQ. However, a weak negative association can be seen between the board independence (-0.051), audit committee independence (-0.027), and the absolute value of discretionary accruals. Thus, it proves when the board and audit committee have more independent directors, financial reporting quality is more likely to improve with lesser board size. CEO duality reports a weak positive relationship (0.040) with discretionary accruals of the study and audit committee size shows a weak significant positive relationship (0.008) with absolute discretionary accruals. Hence, these variables depict a negative relationship with firms' FRQ, as the value of discretionary accruals adversely impacts the quality of financial reporting. Audit committee accounting expertise shows a moderate significant negative relationship (-.0.652) with absolute discretionary accruals, thus depicting a positive impact on FRQ.

Correlation Analysis
Further, audit committee accounting expertise is the only variable that represents a systematic relationship with discretionary accruals and none of the other board and audit committee characteristics has depicted a significant relationship with the absolute discretionary accruals at any of the significant levels of p<0.01, p<0.05, p<0.10. However, considering the control variables, it is noted that the firm size shows a significantly weak negative association with absolute discretionary accruals. Finally, leverage presents a weak positive relationship on the absolute discretionary accruals and thereby creating a negative effect on FRQ.  Table 5 represents the results of multiple regression analysis based on the random effect model as confirmed by the Hausman specification test. The value of adjusted R-squared 0.591 indicates that all the independent variables in the model jointly represent to the extent of 60% variation of the dependent variable, discretionary accruals. Further, the F-statistic value of 21.871 with the probability value of 0.000 indicates the usefulness and statistical efficiency of the model at a 99% significant level to analyze the collected data.

Regression Analysis
The results of the regression analysis show that the board size and audit committee accounting expertise are the only variables that affect FRQ. Audit committee accounting expertise depicts a significant negative association on discretionary accruals with a coefficient value of -0.014 (p=0.002), which in turn shows a significant positive relationship on FRQ. Therefore, the last hypothesis of H6 is supported. However, the board size depicts a significant positive association on discretionary accruals with a coefficient value of 0.007 (p=0.048), which in turn shows a significant negative relationship on FRQ. Thus, H1 is not supported.
No other selected CG variables are significantly related to FRQ (p>0.05) measured through the absolute value of discretionary accruals. Hence, it does not support the hypotheses. These findings are similar to Hasan et al. (2020), Majiyebo et al. (2018), Adewale and Babajide (2019), Khan, Rehman, Zeeshan, and Afridi (2020), Subhasinghe and Kehelwalatenna (2021), Chalaki et al. (2012), Shah et al. (2019), and Uwuigbe et al. (2018). In relation to the control variables, firm size shows a significant negative correlation with a -0.086 value of coefficient (p=0.0370) whereas, leverage represents a significant positive correlation with a 0.095 coefficient (p=0.0000). These findings suggest that the larger firms show a high propensity on increasing FRQ and a firm that records a high degree of leverage is less likely to improve FRQ. The research works of De Silva et al. (2017) and Subhasinghe et al. (2021) have also confirmed similar findings on this in the Sri Lankan context.
According to the regression output, audit committee accounting expertise can be considered as the most important variable to enhance FRQ, as the absence of audit committee accounting expertise might lead to poor FRQ. Similarly, it emphasizes the importance of having a small board size to improve the FRQ. These findings are supported by previous studies in the literature (Alzeban 2019;Hasan et al., 2020). They suggested that, when the audit committee comprises members with a background in accounting and finance, it causes lower levels of discretionary accruals hence, improving FRQ. Further, regression analysis shows that board size negatively affects FRQ through a positive coefficient with ADA. Therefore, it is difficult to prove that the board with a high number of directors enhances the quality of the financial reporting process. Chiedu (2014) and Ibrahim et al. (2018) are some studies that depict similar evidence between board size and FRQ. Nevertheless, some contradictory results were shown by studies of Kankanamge (2015) and Phuong et al. (2020).

Conclusion
The current study examines the association between CG characteristics on FRQ in Sri Lankan listed companies. The study examined six selected CG variables including board size, board independence, CEO duality, audit committee size, audit committee independence, audit committee accounting expertise, and the level of FRQ using the absolute discretionary accruals. Less number of corporate governance-related studies have been carried out in developing economies compared to the developed countries (for example, Alzeban 2019; Amrah et al., 2019;Johl et al., 2013). However, only a few studies have been conducted on the relationship between corporate governance and FRQ in the Sri Lankan context and the evidence of these are inconclusive. Similarly, each study varies one another based on the methodology adopted during the research. Accordingly, this study has become significant to bridge the gap in the literature by providing empirical evidence on the association between corporate governance and FRQ in Sri Lanka using a sample of 209 listed companies in CSE from the year 2017 to 2020.
The output indicates that all the identified corporate governance characteristics except board size and audit committee accounting expertise do not present a significant relationship with absolute discretionary accruals. These findings were rationalized by getting support from the extant literature of previous studies (Chalaki et al., 2012;Khan et al., 2020;Madhurangi et al., 2021;Subhasinghe et al., 2021). Furthermore, it emphasizes the importance of the audit committee participants with membership in a recognized professional accounting body and sufficient finance and accounting literacy. That might decrease the number of fraudulent activities, while also enhancing the effectiveness of the internal control process. Another main argument is that these experts provide appropriate advice on the problems relating to finance and reduce the likelihood of an agency problem forming within the organization, as they have considerable financial acumen and knowledge. Therefore, the quality of the financial reporting process improves (De Silva et al., 2017;Hasan et al., 2020;Kibiya et al., 2016). Similarly, findings of the study demonstrate that a larger board tends to be less vigilant, commit less time and effort to review and decision making. So that, the number of directors on the board may not affect enhancing FRQ.
The current study contributes to the literature by bringing the empirical gap on investigating the CG characteristics on FRQ. The findings of this research are expected to have a significant policy implication for policymakers and regulators in terms of formulating strategies and policies on corporate governance best practices in Sri Lanka. Similarly, the findings of the research study should have an important implication for corporate sectors that aim to maximize the corporate value from effective governance mechanisms. Further, relevant regulatory bodies should deeply emphasize enhancing the financial reporting disclosure requirements in the annual reports to provide good insight to the users of the financial statements. For instance, the investors have a good opportunity to get a clear understanding of firms' performance by studying financial reports before making relevant investment decisions. Our study is subject to limitations of the characteristics of a frontier stock market excluding the banking and finance sector due to unique features. The extension of this type of study to other institutional settings may be a direction for future research.
Appendix A