Corporate Governance Compliance among the Listed Firms in Colombo Stock Exchange: Does Size Matter?

Literature is conflicting on the determinants of corporate governance, particularly relating to emerging financial markets. Therefore, this study examined whether firm size affects corporate governance compliance based on firm-level data from 2016 to 2020 for a sample of 100 firms listed on the Colombo Stock Exchange selected using the systematic random sampling technique. Firm size was measured using the natural logarithm of total assets. A corporate governance index was used to measure corporate governance compliance on 18 board related best practices. Results of three random-effects panel regression models indicated that firm size has a statistically significant positive effect on corporate governance compliance. This implies that larger firms are better motivated or capable of complying with corporate governance due to their more extensive resource. This way they can benefit from the reduced agency cost and increased attractiveness for investors. Smaller firms might have considered compliance with corporate governance as less beneficial given the higher cost of compliance. Therefore, it would be more appropriate to make policies flexible enough in the application based on the firm-specific characteristics, such as the firm size. Even though attempts have been made to assess the effect of firm size on corporate governance using firm size as a control variable, this is the first study dedicated to investigating the size effect on corporate governance compliance in the Sri Lankan context.


Introduction
Owners appoint experienced agents to run the businesses on their behalf. This separation of ownership and control can result in a potential conflict between the interests of owners and managers, which is referred to as agency conflict (Jensen, 1986). Given the extent of information asymmetry, managers of the firms may use firms' assets for their interests instead of maximizing owners' wealth. Corporate governance spells out the rights and responsibilities of different stakeholders including managers, board of directors, and shareholders to streamline the corporate decision making and ultimately assist in attaining corporate objectives (Manawaduge, 2012). Codes of best practices on corporate governance offer self-regulated provisions as a monitoring mechanism to mitigate these inherent principal-agent conflicts (Pritchett, 1983). As a result, a reduction in information asymmetry between managers and shareholders and agency costs are generally expected (Chang, 2018). Moreover, compliance with corporate governance practices also leads to increased investor trust, better performance and higher market value (Brown & Caylor, 2006;Senthuran & Velnampy, 2015).
The first code relating to corporate governance in Sri Lanka was introduced in 1997 based on the British governance code of 1992. Subsequently, this code was amended with the consultation of various stakeholders in 2003, 2008, 2013, and 2017. Despite the vast differences in institutional settings, Sri Lanka still uses corporate governance practices adopted from developed countries (Manawaduge, 2012). As argued by Castelluccio (2005), these codes have ignored the concerns of small firms. Hence, as Hertig (2005) argued, requiring smaller firms to comply with governance practices designed for larger firms can be ineffective.
In Sri Lanka, most of the listed firms are relatively small. For example, the average market capitalization of the 299 entities listed in the Colombo Stock Exchange (CSE) as of 10th May 2021 was LKR 10.6 billion. Only 60 firms (i.e., 20 per cent of the total firms) had a market capitalization above the average, while only 18 firms (6 per cent) had a market capitalization above LKR 50 billion. Recently, the Sri Lanka Code of Best Practice on Corporate Governance 2017 has considered these size differences to some extent and recommended that small companies listed on the Diri Savi board can have a minimum of two non-executive directors compared to three minimum non-executive directors for large companies listed in the mainboard. However, the audit committee and remuneration committee require a minimum of three nonexecutive directors and no alternate provision is there for companies listed on the Diri Savi board.
Further, there are differences in corporate governance compliance among listed firms even though CSE listing rules have mandated compliance with some of the corporate governance practices for listed firms in CSE (Dissanayake et al., 2021). Apparently, the extent to which a firm is implementing corporate governance depends on the decisions of influential shareholders (Aluchna & Kuszewski, 2020). In firms characterized by concentrated ownership, like in Sri Lanka (Hewa Wellalage & Locke, 2014), the central conflict of interest occurs between the majority shareholders and the minority shareholders (La Porta et al., 1998) than between owners and managers. Previous literature emphasizes corporate governance as a way of attracting investors (Chang, 2018). Nevertheless, the controlling shareholders in firms with concentrated ownership may perceive corporate governance as a mechanism that dilutes their power (Chen et al., 2011). This may lead to lower compliance. For example, as Manawaduge (2012) stated, the level of corporate governance compliance in Sri Lanka is unsatisfactory despite the given benefits. Therefore, investigating whether these differences in corporate governance compliance are due to the firm size effect is vital.
Researchers have argued that policymakers consider larger corporations when developing governance codes as governance failures of larger corporations are likely to have a stronger effect on the investors and markets (Hertig, 2005). Hence, smaller firms are facing difficulties in complying with such corporate governance practices. Previous literature also evidenced that firm size affects the level of corporate governance compliance (Haniffa & Cooke, 2002;Samaha et al., 2012). In CSE, most firms are far smaller than the firms in the top 20 percent in terms of market capitalization. Nonetheless, there is a dearth of Sri Lankan studies that focus on the corporate governance compliance differences between larger and smaller firms. Therefore, this paper argues that the codes of best practices on corporate governance formulated in developed countries with a particular focus on large corporations where a substantial diffusion of ownerships is observed turns up a question of whether these codes address the unique concerns of the firms in emerging economies like Sri Lanka where highly concentrated ownership and far smaller corporations are generally observed. Therefore, investigating whether these differences in corporate governance compliance are due to the firm size effect is vital. The contribution of this study is threefold. First, this study remains the first attempt in Sri Lanka to identify the impact of firm size on the level of corporate governance compliance. The findings of the study suggest that compliance with corporate governance reduces when the size of the firm reduces. In empirical studies conducted in other countries, this was mainly attributed to the relatively high corporate governance implementation costs to small firms. Second, this study provides empirical evidence to claim that even though sound corporate governance can be beneficial to smaller firms with high growth opportunities, the smaller firms are less encouraged to comply with governance best practices probably due to the high cost of compliance. These costs might overweigh the benefits in contexts like Sri Lanka. This phenomenon requires further studies. Third, this paper suggests policymakers to develop flexible governance practices considering the firm-specific characteristics such as firm size to achieve a proper balance between costs and benefits of compliance with corporate governance best practices. Existing alternative provisions available for smaller firms are inadequate. This study supports the view that there is no one size fits all approach to corporate governance.

Literature Review
The Agency theory introduced by Jensen and Meckling (1976) remains the foundation for arguments on agency conflict. They have defined a firm as a nexus of contracts. An agency contract can be identified as a contract in which principals engage agents to manage the firm. Here, a separation between ownership and management of firms emerges. Agency theory argues that a conflict of interest between owners and managers can emerge due to this separation of ownership from the firm's control. For example, managers may misuse owners' wealth for their benefit by revealing insider information about the firm (Jensen, 1986), rewarding themselves with excessive salaries and bonuses (Ntim & Osei, 2011;Shleifer & Vishny, 1997), wasting corporate resources by unnecessary consumption of perquisites (Jensen & Meckling, 1976), and making suboptimal decisions (Anderson & Reeb, 2004).
Managing these opportunistic behaviours of agents involves a cost to the firms. These costs are known as agency costs. Agency costs consist of monitoring costs, bonding costs and residual losses (Jensen & Meckling, 1976). By incurring monitoring and bonding costs, principals try to make sure that agents take all the decisions in the best interest of principals. Residual losses occur when monitoring and bonding costs fail to discipline the management behaviour. Agency theory explains that agency costs can be mitigated and the misappropriation of shareholders' wealth can be prevented by making arrangements to align the interests of agents and principals.
Here, corporate governance acts as an arrangement in place across many countries to reduce these opportunistic behaviours of managers, thereby reducing agency costs.
Simply, corporate governance spells out how shareholder property rights can be protected, especially in the context of dispersed ownership structure. However, in some countries like Sri Lanka, ownership is not heavily dispersed. Instead, a concentrated ownership structure can be observed where fewer shareholders hold almost all share capital (Hewa Wellalage & Locke, 2014). In such contexts, the primary concern that corporate governance should address is the potential expropriation by controlling shareholders at the expense of minority shareholders (Chen et al., 2011). This type of controlling shareholder expropriation can occur in the forms of appointing unqualified and overpaid family members or friends as managerial persons, advancing political, personal, and familial agendas that hinder company performance or engaging in self-beneficial trading (Claessens et al., 2002). This can be particularly severe in countries with weaker legal systems where shareholder protection is not strong. In developed countries with better legal systems, controlling shareholder expropriation can be mitigated to a large extent by legal provisions. However, with weak legal and institutional settings in emerging countries, minority shareholders often rely on voluntary corporate governance practices (Claessens et al., 2002).
Corporate governance mechanisms like an active board of directors, separation of the roles of chief executive officer and chairman, and having a majority of outside directors on the board are crucial in protecting the interest of minority shareholders against the controlling shareholder expropriation. For example, Lee et al. (1999) have found that firms with a higher number of nonexecutive directors on the boards result in higher firm performance, particularly in smaller firms characterized by less financial expertise and constrained resources. Similarly, Beasley (1996) has also identified that boards dominated by non-executive directors fulfil their monitoring role more effectively and reduce financial fraud significantly. Fama and Jensen (1983) found that separating the roles of chairman and CEO separates the decision management and decision control and reduces agency problems and leads to enhanced firm performance.
Nevertheless, the nature and extent of agency problems differ across firms. For example, the extent to which the ownership is concentrated or diffused, the existence of controllingshareholders and minority shareholders, and the complexity of firm operations can influence the complexity of agency issues. For example, ample literature has shown that corporate governance best practices designed for developed countries cannot resolve agency issues in emerging countries (Chen et al., 2011). This implies that the benefits and costs of adopting corporate governance best practices can differ across firms and countries. Therefore, the optimal corporate governance for a firm depends on the trade-off between the costs and benefits of corporate governance (Bruno & Claessens, 2010). In some cases, the cost of corporate governance compliance may outweigh the benefits for some firms depending on several firm-specific characteristics.
Politicians can affect how an organization's wealth is distributed using several ways such as regulations, taxes, subsidies, etc. (Watts & Zimmerman, 1978). Therefore, Watt and Zimmerman (1978) have argued that in order to avoid such political pressures companies use various strategies like CSR campaigns, government lobbying, and earning manipulations. According to their view, the political cost is highly related to the firm size as larger firms with high profits often draw the attention of the politicians and the public. Based on this initial study by Watts and Zimmerman (1978), several studies have empirically found that firms make voluntary social disclosures to reduce these political costs (Belkaoui & Karpik, 1989;Lemon & Cahan, 1997). This means, that through the voluntary disclosures and compliance with corporate governance, large firms can reduce the adverse public and political concerns (Mine, 2002).
Concurrently, literature has shown that firm size is often used as a proxy for several firm-specific characteristics such as political costs, information production costs and competitive advantage (Hassan et al., 2006). This is because the firm size or the scale influences many firm-specific characteristics. Therefore, an argument can be made that corporate finance practices differ depending on the firm size. This can be explained using several reasons (Hashmi et al., 2020). First, when firms grow in size, the number of shareholders also grows increasing the diffusion of ownership (Hashmi et al., 2020). Therefore, larger and more balanced boards are required to maintain the board representativeness. Second, larger firms receive greater attention from various stakeholders such as financial analysts, rating agencies, government agencies, and investors (Durnev & Kim, 2005;Hassan et al., 2006). This necessitates, larger firms to be managed under greater scrutiny (Hussainey & Al-Najjar, 2012). Concurrently larger firms require more disclosures and compliance with strict regulations (Hashmi et al., 2020). Third, larger firms are more diversified and geographically spread (Hashmi et al., 2020). As a result, larger firms are characterized by complex operations (Chung & Zhang, 2011;Haniffa & Cooke, 2002;Jensen, 1986). Consequently, this increased diffusion of ownership and complexity of operations lead to higher information asymmetry between owners and managers (Hassan et al., 2006;Jensen, 1986;Nguyen et al., 2020). Nevertheless, information asymmetry between banks and creditors can dissipate with increased reputation resulting from larger size (Nguyen et al., 2020). This can substantially influence the financial policy of the firms (Hashmi et al., 2020). Finally, firms with larger free cash flows are exposed to higher conflicts of interest (Jensen, 1986). Therefore, the agency costs can be higher in larger firms than in smaller firms (Alsaeed, 2006;Himmelberg et al., 1999;Klapper & Love, 2004).
Therefore, on the one hand, larger firms are required to adopt strict corporate governance practices. On the other hand, more stringent corporate governance rules such as information disclosure, maintenance of a large board and sub-committees can be costly to implement (Bruno & Claessens, 2010). Nevertheless, larger firms have more resources to afford the costly implementation of corporate governance practices (Klapper & Love, 2004) such as information disclosure (Hassan et al., 2006;Hussainey & Al-Najjar, 2012). In contrast, implementing such corporate governance practices may not be affordable to the resource-constrained smaller firms (Haniffa & Cooke, 2002). Moreover, some of the strict corporate governance practices such as information disclosure are detrimental to smaller firms because they may create a competitive disadvantage for them (Chhaochharia & Grinstein, 2007). This means the cost of complying outweighs the benefits for smaller firms. In line with these arguments, most of the empirical research has supported the positive association between firm size and corporate governance compliance (Ariff et al., 2007;Bruno & Claessens, 2010;Cho & Kim, 2003;Mallin & Ow-Yong, 2012).
Investors often seek companies with better monitoring mechanisms for investing to ensure their investment is safeguarded (AlHares, 2020). Smaller firms reliant on external financing to finance their growth opportunities tend to use better corporate governance as a substitute for collateralizable assets because better corporate governance can increase investor confidence and thereby act as a signalling device to access external finance (Atanasova et al., 2016;Gompers et al., 2003;Haque, 2015). Therefore, smaller firms can benefit through increased compliance with corporate governance (Klapper & Love, 2004). Total sales Negative Table 1 summarizes the findings of selected studies where firm size has been used as an independent variable together with a dependent variable related to corporate governance compliance. This review of literature first highlights that the association between firm size and corporate governance is quite complex and ambiguous because the empirical evidence is fairly inconsistent. The key reason behind this inconsistency could be the fact that firm size reflects many other firm-specific characteristics. Therefore, the effect of firm size largely depends on the type of proxy used to measure firm size as well as how firm size influences other firmspecific characteristics which in turn is determined by many contextual factors. This inconsistency highlighted in the literature review provides a strong foundation to investigate the effect of firm size on corporate governance compliance in different contexts, particularly in emerging countries.

Research Methods
This study examines the effect of firm size on corporate governance compliance based on the annual report data from 2016 to 2020. The sample consists of 100 firms selected using the systematic random sampling technique out of 283 firms listed in CSE as of 30th April 2021. Nevertheless, a distinction was not made between firms in the financial sector and non-financial sectors. Since some data was missing in some firms for some years, the final data set used for the analysis comprised 488 firm-year observations. Corporate governance compliance was assessed using a Corporate Governance Index (CGI) which was constructed through equally weighting 18 board related recommendations of the Sri Lanka Code of Best Practice on Corporate Governance 2017 as illustrated in Table 2. This index was constructed similar to the literature based on the index approach (Bebchuk et al., 2004;Gompers et al., 2003). A value of one was assigned if a firm has complied with a particular best practice and a value of zero otherwise (Brown & Caylor, 2006;Cremers & Nair, 2005;Gompers et al., 2003). The CGI is the sum of all the scores obtained for each best practice.
Firm size was measured using the natural logarithm of total assets because the majority of the literature reviewed in this study has taken a similar approach. Leverage and firm age were used as the control variables. Here, leverage was measured using the debt-to-equity ratio, and firm age was measured as the number of years since the firms' shares were listed in CSE, similar to Azeez (2015). Based on the Hausman test, a random-effects panel regression model was chosen to assess the association between firm size and corporate governance compliance as specified in equation 1.
In the equation, α and β stand for the intercept and regression coefficients. Moreover, u and ε denote time-invariant error and random error, respectively. Further, i denotes individual firms, and t denotes time. CGI denotes the level of corporate governance compliance, which can vary from zero to 18. FSIZE, LEV, and AGE respectively denote firm size, leverage, and firm age.
Since the data set contained serial correlation and heteroscedasticity, the same random-effects panel regression model was estimated using robust standard errors and bootstrapping. Hence, altogether three models were estimated.
= + 1 + 2 + 3 + + -------(1) Data Analysis Table 1 illustrates the descriptive statistics relating to key variables used in the study. A substantial variation in corporate governance compliance can be observed in the sample. For example, the compliance in some firms is as low as six while some firms have scored 18. This indicates that some firms have complied with all the best practices considered for the study. Moreover, extremely high variation can be observed concerning leverage and firm age. The results of the Hausman test (χ2 = 7.71, p = .052) and Breusch and Pagan Lagrangian multiplier test for random effects (χ̅2 = 458.44, p < .001) indicated that the random-effects model is suitable for the data set over the fixed-effects model and OLS, respectively. Wooldridge test for autocorrelation in panel data (F (1, 99) = 15.541, p < 0.001) suggests the presence of first-order autocorrelation. Therefore, as indicated in the previous section, three random-effects panel-regression models were estimated using STATA 15: Model 1 with default standard errors, Model 2 with robust standard errors, and Model 3 with bootstrapping. The results of these models are shown in table 4.
As illustrated in table 4, all the models were statistically significant. All three models suggest a statistically significant positive association (either at 5% or 1% significance levels) between firm size and the level of corporate governance compliance. This result is consistent with the findings of Chen (2019), Hertig (2005) and Madhani (2016). Moreover, evidence suggests the levered firms tend to comply more with corporate governance best practices even though this coefficient is not significant under Model 3. Nevertheless, any significant association between firm age and compliance level was not detected. Notes: t-statistic is within parentheses; *, **, *** respectively indicate statistical significance at 0.1, 0.05, and 0.01 levels.

Results and Discussion
The positive association between corporate governance compliance and firm size can be attributed to a few factors. First, the cost of implementing corporate governance best practices may not be affordable to smaller firms given their lack of financial and human resources (Haniffa & Cooke, 2002). Firms should have independent directors who possess a balanced set of expertise and skills to exercise unbiased judgements on corporate matters. Hence, corporate governance codes place great emphasis on board independence and board subcommittees in monitoring management. Larger firms are likely to allocate more resources to comply with these practices as they have relatively higher access to resources given their well-established reputation (Hossain et al., 1994). However, hiring a director with adequate competence but remaining independent of controlling shareholders and management can be relatively expensive for smaller firms (Chen, 2019). Especially, outside directors are likely to be reluctant to join as an independent director of a smaller firm with few directors given the higher responsibility than in a larger firm with many outside directors. Hence, the potential for hiring a well-qualified independent director is less for a smaller firm if the candidate is vigilant about the potential responsibilities.
Second, the agency problems of larger firms can be greater compared to smaller firms due to the greater complexity of their capital structures and business operations (Chung & Zhang, 2011;Haniffa & Cooke, 2002). Further, information asymmetry in larger firms is also higher than in smaller firms leading to higher agency costs (Nguyen et al., 2020). Larger firms may have a higher number of minority shareholders, and hence, protecting the rights of these minority shareholders is more complicated than in smaller firms (Chen, 2019). Therefore, larger firms are better motivated to mitigate agency problems even at a higher cost (Laidroo, 2009).
Third, larger firms use corporate governance compliance as a tool to attract investors and to finance their operations through external funding at a lower cost (AlHares, 2020) because investors prefer firms with higher market liquidity (Falxenstein, 1996). Shleifer and Vishny (1997) stated that in the context of concentrated family ownership, block holder expropriation of minority shareholders increases the agency costs of companies. Hence, investors perceive a higher risk in firms with block holders and are less likely to invest because block holders tend to aggravate information asymmetry (AlHares, 2020). As legal protection for investors is low in emerging countries like Sri Lanka, to attract investors at a lower cost, firms have to exercise strong corporate governance practices to show investors that insiders are not exploiting outside funds (Abdullah et al., 2012;Claessens et al., 2002). Gompers et al. (2003) claim that investors require a higher premium for investing in companies with poor governance as they have to tolerate higher risk. In other words, lower compliance with corporate governance best practices indicates a higher potential for opportunistic behaviours by managers and controlling shareholders. It reduces the amounts that investors are willing to invest in firms (Grossman & Hart., 1986). Hence, larger corporations are adhering to corporate governance practices to increase the market liquidity and to reduce the cost of capital (Chung et al., 2010).

Conclusion
Investors pay greater attention to corporate governance compliance by firms, especially in countries with weak legal investor protection, weak institutions, and insufficient transparency standards. Corporate governance practices act as a monitoring mechanism to protect investors, attract investors, and raise funds at a lower cost. Hence, despite the voluntary nature of the corporate governance codes, firms are willing to comply with corporate governance practices to demonstrate that they discharge their accountability towards all the investors and act in a socially responsible manner. However, as firms comply with corporate governance practices at their discretion, vast differences can be seen in the level of corporate governance compliance among firms. For example, as evidenced in this study, larger firms in Sri Lanka seem to comply more with corporate governance best practices to benefit from the reduced cost of capital and agency costs because they can afford compliance with corporate governance best practices better than smaller firms.
This study contributes to the existing literature in two ways. First, this is the initial attempt in Sri Lanka to identify the impact of firm size on the level of corporate governance compliance using firm size as the main independent variable. Second, even though literature suggests that sound corporate governance can be beneficial to smaller firms with high growth opportunities, the empirical evidence generated through this study reveals that the smaller firms are less encouraged to comply with governance best practices probably due to the high cost of compliance. These costs seem to overweigh the benefits in contexts like Sri Lanka. Therefore, this study highlights the necessity of assessing the differences in willingness and capacities to comply and then making appropriate policies flexible enough in the application based on the firm-specific characteristics such as the firm size. This would help achieve a proper balance between costs and benefits of compliance with corporate governance best practices. This study supports the view that there is no one size fits all approach to corporate governance. Nevertheless, an attempt was not made in this study to detect endogeneity issues. Moreover, the corporate governance index used in this study comprises only 18 board related best practices while many of the best practices have been ignored. Future studies on this topic can be extended by addressing this limitation.