Effectiveness of The Board of Directors and Company Performance: Corporate Governance Perspective in Indonesia

Article Information ________________ Article history: Accepted: September 2020 Approved: January 2021 Published: March 2021 ________________


INTRODUCTION
Corporate governance is important in business (Shahid et al., 2020). According to Zabri et al. (2016), the issue of corporate governance in Asia became a concern after the Asian crisis in 1997. This problem is also increasingly interesting for academic research, after the Enron case in 2008. Shahwan (2015) and Jo & Harjoto (2011) argue that corporate governance can affect company performance. The quality of corporate governance is based on the principles of transparency, stakeholder relations, the board of directors, and the company's ownership structure. According to Larcker (2007) in Arayssi & Jizi (2019) good corporate governance shows an indication of their transparency and responsibility. This means that a company that has good management will produce a good performance.
The implementation of corporate governance can be seen from the perspective of agency relations. In an agency relationship, there is a separation of interests between owners and managers (Jensen dan Meckling, 1976). Agency conflicts will occur because company owners and managers have different interests. Agency theory explains that companies with strong governance will do better than companies with weak governance (Bhatt & Bhatt, 2017). Zabri et al. (2016) stated that there are two governance mechanisms, namely internal and external mechanisms. Internal mechanisms include the size of the board and the independent board of commissioners. External mechanisms include market competition and the labor market. According to Tristiarini et al., (2017); Setiawanta & Purwanto (2019); Krishnan et al., (2017); Utama et al., (2017); Nurazi et al., (2020) ;Jaffar & Abdul-Shukor (2016); Kao et al., (2019) corporate governance mechanisms can improve company performance.
The involvement of the board of directors consistently will affect company performance (Bhagat & Bolton, 2019;Goyal et al., 2019). This study uses internal mechanism indicators, namely the effectiveness of the board of directors, which includes board size, board independence, and board activity. Research on corporate governance has been carried out in several countries with different findings. Research by Zabri et al. (2016) found a negative influence between board size on company performance as measured by ROA. Arayssi & Jizi (2019), in their research using board characteristics and ownership structure in controlling and analyzing company performance. The result found is that concentrated family ownership can replace board independence by forming a governance committee. In contrast to research conducted by Narwal & Pathneja (2016), who found no effect between board size and the number of board meetings on banking performance. Similar results were also found by Shahwan (2015), in his research, there is no influence between corporate governance and company performance in Egypt.
Indonesia is one of the countries that supports the implementation of good corporate governance (OECD, 2019 and Siagian et al., 2013). This study will examine board effectiveness and company performance from the perspective of corporate governance practices in Indonesia. Specifically, this research aims to answer the relationship and influence of corporate governance as measured by the number of boards, board independence, and board activity on company performance in Indonesia.

Hypotheses Development
Related to agency theory, a larger board size will have a good effect on firm performance. This is because the larger the board size will have the ability to pressure management to minimize debt costs and improve company performance (Ofoeda, 2017). Zabri et al., (2016) and Orozco et al., (2018) stated that the size of the board of directors is very important in managing a company. The standard size of the board of directors from one country to another will be different. This is due to the different cultures in each of these countries. According to previous researchers, board size will affect the quality of corporate governance, so that it will have an impact on company performance. Leblanc & Gillies (2003) argue that about eight to eleven people on the board of directors are the optimal size. Ofoeda (2017) and Johl et al., (2015), in their research, found evidence that board size has a positive relationship with company performance. More and more boards of directors will have a variety of expertise that will support corporate governance, to improve company performance. This description can support the hypothesis of this study. H1a: Board size has a positive effect on ROA. H1b: Board size has a positive effect on ROE. Jensen & Meckling (1976) explained that in agency theory, the interests of managers will be achieved by ignoring the interests of owners. Agency problems can be reduced by the independence of the board of directors. Corporate governance mechanisms related to board independence are measured by the composition of the independent board of directors. According to Prabowo & Simpson (2011), Board independence is a percentage of the ratio of the number of independent boards of directors to the entire board of directors. This means that the more independent boards of directors, the more independence that can be achieved (Zabri et al., 2016). Based on this description, the following hypothesis can be formed: H2a: Board independence has a positive effect on ROA. H2b: Board independence has a positive effect on ROE.
Good corporate governance mechanisms in the company are shown by the efficiency of the board of directors. According to Ofoeda (2017), the intensity of the meetings can show the effectiveness of their performance in monitoring managers, so it will have an impact on company performance. Board activity can be measured using the frequency of meetings held (Brick & Chidambaran, 2010). Research on the effect of board activity on company performance was conducted by Ntim & Osei (2011) in South Africa and Brick & Chidambaran (2010). Their research found that a board that meets frequently will result in high company performance. H3a: Board activity has a positive effect on ROA H3b: Board activity has a positive effect on ROE.

METHOD
The approach used in this study is a quantitative method, which shows a causal relationship. This study uses three independent variables and two dependent variables. The independent variable used is corporate governance as proxied by board size, board independence, and board activity. The dependent variable used is company performance as measured by Return on Assets (ROA) and Return on Equity (ROE). The conceptual framework in this study is described as follows:

Figure I. Conceptual Framework
The population used is issuers on the Indonesia Stock Exchange in 2017-2018, which includes companies in the manufacturing sector. Through the sample selection with the purposive sampling method, there were as many as 146 companies. This study obtained data sourced from the annual report, which is displayed on the website www.idx.co.id. Measurement of variables in this study is described in the following

ROE Board Activity
The statistical method used is quantitative statistics. This study uses multiple linear regression to test the hypothesis. This research model is shown in the following equation:

Descriptive Statistics
Descriptive test results are shown in the following table: The table shows the mean board size score of 5.89 (approximately. 6 people), with a lower standard deviation of 1.923 (approximately 2 people). A lower standard deviation of the mean indicates that the board size data are closer to the mean. The average value of board independence is 0.4012 with a smaller standard deviation. Looking at this comparison, it can also be concluded that the condition of the board independence data is close to the average value. The mean values of the board activity variables, ROA and ROE are also each greater than the standard deviation. A standard deviation value that is smaller than the mean indicates that the data in the study are close to the average.

Classic Assumptions
The first classical assumption test is the normality test with the following test results: Source:.data processing The table above shows the results of the normality test in equations I and II. The significance value of the two Kolmogorov-Smirnov tests, each is greater than the significant value of 0.05 (5%), so it can be concluded that the data in this study were normally distributed. The next classic assumption test is the heteroscedasticity test. The following is a table of heteroscedasticity test results: Source:.data processing The results of the heteroscedasticity test of the two equations are shown in the table above. Based on the table, it can be seen that the significance value of all variables in equations I and II is above 0.05 (5%). Thus, it can be interpreted that the two regression models in the study did not have heteroscedasticity symptoms. The third classical assumption test performed in this study is the multicollinearity test. The test results in the below table concluded that there was no strong correlation between the independent variables used in this study. This conclusion is based on a tolerance value for each variable more than 0.1 and a VIF value less than 10. The last classic assumption test is the autocorrelation test. The Durbin-Watson test values in the two equations are respectively free of autocorrelation. The test results are shown in the following table:

Coefficient of Determination
The ability of the independent variable in explaining the dependent variable can be tested by using the coefficient of determination test. The coefficient of determination is shown by the Adjusted R Square value as follows: Both tables show that the significance values in the first and second models are 0.000 and 0.019. This value is smaller than the real level of 0.05 (5%). All independent variables in the equation I jointly affect the ROA variable. All the independent variables in the second equation also jointly affect the ROE variable. Testing on equation I can answer the hypothesis H1a, H2a, and H3a. The significance value of the variable board size is 0.009, significant at 5%. This means that board size has an effect on company performance. The board independence variable is also proven to have an effect on corporate performance. However, the variable board activity has no effect on corporate performance, because it has a significant value above 5%.

Results for Hypothesis 1
H1a in this study states that board size has a positive effect on ROA. This study found evidence that board size affects company performance as measured by ROA. Viewed from the point of view of agency theory, the greater the size of the board, the better it will be able to oversee the performance of managers so that they can make decisions that are in line with improving company performance. This study found a negative correlation between board size and ROA. This means that the smaller the board size, the more effective it is in overseeing the use of assets to generate profits. The smaller the board size will have the more appropriate contribution in increasing performance by optimizing the use of assets. The results of this study support the research conducted by Zabri et al., (2016), Nyamongo & Temesgen (2013), and Byard et al., (2006), who found that smaller board size is more effective in influencing ROA. Guest (2009) also found that board size has a negative effect on profitability in the UK. But, the results of this study are different from those found by Ofoeda (2017) and Narwal & Pathneja (2016). Ofoeda (2017) found that board size has a positive effect on ROA, and Narwal & Pathneja (2016) found that board size has no effect on company performance.
H1b in this study states that board size has a positive effect on ROE. This study finds evidence that board size is positively correlated with corporate performance as measured by ROE. The higher the board size, the more optimal it is supervising managers in the efficient use of capital to generate profits. The data in this study show that the average number of boards in each company is six people. Most of the sample firms have board numbers close to average. This study found that an average board size of six people can influence the achievement of company performance as assessed by ROE. Board size can control managers in optimizing the use of capital to generate profits. The results of this study do not support Zabri et al., (2016) and Nyamongo & Temesgen, (2013), who found a negative correlation between board size and corporate performance as measured by ROE. The results of this study support Johl et al., (2015), who found that board size has an effect on company performance. Leblanc & Gillies (2003) state that the more the size of the board, the more optimal, about eight to eleven people on the board of directors are the optimal size.

Results for Hypothesis 2
The agency conflict proposed by Jensen & Meckling (1976) can be reduced by board independence. The higher the independence of the board, the higher the supervised performance of managers, so that the more trying to improve company performance. This description is the basis for hypothesis 2a which states that the independence of the board of directors has a positive effect on ROA. This study found that the greater the percentage of the composition of the independent board of directors will increase the company's ROA. The more independent directors who are on the board, the more managers will be in achieving better company performance. The independence of the board can support managers to optimize the use of assets to generate maximum profits. These findings support research from Ofoeda (2017) which found that board independence has a positive effect on company performance as measured by ROA. However, these results are not following the results found by Zabri Hypothesis testing of H1b, H2b, and H3b can be answered through the output results in the table above. The significance value of the variable board size is significant at the 5% level of significance, while the variables of board independence and board activity are not significant at the 5% level. Thus, it can be concluded that in equation II, the variable board size has an effect on ROE, while the variable board independence and board activity have no effect on ROE. et al., (2016) which states that board independence has no effect on ROA. The results of this study are also not following the results found by Arayssi & Jizi (2019) which states that board independence has a negative effect on ROA.
H2b states that board independence has an effect on company performance as measured by ROE. The greater the independence of the board, the greater the ability to oversee the manager's performance in maximizing profits using the equity held. However, the results of the study show no influence between board independence on ROE. The existence of an independent director on the board does not have a significant effect on the achievement of ROE. These results are in line with research conducted by Zabri et al., (2016) and Adams & Mehran (2012), which found no effect of board independence on company performance. Achievement of company performance as measured by the company's ability to optimize equity in generating profits is not necessarily due to the influence of the number of independent commissioners on the board.

Results for Hypothesis 3
Hypothesis H3a states that board activity affects ROA. More meetings held by the board of directors will contribute to the manager's decision-making to improve company performance. This study found that board activity had no effect on ROA. The number or frequency of meetings does not motivate managers to maximize ROA.
The results of this study also reject H3b which states that board activity has an effect on ROE. This means that the frequency of board meetings cannot encourage managers to maximize profits by using the equity. The results of this study are inconsistent with the results of research conducted by Narwal & Pathneja (2016) and Ofoeda (2017) which found a positive influence between the frequency of board meetings on company profitability in the banking sector.

CONCLUSION AND RECOMMENDATION
This study examines the effectiveness of the board of directors as part of a corporate governance mechanism in influencing company performance. Board effectiveness is measured by board size, board independence, and board activity. Manufacturing companies in the Indonesia Stock Exchange have an average of 6 boards of directors, where this number can influence company performance as measured by ROA and ROE. The second mechanism is board independence. Board independence, indicated by the presence of independent directors on the board, is proven to be able to influence the company's ROA, but it has no effect on ROE. The independence of the board is not able to influence managers' decisions in equity management that can generate profits. The third governance mechanism is board activity. This study found that the frequency of board meetings has no effect on manager performance achievement.
The limitation of this study is that the data used ends in 2018. The concern experienced is that the end of 2019 will be the initial year of the COVID 19 pandemic, where the pandemic has an impact on the financial conditions of all companies in the world. Of course, many factors can be examined again in influencing company performance, one of which is external factors such as an economic recession. Thus the accuracy of the research data will be better by considering other factors. Based on these limitations, further research is expected to be able to explore other factors that are relevant in influencing company performance during and after the COVID 19 pandemic, for example, namely external factors.