Board Characteristic and Financial Restatement

This study explores the role of board characteristics, namely the Board of Directors (BoD) and the Board of Commissioners (BoC), on the possibility of financial restatement. BoD characteristics were analyzed based on BoD size, female, and overconfidence. Meanwhile, BoC characteristics were analyzed based on BoC size and independent BoC. The population is all manufacturing companies on the Indonesia Stock Exchange for the 20172019 period. Determination of the sample using the purposive sampling method. The number of samples used is 32 manufacturing companies or 96 observational data. This study using logistic regression to hypotheses testing. The results showed that two variables, namely female of BoD and independent of BoC, reduced the possibility of financial restatement. However, three other variables, including BoD overconfidence, BoD size, and BoC size, do not affect the possibility of financial restatement.


INTRODUCTION
Financial restatement is a phenomenon that is appealing to investor attention today. One of the companies that make financial statements in Indonesia is PT. Garuda Indonesia for financial statements published in 2018. This case appears when the stock exchange authority found irregularities in the presentation of earnings. There has been a decline in profits when presenting information to the public (Kompas.com, 2019). Another company that also performs financial restatement is PT. Perusahaan Gas Negara due to currency changes in financial statements (Cnbcindonesia.com, 2019). These financial restatements raise significant questions regarding the role of the board in financial reporting.
The phenomenon of financial restatement has been discussed in the accounting literature. Referring to Agency Theory, corporate governance plays a role in creating transparency on financial statement reporting (Jensen & Meckling, 1976). The financial restatement makes the public question the role of BoD and BoC (Hasnan et al., 2017). The presence of female BoD also affects the dynamics of financial reporting. The higher of female members, the lower the likelihood of a financial restatement (Puspitasari & Januarti, 2014). Managerial overconfidence influences a managerial's behavior in presenting financial statements. Besides, the number of BoC members and the independent member also determines financial information quality (Butar, 2018;Yuristisia & Lukviarman, 2008).
Based on this phenomenon, this study has three research motivations. First, the financial restatement is considered company misconduct (Connelly et al., 2016). This practice raises big questions regarding the role and adequacy of the BoD in financial reporting. The companies that commit financial restatement are seen as having low governance credibility (Puspitasari & Januarti, 2014). However, there is still debate regarding the relationship between the BoD characteristic and financial restatement. Therefore, there has been no conclusive conclusion on these empirical findings (Altarawneh et al., 2020). This study explores three characteristics of BoD, namely size, presence of a female, and overconfidence in a financial restatement.
Second, not many scholars have discussed the consequence of BoD overconfidence on the financial restatement. Most of the previous findings discuss the impact of managerial overconfidence on accounting conservatism (Hsu et al., 2017), accounting misstatement (Azhari et al., 2020), earning management (Berry-Stölzle et al., 2018;Sutrisno, 2019), and financial statements comparability (Almaleki et al., 2021). However, board overconfidence has two opposite consequences. On the one hand, overconfidence encourages the managerial to create corporate value (Reyes et al., 2020;Wong et al., 2017;You et al., 2020). On the other hand, overconfidence generates optimism in accounting recording. As a result, the financial statements contain unrealistic information (Habib & Hossain, 2013;Schrand & Zechman, 2012). Initially, BoD overconfidence leads to immaterial misstatements. Furthermore, overconfidence creates larger misstatements, which conduce to financial restatement (Presley & Abbott, 2013). Thus, this phenomenon is essential to be studied in a financial restatement context. Third, the practice of corporate governance in Indonesia adheres to a two-tier board system. The BoD plays a crucial role in the company's operational activities while the BoC is in charge of directing and monitoring management performance (Arifai et al., 2018;Nugraheni & Muhammad, 2019), including in financial reporting. However, there are still limitations of empirical research that discusses the BoC's role in a financial restatement. The majority of previous studies only identified BoD's role in financial restatement (Azhari et al., 2020;Cai & Chang, 2014;Hasnan et al., 2017;Ma & Chang, 2015). Only a few studies examine the BoC characteristics and financial restatement (Butar, 2018;Siregar & Rahayu, 2019;Yuristisia & Lukviarman, 2008). Considering that Indonesia adheres to a two-tier system, this study analyzes the role of the BoD and the BoC in financial reporting activities, particularly in reducing the possibility of financial restatements.
This study identifies the role of board characteristics, namely the Board of Directors (BoD) and the Board of Commissioners (BoC), on the financial restatement. BoD characteristics were analyzed based on BoD size, female BoD, and BoD overconfidence. Meanwhile, BoC characteristics were analyzed based on BoC size and independent BoC. The results showed that two variables, namely female of BoD and independent of BoC, reduced the possibility of financial restatement. However, three other variables, BoD overconfidence, BoD size, and BoC size, do not affect the possibility of financial restatement.
This research has theoretical and practical contributions. Theoretically, this study confirms the role of Agency Theory which emphasizes good corporate governance practices to reduce corporate misconduct, such as financial restatement. Practically, this finding confirms that the company's management must have a competent BoD and BoC structure to maintain the quality of the presentation of financial statements. Also, these findings provide additional insight to financial statement users to always pay attention to corporate governance when assessing the company's financial statements.
The rest of this paper is organized as follows. The second section formulates five research hypotheses. The third, this paper describes the research method, followed by a discussion of the research results in the fourth section. The final section discusses the conclusions, implications, and limitations of the study.

BoD Size and Financial Restatement
The practice of corporate governance in Indonesia adopts a two-tier board system. Several other countries also adopted this system, including China, Taiwan, Japan, Dutch, Denmark, Germany, and France (Arifai et al., 2018). This system positions the BoD as one of the company's organs with a central role (Hidayat & Utama, 2017). BoD acts as an executive responsible for the company's operational activities (Harymawan, 2018;Hasnan et al., 2020). Besides, BoD provides experience, knowledge, and business expertise (Hasnan et al., 2017;Nugraheni & Muhammad, 2019). In Indonesia, BoD is generally led by a president director assisted by independent directors and other staff directors in finance, operations, marketing and sales, human resources, and general affairs (Joni et al., 2020). Nevertheless, the composition and number of BoD in each company are not the same. The company size or other characteristics determine this condition.
In the practice of corporate governance, BoD has the authority to determine policies and manage company resources. The BoD's decision is considered to represent the corporate's policies, particularly in providing financial information. The BoD is responsible for the high-quality financial reports to stakeholders (Brochet & Srinivasan, 2014;Omer et al., 2019). BoD size is considered to determine the quality of financial information submitted to the public (Altarawneh et al., 2020). The larger size is considered to have higher reporting quality (Hasnan et al., 2020). The larger the number of BoD, the more influential the financial reporting process. However, previous studies have shown conflicting results (Hasnan et al., 2020). Several researchers found that the BoD size had a negative effect on financial restatement (Hasnan et al., 2017(Hasnan et al., , 2020Puspitasari & Januarti, 2014). Thus, this study formulated the following hypothesis: H1: BoD size reduces the possibility of financial restatement.

Female BoD and Financial Restatement
The gender heterogeneity of BoD members also affects the effectiveness of corporate governance (Abbott et al., 2012). Heterogeneity has several advantages, one of which is avoiding groupthink. BoD members that only consist of men (homogeneous) tend to be reluctant to express opinions or creative ideas and produce inaccurate decisions. Generally, homogeneous members tend to avoid disputes (Umans et al., 2008). Contrary, women are more independent, flexible, broad-minded, and cooperative. Female members also have independent thinking and provide different views (Putri et al., 2021). The existence of womanly members increases the quality of decision-making (Abbott et al., 2012). Therefore, regulators expressed ideas to increase gender diversity in overcoming financial reporting failures (Wahid, 2019). The increasing number of women directors afford to improve the quality of financial reporting (Mordi & Obanya, 2014).
In the financial restatement context, gender heterogeneity stimulates discussion activities resulting in high-quality decision-making (González-Moreno et al., 2018). Female members have a more proactive and conservative attitude towards their duties (Abbott et al., 2012). Women have thoroughness and discipline, and they can maintain the work quality. A woman also tends to avoid risk, trying to minimize mistakes (Altarawneh et al., 2020). Therefore, at least one female director directs their group on process and control-oriented. In this situation, gender diversity increases vigilance in presenting financial statements. Thus, the woman director effectively reduces the financial restatement (Abbott et al., 2012;Puspitasari & Januarti, 2014;Wahid, 2019). Thus, this study formulated the following hypothesis: H2: Female BoD reduces the possibility of financial restatement.

BoD Overconfidence and Financial Restatement
Since the 1960s, the term overconfidence is generally used in the discipline of psychology. In the early 1990s, overconfidence has been widely used to describe irrational behavior in an economic context. Generally, overconfidence is a behavioral bias on positive beliefs that are unrealistic or exaggerating (Skala, 2008). BoDs that have an overconfidence attitude tend to make inaccurate estimates or estimates. This action creates misreporting of accounting records (Azhari et al., 2020). BoD overconfidence increases the potential for overstatement, and it is not consistent with the conservatism principle (Ramsheh & Molanzari, 2014). The financial statements are optimistic and aggressive. As a result, BoD tends to invest actively, but the outcomes are challenging to realize (Presley & Abbott, 2013).

BoC Size and Financial Restatement
The BoC is the principal's representative who has the authority to monitor management performance. Considering that BoC is the primary internal control board, the BoC size is crucial in decision-making. Therefore, BoC, which has a larger size, is assumed to perform inspections and monitor effectively (Hasnan et al., 2020). Besides, a more significant number of BoC members can reduce the dominance of the BoC over company resources (Al Azeez et al., 2019).
In financial reporting, the larger BoC consists of more business knowledge, experience, and expertise. The number of members makes it easier for the board to oversee financial reporting by company management (Siregar & Rahayu, 2019). The larger the number of BoC members, the more influential the monitoring process on BoD performance will be. Thus, this study formulated the following hypothesis: H4: BoC size reduces the possibility of financial restatement.

Independent BoC and Financial Restatement
Independent BoC is a member that is not affiliated with the BoD, other BoC members, or the controlling shareholder. This membership is also independent of any business relationship that affects the board's ability to be independent. While the independent member is essential, the BoC's independent activities must align with the corporate's interests. The presence of independent members increases the effectiveness of supervision and reduces conflicts of interest between principals and agents (Al Azeez et al., 2019). In the corporate governance function, independent commissioners are responsible for controlling the managerial policies. Thus, independent members protect minority shareholders in strategic decision-making (Nugroho & Eko, 2012).
Concerning the financial reporting function, independent commissioners are considered capable of improving the quality of financial reports. Several findings reveal that independent members affect earnings informativeness (Holtz & Sarlo Neto, 2014) and control earnings management (Talbi et al., 2015). The presence of independent members also reduces the potential for financial restatement (Hasnan, 2020). Furthermore, independent board members play an influential role in the supervisory function to increase company transparency and accountability. The more significant the independent commissioner's proportion, the more effective supervision quality to minimize fraud (Butar, 2018;Siregar & Rahayu, 2019). Thus, this study formulated the following hypothesis: H5: Independent BoC reduces the possibility of financial restatement.

METHOD, DATA, AND ANALYSIS
The research population is all manufacturing companies on the Indonesia Stock Exchange for three-period, including 2017-2019. The sample was selected using a purposive sampling method. They are three criteria: 1) publishing financial statements consecutively during the study period; 2) the company has at least one female director; and 3) having the data needed for this study. Based on these criteria, 32 companies become the research sample. Therefore, this study uses 96 data for three years of observation. This study uses financial restatement as the dependent variable. The independent variable consists of five variables: BoD size, female BoD, BoD overconfidence, BoC size, and independent BoC. Information related to variable measurement is presented in Table 1. The measurement of the variable uses a dummy variable. Namely, the number one (1) indicates the company is conducting financial restatement while the number zero (0) indicates that the company does not perform financial restatement (Agrawal & Chadha, 2005;Presley & Abbott, 2013). Board of Director Size (BoD_Size) The measurement uses the number of directors in a company (Hasnan et al., 2017(Hasnan et al., , 2020.

Variable
Measurement Female Board of Director (Fem_BoD) The measurement using the percentage of female directors in the BoD (Lanis et al., 2017) BoD Overconfidence (BoD_Ovc) Measurements used the following method (Kouaib & Jarboui, 2016;Sutrisno & Karmudiandri, 2020), namely: 1. Measurement of overinvestment to find out the company's investment policy, using the formula: ∆asseti,t /Asseti,t-1 = α0 + β1∆Salesi,t/Salesi,t-1 + ε……… (1) The researcher conducted a regression test to obtain the residual value. Next, the residual value is reduced by the median value. If the residual value is higher than the median value, it is given a value of 1. Conversely, if the residual value is lower than the median value, it is given a value of 0. 2. Measurement of the Debt to Equity Ratio to find out the company's financing policy, using the formula: Total Debt ……………………………………………….. (2) Total Equity If the Debt to Equity Ratio is higher than the median value, it is given a 1. Conversely, if the Debt to Equity Ratio is lower than the median value, it is given a value of 0. 3. Dividend yield ratio to measure the company's dividend policy. Dividend policy depends on the company's investment policy. If the company wants to invest in the future, the company tends not to distribute dividends, and it will be used as cash reserves (Deshmukh et al., 2013;Sutrisno & Karmudiandri, 2020). If the company does not pay dividends, it is given a value of 1. Conversely, if it makes dividend payments, it is given a value of 0. These three components determine the determination of BoD overconfidence. If two of the three components have a value of 1, the board is considered overconfident.
If not, it is assigned a value of 0. Board of Commissioner Size (BoC_Size) Measurement of variables using the number of commissioners, both internal and external (Butar, 2018;Siregar & Rahayu, 2019). Independent Board of Commissioner (Ind_BoC) Measurement of variables using the percentage of independent commissioners in the BoC (Puspitasari & Januarti, 2014

RESULTS
This study uses 96 observational data for three years of research. Table 2 presents the minimum, maximum, mean, and standard deviation values. The descriptive statistical calculations in Table 2 present the average number of BoD membership is 4-5 persons. Based on the heterogeneity level, the presence of female BoD is 30.43%. This value is adequate to represent gender diversity in top management positions. Research data shows the average level of BoD overconfidence is 0.49. Based on BoC characteristics, the sample companies have an average of 3-4 BoC members. This figure is lower than the number of BoD members. Meanwhile, the average number of independent commissioners is 41.90%. This amount has complied with the regulatory requirements that require each company to have at least 30 percent of the total BoC members.
This study using logistic regression to hypothesis testing. The feasibility test of the model uses Hosmer and Lemeshow's Goodness of Fit Test, presented in Table 3.  Table 3 presents the Chi-Square value of 2.721 and a significance value of 0.951 > 0.05. Thus, this model is accepted because does not significant difference between model and observed values. Next, the study assesses the overall regression model using numbers in -2 log-likelihood, -2LL block number = 0, and -2LL block number = 1. The test results are shown in Table 4.  Table 4 shows the initial number -2 log-likelihood block number = 0 is 64.255. After the test includes five independent variables, the value of -2 log-likelihood block number = 1 decreased to 18.877. These result confirmations the regression model fits the data. The next test calculates the coefficient of determination using Nagelkerke R Square, shown in Table 5.  Table 5 shows the Nagelkerke R Square value of 0.771. The figure indicates that 77.10% of the variation in financial restatement variables can be explained by variations in independent variables, namely BoD size, female BoD, BoD overconfidence, BoC size, and independent BoC. Other factors outside the research model explain the remaining 22.9%. The next test examines the classification matrix to determine the predictive power of the regression model, presented in Table 6.  (2021) The results of the classification table in Table 6 show the predictive power or accuracy of the model in classifying observations of 95.8%. This study also conducted a multicollinearity test to determine the magnitude of the correlation between independent variables, which is presented in Table 7.  Table 7 presents the value of the correlation coefficient between the independent variables less than 0.8. Thus, there is no multicollinearity between the independent variables. Therefore, the next step is to test the research hypothesis, which is presented in Table 8. Source: data processed (2021) The hypothesis test results in Table 8 show that the BoD size variable has a significance value of 0.058. These results indicate that the BoD size does not affect financial restatement. Therefore, H1 is rejected. The test results for the female BoD variable have a negative regression coefficient and a significance value of 0.016. This figure shows that the female directors are effectively reducing financial restatement. Thus, H2 is accepted. The BoD overconfidence has a significance value of 0.443. It can conclude that the overconfidence variable has no effect on financial restatement, or H3 is rejected. Consistent with the test results on the BoD size, the BoC size test result has a significance value greater than 0.05, which is 0.848. Therefore, the BoC size does not affect the financial restatement, or H4 is rejected. Finally, the results for independent BoC have a negative regression coefficient and a significance value of 0.046. It means that independent commissioners reduce the possibility of financial restatement. Thus, H5 is accepted.

BoD Size and Financial Restatement
Testing the first hypothesis reveals that the BoD size does not affect the financial restatement. This finding may be due to the lack of BoD effectiveness in providing financial information (Hasnan et al., 2020). The more significant number of BoDs may make it difficult for them to make decisions. As a result, BoD members cannot provide additional business knowledge and experience to improve the high quality of financial statements (Hasnan et al., 2017). Besides, there may be inhibiting factors that reduce board performance, such as corporate culture and industry structure (Habib & Hossain, 2013). This condition causes the BoD size to be unable to reduce financial restatement. The results are consistent with the previous finding that BoD size does not affect financial restatement (Cai & Chang, 2014;Chandra, 2020;Hasnan et al., 2020;Ma & Chang, 2015;Soroushyar & Ahmadi, 2016). Meanwhile, this finding does not support other findings that the BoD size potentially reduces financial restatement (Hasnan et al., 2017;Puspitasari & Januarti, 2014).

Female BoD and Financial Restatement
Based on the hypothesis 2 test result, the presence of female board members is effective in reducing financial restatement. These results indicate the importance of heterogeneity in the company's top management positions. These results support the regulator's statement that gender diversity in board directors reduces financial violations that lead to financial reporting failures (Wahid, 2019). Financial reporting is an activity that requires high cognitive independence. The presence of women on the BoD triggers the board's cognitive conflict to improve the quality of financial reports. Also, female board members generally have higher monitoring rates than male members to effectively succeed in their function (Habib & Hossain, 2013). Thus, female directors reduce the potential for financial restatement. These results support the previous findings (Abbott et al., 2012;Puspitasari & Januarti, 2014;Wahid, 2019), revealing a negative association between female directors and financial restatement. Nevertheless, the finding does not support Hasnan et al.'s (2017) finding of no relationship between female directors and financial restatement.

BoD Overconfidence and Financial Restatement
The third hypothesis test result reveals that the BoD overconfidence does not affect financial restatement. Based on descriptive statistical analysis, the average level of board overconfidence is less than fifty percent. This figure indicates that the majority of BoD s apply the precautionary principle in accounting records. In this case, the BoD implements the practice of conservatism to make realistic estimates or estimates. Therefore, the accounting records do not contain material misstatements, and that the company does not need to do financial restatements. This finding supports Shekarkhah et al. (2019), which failed to prove the association between executive overconfidence and financial restatement. Nevertheless, the finding does not support previous findings that have succeeded in proving the role of CEO overconfidence in increasing financial restatement (Lee, 2016;Lestari & Faisal, 2019;Presley & Abbott, 2013;Sutrisno & Karmudiandri, 2020).

BoC Size and Financial Restatement
The statistical results of the fourth hypothesis fail to prove the role of the BoC size on the financial restatement. In the corporate governance context, larger BoCs have better knowledge, experience, and ability to manage a business. The more prominent members make it easier to oversee management. The greater the number of BoC members, the more influential the monitoring process is on BoD performance (Siregar & Rahayu, 2019). Nevertheless, this result failed to verify the assumption. The ineffectiveness of the BoC's role may be due to difficulties in uniting different interests so that it is vulnerable to causing internal conflicts. This condition causes the BoC size to fail to improve the financial statement quality, particularly related to a financial restatement. The finding supports previous results (Siregar & Rahayu, 2019) that were not a relationship between BoC size and financial restatement. However, this finding contradicts Yuristisia and Lukviarman's (2008) research that proves a positive relationship between board size and accounting restatement.

Independent BoC and Financial Restatement
The fifth hypothesis successfully proves that independent commissioners are proven to reduce the possibility of financial restatement. This finding indicates that the independent commissioners have performed the management oversight function well. Independent commissioners can work neutrally without pressure or intervention from management (Siregar & Rahayu, 2019). They also have strong incentives to perform their monitoring function effectively to maintain their reputation as independent parties. Also, they will try to transfer their skills and experience from their previous work into the company (Butar, 2018). This finding also indicates that substantially the company has complied with the provisions of the regulator to streamline the role of independent commissioners in corporate governance (Hasnan et al., 2020). Thus, independent commissioners reduce the possibility of companies doing financial restatements. This finding does not support previous empirical evidence that shows a positive association between board independence and accounting restatement (Yuristisia & Lukviarman, 2008). Also, this result does not support previous findings (Butar, 2018;Siregar & Rahayu, 2019) that independent commissioners do not affect the financial restatement.

CONCLUSION, LIMITATIONS, AND SUGGESTIONS
This study explores the role of board characteristics on the financial restatement. BoD characteristics were analyzed based on BoD size and female BoD. Meanwhile, BoC characteristics were analyzed based on BoC size and independent BoC. Today, there is still a debate regarding the result finding of the BoD characteristic and financial restatement. Also, not many scholars have discussed the consequence of board overconfidence on the financial restatement. Therefore, this study examines the role of the BoD and BoC in reducing the likelihood of financial restatements. The statistical results reveal that female directors and independent commissioners reduce the probability of financial restatement. Nevertheless, this study failed to prove the role of three other variables, namely BoD size, BoC size, and BoD overconfidence, on the financial restatement.
The results provide theoretical and practical implications. Theoretically, the results imply the importance of corporate governance practices in creating quality financial reports. However, not all corporate governance indicators can play an influential role in reducing financial restatement. Therefore, further research is needed to identify other characteristics in the context of financial reporting. Based on practical and social perspectives, the results imply that financial restatement indicates the failure of the BoD and BoD in creating good governance practices. Therefore, companies must maximize the functions of both boards through gender diversity and the presence of independent external parties. The existence of women and independent members provides tangible benefits in improving the quality of financial reporting.
Based on statistical results, this study's limitation is the small number of samples because many companies do not have female directors. These results provide an opportunity for further researchers to explore other companies' sectors to increase the number of samples. Another limitation, this study does not support the literature on overconfidence and its impact on the financial restatement. Given that the concept of overconfidence is multi-dimensional (Shekarkhah et al., 2019), further research needs to consider other indicators. Thus, overconfidence assessment can measure correctly.