Chief Executive Officer Characteristics and Financial Restatements in Malaysia

The purpose of this paper is to investigate whether the Chief Executive Officer (CEO) characteristics affect the occurrence of financial restatements in Malaysian firms. The CEO characteristics used in this study were tenure, honorific title, gender, expertise, and age. In addition, the financial restatement has been measured as a dummy variable as to whether companies restate their financial statements or not. The sample of this study comprised 442 companies listed in the main market of Bursa Malaysia during the period 2012–2016. The panel data method was utilised to analyse the data. This study employed a logistic regression analysis. The results of this study revealed that there is a positive and significant relationship between CEO tenure and CEO gender with financial restatements. In addition, this study found a negative and significant relationship between CEO honorific title and financial restatements. However, the results found insignificant relationships between CEO expertise and age with financial restatements. This study highlighted the importance of considering CEO characteristics as one of the influential determinants of financial restatements in Malaysian companies.


Introduction
The events of high profile financial related cases, which have caused investors to incur losses, have increased the attention of many to the subject of financial restatements. Amongst the notable cases was that of Enron which had previously declared a US$618 million loss in its 2001 third quarterly report, before admitting to restating its earnings for a number of years not long after that (Sridharan, Caines, McMillan, & Summers, 2002). Such a confession caused Enron to suffer a sharp plummet in its shares from only a few cents, from its highest value of 90 USD per share (app. 60 USD billion in total), which caused its shareholders to suffer a paper loss of 90 USD billion.
Relevantly, in the American context, the Government Accountability Office (GAO) (2004) reported the loss of 100 USD billion during the period of January 1997 to June 2002 by accounting restatements. Then, an additional 36 USD billion in the market capitalisation loss was reported for the July 2002-September 2005 period, during the days around the early announcement of the restatement (GAO, 2006;Abdullah, Mohamad Yusof, & Mohamad Nor, 2010). In addition, Hashim (2009) mentioned that in the capital market of East Asian countries, including Malaysia, investors" confidence had deteriorated due to the financial reporting quality problem which required an in-depth examination. Further, high profile cases involving the well-known Malaysian firms, such as Transmile Berhad, NWP Berhad, and Bina Darul Aman Berhad, denoted an urgency for more studies in the Malaysian context. The regulators appear to have blamed the management as the preparers of the financial statements for producing low quality financial reporting. Specifically, the regulators have been concerned with the effectiveness of CEOs in Malaysia due to the rising corporate scandals that have continued to shock the business community. The reason is because, the CEO acts as an executive manager who will manage, lead, and verify the financial results (Vafeas, 2009). The CEOs often determine the kind of information that needs to be revealed because they have authority over several decisions (Cheng & Lo, 2006).
The case of Transmile Berhad is an example of CEOs" wrongdoings in managing their reported income. Gan Boon Aun, the CEO of the company was charged with intentionally authorising the manipulation and misleading financial statements to Bursa Malaysia which caused financial restatements (18 Mar., 2017, The Star Online). Moreover, a few firms listed on Bursa Malaysia, such as Bina Darul Aman and NWP Holdings, had restated their financial statements For the remaining portion of this study, the following section discusses the previous literature and the development of the study hypotheses. Section 3 presents the methodology of the study. In section 4 the results and the discussion are provided. The last part of this study highlights the conclusion derived from the obtained results.

Literature Review and Hypothesis Development
The financial restatement involves a process of revising and amending the financial statement, which was issued beforehand for the purpose of rectifying errors and non-compliance with generally accepted accounting principles (Abdullah et al., 2010). Accordingly, the General Accountability Office (GAO) (2002) in the United States explained that "A financial restatement occurs when a company, either voluntarily or prompted by auditors or regulators, revise public financial information that was previously reported".
As reported by Abdullah et al. (2010) there are three main causes of accounting errors: problems in applying the accounting rules, human and system errors, and fraudulent behaviors. It is argued that the restatement reasons are due to revenue recognition, reserves, equity accounting, accruals, and contingencies. In addition, Kinney and McDaniel (1989) show firms that previously corrected their quarterly reported earnings are found to be, smaller, with high debt levels, less profitable, slower growth, and facing more serious uncertainties. This is similar to the results reported by Ismail and Abdullah (2009) who found that firms that defer the recognition of the exceptional items, a tool used to manage quarterly earnings to the fourth quarter tend to have less profitable and to be smaller. Additionally, Efendi, Srivastava and Swanson (2007) and Persons, (2005) argued that firms with CEO who also serves as board chairman are more likely to have misstatements.
Previous studies revealed that there is a large negative market reaction following the earnings restatements announcement such as (Richardson, Tuna, & Wu, 2002;Akhigbe, Kudla, & Madura, 2005). Restatements have caused concern regarding financial reporting quality (Levitt, 1998;Palmrose, & Scholz, 2000). However, in the face of the negative publicity from misstatement, it is noted that restating companies do not seem to adopt a more conservative financial reporting strategy following restatement (Moore, & Pfeiffer, 2004). Likewise, a deterioration in the confidence of investor regarding financial reporting following restatements is only short-term; and that suspicion regarding the information loss of post-restatement earnings in the long-term is unfounded (Wilson, 2008). Furthermore, as reported by some prior studies, managers resort to perform financial restatements because of several factors including usage of International Financial Reporting Standards (Tendeloo & Vanstraelen, 2005), CEO duality (one person who holds the CEO position and other leadership position simultaneously) (Davidson, Jiraporn, Kim, & Nemac, 2004), the appointment of senior executives from the hired external accounting firms (Geiger, North, & O"Connell, 2005), the quality of the accounting figures (Davidson et al., 2004), tax incentives (Dhaliwal, Gleason, & Mills, 2004), and the spread of the company"s ownership (Hsu & Koh, 2005).
The act of financial restatements presents extreme dangers to a firm, especially when such practices are discovered, which would cause the trust and confidence of the investors in the financial reporting quality of the firm to disintegrate (Jackson & Pitman, 2001). The authors further added that in the financial market, financial restatements could obstruct the functional flow of capital. Therefore, due to the financial restatement practices which cause the figures to not signify the real performance of the firm, the reliability of the reported earnings are tarnished. Moreover, because of the various failures of the firms around the world due to financial restatements, more studies of the financial restatement issue should be carried out (Arya, Glover, & Sunder, 2003;Shafie & Zainal, 2016).
Several subsequent studies have examined the role of CEO and restatements such as (Efendi et al., 2007;Persons, 2005: Tendeloo & Vanstraelen, 2005Abbott, Parker andPresley, 2012: Huang, Rose-Green andLee, 2012;Bhandari, Mammadov, Shelton, & Thevenot, 2018). However, the results are mixed and extant evidence linking the CEO characteristics and financial restatements is not conclusive. It is also argued that the boards of directors, especially the CEOs, are responsible for delivering a high quality of financial information to the shareholders who is in their best interest (Brochet & Srinivasan, 2014;Omer, Shelley, & Tice, 2014). In fact, CEO characteristics are considered as an important component for controlling the decisions of the management (Goodman, Neamtiu, Shroff, & White, 2013) including the quality of the financial reporting of the firms. In summary, literature has mostly focused on the role of corporate governance and restatements such as (Klein, 2002;Aier, Comprix, Gunlock, 2005;Srinivasan, 2005;Marciukaityte, Szewczyk, & Varma, 2009: Abdullah et al. 2010. However, there is a lack of debate on the effect of CEO characteristics and financial restatements to date. Hence, the present study seeks to examine the CEO characteristics and financial restatements in Malaysian contexts.

CEO Tenure
As projected in Hambrick and Mason (1984), the tenure of the CEO could have an important effect on the types of decisions made by the CEO or the top management team. Previous studies have provided evidence for the firms to possess a top management team that is long-tenured because it is allied with team familiarity and the consistency of the firm"s internal business (Michel & Hambrick, 1992;Smith et al., 1994). Relevantly, the concept of CEO tenure is amongst the most studied CEO characteristics in the upper echelons theory research (Finkelstein, Hambrick, & Cannella, 2009;Besar, Ali, & Ghani, 2017). Additionally, the literature might propose a reason beyond the concern of long-tenured CEOs for status to provide high quality financial reporting. This implies the impact of the CEO"s knowledge on the process of financial reporting, hence it increases the quality of financial reporting (Feng, Ge, Luo, & Shevlin, 2011;Vafeas, 2009).
Prior research stated that the short-term earnings targets are considered by the CEOs as a critical factor to maintain a positive image for themselves and their firms (Graham, Harvey & Rajgopal 2005). Additionally, Davidson, Jiraporn, Kim and Nemec (2004) indicated that newer CEOs "particularly those who also hold the role of board chair" are more expected to engage in manipulation practices to show a positive impressions during their tenure.
In contrast, the CEO accumulates more power and lower level of monitoring on his/her actions due to a long tenure period. In such a situation, CEOs would be free of the scrutiny that would prevent the engagement in deceptive behaviours (Ndofor, Wesley, & Priem, 2015). Besides that, Davidson, Xie, Xu and Ning (2007) stated that a CEO with a long tenure may face the "horizon problem" where the CEO takes actions to enhance the short-term performance rather than shifting towards long-term investment. Hence, the research hypothesis is as follows: H1: There is a significant relationship between CEO tenure and financial restatements.

CEO Title
A CEO honorific title is considered as a signal of a CEO"s political connections, and amongst the Malaysian titles includes Datuk, Datuk Sri, Tan Sri or Tun (Amran, Ishak, & Abdul Manaf, 2016). The awarded title also denotes the CEO"s existing or previous position as a government public servant; for example, the CEO could be an existing or past officer of the local or central governments or the military. It is equally a measure for government interference in the firm. Political connections are an extensive occurrence in emerging and transitioning nations, and their impacts are increasingly an interesting subject of scrutiny amongst scholars. As discovered in Faccio (2006), the political connections" benefits appear to be more significant amongst nations with highly interventionist governments and weaker property rights protection.
Chaney, Faccio and Parsley (2011) asserted that firms with political associations are less expected to exhibit reliable financial disclosures, as opposed to those which are non-connected, owing to the protection that these firms are attaining following the establishment of the aforesaid connections. However, prior empirical research by Hussin, Hasnan and Sanusi (2014) concluded that there was a negative significant linkage between politically associated firms and the probability of accounting misstatements. Moreover, CEOs who have the honor name are more trustworthy and they enjoy credibility and transparency, and usually do not resort to manipulations that cause the retrieval of financial statements. Hence, based on the arguments and prior studies" results, the following research hypothesis is proposed:

CEO Gender
Previous studies have shown that managerial characteristics are recognised as an important factor in financial restatements as well as to the quality of financial reporting. Recently, corporate finance literature posited that the financial reporting quality may be affected by the gender of the firm"s executives (Peni & Vä hä maa, 2010). Krishnan & Parsons (2008) found that the quality of reported earnings increases when there is gender diversity in the senior management. Similarly, Shawver, Bancroft, and Sennetti (2006) posited that females were less likely to be involved in earnings management compared to males. The argument behind this is that women are more likely to exhibit ethical behaviour even when they can personally benefit from an unethical behaviour (Betz & Skepard, 1989;Bernardi & Arnold, 1997). Krishnan and Park (2005) argued that women have more effective skills in leadership. The distinctive features of women require influencing the strategic direction of a firm positively and contributing to the monitoring of the financial reporting process. Consequently, their presence is positively associated with the quality of financial reporting (Obanya & Mordi, 2014).
Prior studies have argued that females are less likely to be involved in unethical behaviour to obtain financial rewards as they are more ethical in the workplace. This is because females tend to take fewer risks, as they are given less room for error (Betz & Skepard, 1989;Khazanchi, 1995). Furthermore, Abbott et al. (2012) found a negatively significant relationship between female and the probability of financial restatements.
On the other hand, Debnath, Patnaik and Satpathy (2019) found that the proportion of females is associated positively with earnings management. Hence, firms are more likely to be involved in earnings management if the board has more representation from females which in turn reduces the quality of financial reporting. A similar result was found by Byrnes, Miller and Schafer (1999); Arun, Almahrog and Aribi (2015). Based on this view, the following hypothesis is formulated: H3: There is a significant relationship between CEO gender (female) and financial restatements.

CEO Expertise
The CEO expertise and financial restatement relation has been argued based on the upper echelon theory. The theory predicts that a CEO with experience will be able to develop higher quality financial reports (Matsunaga & Yeung, 2008). The theory suggests that the personalities of managers, such as expertise, can affect their interpretations of the problems and situations they have to deal with (Hambrick & Mason, 1984;Hambrick, 2007). Furthermore, when making accounting decisions, managers with expertise will rely on their previous working experience (Pham, 2016). Hence, the expectation of this study has been that financially expert CEOs will have more ability to monitor the accounting process and provide high quality financial statements. In addition, this study has argued that the CEOs who have the financial expertise in their past careers will have a deeper understanding of financial and accounting concepts and structures, and will have more technical training which, in turn, will help the CEOs to provide a higher quality of information and decrease the likelihood of financial restatements. Based on the prior research arguments, the research hypothesis is developed as follows: H4: There is a significant negative relationship between CEO expertise and financial restatements.

CEO Age
The upper echelons theory suggests that older CEOs exhibit a greater standard of ethical beliefs, and they are more likely to draw inferences more consistently than the younger ones, with the trait-diagnostic inferences of ethical behaviour. Extensive research, for example, the one by Huang et al. (2012), has concluded that there is a negative significant linkage between the age of the CEO with both the inclination towards practicing financial restatements and the probability of beating analyst"s forecasts of earnings. Likewise, Huang et al. (2012) provided empirical evidence supporting the connection existing between CEO age and the quality of financial reporting.
Such a result was obtained by Peterson, Rhoads and Vaught (2001) and Sundaram and Yermack (2007) who showed that individuals become more conservative and ethical as they become older. Accordingly, the present paper has expected to observe a negative significant linkage between the age of the CEO and financial restatements. Hence, the study hypothesised that:

Sample Duration, Selection, and Data Collection
The sample employed in this study contained all of the 800 listed firms in the main market of Bursa Malaysia from the years 2012 to 2016. In accordance with previous studies, the sample excluded financial firms because of the unique characteristics and the specific regulations and rules imposed on them by the Malaysian Central Bank. Data on the CEO characteristics were hand-collected from the annual reports that were available on Bursa Malaysia"s website. In terms of financial restatements, the study obtained the data from Datastream, whilst the types of restatement classifications were derived from the annual reports of these listed firms. The final sample also excluded the firms where the financial data were not fully available, they had changed their CEO, changed fiscal year end and those delisted from Bursa Malaysia. The final sample of this study was 442 firms.
Based on the annual reports" review, the financial restatements in this study were classified into several categories as follows: misrepresentation, accounting rule application failures, and irregularities (Paterson & Valencia, 2011). A similar procedure was followed by Abdul- Wahab, Gist and Majid (2014). The sample firms which had restated their financial outcomes in the annual reports were identified using the following keywords: "restate", "restated", "restatement", and "prior year adjustment".

Data Description
The total sample of firm observations during the period (2012-2016)  However, as shown in Table 1, the non-restatement distribution of the sample period was 1760 (82.67% of total observations) and the yearly distribution for the non-restatements was as follows, 349 (  Where, RESTATE = Financial restatement is a dummy variable implied "1" if the sample firm is a restatement firm and "0" otherwise. CEOTENURE = The number of years that a CEO continuously holds this position in a firm. CEOTITLE = A dummy variable coded "1" if the CEO titled as Datuk, Datuk Sri, Tan Sri or Tun and "0" otherwise. CEOGENDER = A dummy variable equal to "1" when the CEO is a female and "0" otherwise. CEOEXPERT = A dummy variable coded "1" if the CEO has a professional qualification or qualified as accounting or finance expert, "0" otherwise.

Descriptive Statistics
The descriptive statistics in Table 2 depicts the mean, standard deviation, minimum, and maximum values. Table 2 also represents all the study variables including the financial restatement as the dependent variable, and CEO tenure, CEO title, CEO gender, CEO expertise, and CEO age as the independent variables. In addition, this study used firm size, firm age, leverage, Big 4, ROA, sales growth, board meeting, subsidiaries, MTB, and loss as the control variables. Table 2 shows that the total number of observations is 2129. In addition, Table 2 presents the mean value of RESTATE as 0.1723, with the minimum value of 0 and the maximum value of 1. In addition, the mean value of CEO tenure is 12.230 with the minimum of 1 and maximum of 45. Moreover, for CEO title, the mean value is 0.4095 with the minimum and maximum of 0 and 1, respectively. As for the average value for CEO gender presented in this study is 0.0375 and minimum of 0 and maximum of 1. Furthermore, the mean value of CEO expertise stated in this study 0.1845, with the minimum and maximum of 0 and 1, respectively. Moreover, the mean of the log CEO age stated in this study is 1.7382 and the minimum and maximum of 1.3617 and 1.9294, respectively.
With regards to the control variables, Table 2 reveals that the value of the mean for firm size is 5.6725, ranging from 4.3452 to 8.1234. The mean value of firm age stated in the study is 1.4491, with the minimum and maximum 0.3010 and 2.2671, respectively. The mean value of the leverage ratio is 0.1903, with a minimum (maximum) value of 0 (0.7804) years. Furthermore, the average of Big 4 firms is 0.4621, with a minimum of 0 and maximum of 1. Moreover, the mean proportion of the return on assets is 0.0511, ranging from -1.165 to 6.3378. Furthermore, the average of sales growth is 0.1141, with the minimum and maximum of -0.9954 and 30.7648, respectively. With respect to the frequency of board meeting, the descriptive statistics indicated that the average number of board meeting is 5.2404, with a minimum of 2 and a maximum of 21. In addition, Table 2 shows that the mean value of subsidiaries is 1.1019, ranging from 0 to 2.6739. With regards to the market to book value, the mean value is 1.1817, with a minimum of -4.33 and maximum of 28.47. Finally, the average (mean) value of loss is 0.1808, assuming that 18.08 % of firm-years observations were facing losses, and the minimum and maximum values of 0 and 1, respectively.

Pearson Correlation Matrix
The Pearson Correlation Matrix analysis was conducted to elucidate and assess the relationship strengths amongst the research constructs. Hair et al. (2010) accordingly recommended that the correlation value of 0 shows no relationship, whereas the correlation of ± 1.0 proves a perfect relationship. Alternatively, the correlation could be inferred with values within 0 and 1.0 whereby the correlation (r) of between ± 0.1 and ± 0.29 indicates a weak relationship, whilst that between ± 0.30 and ± 0.49 shows an average relationship, and that which is more than ± 0.50 demonstrates a solid relationship. Further, Table 3 displays the Pearson Correlation Matrix results between the independent and control variables of the sample. Accordingly, the highest level of correlation was between firm size and subsidiaries with an obtained value of 0.655. Hence, the current study has no serious issues in terms of multicollinearity.
In addition, the result from Table 3 confirms that board meeting, subsidiaries, and market to book value have a positive correlation with financial restatements at 1%. Similarly, sales growth is found to have a positive association with financial restatements at 5% significant level. Furthermore, CEO title, CEO age, firm size, firm age, and subsidiaries are positively associated with CEO tenure at 1%. Whereas, CEO gender, CEO expert, sales growth, board meeting, and market to book value have a negative association with CEO tenure at 1%. Additionally, the loss has a negative relationship with CEO tenure at 5%. Besides, the analysis of the result from the correlation matrix as represented in Table 3 revealed a positive relationship exists between CEO age, firm size, firm leverage, Big 4, board meeting, and subsidiaries with CEO title at a 1% significant level. CEO expert has a positive relationship with CEO title at a 5% significant level. However, ROA found to have a negative relationship with CEO title at a 1% level of significant.
CEO gender is found to have a positive significant relationship with CEO expert, firm leverage, Big 4, and market to book value at a 1% significant level, and a positive significant relationship with loss at a level of 5%. In addition, CEO gender negatively correlated with CEO age at a 1% significant level, it is also negatively correlated with sales growth and board meeting, but at level of 5%. Furthermore, CEO expert has a significant positive association with firm age, firm leverage and subsidiaries at a 1% significant level. Similarly, the relationship between CEO expert and firm size is significantly positive at 5%. On the other hand, CEO expert has a negative significant relationship with CEO age and ROA at level of 1%. Moreover, CEO age is positively correlated with firm size, firm age, Big 4, and subsidiaries at a significant level of 1%. Moreover, firm size has a positive significant relationship with firm age, firm leverage, Big 4, ROA, board meeting, subsidiaries, and market to book value at 1% level of significance. Firm size negatively associated with loss at a level of 1%.
As presented in Table 3, firm age has a positive correlation with subsidiaries at 1%. Likewise, firm age found to have a positive association with Big 4 at a 5% significant level. In addition, firm age is associated negatively with ROA, sales growth, board meeting, and market to book value at 1%. Furthermore, firm leverage positively correlated with Big 4, board meeting, subsidiaries, and loss at 1% and with sales growth at 5%. Whereas, firm leverage has a negative relationship with ROA at 1%. The result from the correlation in Table 3 shows that a positive relationship exists between ROA, subsidiaries and market to book value with Big 4 at a 1% significant level. Big 4 is also found to have a significant negative relationship with loss at a 1% significant level.
In addition, ROA has a positive significant relationship with sales growth and market to book value at 1%. Whereas, ROA has a negative significant relationship with subsidiaries and loss at a level of 1%, as well as a negative significant relationship with board meeting at a level of 5%. The results from Table 3 revealed that sales growth positively and significantly correlated with the market to book value at a 1% level of significance. Whilst sales growth negatively and significantly associated with loss at a 1% level of significance. Furthermore, board meeting found to have a positive significant relationship with subsidiaries at a 1% level of significance, and positively and significantly associated with market to book value at a 5% level of significance. Finally, market to book value is negatively correlated with loss at a significant level of 1%. Table 3. Correlations matrix of the study variables *. Correlation is significant at the 0.05 level (2-tailed).

Logistic Regression Results
A logistic regression analysis is utilised in the scrutiny of the relationships of CEO tenure, CEO title, CEO gender, CEO expertise, and CEO age with the existence of financial restatements. Table 4 shows that the result of the R 2 for the study model is 0.0416 and the model is significant at level of (p<0.0l) which showed the validity of the study model.
In addition, Table 4 indicates that the relationship of CEO tenure with financial restatement is positively significant and the z-value is 1.64 and value of p<0.05 which indicates that financial restatements are highly related to CEO tenure. Thus, the result is in line with hypothesis 1, which predicts a significant linkage between CEO tenure and financial restatements. This means that a CEO with long tenure takes more actions to improve the short-term performance than the long-term performance. Moreover, the result is consistent with Lovata, Schoenecker and Costigan (2016) who found that CEO tenure is related to the likelihood of a firm reducing the financial reporting quality.
Furthermore, as shown in Table 4, CEO title and financial restatement have a negative and significant relationship. The z-value for the CEO title is -1.91 and the p-value at a significant level of 0.05%. This means that CEO title has an effect on financial restatements and is consistent with hypothesis 2, which expected a significant association between CEO title and financial restatement. This result indicates that CEOs with honorific titles care about their reputations and they are trustworthy, hence they are less likely to be implicated with financial restatements and provide higher quality financial reporting. Moreover, the results also show that the test of the hypothesis indicated a positive significant relationship between CEO gender and financial restatement; the z-value is 1.43, showing the possibility of the existence of a relationship between CEO gender and financial restatements where the value of p<0.10. Therefore, this result is in line with hypothesis 3. In addition, the result is inconsistent with Abbott et al. (2012) who show a negative significant association between female and the probability of financial restatements. A possible justification may be offered as potential explanations for this situation that females in an absolute minority may not be capable or willing to reduce the manipulations which cause financial restatements, since, only 0.037 per cent of the CEO female. In addition, if they were appointed only to fulfill the gender quotas, they would be unsuccessful at limiting the manipulations.
In addition, Table 4 shows insignificant relationship between CEO expertise and financial restatements and the z-value is -0.99, with the value of p>0.10. Hence, hypothesis 4 is rejected. The result is not consistent with Baatwah, Salleh and Ahmad (2015) as they stated that CEO has a significant influence on the quality of financial reporting; as well, these characteristics have an impact on the CEOs" behaviours.
Likewise, the results indicate that there is insignificant relationship between CEO age and financial restatements with z-value of -0.90 and value of p>0.10. Hence, this result rejected hypothesis 5, which predicts that CEO age decreases the level of financial restatements. The result is not in line with Huang et al. (2012) who found that the age of the CEO is negatively related with financial restatements.
The result of the control variables are shown in Table 4. A negative insignificant relationship is shown between firm size and financial restatements, and firm age with financial restatements. The z-value for firm size in the model is -1.13 and p-value > 0.10; also, the z-value for firm age in the model is -1.07 and p-value > 0.10. Moreover, the results found a positive and significant relationship between leverage and financial restatements with a z-value of 1.73 and p-value of 0.05. In addition, Big 4 auditors found to be negatively and significantly associated with financial restatements. It is argued that firms audited by one of the Big 4 are less engaged in financial restatements compared to firms audited by non-Big 4 firms; the z-value for Big 4 is -2.27 and significant at a level of 0.05%. Furthermore, the result of ROA found to be positive and significantly related to financial restatements, as shown in Table 4, the z-value for ROA is 1.70 and significant at a level of 0.05%. In addition, the results showed a positive significant relationship between sales growth and financial restatements and the z-value for sales growth is 1.36 at a level of 0.10%. The results of this study show that the frequency of the board of directors meeting is positively and significantly related to financial restatement and the z-value is 3.10, where the value of p<0.01. Furthermore, the table displays a positive significant association between subsidiaries and financial restatements with a z-value of 3.87 at a level of 0.01% of significant. As shown in Table 4, there is a significant positive relationship between MTB and financial restatements and the z-value is 3.19 at a significant level of 0.01. Finally, Table 4 displays that the z-value of LOSS is 2.11 and has a significant and positive association with financial restatements at a level of 0.05%. FIRMSIZE is the firm siza, FIRMAGE, is the firm age, LEV is the leverage; BIG4 is the top four audit firms; ROA is the return on assets, SGROWTH is Sales growth, BMEETING is the board meeting, SUB is the subsidiaries, MTB is the market-to-book ratio, Loss is the net loss of firms.

Conclusions
The current study attempted to investigate the impact of CEO characteristics and financial restatements in Malaysia. Specifically, the study covered Malaysian listed firms in Bursa Malaysia during the period of 2012-2016. Five CEO characteristics, namely tenure, title, gender, expertise, and age, were employed in this study.
The findings generated by this study add to the knowledge of the link between CEO characteristics and financial restatements, particularly in relation to the factors of CEO tenure, CEO title, CEO gender, CEO expertise, and CEO age. In this study, a significant positive relationship found between CEO tenure and financial restatements. This shows that CEO tenure affects the likelihood of financial restatements. Additionally, CEO title has a significant negative relationship with financial restatements. This indicates that CEOs who have the honorific title are not involved in financial restatements because they are trustworthy. Furthermore, a significant positive relationship found between CEO gender and financial restatements. This suggests that female CEOs engage with financial restatements in Malaysian firms. However, an insignificant negative relationship appeared between CEO expertise and CEO age with restatements of financial statements.
Furthermore, this study has made several contributions. Firstly, this study contributes to the literature and to the body of knowledge, by providing an examination of CEO characteristics and financial restatements in Malaysia. Second, this study will be valuable for the future researchers who are interested in conducting research such as this study in terms of other CEO characteristics and financial reporting quality; as well as this study will help them to evaluate the factors that affect the financial restatements. Similarly, this study provides evidence of the relationship between CEO tenure, CEO honorific title, CEO gender, CEO expertise, and CEO age with the incident of financial restatements in Malaysia. Third, the study is very significant to the practitioners, such as shareholders, investors because this study highlights the factors that can affect the financial restatements. From the results of the study, shareholders and investors can get an overview of the factors amongst the characteristics of CEO that should be emphasised more for bringing high financial reporting quality to the firms. Hence, this study is expected to deliver feedback to several concerned parties on the financial reporting of Malaysian listed firms. This study may help Malaysian regulators, such as Bursa Malaysia, the Audit Oversight Board (AOB), and the Securities Commission, to review the factors that affect the financial restatements and at the same time the quality of financial reporting of Malaysian listed firms, specifically the CEO characteristics.
This study has several limitations. First, financial restatement can be affected by other factors and characteristics that are not included in this study since this study focuses on only five CEO characteristics. Second, the financial restatement classification in this study primarily depends on three reasons for the restatements which are misrepresentation, accounting rule application failures, and irregularity categories. Third, the current study results may be not applicable to the other developing countries with a different structure of regulations. These limitations open up avenues for future studies on this issue that could be undertaken in the future, for example, to investigate other characteristics that would affect financial restatements, such as CEO power, CEO founder, CEO ethnicity, CEO compensation, CEO ownership, and CEO religiosity. Furthermore, despite all practical efforts that have been taken by this study into account to identify the restatement firms, it is possible for future studies to use all reasons and categories in identifying the financial restatement firms in Malaysia to see the impacts of restatement announcement based on different reasons. Finally, future studies may also examine the likelihood of restatements in quarterly financial reports.