THE ROLE OF CORPORATE GOVERNANCE IN PREVENTING MISSTATED FINANCIAL STATEMENT

The purpose o f our study is to empirically investigate the relation between certain corporate governance mechanisms and the likelihood o f a company having accounting problem, as evidenced by a misstatement o f its earnings. We use public listed firms in JSE as our sample, with total samples o f 160 firms-years. Our study finds that proportion o f independent board, proportion o f institutional ownership, and audit quality is significantly have negative relationship with the probability o f restatement. These results suggest that those governance mechanisms are able to prevent misstated financial misstatement. But, we find that three governance mechanisms board size, existence o f audit committee, and block holders do not have significant relationship with the probability o f restatement.


INTRODUCTION
Financial statements are perceived as means to reduce information gap between management and externa! stakeholders, especially investors and creditors. In order to achieve this purpose, valuable financial statements must comply with generally accepted principles, which guided the overall recognition, valuation and presentation o f financial statements. However, the generally accepted principles often provide options, which in turn create management discretion towards the selection o f accounting principles in presenting their financial statements.
As many options provided in the accounting principles, nevertheless; there is always probability o f management providing non-generally accepted accounting principles-financial reports to stockholders. Following this statement, a restated financial statement was published. At a minimum, this admission creates the appearance of executive who are at odds with their stockholders and this action was costly to stockholders. The GAO estimates that accounting restatements announced between January 1997 and June 2002 caused market capitalization losses o f about $ 100 billion and substantially reduced public confidence in the business community and capital markets (GAO 2000, 26, 32-41) The purpose o f this study is to examine restatement announced in 2002 to 2003. Specifically we examine the role o f corporate governance mechanism in preventing misstated financial statements, since corporate governance mechanisms are aimed to provide financial statements in a timely manner. The corporate governance framework should ensure that timely and accurate disclosure is made on all material matters regarding the corporation, including the financial situation, performance, ownership, and governance o f the company. We empirically investigate the relation between certain corporate governance mechanisms and the likelihood o f a company having accounting problem, as evidenced by a misstatement o f its earnings.
Understanding governance factors is necessary to take steps to prevent a recurrence. The result of this study should be o f interest to members o f the accounting and auditing community who want to avoid the legal obligations and loss o f prestige that accompany restatements; investors who lost substantial sums o f money while relying on misstated financial statements; regulators who had been criticized for negligent oversight; and the business community which lost much o f the public's trust (Alsop, 2004).
As to our knowledge, this is the first empirical study in Indonesia to analyze the relation between corporate governance mechanisms and the incidence o f earnings restatements. We hope the study will provide insight understanding about the role o f corporate governance mechanism in preventing restatement announcement. We find that several corporate governance mechanism have significant relation to probability o f restatement o f financial statement. Consequently this result might be a fundamental reason to enhance the practice o f good governance by the regulators. At the same time, we hope that this study will benefit potential and current investors and creditors in assessing the relevance and reliability o f financial statement through the practice o f corporate governance adopted by the firms.

PREVIOUS RESEARCH
Several papers examine the negative consequences o f an accounting restatement. All the studies examining market reactions to the announcement o f a restatement provide evidence of adverse consequences. Turner, et al. (2001) report negative returns o f 12 percent for revenue restatements and 5 percent for restructuring, impairments and other misstatements. Palmrose et al (2004) report a decline o f 9 percent, Anderson and Yohn (2002) find a decrease o f 3.9 percent. Wu (2002) reports a decline o f 12 percent for a sample o f companies either revising an earnings announcement or announcing a restatement. While the stock declines vary across these studies (due to different time periods, return windows, and othej. design issues), the overall conclusion is that restatements are costly to stockholders.
Other research includes Kinney and McDaniel (1989) who examine market reaction to restated earnings. They find, on average, negative return between the date when misstated quarterly statement were issued and the date when the statements were corrected. Palmrose (2004) also fines negative abnormal return on the date o f the announcement o f restated financial statement. On average market reaction is even higher when there is possibility o f management fraud, restatement o f material amount and auditors' restatement. Anderson and Yohn (2002) examine firms issuing audited restated financial statement. They find that the median o f firms stock price decline around the restatement date, the decline is even higher for firms with restated revenue.
Earning manipulation often leads to financial restatement. Richardson, Tuna and Wu (2002) find that that restatement firms are engaging in aggressive accounting to maintain a pattern o f (i) consistently meeting analyst earnings targets and (ii) consistently reporting increases in quarterly earnings. Therefore, their study shows that that total accruals are positively associated with the likelihood o f observing an earnings restatement.
Other researches investigate factors associated with the occurrence of accounting restatements. Agrawal and Chadha (2004) examine whether certain corporate governance mechanism are related to the probability of a restatements. Using hand-collected data for 159 companies announcing a restatement in 2000 or 2001 and an equal number of industry-and size-matched control firms, they find the presence o f independent directors with financial expertise reduces the likelihood o f a restatement. They also find that firms whose CEO comes from founding family have higher probability to restate their financial statements comparing to firms whose CEO comes from outside parties.
Audit committee is formed to monitor the process o f presenting reliable and relevant financial statements. Klein (2002) finds negative correlation between the existence of independent audit committee and earnings management. This finding is consistent with the idea that lack of independency will reduce board and audit committee's ability to monitor managements' action.
Abbot et. all (2002) find that the audit committee independence and activity (whether the committee meets at least four times per year) exhibit a significant and negative association with the occurrence o f financial reporting restatements. They also find a significant and positive association between the occurrence of financial reporting restatement and the lack of financial expertise (the presence of a financial expert) on the audit committee.
However, audit committee is often inactive. They do not meet intensively in a year,.Therefore, even though audit committee consists of independent members, there is little possibility that they could detect accounting "scandal" in big and complex firms in a relatively short period o f time due to their lack of activities (Agrawal and Chadha, 2004). Consistent with this idea, Beasley (1996) finds no significant difference in audit committee composition between fraud and non-fraud sample companies. Similarly, even though a typical board meets more frequently (usually about six to eight times a year) than the audit committee, it has a variety o f other issues on its agenda besides overseeing the financial reporting o f the firm. The board is responsible for issues such as the hiring, compensation, and firing o f the CEO and overseeing the firm's overall business strategy, including its activity in the market for corporate control. So it is possible that even a well functioning, competent, and independent board may fail to detect accounting problems in large firms. Accordingly, Chtourou et al. (2001) find no significant relation between board independence and the level of earnings management. Larker et al. (2004) examines the effect of various corporate governance factors to various measurements of managerial behavior and firm's performance. They find that of 14 corporate governance factors are able to explain from 0.6% to 5.1 % of cross sectional variation using regression analysis. In addition to that, the sign on corporate governance factor coefficients are often contrary to expectation. They assume that corporate governance indicators used for various academic researches have limited power in explaining managerial behavior and firms performance.

RESEA RCH M ETO D O LO G Y Hypotheses Development
We investigate the relation between certain corporate governance mechanism and the likelihood o f restatement. The specific corporate governance issues that we analyze are: the existence o f independent board members, size o f the board, the existence of audit committee, numbers of block holders or parties with more than 5% ownership, and the percentage owned by institutional ownership. We also interested in investigating the role o f external auditor in preventing restatement. As o f control variables we examine the relationship between leverage, interest coverage, size and the existence o f new debt and equity with the probability o f restatement.
Considering the corporate governance framework as to ensure that timely and accurate disclosure is made on all material matters regarding the corporation and the result of previous research, we state our hypotheses as follows: Firms with higher proportion o f independent board members have lower probability o f accounting fraud and earnings management (Beasley (1996), Dechow et al. (1996), and Klein (2002).
Hypothesis 1: Proportion o f independent board members have negative relationship with the probability o f restatement. Klein (2002) finds a negative relation between audit committee independence and earnings management. This finding is consistent with the idea that lack o f independence impairs the ability o f boards and audit committees to monitor management. Agrawal and Chadha (2004) find that probability of restatement is significantly lower in companies whose boards or audit committees include an independent financial expert.
By 2000, Indonesian stock markets started requiring each listed firm to have an audit committee. BEJ requires that at least a member o f audit committee should possess financial expertise. We use audit committee list provided by BEJ to justify the existence of audit committee in sample firms. Therefore, our second hypothesis is:

Hypothesis 2:
The existence of audit committee has negative relationship with the probability o f restatement. Yermack (1996), as stated in Larcker et al. (2004), finds that firms with more members in boards show higher performance compare to firms with fewer board members. Nevertheless, increase number o f board members will increase arguments or even intrigues among board members. Also there are possibilities o f free riders in board members. Therefore we predict that board size might have positive or negative relation to probability of restatement.

Hypothesis 3:
Board size relate to probability of restatement Jensen (1993) and Shleifer and Vishny (1997) argue that block holders or institutional investors that hold large debt or equity positions in a company are important to a well functioning governance system. Institutional investors are expected to have close monitoring over management decision-making process, including selection of accounting methods applied in presenting financial statement. This constructs our fourth and fifth hypothesis as follows: Hypothesis 4: Proportion o f institutional ownership has negative relation with the probability o f restatement.
Hypothesis 5: Number of block holders has negative relation with the probability of restatement.
T heexternal audit is intended to enhance the credibility o f financial statements of a firm. Auditors are supposed to verify and certify the quality of financial statements issued by management. However, auditors often provide consulting and business services to their audit clients. Auditors argue that providing consulting services to audit clients increases their knowledge and understanding o f the client's business, which leads to improvement in the quality of their audits. Investigating the range o f service provided by external auditors will be very rewarding; yet, the data is difficult to obtain since we use handcollected data. Therefore, in this study we only examine whether selection of audit firm might correlate with the probability o f restatement.

Hypothesis 6: Firms audited by big 4 audit firms have lower probability o f restatement comparing to firms audited by other audit firms
As control variables, we examine the effect of leverage, interest coverage, size, and the existence of new debt or equity.
In studies involving financial errors and fraud risk, Kreutzfeldt and Wallace (1986), find a positive association between performance pressures and the risk o f financial misstatements. They find that firms with liquidity and profitability problems were more likely to make more errors. Likewise, firms subject to capital market pressures were at a greater risk o f financial misstatements. Therefore, we expect that leverage, interest coverage and the issuance o f new debt or equity will have positive correlation with the probability of restatement.
Company size is used as a proxy for information asymmetry in the predisclosure information environments, as managers o f small companies are able to retain their private information more successfully than are their counterparts o f large companies (Lee & Choi, 2002). Information on large firms usually is more publicly available and could be obtain with lower costs than information on small firms (Bhattacharya 2001). Accordingly, firm size is often used as a proxy o f information availability in the market. Hence, it is more possible for investors to detect misstated financial statement in bigger firms than in smaller firms. So, we expect that size has negative relationship with probability o f restatement.

R esearch Model
Using the above variables, we state our regression model as follows:

Restatement
Using hand collected randomly sampled data for 160 financial statements reported in 2001 and 2002, we applied subjective judgment to differentiate restatement and non restatement firms. We applied definition provided by General Accounting Office (GAO). The GAO report (2002) differentiates restatement that result from accounting irregularities and restatement that result from normal corporate activity or simple presentation issues.
GAO report stated ' '' 'accounting irregularity is defined as an instance in which a company restates its financial statements because they were not fairly presented in accordance with generally accepted accounting principles (GAAP). This would include material errors and fra u d ". (GAO report 2002, page 2) Examples o f restatement that is not irregularity and resulted from normal corporate activity or simple presentation issues includes: restatements due to mergers and acquisitions, general accounting changes under new GAAP, and restatement for presentation purpose.
We examine the reason for restatement or reclassification by reviewing the notes to financial statements. We adopt GAO classification and excluding restatement due to mergers and acquisitions, general accounting changes under new GAAP, and restatement for presentation purpose.
Thp-sample for the study is comprised o f all firms listed on Jakarta Stock Exchange that is non-regulated and non-financial in nature. In addition, the firms From criteria listed above, we have 160 sample firm-years. We classify our sample into restatement and non restatement firms as follows:

Board Structure
We examine two variables related to board structure, which include proportion of independent board and board size. The number o f independent board members is obtained through Stock Exchange websites. Where proporsion of independent board is computed by dividing the number independent board with the total o f board members.

Ownership structure
We examine the relation o f institutional ownership and block ownership. Institutional ownership is defined as percentage o f shares owned by financial institution. While block ownership is defined as number o f owners having ownership more than 5%.

Audit Committee
The existence o f audit committee is obtained through Jakarta Stock Exchange websites announcement. The existence of audit committee should comply to BEJ regulation which stated that at least one member o f the committee have financial expertise. Therefore we only include audit committee which has fulfilled the requirement o f having financial expertise. We use list o f firms provided by JSX website, and thus, excluding audit committee which has not fulfilled the regulation.

Auditors
Selection of auditors is classified into categories o f Big 4 audit firms and other audit firms. We use dummy variable, where 1 for firms audited by big 4 audit firms and 0 otherwise. Descriptive statistics for major components o f corporate governance mechanism are presented in Table 1. The mean for institutional ownership is only 0.079 indicating that on the average institutional ownership is less than 10 percent. Turning to block ownership, the average shows 2.53 indicating that on the average, our sample firms almost have 3 parties whose ownership is more than 5 percent.
Turning to board structure, the table shows that the average o f board size is 4,41 indicating that on average, our sample firms has board o f directors consisting o f 4 persons.
Descriptive statistics also shows that on the average our sample firms appears to be able to fulfill their debt related obligation as shown by average interest coverage o f 5,4. Average leverage shows 66,4%, which shows that average o f our sample firms are not heavily burdened by debt. Based on the regression result in Table 2, coefficient o f BOD is negative and significant (HI is supported). This suggests that independent board effectively prevent mistated financial restatement. This result is consistent with result found in Beasley (1996), Dechow et al. (1996), and Klein (2002).
AUDCOM is negative is predicted, but the result shows that it is insignificant. This indicates that altough the existence o f audit committee decrease the probability o f restatement but its influence is not significant.
BODSZ is insignificant too. The positive coefficient shows that bigger board size is better to prevent mistated financial mistatement.
INST is negative and significant at 10% (H4 is supported). It suggest that higher proportion o f institutional ownership will lower the probability o f restatement. This finding is consistent with Jensen (1993) and Shleifer and Vishny (1997), who argue that institutional investors that hold large debt or equity positions in a company are important to a well functioning "governance system.
BLOCK has negative coefficient (as predicted), but it is statistically insignificant. It seems that concentration o f ownership in institutional investors serve as better governance system than concentration o f ownership in block holders.
AUDIT is negative and highly significant (H6 is supported). It indicates that big 4 could provide higher audit quality, which lower the probability of restatement, than that of non big 4.
All o f our control variables are insignificant. This results may indicate that leverage, interest coverage, new financing, and size do not significantly affect the probability o f restatement for our sample firms.

Conclusion
Our study finds that proportion of independent board, proportion of institutional ownership, and audit quality is significantly lower the probability of restatement. These results suggest that those governance mechanisms are able to prevent misstated financial misstatement. But, we find that three governance mechanisms -existence o f audit committee and block holders -is not significantly lower the probability o f restatement.

Limitation o f the study
1. Due to the availability o f time and data, we are only able to collect 160 firms as our samples. Using additional data will improve the generalization o f the study. 2. We apply subjective judgment to classify restatement and non-restatement firms.* 3. Due to lack o f data on corporate governance index, we use only independent board, audit committee, institutional ownership, blockholders, and audit quality, to measure corporate governance practices in public listed firms. Nowadays, there is no institution in Indonesia that has developed corporate governance index for all public listed firms in JSE. One o f CG index available is Corporate Governance Performance Index (CGPI), which was issued by IICG. But it only surveys limited number of firms and only index for top 10 firms are published in SWA magazine where index for other firms is held confidential. Arsjah (2004) have done a survey on corporate governance and determine CG score for many companies listed in JSE. But her survey was done in 2004 and to use it in our study we need 2005 financial statementwhich has not been issued at this time.