DOI: http://dx.doi.org/10.25105/imar.v19i2.7113 Indonesian Management and Accounting Research http://www.trijurnal.lemlit.trisakti.ac.id/imar Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Received: 18 Jan 2020 Revised: 01 May 2020 Accepted: 27 Jun 2020 Sekar Mayangsari Universitas Trisakti Email: sekar_mayangsari@trisakti.ac.id Abstract This research investigates the variables that influence manipulated financial statements. There are several fraud theories explaining these conditions, including fraud pentagon which shows six factors that might influence manipulated financial statements. This study uses state-owned entities as samples with a total of 96 non-financial firms listed during 2013-2018, all disclosing audited financial statements. The results of the study show that capability is the only factor with a positive effect on manipulated financial statements. Furthermore, the audit quality has the potential to weaken the effect of capability on manipulated financial statements. Keywords: fraud pentagon, audit quality, state-owned entities, manipulated financial statement Indonesian Management and Accounting Research ISSN: (e) 2441-9724 (p) 1411-8858 Volume 19 Number 02 July 2020 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari INTRODUCTION There is a need for public sector accountability for good governance in Indonesia, which relates to transparency and the provision of information to fulfill the wider community's rights. Three main aspects support the creation of good governance, including supervision, control, and audit. Auditing ensures that the management of funds is in line with the regulations outlined by the government. Therefore, one of the contents of the package of laws on state finances is the provision of the audit function, precisely Law Number 15 of 2004. It includes internal government and external audits by the Audit Board of the Republic of Indonesia (BPK RI). BPK RI is mandated by Law No. 15 of 2004 on the Inspection of State Financial Responsibility to conduct audits of Government Financial Statements, including state-owned enterprises. The activities carried out provide opinions and show various findings that explain the weaknesses of internal control and disobedience to laws and regulations. They also reveal information on potential losses resulting from misuse and inefficiency in the use of APBN (State budget) / APBD (Regional Budget). After obtaining findings, investigations on the audit begin, and where necessary, graft cases are initiated. Earnings quality is divided into three, including earnings character, investor response, and external misstatement indicators. It has other characteristics, such as persistence and accruals, smoothness, and time inaccuracy when acknowledging the loss. The investor response to earnings, including the R2 earnings-returns model, relates to the earnings response coefficient, which is also used as an audit quality construct. The last category is external indicators, such as earnings misstatement or earnings management (Dechow, Ge, & Schrand, 2010). However, to understand audit quality, it is necessary to add auditors' ability to detect misstatements and report them (DeFond & Zhang, 2014). This quality needs to be built from the beginning of the audit to reporting and provision of recommendations. Cases of fraudulent financial statements in Indonesia, which declines investor confidence in management, have occurred several times. For instance, there was an incidence of fraud at PT Kimia Farma, which was listed on the Indonesia Stock Exchange (IDX) in 2001. Based on an examination conducted by the Ministry of SOEs and the Capital Market Audit Agency (Bapepam), currently known as the Financial Services Authority (OJK), there was an overstatement of the net income of PT Kimia 102 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 Farma, Tbk in 2001. The overstatement involved manipulating sales and inventory accounts in three business units by increasing the inventory price as authorized by the Production Director. Additionally, the management also doubles listed sales in two business units. It was established that the company uses return on assets as a means of manipulating financial statements—the stock price of PT Kimia Farma declined sharply when the misstatement was reported to the public (Martantya & Daljono, 2013). In practice, fraud can occur in various sectors, including SOEs. In another case, PT Waskita Karya, in 2009, overstated net income between 2004 and 2007. This condition was disclosed when the substitute Director re-audits the financial statements for issuance of the prior year's initial public offering. The audit results found an over-statement in profit recording of around Rp. 400 billion. Figure 1. Industry of victim organizations. Source: Association of Certified Fraud Examiner (2014) Consequently, the initial offering of PT Waskita Karya shares was postponed until PT Perusahaan Pengelola Aset (PPA) completed the restructuring case, which was estimated to take two years. At that time, PPA provided an additional fund of IDR 200 billion to improve the survival of PT Waskita Karya (WIKA). According to a survey by the Association of Certified Fraud Examiner (ACFE, 2014), the government sector, including state-owned enterprises, commits much fraud than other industries, as shown in the diagram below. 103 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari The novelty of this research, we use state-owned enterprises (SOEs), which are listed in the Indonesia stock market. Compared with other firms, SOEs have a purpose, mission, and objectives related to some aspect of public service and social outcomes. State ownership has been regarded as an instrument through which governments can regulate natural monopolies, public goods provision, regional policies, employment or social issues, and reduce market failures. Their ultimate goal is not only to maximize profit. Nonetheless, the contrasting views are that state ownership is mainly used for the interests of the ruling elite and that even in the case of market failure, state ownership proves to be inefficient. LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT Agency Theory This study uses agency theory as a theoretical basis (Jensen & Meckling, 1976). The theory explains various matters related to the characteristics of a company, such as (1) a manager in a company with a diverse financial structure choose a series of activities that lessens the value of the company (2) the background of management failure maximizing company value is consistent with efficiency, (3) the reason why shares are a significant financial source even though managers are unable to optimize the firm’s value, (4) why debt is relied on as a source of capital even though it does not provide tax benefits, (5) reasons for the issuance of preferred shares, (6) why audited financial statements need to be given to creditors and shareholders voluntarily, (7) why creditors often place restrictions on the activities of debtor companies, and (8) why highly regulated industries such as utility companies or banks have higher debt-equity ratios than nonregulated firms. Furthermore, the agency is a contract between company owners and management (Jensen & Meckling, 1976). The agency problem is caused by a conflict of interest between the company owner and management since they have different maximum utility. Based on agency theory, the emergence of manipulator behavior is driven by conflicts of interest. For this reason, the management tries to maximize its interests. The urge to manipulate financial statements is also attributed to the selfinterest of human beings. According to Bosse and Phillips (2016), selfinterest is based on reciprocal norms between positive-negative behavior and fairness. For example, in case the management demands are large, the reciprocity is negative behavior. 104 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 Fraud Fraud is defined as a deliberate behavior to manipulate or any act that is not reasonable for one's interests at the other parties' expense. It might also involve taking money, property, or the legal rights of others. Corporate fraud is a fraudulent behavior that causes errors in presenting financial statements (AICPA, 2019). It can be classified into two, including fraud in the presentation of financial statements and the use of assets. Furthermore, corporate fraud is a dishonesty activity that affects other parties (CIMA, 2017). These activities include stealing, corruption, money laundry, conspiracy, bribery, embezzlement, and extortion. In the company, fraud can be committed by stakeholders and the government, fraudulent financial statements, and violations of various existing regulations (Agrawal, Jaffe, & Karpoff, 1999). In general, companies in a dynamic environment or a government vortex commit fraud easily (Abdullahi & Noorhayati, 2015). Political interests and the demands of the authorities sometimes force the management to commit fraud. Company fraud research was first conducted in the 1980s on manipulating earnings by company executives to achieve the company targets (Vassiljev & Alver, 2016). The executive's motivation to manipulate earnings is earning obtained in case the company reaches the target. This manipulation is mainly associated with corporate governance and financial ratios. The research scope for corporate fraud has also changed in decades. Only financial statement fraud was initially discussed but later progressed to stakeholders, regulatory violations, and the government sector. Fraud research is often carried out using several financial ratio variables (Brazel, Jones, & Zimbelman, 2009), (Dechow, Ge, Larson, & Sloan, 2011), (Nelson, 2012b), (Nelson, 2012a), (Schrand & Zechman, 2012) and (Shih, Cheng, & Wang, 2011). For instance, recent research on earnings manipulation is associated with earnings management, income smoothing, and corporate fraud. The theory of fraud explains the reason a corporation commits this violation and has been developing for decades. For instance, the fraud theory literature first developed was a fraud triangle (Cressey, 1953). It stated three main components of fraud, including pressure, opportunity, and rationalization. The theory was then developed into a fraud diamond by adding capability as a new component (Wolfe & Hermanson, 2004). The last model developed was fraud pentagon, which incorporated arrogance (Marks, 2012). 105 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari Beneish theory is a financial statement analysis technique that can be applied to detect fraudulent financial statements based on earnings manipulation. Beneish (1999) studied quantitative differences between companies involved in earnings manipulation and companies that are not identified as doing earnings manipulation. Beneish uses the company's data and calculates the financial ratios to determine the availability of conditions that encourage earnings manipulation. According to the study, earnings manipulation is indicated by an increase in revenue or a decrease in company expenses significantly from the year (t) to the previous year (t-1). From this, a ratio related to changes in assets and sales growth was formulated in the M-Score to reflect earnings manipulation. Some financial ratios used to detect earnings manipulation include a. Days Sales in Receivables (DSRI) Index b. Gross Margin Index (GMI) c. Asset Quality Index (AQI) d. Sales Growth Index (SGI) e. Depreciation Index (DEPI) f. Sales and Administration expense index (SGAI) g. Leverage Index (LVGI) h. Total Accruals on Total Assets (TATA) Audit Quality Audit quality occurs if an audit results produce published earnings free of material misstatement and are relevant and reliable. Conceptual Framework The purpose of this research is to detect the alleged manipulation in the financial statements, which is conducted using the fraud pentagon model (Marks, 2012). The model shows that fraudulent behavior is caused by pressure, rationalization, opportunity, fraudsters' ability, and the arrogant attitude of leaders. Although a company has the urge to commit fraud, with efficient external party control, fraud can be reduced. 106 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 Figure 2. Conceptual Framework Pressure Rationalization Manipulated Financial Statements Opportunity Competence Arrogance Audit Quality Hypothesis Development According to Summers and Sweeney (1998), financial targets between non-fraud and fraud companies are significantly different. One form of pressure that trigger fraud is the financial target. They lead to excessive pressure on management to achieve the targets set, including revenues and profits. According to Skousen, Smith, and Wright (2009a), financial targets are generally used to measure the manager's performance and determination for earnings. The higher the financial targets set, the higher the pressure felt by management. This target increases the possibility of management to manipulate earnings. Therefore, the hypothesis developed is H1: Financial pressure has a positive effect on the manipulation of financial statements. Studies show that receivables and inventories accounts require a subjective valuation to estimate uncollectible amounts and obsolete stock (Summers & Sweeney, 1998; Hammersley, 2011). Under these conditions, management may use them as tools to manipulate financial statements. Furthermore, receivables and inventories also show the characteristics of a particular industry (Summers & Sweeney, 1998). In the financial statements, there are certain accounts whose balances are determined by 107 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari the company based on an estimate, including uncollectible receivable and obsolete inventory accounts. The management uses these accounts to manipulate financial statements, including earnings (Beneish, 1999). Therefore, the second hypothesis is: H2: Industry characteristic has a positive effect on the manipulation of financial statements Rationalization is the main factor causing manipulation of financial statements, with fraudsters seeking justification for their actions. This attitude encourages individuals to commit fraud. Therefore, management integrity is needed to eliminate the desire to manipulate. The honesty of managers determines the reliability of financial statements. Someone with a dishonest attitude easily justifies manipulative behavior. Contrastingly, someone with high moral standards cannot participate in manipulative actions. Fraudsters always look for reasons that justify fraudulent deeds. Therefore, the hypothesis developed is H3: Rationalization has a positive effect on the manipulation of financial statements. According to Wolfe & Hermanson (2004), capability is a qualitative complement to the fraud triangle model (Cressey, 1953). Many capability components might encourage someone to commit fraudulent financial statements, including position, reason, selfishness, depression, and stress. In this study, directors' changes are a form of capability or competence and depend on political content and interests of certain parties, which often causes conflicts of interest. Capabilities due to changes in directors indicate the occurrence of fraud (Wolfe & Hermanson, 2004). The change in directors in state-owned companies was motivated by political reasons or other subjective factors and may hurt the company. Directors are often changed to improve the performance of previous management. However, the change is sometimes a company's attempt to get rid of directors knowing the fraud committed by the company (Schrand & Zechman, 2012). Therefore, the fourth hypothesis is H4: Capability has a positive effect on the manipulation of financial statements Arrogant attitude is one of the factors that trigger fraud in financial statements (Marks, 2011). CEOs' arrogant attitudes trigger fraudulent behavior, including egos, underestimating other people, autocratic style, fear of losing position, and desire to do all activities without being 108 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 supervised. These five factors individually and collectively form an arrogant attitude that influences the desire to commit fraud and maintain the position. One proxy that is often used to measure arrogant attitudes is how often photos of company leaders appear in the public sphere (Marks, 2012). Furthermore, the appetite CEO's frequency encourages a leader to maintain the quality of personal appearance. The leader belief often that the image of the company depends on the appearance of the leader. Previous works show the arrogant attitude of CEOs in the form of repeated explanations related to their appearance in the companies' annual report (Yusof, Khair, & Simon, 2015). Leaders tend to show the power and career possessed in the company to maintain their status. This arrogant attitude leads to fraudulent financial statements through the use and utilization of the authority possessed. Any internal party cannot limit the actions and behavior of the leaders because of the powers possessed. According to Yusof et al. (2015) and Tessa and Harto (2016), the frequency of CEOphotos' appearances is related to arrogance relates to fraud in financial statements. Therefore, the fifth hypothesis is, H5: Arrogance has a positive effect on the manipulation of financial statements Agency theory states that there are differences in interests between management and company owners. Therefore, the company owner needs to create a mechanism to control management behavior to be in line with their needs (Jensen & Meckling, 1976). Setting financial targets is among the tools used by company owners to control management behavior. Agents and principals hope to fulfill each other's interests. In this regard, there is a desire for management to get a bonus for the results of their performance in the fulfillment of the principal's wish, specifically profit maximization. Profit is one of the strategies used for manipulating financial statements because it is a performance measure used as the basis for determining bonuses (Skousen et al., 2009). The auditor is responsible for carrying out audit activities to obtain adequate confidence about the financial statements' fairness and being free from material misstatement. The auditor needs to have sufficient knowledge of audit techniques and understand the criteria used. This knowledge can be obtained through formal and informal education, as well as experience in conducting audits. The increasing competition today makes it more difficult for public accountants to behave professionally. Consequently, many public accounting firms are more interested in 109 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari retaining clients and large profits. However, demands for quality audits suppress opportunistic attitudes of management. H6: Audit quality weakens the effect of financial targets on the manipulation of financial statements Summers and Sweeney (1998), Loebbecke et al. (1989) stated that accounts receivable and inventory require subjective judgment in estimating uncollectible receivables and obsolete inventories. Since there are subjective judgments in determining the account's value, management may use it as a tool to manipulate financial statements. Good quality of financial statements can be achieved if the process carried out by the auditor proceeds effectively. This quality means the implementation of the audit achieves predetermined objectives. This quality needs to be built from the beginning of the audit to reporting and provision of recommendations. H7: Audit quality weakens the effect of industry characteristics on the manipulation of financial statements Lack of integrity causes one or more individuals to commit fraud rationally and determines financial statements' quality. When the integrity of managers is questioned, the reliability of financial statements is doubtful. For those who are generally dishonest, it might be easier to rationalize fraud. In contrast, it might not be too easy for those with higher moral standards to commit fraud. In auditing financial statements, each public accountant needs to fulfill professional responsibilities with the highest possible integrity level. Accountability possessed by an auditor improves the cognitive process in making decisions that affect financial statements' quality. H8: Audit quality weakens the effect of rationalization on the manipulation of financial statements There are six components of capabilities: position, intelligence, confidence, coercion, fraud, and stress management. The substitution of directors shows the ability to carry out stress management. Public accountants or independent auditors need to have certain professional principles and quality guidelines. Arens (2017) and Daljono (2013) stated that directors' change is often used as an agenda to manipulate financial statements with specific objectives. Good quality of financial statements should not be affected by the change of directors. H9: Audit quality weakens the effect of capability on the manipulation of financial statements 110 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 Company leaders often want to be respected because of their status in the company. This argument is consistent with one of the elements presented by (Tessa & Harto, 2016) and (Laffin & Gomes, 2013), specifically arrogance. However, it is quite challenging to measure a leader's arrogance. Research shows that the frequent appearance of toplevel management in the media is an indication of arrogance. This indication is shown by the frequency of the CEO's picture or the number of photos displayed in the company's financial statements (Oktarigusta, 2017; Faradiza, 2019). The number of photos displayed in a company's financial statements may represent the level of arrogance or superiority possessed by the CEO. Financial statements' quality is a joint probability where an auditor finds and reports violations in the client's accounting system. The probability that the auditor might find misstatements depends on the quality of the auditor's competence. However, the act of reporting misstatements depends on the independence of the auditor. The quality of financial statements is critical because high quality means they can be trusted as the basis for decision making. H10: Audit quality weakens the effect of arrogance on the manipulation of financial statements RESEARCH METHODS Research Design This study used a hypothesis test design to determine the effect of the fraud pentagon indicator on manipulating financial statements using the Beneish M-Score model. This study used secondary data and panel data. Fraud pentagon variables used include pressure factors with financial target based on return on assets (ROA) variable, opportunity factors with industry characteristics categories determined receivable, rationalization factors measured by total accruals (TACC), capability factors based on changes in directors (DCHANGE), arrogance, and audit quality as a moderating variable. This study was conducted on Non-Bank SOE companies listed on the Indonesia Stock Exchange in 2013-2018. The sampling method utilized was purposive, while the analysis unit was based on entities. 111 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari Operational Definition of Variable The operational definition of each variable is as follows. Table 1. Variable Measurement No . 1 Variable Proxy Measurement FRAUD Beneish M-Score Model 4 Pressure Industry characteristics Rationalization ManipulatedFina ncial Reporting ROA (X1) RECEIVABLE (X2) TACC (X3) 5 Competence DCHANGE (X4) 6 Arrogance 7 Audit Quality 2 3 CEOPIC (X5) KA (Z) Net income / Total asset (Receivables/Sales) (Receivables/Sales) Total Accrualst / Total Assets Dummy variable 1 = in case of a change in company directors. 0 = in case there is no change in company directors This variable is measured by counting the number of CEO photos contained in the financial statements. Dummy Variable 1 = if BIG4 0 = if Non-BIG4 Hypothesis Test The analysis used in this study was the logit regression analysis, with the following regression equation: FRAUD=α+β1ROAit+β2Receivableit+β3TACCit+β4DCHANGEit+β5CEOPICit+β 6AQit+β7ROA*AQit+β8Receivable*AQit+β9TACC*AQit+ β10DCHANGE*AQit+ β11CEOPIC*AQit+ε.... (1) Description: FRAUD α ROAit TACCit DCHANGEit CEOPICit AQit εit : Manipulated financial statements : Constant : Return on Asset Receivableit: Receivable : Total accruals : Change of Directors : Number of CEO's photos : Audit quality : error term 112 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 ANALYSIS AND DISCUSSION Description of Data / Research Object The study used secondary data available on the Indonesia Stock Exchange (IDX) during the period 2013-2018. The research object was all Non-Bank State-Owned Enterprises (SOEs) in Indonesia, which were listed during 2013 - 2018. Also, the study used a purposive sampling method based on the following criteria 1) Non-Financial State-Owned Enterprises (SOEs) listed on the Indonesia Stock Exchange (IDX). The financial industry is not used since this industry is highly regulated, and some accounts are different compared to other sectors. 2) Non-Financial State-Owned Enterprise (SOE) reports their complete financial data on the Indonesia Stock Exchange from 2013 through 2018. 3) Non-Bank State-Owned Enterprises (SOEs) are listed until 2018. This study used the panel data method, which is a merger between the cross-section and time-series data. The observation period was 20132018 with 16 Non-Financial SOEs in one year and a five-year research period between 2013-2018. Therefore, a total of 96 company data were obtained. Table 2 shows the stages in the sampling of Non-Bank SOE companies on the IDX. Table 2. Sampling Chronology Description Number of Number of Samples Observations Listed SOEs 20 Financial (4) firms Number of samples 16 96 studied (four years) Source: IDX data processed SOE companies listed on the Indonesia Stock Exchange totaled 20 companies, including four banking and 16 non-banking. The 16 companies were from several sectors, including energy, transportation, mining, construction, telecommunications, pharmaceutical, and cement. However, not all these companies reported their complete financial statements from 2013-2018. The purposive sampling was used. It was based on the following criteria, 1) Non-Financial State-Owned Enterprises (SOEs) listed on the Stock Exchange Indonesia (IDX). This study did not use banking 113 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari companies because several research variables are not in banks' financial statements. 2) Non-Bank State-Owned Enterprises (SOEs) reported their complete financial data on the Indonesia Stock Exchange during 20132018. 3) Non-Bank State-Owned Enterprises (SOEs) are listed until 2018. This study used secondary data from the 16 annual financial statements of non-financial listed SOEs during 2013-2018. Therefore, the total sample is 96 companies. Descriptive Statistics Based on the results of descriptive statistical tests, the multiplication of the number of samples by research year periods (six years) was 96. Descriptive statistics explain the data description of all variables included in the research. Table 3 shows the descriptive statistics of the variables used. Table 3. Descriptive Statistics Data Variable N Mini mum Maxi mum Mean Fraud ROA Receivable TACC Dchange CEOPIC 96 96 96 96 96 96 -7,73 0,00 0,00 -0,21 1 1 8,31 0,20 0,90 0,26 2 3 0,24 0,07 0,07 -0,01 1,52 1,32 Std. Deviati on 12,03 0,04 0,15 0,07 0,503 0,522 Source: Processed Data Based on the descriptive analysis results above, the Fraud has the highest value of 8.305 for PT Perusahaan Gas Negara in 2014, while the lowest value was -7.7251 for PT Garuda Indonesia in 2015. The company with the most fraud was PT Semen Baturaja in 2013, while the one with the least fraud was PT Garuda Indonesia in 2015. The mean value was 0.243013, where the average company committed fraud and the standard deviation was 12.0328. The highest value of ROA was 0.1959 for PT Perusahaan Gas Negara in 2014, while the lowest was 0.0021 for PT Aneka Tambang in 2016. The company with the highest profit was PT Perusahaan Gas Negara in 2014, while the one with the lowest profit was PT Aneka Tambang in 2016. The mean value was 0.066180, which means that the average company has a profit of 0.0066180, while the standard deviation was 0.0448755. For Receivable, the highest value was 0.8978 for PT Perusahaan Gas Negara is 2013, while the lowest value was 0,000 for PT Krakatau Steel in 2017. The higher the value, the greater the risk faced by a company. This 114 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 value requires significantly higher estimates and considerations. The mean value was 0.074481, which means that the average company has a risk of 0.074481 with a standard deviation of 0.1540516. For Total Accruals, the highest value was 0.0698871 for PT Krakatau Steel in 2015, while the lowest was -0.2075 for PT Indo Farma in 2016. The higher the value, the greater the rationalization by management to commit fraud. The mean value was -0.013947, with a standard deviation of 0.0736296. DCHANGE is a dummy variable used to measure capability. In case the company has management or directors with capabilities, it will not change its directors. If company changes directors from the previous year, it is coded 2, and if not, it is coded 1. CEOPIC is a dummy variable used to measure the level of arrogance of a company's management. Audit Quality is determined by the dummy variable using BIG4 audit firm. A company that uses a BIG4 audit firm is coded by 1, and if not, it is coded 0. The BIG4 audit firm is used as a proxy for audit quality because it is perceived to have good quality, competent, and well known to the public. Therefore, the audit quality can be reliably compared to the published financial report audited by Non-Big4 audit firm. Table 4. Prediction Accuracy N % Predicted fraud Non-Fraud 57 59,38 Fraud 39 40,62 Total 100% Source: Processed Data Data analysis performed using logistic regression showed that the prediction accuracy for entities categorized as fraud was approximately 41%, while the non-fraud was 59% correct. The test results of ROA, receivable, TACC, DCHANGE, CEOPIC, and audit quality as a moderating variable on fraudulent financial statements are shown in Table 5. below. 115 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari Table 5. Hypothesis Testing Results ROA Receivable DCHANGE CEOPIC TACC ModROA ModReceiv ModTACC Modchange Modceopic AQ Constant β 5,573 -3,071 2,159 -0,935 11,331 -10,117 -2,995 -2,402 -1,615 1,447 -2,634 2,079 Sig. (one tail) 0,337 0,31 0,008** 0,212 0,037** 0,255 0,373 0,401 0,089* 0,151 0,117 0,112 **Sig.level 5%; *Sig.level 10% Hypothesis testing is carried out by comparing the level of significance with an error rate of 5%. Based on the results of the statistical tests, the financial target shows a regression coefficient (β1) of 0.337. Since the significant value is greater than α = 0.05 or 0.3365> 0.05, H1 is rejected. The hypothesis stating that financial targets have a positive effect on fraudulent financial statements is rejected. From statistical tests, the receivable has a significance of 0.310. Since the value is greater than α = 0.05 or 0.310> 0.05, H2 is rejected. Therefore, the hypothesis stating that receivable has a positive effect on fraudulent financial statements is rejected. Additionally, TACC has a significance value of 0.037, which is significant. For this reason, the third hypothesis stating that opportunities have a positive effect on fraudulent financial statements is accepted. From statistical tests, DCHANGE shows a significance of 0.008. Therefore, the change of the director harms a fraudulent financial statement. CEOPIC variable has a significance value of 0.05. Therefore, the hypothesis stating that CEOPIC has a positive effect on fraudulent financial statements is rejected. Other results show that audit quality cannot moderate the relationship between financial targets and fraudulent financial statements. Furthermore, the hypothesis stating that audit quality strengthens industry characteristics on fraudulent financial statements is not supported. The role of audit quality in strengthening the effect of TACC and arrogance on fraudulent financial statements is rejected. However, the moderating role of audit quality on arrogance and fraudulent financial statements is not supported. 116 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 Discussion Based on the results, return on assets has not a positive effect on fraudulent financial statements. Therefore, the first hypothesis was rejected. Skousen et al. (2009) stated that firm performance is often used in assessing a manager's performance and determining bonuses and wage increases. The higher the performance, the more vulnerable the management do earnings manipulation. This result is against the pentagon theory. The industry characteristics do not affect the fraudulent financial statement. The hypothesis stating that receivables have a positive effect on fraudulent financial statements is rejected. According to Summers & Sweeney (1998), receivables and inventories require subjective judgment in estimating uncollectible receivables and obsolete inventories. Furthermore, managers focus on both accounts in case they intend to manipulate financial statements. Also, Summers and Sweeney (1998) show that the ratio of changes in accounts receivable has a positive effect on the manipulation of financial statements. The result is against fraud pentagon theory, which is suggested that financial targets will cause fraud. Changes in company leaders do not always have a good effect on the company. Sometimes, it could be an attempt by the company to improve the previous directors' performance by changing their composition of the directors or recruiting more competent individuals. However, the change may be the company's attempt to get rid of directors who know the company's fraud. Since the changes require adaptation time, the initial performance is not optimal because they do not want to lose their status or position in the company's management. Arrogance is yet another factor that triggers fraud. Dietrich & Amrein (2017) and Tessa & Harto (2016) established that the frequent number of CEOs' pictures related to the arrogance that influenced fraud. The test results also show that audit quality moderates the effect of financial targets on manipulated financial statements. Therefore, the hypothesis stating that audit quality strengthens financial targets' effect on manipulated financial statements is accepted. Because with high-quality audits, even though management is depressed by these targets, it is not easy to commit fraud because high audit quality is generated from a qualified audit process. This result support agency theory The audit quality does not weaken the effect of receivables on manipulated financial statements. The hypothesis stating that the audit quality strengthens the effect of opportunity on manipulated financial 117 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari statements is rejected. According to Skousen, Smith, & Wright (2009b), accounts receivable and inventory require subjective judgment in estimating uncollectible receivable and obsolete inventory. Since there is a subjective judgment in determining the value of the account, management can use it as a tool to manipulate financial statements. In Summers and Sweeney (1998), the proxy used for receivables' industry characteristics is the ratio of changes in accounts receivable. Quality financial statements can be achieved if the auditing proceeds effectively, hence achieving the predetermined goals. Importantly, quality needs to be built from the beginning of the audit to reporting and provision of recommendations. The result against agency theory, which stated that an audit could influence the opportunities behavior of management. This influence is not surprising because auditors are under pressure management in certain conditions, especially for a deep pocket. Another reason, detecting fraud is neither the purpose nor the focus of an external audit. Based annual report of SOEs has some of the essentials in place to prevent fraud. They have fraud policies and codes of conduct and encourage staff to raise concerns. The hypothesis stating that audit quality strengthens the effect of TACC on manipulated financial statements is not supported. Management integrity is the main determinant of the quality of financial statements. When manager integrity is questioned, the reliability of earnings is also doubtful. For generally dishonest, it might be easier to rationalize fraud. In contrast, managers with higher moral standards cannot commit fraud. In auditing financial statements, each public accountant needs to fulfill professional responsibilities with the highest possible integrity. Accountability of an auditor improves the cognitive process in making decisions that affect the quality of financial statements. Some of the results do not support agency theory. The study results are consistent with the assumption that the audit process on SOEs is less effective due to various reasons, such as politics and economics. The auditors are not strong enough to reduce agency conflicts between the government as the majority owner and SOEs. The political pressure often makes SOEs unable to take professional action. 118 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 CONCLUSIONS, LIMITATIONS, IMPLICATIONS Conclusion The test results prove that financial targets, arrogance, industry characteristics, and opportunities do not influence manipulated financial statements. However, the change of directors has a positive effect on manipulated financial statements. Also, audit quality weakens the positive influence of capability on manipulated financial statements. Limitation There are limitations on data access for state-owned firms affecting the number of study samples. Suggestions for Further Research These empirical results can be expanded using a sample of non-stateowned companies in the financial sector to determine the research results' robustness. Furthermore, it can be tested using non-state-owned companies to improve the generalization of the results. Future studies need to add other variables, such as a positive accounting hypothesis, which states that bonus plans, debt contracts, and company size influence financial statements' preparation. 119 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari REFERENCES Abdullahi, R., & Noorhayati, M. (2015). Fraud Triangle Theory and Fraud Diamond Theory. Understanding the Convergent and Divergent For Future Research. International Journal of Academic Research in Accounting, Finance and Management Sciences, 5(4), 38–45. https://doi.org/10.6007/IJARAFMS/v53/1823 Agrawal, A., Jaffe, J. F., & Karpoff, J. M. (1999). Management turnover and governance changes following the revelation of fraud. Journal of Law and Economics, 42(2), 309–342. https://doi.org/10.1086/467427 AICPA. (2019). Consideration of Fraud in a Financial Statement. Audit and Accounting Guide – Construction Contractors, 2019, 193–206. https://doi.org/10.1002/9781119679295.ch12 Beneish, M. D. (1999). The Detection of Earnings Manipulation. Financial Analysts Journal, 55(5), 24–36. https://doi.org/10.2469/faj.v55.n5.2296 Bosse, D. A., & Phillips, R. A. (2016). Agency theory and bounded self-interest. Academy of Management Review, 41(2), 276–297. https://doi.org/10.5465/amr.2013.0420 Brazel, J. F., Jones, K. L., & Zimbelman, M. F. (2009). Using nonfinancial measures to improve fraud risk assessments. Journal of Accounting Research, 6(1). https://doi.org/10.2308/ciia-50168 CIMA. (2017). Corporate fraud. https://doi.org/10.4324/9781315259260 In Corporate Fraud. Cressey, D. . (1953). Other’s People Money: A Study in The Social Psychology of Embezzlemet. The American Journal of Sociology. Dechow, Ge, W., Larson, C. R., & Sloan, R. G. (2011). Predicting Material Accounting Misstatements. Contemporary Accounting Research, 28(1), 17–82. https://doi.org/10.1111/j.1911-3846.2010.01041.x Dechow, P., Ge, W., & Schrand, C. (2010). Understanding earnings quality: A review of the proxies, their determinants and their consequences. Journal of Accounting and Economics, 50(2–3), 344–401. https://doi.org/10.1016/j.jacceco.2010.09.001 DeFond, M., & Zhang, J. (2014). A review of archival auditing research. Journal of Accounting and Economics, 58(2–3), 275–326. https://doi.org/10.1016/j.jacceco.2014.09.002 Dietrich, A., & Amrein, S. (2017). Crowdfunding Monitoring Switzerland 2017. Faradiza, S. A. (2019). Fraud Pentagon Dan Kecurangan Laporan Keuangan. EkBis: Jurnal Ekonomi Dan Bisnis, 2(1), 1. https://doi.org/10.14421/ekbis.2018.2.1.1060 120 Indonesian Management and Accounting Research ISSN 2441-9724 (Online) ISSN 1411-8858 (Print) Volume 19 Number 02 July 2020 Hammersley, J. S. (2011). A review and model of auditor judgments in fraudrelated planning tasks. Auditing: A Journal of Practice & Theory, 30(4), 101– 128. https://doi.org/10.2308/ajpt-10145 Jensen, M. C., & Meckling, W. H. (1976). THEORY OF THE FIRM: MANAGERIAL BEHAVIOR, AGENCY COSTS AND OWNERSHIP STRUCTURE. Journal of Financial Economics, 3, 305–360. Laffin, M., & Gomes, T. (2013). The Prevention of Error and Fraud in Accounting. International Journal of Advances in Management and Economics. Marks, J. (2011). Why the Fraud Triangle is No Longer Enough. In Www.Crowe.Com. Marks, J. (2012). The Mind Behind The Fraudsters Crime : Key Behavioral and Environmental Elements. Retrieved from www.crowe.com Martantya, M. R., & Daljono, D. (2013). PENDETEKSIAN KECURANGAN LAPORAN KEUANGAN MELALUI FAKTOR RISIKO TEKANAN DAN PELUANG (Studi Kasus pada Perusahaan yang Mendapat Sanksi dari Bapepam Periode 2002-2006). 2, 700–711. Nelson, S. P. (2012a). Post fraud: An empirical study of firms in Malaysia. International Journal of Management and Business Studies, 2(3), 59–065. Retrieved from http://internationalscholarsjournals.org Nelson, S. P. (2012b). PRE FRAUD: AN EMPIRICAL IN MALAYSIA Sherliza Puat Nelson. International Journal of Economics and Finance Studies, 4(1), 143– 151. Oktarigusta, L. (2017). Terjadinya Financial Statement Fraud Di Perusahaan ( Studi Empiris Pada Perusahaan Manufaktur Yang Terdaftar Di Bei Tahun 2012-2015). Jurnal Daya Saing, 19(2), 93–108. Schrand, C. M., & Zechman, S. L. C. (2012). Executive overconfidence and the slippery slope to financial misreporting. Journal of Accounting and Economics, 53(1–2), 311–329. https://doi.org/10.1016/j.jacceco.2011.09.001 Skousen, C. J., Smith, K. R., & Wright, C. J. (2009a). Detecting and Predicting Financial Statement Fraud: The Effectiveness of the Fraud Triangle and SAS No. 99. In Corporate Governance and Firm (Vol. 3). https://doi.org/10.2139/ssrn.1295494 Skousen, C. J., Smith, K. R., & Wright, C. J. (2009b). Detecting and Predicting Financial Statement Fraud: The Effectiveness of The Fraud Triangle and SAS No. 99 in Corporate Governance and Firm Performance. In International Journal of Quality & Reliability Management (Vol. 32). https://doi.org/10.1108/S1569-3732(2011)0000014001 121 Indications of Manipulated Financial Statements: Evidence from Indonesia State Owned Enterprise Sekar Mayangsari Summers, S. L., & Sweeney, J. T. (1998). Fraudulently misstated financial statements and insider trading: An empirical analysis. Accounting Review, 73(1), 131–146. Tessa, C. ., & Harto, P. (2016). Pengujian Teori Fraud Pentagon Pada Sektor Keuangan Dan Perbankan Di Indonesia. Simposium Nasional Akuntansi, 1– 21. Retrieved from file:///C:/Users/ASUS/Downloads/Pengujian Teori Fraud Pentagon Pada Sektor Keuangan dan Perbankan di Indonesia.pdf Vassiljev, M., & Alver, L. (2016). Conception and Periodisation of Fraud Models: Theoretical Review. (Icaat), 473–480. https://doi.org/10.2991/icaat16.2016.47 Wolfe, D. T., & Hermanson, D. R. (2004). The Fraud Diamond : Considering the Four Elements of Fraud: Certified Public Accountant. The CPA Journal, 74(12), 38–42. https://doi.org/DOI: Yusof, M., Khair, A., & Simon, J. (2015). Fraudulent Financial Reporting: An Application of Fraud Models to Malaysian Public Listed Companies. A Multidisciplinary Journal of Global Macro Trends, 2(4), 144–160. 122