Determinants of Fraud Financial Statements on State Owned Enterprise

Indonesia (PSAK). Penelitian


INTRODUCTION
Fraud is a deliberate omission. In accounting, the fraud conception is action outside the accounting procedures that should be enforced to an organization. The information presented is not under the company's circumstances if it is a misstatement of information of material nature (Lestari & Bernawati, 2020) (Lestari & Bernawati, 2020). The financial losses incurred by PT Pos Indonesia (Persero) are caused by errors in the additional expenditure of social protection card (KPS) shipments. Agency theory is a basic theoretical perspective that is acclimated to understand the problems of internal control and organizational culture. Previous study mentions the agreement between business owners and management to practice services for their needs by transferring the authority of decision-making to management (Jensen & Meckling, 1976). The purpose of internal audit and internal control is to block fraud. When internal audits of the company have qualified and competent resources, and the discharge of the company's internal control by the established system, this can reduce the possibility of fraud (Maliawan et al., 2018;Yelvionita et al., 2020).
Condition when the misalignment of information between the party who owns or provides the information and the party who needs the information is called information asymmetry. Previous research mentioned two information asymmetries, namely: 1) Adverse selection, occurring because some people such as company managers and other insiders are aware of the current conditions as well as prospects of the company. 2) Moral Hazard, can occur due to separation of ownership and control (Drogalas et al., 2017;Pratiwi & Setiyawati, 2021). In theoretical institutions produce asymmetric relationships between owners, managers, and customers to avoid the relationship that asymmetry requires concepts of internal control and organizational culture aimed at making the company healthier. Management is more aware of the state of the company than the business owner (Setyaningsih & Nengzih, 2020;Wulandari & Nuryanto, 2018). This can lead to fraud, so it requires other parties who are independent and have an essential role in auditing contracts and minimizing the risk of fraud.
The control system should be able to be the ground of the company's development plan. The control system in the company will not be enough if the internal system is not effective and efficient. Good management with effective internal control can limit the acts of fraud that may occur. Internal control subsists of mechanisms and covers all activities and ensures efficient and effective operations (Munirah & Nurkhin, 2018;Oktavia, 2017). A procedure built to regulate an activity that is used as an instruction in carrying out the functioning activities of an organization is internal control. Organizations generally use this system so that fraud can be prevented. Some of the factors common to any organization in establishing an internal control system for the greater part of organizations are organizational systems, accounting systems, internal audits, regeneration, balance sheets, product, and service quality regulation (Luthfi et al., 2019;Yuniarti & Ariandi, 2017).
Organizational culture has a significant effect on the skewness of accounting fraud. This implies that the institution must continue to improve the existing organizational culture to establish good behavior or habits in employees so that the agency's goals can be achieved. Effective internal control has a negative impact on the desire to do accounting fraud (Pratiwi & Setiyawati, 2021;Setiawan, 2018). The quality of the supervisory team has a positive effect on the desire to commit accounting fraud. Morality does not affect the desire to commit accounting fraud. Internal control has a negative effect on the tendency of accounting fraud (Budiantari et al., 2018;Yelvionita et al., 2020). The outcomes of another research are inversely proportional, namely that the internal control system has a positive and significant effect on the prevention of fraud in OPD Deli Serdang Regency (Bukit & Nasution, 2015;Fernandhytia & Muslichah, 2020).
In line with previous study internal audit factors have a positive impact on fraud prevention variables at state-owned banks in Denpasar (Mahendra et al., 2021). Internal Audit on Fraud Prevention in financial statements has an influence of 38.3% and the remaining 61.7% is influenced by other things that are not studied. There are also other study that examine the effectiveness of internal control was also found to have an impact in fraud prevention in the Telkom Foundation case study (Nurani & Octavia, 2016). The results showed that the higher probability of the effectiveness of the company's internal control, the higher the prevention of fraud is.
The organizational ethical culture does not affect the tendency of fraudulent activities. However, better perceptions of ethical culture are negatively associated with the perceived frequency of job fraud with statistical significance (Akhsani, 2018;Ramantha, 2020). These findings make it clear that investing in an ethical culture is more effective in preventing job fraud. Fraud involves any unauthorized act characterized by deceit or breach of trust. Fraud can be practiced by individuals or organizations to earn money, wealth, avoid losses by engineering, or secure business profits for personal gain (Fachruroji, 2020;Suh et al., 2018). Organizational ethical culture is a member-owned procedure that distinguishes each organization. Organizational culture is the result of the repetitive behavior of its members. If the behavior of members of the organization consistently exhibits ethical behavior then it is a good culture and vice versa. Therefore, a good organizational culture will force members to stay away from fraudulent practices (Albar & Fitri, 2018;Marks, 2009;Wulandari & Nuryanto, 2018).
Research on the ethical culture of organizations discusses honesty and ethics of high value to anticipate fraud. Organizational culture reflects the personality of the organization and shows beliefs, habits, and the end of the result. Therefore the aims of this study is to inspect the determinants of financial statement fraud with case studies at state-owned-entreprise PT POS Indonesia Regional II.

METHODS
A research framework is a design that shows the relationship between theory and various variables identified as research problems in the form of graphs. Pentagon Fraud theory states that fraud can occur if supported by pressure, opportunity, rationalization, competence/ability, and arrogance (Fernandhytia & Muslichah, 2020;Sugiyono, 2018). From the formulation of existing problems, research objectives, supporting theories, and previous research results, all organization requires an internal audit for business readiness in the sense of good business management, provide an assessment of decisions taken by the leadership that ensure the normal and adaptability functioning of activities, aiming to create added value and detect the possibility of fraud (Astuti et al., 2018;Tripermata et al., 2021).
The population of respondents to this research is employees of PT. Pos Indonesia (Persero) works in the Internal Control Unit section of the Jakarta Regional II working area. Because it uses saturated sampling methods, the population and sample of this study are the same, namely 35 internal control unit auditors in the Regional II working area of Jakarta. The saturated sampling method is used by distributing questionnaires to all population members to SPI Representatives at PT Pos Indonesia in the Regional II working area of Jakarta.
Submission of this questionnaire requires a trial to see if the subject can understand the given case. The trial is intended to assure that all participants understand not only every question or statement of the instrument but also the entire instrument with the same mindset. Data analysis using the Partial Least Square (PLS) approach using the help of the SmartPLS program. PLS is a model of Structural Equation Modeling (SEM) equations based on components or variants.

Respondent Overview
The characteristics of the respondents who are employees who are in the internal control division of PT Pos Indonesia Regional II Jakarta are divided into several demographic aspects. These demographics are the gender, age, and last education obtained. As for the exposure of its characteristics, it can be drawn on Table 1. Researchers use the mean value to see how agreed and disapproved respondents to the indicators submitted in variable measurements so that it was concluded that the auditor strongly agreed that carrying out testing and evaluating information in transactions is very important. Respondents also agreed that the internal control system can be carried out by providing numbers on each documentation in order so as not to cause double recording. But not all employees who serve in the control division know business risks.
Respondents agreed on the statement of the company where they worked, the leadership gave clear work directions to new employees very counter to the statement that in the workplace, decisions are more often made by the lower level. It is less ethical if in decision making by subordinates because the full responsibility lies with the superior.
Another finding of the high mean value is the agreement with respondents regarding the preparation of the Budget Realization Report as real conditions because the leadership is threatened by the sanctions of the Law. This is contrary to the disapproval of employees on the statement is not a problem if there is often a shortage of goods received from the purchase of office equipment. Construct variable structural models in smartPLS is show in Figure 1.

Validity Test
The validity test uses convergent validity with indicator criteria considered valid when it has an outer loading value greater than 0.5 and result of validity test presented on Table 2.   Table 3 show outer loading of deleted indicators result. The result found that all of the indicators found invalid result. The expected value of average variance is extracted >0.5 but the result show is not valid. It almost same as the result of average variance extracted as show in Table 4.  Table 4, the expected value of average variance is extracted >0.5 but the test results are not valid. So the indicators of each variable must be checked and the indicators that have the lowest Loading Factor will be removed for retesting. The result of outload value after indicator deleted is show in Table 5.  Table 5, the result of outload value after indicator deleted is declare as valid in all of the variable. Then for the correlation between construct and AVE value result is show in Table 6.  Table 6, the correlation between construct and AVE value is reveal several items is declare valid according to test result.

Reliability Test with Composite Reliability
Reliability tests using composite reliability have a criterion of 0.8. If there is a variable that has a composite reliability value below 0.8, the indicator must be checked again before continuing with the next test. Reliability test results using composite reliability is show in Table 7. Outer loading values to be removed. The following is a reliability test with Composite Reliability after removing the indicator and displayed in the Table 8 and Table 9.  Table 9 Cronbach's Alpha value for all constructs is above 0.6 and the lowest value is in the Internal Audit Variable (X1) with a value of 0.749 so it can be stated that the measurements used in this study have good rehabilitation and high value. However, it must be removed by indicators that have the smallest outer loading value. The following is Cronbach's Alpha value for all constructs after 10 indicators are removed in the Table 10.  Table 10, reliability test after multiple indicators removed is show all of the tested items is show reliable and high reliability. Designing the inner model of R-square before indicator is removed is show in Table 11.  Table 11, R-Square can explain that the internal Audit variable (X1), the effectiveness of the internal control system (X2), the Organizational Ethical Culture (X3), affect the financial statement fraud variable (Y) by 47.1%. The remaining 52.9% was explained by other variables outside the study. From this number, it can be categorized that dependent variables can be interpreted by independent variables on a moderate scale because 47.1% because it is greater than 33% as the limit of the criteria for determining moderate or not the influence of dependent variables.
The adjusted value R-square has an interval value between 0 to 1. If the adjusted value of R-square is close to 1, it indicates that the independent latent variable (X) explains the variation of the dependent latent variable (Y) well. This research, obtained an adjusted value of R-square of 0.420 or 42.0% so it can be summarized that 42.0% of the variations that occur in variable Y can be explained by independent latent variables, while the remaining 58.0% is explained by other variables outside the study. R-square inner model measurement is show in Table 12.  Table 12 explain that the internal audit variable (X1), the effectiveness of the internal control system (X2), Organizational Ethical Culture (X3), affect the financial statement fraud variable (Y) by 41.2%. The remaining 58.2% was explained by other variables outside the study. From this figure, it can be categorized that dependent variables can be explained by independent variables with a moderate scale because 41.2% is greater than 33% as the limit of the criteria for determining moderates in the influence of dependent variables.
From the adjusted value R-square has a value with an interval between 0 to 1. If the adjusted value of R-square is getting closer to 1, it indicates that the independent latent variable (X) explains the variation of the dependent latent variable (Y) well. This study, obtained an adjusted value of R-square of 0.355 or 35.5% so that it can be concluded that the variation of 35.5% variation that occurs in variable Y can be explained by independent latent variables, while the remaining 64.5% is explained by other variables outside the study.

Discussion
Independent variable named internal audit variable has a significant influence on the fraud of financial statements with negative influence. The higher the internal audit rate, the lower the fraud of financial statement. Another result shows that second and third hypothesis are rejected. Independent variable named effectiveness of the control system and the influence of the organization's ethical culture have insignificant influence on the fraud of financial statement. Effectiveness of the control system has positive impact to fraud of financial statement with meaning that the higher the level of effectiveness of internal control system, the higher level of effectiveness of the internal control system the higher fraud prevention. Or in another form, the lower of effectiveness of internal control system the more fraud possibility to happen. To add the results above, the organization's ethical culture has negative influence which means the higher the level of organization's ethical culture the lower the fraud of financial statements.
Cressey's development of the triangle theory of fraud is Fraud Pentagon Theory. A partner-incharge at Crowe Horwath LLP added two elements that drive fraud. This theory is later known as Crowe Horwath's Fraud. Here's the Pentagon fraud theory. Pressure. A person commits fraud in response to financial pressure. According to fraud experts, four pressures can cause fraud, namely: (1) financial pressure, (2) bad habits, (3) pressure from work relationships, and (4) other pressures. Opportunity. Bisnis patterns often provide opportunities for fraudsters so that there is theft or misuse of assets. There are three categories of conditions that create opportunities for financial fraud, namely the nature of the industry, ineffective supervision, and organizational structure (Hamdani & Albar, 2016;Setyaningsih & Nengzih, International Journal of Social Science and Business, Vol. 7, No. 1, 2023, pp. 1-10 8 IJSSB. P-ISSN: 2614-6533 E-ISSN: 2549-6409 2020). Improved internal control systems will reduce fraud. Good internal control can improve employee compliance with existing rules to minimize fraudulent practices.
Rationalization, employees who are accustomed to cheating, will not see if the action is criminal behavior. Rationalization is the existence of an attitude, character, or ethical value that allows a person to perform an act that is dishonest or is in an organizational environment or culture that exerts sufficient pressure to justify the action. Competence, the existence of individual quality or competence to override the control in his company, create strategies to cover up fraud committed and control social situations for personal gain (Ramantha, 2020;Zelmiyanti & Anita, 2015). The capacity of an employee is not only good for the company but can have a bad impact. Employees believe that the rules or procedures contained in the company do not apply to employees. The perpetrators of this scam are completely unaware of the consequences of their actions.
In line with previous study which aims to get empirical proof relating to the results of fraud determinant supported fraud pentagon (Harman et al., 2020). The findings showed that financial stability, external pressure, ineffective monitoring, nature of industry, competence, arrogance, and a frequent number of CEO's picture have no impact on financial statement fraud. Then other study showed that financial stability, the quality of external auditor had a positive effect on the prediction of fraudulent financial reporting, while financial targets, liquidity, institutional ownership, monitoring effectiveness, replacement of external auditors, and changes of corporate directors had no significant effect on the prediction of fraudulent financial reporting (Apriliana & Agustina, 2017). The conclusion of this study indicates that unstable financial condition, the quality of corporate audit can influence the occurrence of fraudulent financial reporting.
Based on the existing conclusions of this study, it is advised that the next research use arrogance as an indicator to see variations in research results in the same dependent variable and add indicators that can facilitate financial statement fraud research in other SOEs. This research shows that several indicators can be taken into consideration for the future such as changes determined based on the leadership's decree, avoiding every transaction has double supporting evidence, maintaining outstanding employees, and business risks that have been analyzed. What can be used as another consideration is taking data offline and using questionnaires or open questionnaires to dig up criticism of suggestions from speakers and the low value of the model obtained, the factors used in this study show the lack of determinant depictions of financial statement fraud. Researchers suggest changing the respondents in the study and adding variables.

CONCLUSION
Internal audits have a significant effect on financial statement fraud. In its implementation, the audit team conducts testing and evaluating transaction information that is at the core of financial statements. The presence of information discrepancies can affect the final result or output. This discrepancy is the beginning of financial statement fraud. One of the efforts to reduce this discrepancy is to emphasize thoroughness in the audit process. Competence in the field of auditing, both in terms of education and professional experience, can reduce the risk of fraud.