THE EFFECT OF FRAUD DIAMOND ON FRAUDULENT FINANCIAL STATEMENT IN ASIA PACIFIC COMPANIES

- This study aims to determine how the influence of fraud diamonds in detecting fraudulent financial statements in companies in Asia-Pacific listed in S&P in 2017-2019. The study used a purposive sampling method with a total population of 228 companies with a sample of 78 companies and used the STATA program for analysis. The results of the hypothesis testing carried out were that the pressure on financial stability had an effect on fraudulent financial statements with a significant level of 0.000 which was smaller than the significant level in this study which was 0.0101 so that it is accepted with the understanding that financial assets in the company have an effect on committing fraud, the opportunity for the nature of industry to affect the fraudulent financial statement with a significant level of 0.001 where smaller than the significant level in this study which is 0.01 so it is accepted with the understanding that warehouse inventory can be a trigger for someone to commit fraud, the rationalization of total accruals has an effect on fraudulent financial statements with a significant level of 0.003 which is smaller than the significant level in this study which is 0.01 so it is accepted with the understanding that the total accruals owned by the company have an impact because someone can commit fraud by falsifying total accruals, and the ability to use a dummy has no effect on fraudulent financial statements with a significant level of 0.295 which is greater than the significant level in this study which is 0.01 so that it is rejected with the understanding that the change of directors is not a factor in someone committing fraud.

financial position, notes and other reports, and explanatory material that is still an integral part of a financial report. Meanwhile, according to PSAK No.1 (2015: 1) says that the financial report is a structured presentation of the financial position and financial performance of an entity. Each company will produce financial reports that aim to make it easier for the company, provide guidance within the company for forecasting business growth, and provide guidelines in assessing funding and operational activities. The function of financial reports is to be used as review material, guidelines in making decisions, creating new strategies, and increasing company credibility. There are 4 tigers of financial reports, Profit and Loss Statements, Change in Capital Reports, Balance Sheet Statements, and Cash Flow Statements.

Fraud
Fraud (fraud) is an intentional act committed by one or more individuals in management or those responsible for governance, employees, and third parties that involve the use of deception to gain an unfair advantage or break the law (IAPI, 2013). According to Rozmita (2013), fraud is a deviation, error and irregularity in financial problems (irregularities). According to The Association of Certified Fraud Examiners (ACFE), which is a professional organization engaged in fraud examination. They classify fraud into three levels called the Fraud Tree, including: Deviation of assets, false or misstated statements, and corruption. COSO issues five aspects of control to prevent fraud, namely the control environment, risk assessment, control standards, information and communication, and monitoring.

Fraud Triangle
The fraud triangle is a theory that has been stated by Donald R. Cressey after he did more in-depth research in 1950. Cressey said that someone can commit fraud when the person has financial problems so that the person thinks that financial problems can be settled by their position in which the position is trusted by the company to hold the assets. The fraud triangle was first pointed out in SAS No. 99 which is standardized in the United States so that according to SAS No.99 the fraud triangle consists of pressure, opportunity, and rationalization. These three components encourage someone to commit cheating. Pressure related to management and other employees has pressure in committing fraud, opportunity is an opportunity that arises before there is fraud, and justification for action is a justification for fraud committed by the perpetrator.

Fraud Diamond
Fraud diamond is a new model for improving the fraud triangle theory proposed by Wolfe and Hermanson in 2004 by adding a capability element to enhance this model. They argue that a lot of fraud that is carried out by withdrawing large amounts of money will not occur if you do not have special abilities in the company. The fraud diamond theory has four important elements, including pressure which has four conditions, namely financial stability, external pressure, personal financial need, financial targets, opportunity which has two conditions, namely Nature of industry and effective monitoring, justification for actions (rationalization) and capability

Fraudulent Financial Statement
According to Karyono (2013: 17), fraudulent financial statements are the overstatement or understatement of the company's financial statements. By increasing the financial statements, you will benefit from the sale of shares and it is not difficult to get financing. Meanwhile, when the financial statements are published, it aims to avoid taxes so that they are not too high. International Standard Auditing states that there is fraud in financial statements that results in intentional misstatement, negligence in presenting amounts (Tuanakotta, 2014: 204). According to IAI (2001), there are two misstatements, they can be due to errors or fraud. Intentional misstatement in financial statements aims to gain personal gain.

Fraud Score (F-Score)
The F-score was first introduced by Dachow et al., In 2009, according to Sukrisnadi (2010, saying that the F-score is a form of measurement that can be claimed as a tool to detect material misstatements in financial reports. According to Dechow et al., (2009) in their research, he said that this model has an accuracy rate of 68-70% but this accuracy returns to the ongoing fraud case. The F-score has two components, namely accrual quality and financial performance which are defined as follows: accrual quality and financial performance. The purpose of developing the F-Score is to be able to calculate directly with the information in the financial statements and make it easier for readers to find out which financial statements are misstated and not misstated. The guidelines used to assess the F-Score in measuring the risk of material misstatement in the financial statements include:

Framework
This study uses a fraud diamond as a fraud detector in financial statements where the fraud diamond was initiated by Wolfe and Hermanson in 2004 which has four elements, namely pressure, opportunity, rationalization, and capability.

RESEARCH METHOD 2.1 Population, Sample, Data Source
The population selected by the researcher were all manufacturing companies registered with the S&P. In selecting the sample to be used, the researcher chose to use purposive sampling. According to Jogiyanto (2010), purposive sampling is taking samples from the population based on certain criteria. The method used by researchers to collect data is the quantitative method, by looking at the financial statements of companies in Asia-Pacific on the S&P website (https://platform.marketintelligence.spglobal.com).

Fraudulent Financial
Statement (

Definition of Operational Variables • Fraudulent Financial Statement (FFS)
The formula for the fraudulent financial statement ratio:

• Financial Performance = Change Receivable + Change in Inventory + Change in Cash Sales + Change in Earning
• Financial Stability (FS) The ratio formula for financial stability uses the ratio of changes in assets as follows: • Nature Of Industry (NOS) Nature of Industry uses the following formula: The formula for rationalization is: • Capability: Change of Directors Based on research conducted by Sihombing and Rahardjo (2014), they say that the change of directors is with dummy variables, where 1 = there is a change of directors for 3 years and 0 = no change of directors for 3 years.

Data analysis method
The method used in this research is to use the application of multiple regression models with the Stata software program. Researchers use stata because they provide features of analysis and data processing that are more complete and efficient than other programs. The analysis technique used is descriptive statistics, correlation test, classic assumption test in the form of normality test, multicollinearity test, heteroscedasticity stest, determination coefficient test, simultaneous significance test (F statistical test), and t statistical test.

Description of Research Objects
The object of research taken is in the form of companies operating in Asia-Pacific and registered with S&P. This study used a purposive sampling method based on predetermined criteria for data collection. From the results of data collection in the S&P, there were a sample of 228 that met the predetermined criteria. The number of samples taken has met the criteria for the central limit theory which states that a large number of samples can be said to be normally distributed if the sample is 30 unless there is a limited sample. There are several characteristics used in this study so that the research results are more accurate, including companies operating in Asia-Pacific that are registered with S&P and have been operating for at least 5 years and the financial statements used for 2017-2019 are in English/Indonesian. The table above is an overview of the five variables used in conducting this research using 178 financial reports for the 2017-2019 period. It can be interpreted as follows: Financial Statement Fraud (ffs): the mean value of the Financial Statement Fraud (ffs) of 228 data for the 2017-2019 period is 0.0816134 which has a standard deviation of 0.4291384. The maximum value of this variable is 1.103613 and the minimum value, the minimum value is -2.911279. Financial Stability (fs): the mean value of Financial Stability (fs) is 0.0909252 which has a standard deviation obtained is 0.2038128. The maximum value of this variable is 2.029183 and the minimum value is -0.3230963. Nature of Industry (nos): the mean value obtained from Nature of Industry (nos) is 0.0259506 which has a standard deviation obtained in this variable is 0.2340939. The maximum value obtained is 1.946091 and the minimum value is -1.395996.Razionalization (ta): the mean value of rationalization (ta) is -0.0146818 which has a standard deviation of 0.099817. The maximum value obtained is 0.2803544 and the minimum value is -0.6581951 Capability (cap): the mean value obtained from this Capability (cap) variable using the dummy is 0.5789474 which has a standard deviation obtained is 0.4948143, meaning that there is ± 49.48143% there is a deviation. The maximum value obtained is 1 and the minimum value is 0. In the table above, you can see that there is an asterisk (*) on the correlation test between financial stability (fs) and fraudulent financial statement (ffs) resulting in a coefficient of 0.4457. Meanwhile, nature of industry (nos) with fraudulent financial statement (ffs) produces a coefficient of -0.5070. The result of the correlation test between rationalization (ta) and fraudulent financial statement (ffs) produces a coefficient of 0.4851. The result of the correlation test between rationalization (ta) and financial stability (fs) produces a coefficient of 0.2114. In the results of this correlation test, there is no indication of multicollinearity because the coefficient on each variable is not greater than 0.8 and insignificant because it is <0.1, so it is said not to experience multicollinearity.

Variabel
Obs Pr(Skewness) Pr(Kurtosis) adj. Chi2(2) Prob>Chi2 ffs 228 0.0000 0.0000 . 0.0000 Based on the normality test carried out, it can be seen that the data used has not been normally distributed with Prob> z of 0.00000 so that this data needs to be treated so that it can be normally distributed using the skewness-kurtosis test.

Variabel
Obs Pr(Skewness) Pr(Kurtosis) adj. Chi2(2) Prob>Chi2 bc_ffs 160 1.0000 0.7185 0.13 0.9371 From the results of treatment it can be said that the data used in the study are still not normal, but with data of more than 200 observations the researcher can ignore the results of the treatment and the data is considered normal. So the data that have not been distributed normally after the skewness-kurtosis test treatment is normal.
Based on Figure 4.1. Related to the results of the normality test in the form of a histogram graphic, it gives a pattern that is close to normal distribution, but if you pay attention to the figure, there are some data that are far on the left side of the distribution pattern, while in Figure 4.2. Related to the results of the normality test in the form of a normal probability plot graph, it shows the distribution points on the diagonal line and the distribution follows the direction of the line. So it can be concluded that the regression model for this study meets the normality assumption test. The multicollinearity test serves to find out whether there is a correlation between the independent variables with one another by looking at the Tolerance or Variance Inflation Factor (VIF) value. If the Tolerance value > 0.1 or VIF < 10 then there is no multicollinearity problem. Based on the test results in Table 5, it can be seen that there is no high multicollinearity problem between the independent variables because the value generated in this test is a mean VIF value of 1.03 where the anchor is less than 10. The heteroscedasticity test has the aim of finding out whether the residual variance from one observation with other observations is fixed or not. Based on the results of this test, it can be seen that there is no problem in heteroscedasticity because the value of chi2 (1) is 0.39 smaller than the value Prob> chi2 with a value of 0.5335 and the probability value is greater than 0.

Determination Coefficient Test
The coefficient of determination serves to show the magnitude of the model's ability in research to explain the independent variables. The coefficient of determination test has a weakness in the use of R-Square, namely the emergence of bias in the number of independent variables and how to overcome the problem of bias that arises from this R-Square, so this test will focus on the Adjusted R-Square as a complement.Based on table 7, the Adjusted R-Square value is 0.2197 which means that the total of fraudulent financial statement variables can be explained by the variables of financial stability, nature of industry, rationalization and capability of 21.97% and the remaining 78.03% is explained by other external variables. this research model.

Simultaneous Significance Test (Test Statistic F)
The F statistic test aims to find out whether the independent variable in the study affects the dependent variable.Based on table 7, the calculated F value is 12.19 with Prob> F of 0.0000 which means that the variables financial stability, nature of industry, rationalization, and capability simultaneously affect the fraudulent financial statement variable. With a probability value of 0.0000 <0.05, H0 is rejected so that the independent variable can simultaneously influence the dependent variable.

Multiple Linear Regression Analysis
Based on Table 7, the regression model used is as follows: FFS = α0 + α1FSit + α2NOSit + α3TAit + α4CAPit + eit = -1.348837 + 0.9674339 FSit -0.7859848 NOSit + 1.52573 TAit -0.0771113 CAPit + eit The definition of the regression model above is: 1. A constant value of 0.0749752 with a negative direction means that if FS, NOS, TA, and Cap are not valued or zero, then the FFS will be worth -1.348837. 2. The FS regression coefficient value is 0.9674339 with a positive direction, so if the other independent variables still mean that each FS increment is 1 unit, then there is a potential increase of 0.9674339 units in the fraudulent financial statement. 3. The NOS regression coefficient value is 0.7859848 with a negative direction. If the other independent variables still mean that each increase in NOS is 1 unit, there is a potential decrease of 0.7859848 units in the fraudulent financial statement. 4. The TA regression coefficient value is 1.52573 with a positive direction, so if other independent variables still mean that each TA increase is 1 unit, there is a potential for an increase of 1.52573 units in the fraudulent financial statement. 5. The CAP regression coefficient value is 0.0771113 with a negative direction, so if the other 6. independent variables still mean that each increase in CAP is 1 unit, then there is a potential decrease of 0.0771113 units in the fraudulent financial statement.

Statistical test t
The t statistic test aims to show how much influence the independent variable has in explaining the dependent variable. Based on the coefficient and t-count in table 7, it is found that prob> F is 0.000 with a significance of 0.01 or 1 percent with a t-test of 2.576. If seen in the table, the FFs variable gets the number 0.000, Nos gets the number 0.001, ta gets the number 0.003, and the cap gets the number 0.295 so that only the capability variable is rejected because the value is greater than 0.05 and the rest is accepted with the understanding that there is a relationship between the independent variable and the variable dependent.

The Effect of Financial Stability on the Fraudulent Financial Statement
Based on the results of the regression analysis on the financial stability variable using changes in total assets as a proxy in this study, it has a coefficient of 0.9674339 and a significant level of 0.000 <0.01. The test results mean that financial stability has an effect on fraudulent financial statements so that the higher the change in total assets of a company, the value of asset growth of the company also increases, and the lower the potential for fraudulent financial statements. So based on the results in table 7, hypothesis 1 is accepted.