DETECTING OF INDICATION FINANCIAL STATEMENT FRAUD

This study aims to determine the percentage of manufacturing companies categorized as manipulators, non-manipulators, or gray companies from 2016 to 2020. This study uses the Beneish M-Score index ratio as a data analysis technique. The population comprises 198 manufacturing companies listed on the IDX during that period, with a final sample size of 90 companies. Simple random sampling is used for the research sample. The findings reveal the following percentages for each category: In 2016, manipulators accounted for 19.05%, non-manipulators for 80.95%, and there were no gray companies. In 2017, manipulators constituted 42.86%, non-manipulators 57.14%, with no gray companies. For 2018, manipulators were 47.62%, non-manipulators 52.38%, and no gray companies. In 2019, manipulators were 26.19%, non-manipulators 73.81%, and no gray companies. Finally, in 2020, manipulators accounted for 30.95%, non-manipulators for 66.67%, and gray companies for 2.38% of the sample. These findings provide valuable insights into the distribution of manipulator, non-manipulator, and gray companies within the manufacturing sector over the specified five-year period.


INTRODUCTION
Fraud is a feature of every organized culture in the world and it affects many organizations, regardless of size, location, or industry (Skalak et al., 2011). Association of Certified Fraud Examiners (2020) in the Report to The Nations 2020, estimated 5 percent of revenue in organizations was lost each year, approximately more than US$ 3.6 billion of 2,504 cases from 125 countries. In Indonesia itself, there are 239 fraud cases with approximately lost to Rp 873 billion (ACFE Indonesia Chapter, 2020). In cases of fraud that have occurred, financial statement fraud is the least common case with 10 percent occurring but the median loss affected is most costly up to US$ 954 thousand (Association of Certified Fraud Examiners, 2020). ACFE Indonesia Chapter (2020) examined that financial statement fraud is one of the fraud schemes that commonly occurred in Indonesia. In their report, ACFE Indonesia Chapter recaps 22 cases of financial statements fraud which are 9.2 percent of the whole fraud cases with a total loss of up to Rp 242 billion.
Manufacturing companies are one of the industries sectors that most commonly occurred in financial statement fraud (Association of Certified Fraud Examiners, 2022). Recently in Indonesian manufacturing companies, financial statement fraud occurred in PT. Tiga Pilar Sejahtera (TPS) for 2017. The manipulation of the TPS Food Financial Statements in 2017 was carried out by inflating the receivables of six affiliated distributors from the actual Rp200 billion to Rp1.6 trillion.
In 2001, financial statement fraud also occurred in PT. Indofarma and PT. Kimia Farma. Based on The Capital Market Supervisory Agency (Bapepam, now OJK) investigation, it found that Indofarma overstated the value of work in process. As a result, the cost of goods sold is understated and net income is overstated by the same value. Meanwhile, in the same year, Kimia Farma reported a large net profit which caused the management to carry out a re-audit on October 3, 2002, on Kimia Farma's 2001 financial statements restated, because there is a fairly basic error that was found. In the new financial report, it turns out that the company's profit was only Rp 99.56 billion, Rp 32.6 billion lower, or a decrease of 24.7 percent from the initially reported profit.
The financial statement is one of the most important pieces of information needed by users in considering decisions about their business. Therefore, the presentation of financial statements that can attract the attention of other parties such as investors is an important thing that must be done by companies. However, to achieve those goals several companies sometimes commit fraudulent financial statements by manipulating the numbers in their financial statement.
Hence, because there is a possibility of manipulation in financial statements, a technique for detecting financial statement fraud is needed. Beneish (1999) conducted a study to detect the indication of manipulation known as Beneish M-Score. Beneish Mscore is a mathematical model that uses eight financial ratios to identify manipulated earnings and to detect financial statement fraud. The variables are constructed from the company's financial statements and a score is derived from the model to describe the degree to which the earnings have been manipulated. However, the Beneish M-Score has limitations to detect the manipulation that uses scheme understatement profit. Besides that, due to the Beneish M-Score being a probabilistic model, the result cannot detect fraud perfectly a hundred percent.
This study is a replication of previous research conducted by Christy & Stephanus (2018). The main difference from the study is the company sector to be addressed. This study will be addressed to the manufacturing sector listed in IDX for 2016 -2020. Meanwhile, the previous research addressed the banking sector listed in IDX for 2014 -2016.

LITERATURE REVIEW Financial Statement
According to PSAK 1, a financial statement is a report that is intended to meet the needs of users who are not in a position to request specific financial statements to meet the information needs of these users. The objective of financial statements is to provide information about an entity's financial position, financial performance, and cash flow to make decisions (Ikatan Akuntan Indonesia, 2018).

Fraud
AICPA (2021) formally defined fraud as an intentional act by one or more individuals among management, those charged with governance, employees, or third parties, involving the use of deception that results in a misstatement in financial statements that are the subject of an audit.
Subsequently, the ACFE classifies fraud into three types of occupation, namely; corruption, asset misappropriation, and financial statement fraud (Association of Certified Fraud Examiners, 2022).

Fraud Detection
Fraud detection is an act to obtain the occurrence of fraud, who is the perpetrator and the victims, and what caused it. According to Skalak et al. (2011), the key to fraud detection is the ability to spot errors and irregularities.
To detect the occurrence of fraud, an understanding of the fraud scheme is one of the important things that must be attached. According to Wells (2017), financial statement fraud schemes fit into five broad classifications, which are: (1) fictitious revenues, (2) timing differences, (3) concealed liabilities and expenses, (4) improper disclosures, and (5) improper asset valuation. Each type of fraud needs different ways to detect it. For instance, Beneish Mscore will be used to detect a fraud occurrence in the financial statement.

Beneish M-Score
Beneish M-score was developed by Professor Messod Beneish in 1999. It is a mathematical model that uses eight or five financial ratios to identify manipulated earnings and to detect financial statement fraud. The variables are constructed from the company's financial statements and a score is derived from the model to describe the degree to which the earnings have been manipulated. Furthermore, the eight variables of Beneish M-score are calculated using the following formula: The previous formula is derived from the probability of occurrence manipulation that is characterized by a remarkable increase in accounts receivable, worsening gross profit, asset decline, sales growth, and an increase in accruals. Specifically, the eight variables of Beneish M-score are explained below.

Day Sales in Receivables Index (DSRI)
DSRI is calculated by comparing receivables in the first year that the manipulation is discovered (year t) to the same measure in year t-1 based on sales. It assesses whether receivables and revenues are in balance over two years (Aghghaleh et al., 2016).

Gross Margin Index (GMI)
The GMI is calculated as a ratio of total sales revenue minus the cost of goods sold divided by sales in year t-1 compared to the measurement in year t. A GMI value greater than one indicates a decline in gross margins, which is associated with worsening business prospects and a higher chance of manipulation.

Asset Quality Index (AQI)
The AQI is calculated by dividing the percentage of assets that are intangible this year by the same calculation last year. An increase in AQI may reflect higher expenditures that are capitalized to maintain profitability, which is expected to enhance the potential for manipulation (Warshavsky, 2012).

Sales Growth Index (SGI)
The SGI is a ratio of revenue growth from one year to the previous year. In the year under review, an index greater than 1.0 indicates positive growth, while less than 1.0 indicates negative growth.

Depreciation Index (DEPI)
The DEPI ratio is calculated by dividing the depreciation rate in year t-1 by the depreciation rate in year t, with the reasoning that reduced depreciation expenditures result in more discretion over revenue, and hence a larger risk of manipulation.

Sales, General and Administration Index (SGAI)
SGAI is calculated by comparing sales in year 1 to the same measure in year t-1. A greater SGAI is considered to enhance the chance of manipulation.

Leverage Index (LVGI)
The LVGI ratio compares total debt to total assets. An index larger than 1.0 is viewed as an increase in the company's leverage and, as a result, is vulnerable to manipulation (Mahama, 2015).

Total Accruals to Total Assets Index (TATA)
TATA is used to determine how much of a company's sales are done on a cash basis. It is an indicator of the company's cash flow quality. Total accruals are calculated as the change in net income less cash flow from operating activities.
Furthermore, the formula to calculate each ratio index used in Beneish M-score is described in Table 1.

RESEARCH METHODOLOGY Population and Sample
The population in this study is 198 manufacturing companies listed on Indonesian Stock Exchange (IDX). To determine the population that fulfills the research objectives, the population that has the following characteristics are excluded as described in Table 2.
3 Company unissued its financial statement during 2016 -2020. 4 Company not generating a profit during 2016 -2020 The population that meets the criteria 90 Years of observation 5 The total number of the target population 450 The sampling technique used in this study is probability sampling using simple random sampling. To draw the amount of selected sample, this study used the Slovin formula with a margin of error of 5 percent, thus the amount of sample used in this study is 211.76.
After the amount of sample observation found, it is divided into five years which resulted in 42 companies as research samples.

Data Sources and Data Collection Method
This study uses secondary data as the data source. Secondary data means that the data is collected by others for different purposes from the current study (Sekaran & Bougie, 2016). Thus, the secondary data used in this study are the published financial statements of manufacturing companies listed on the Indonesia Stock Exchange (IDX) for the 2016-2020 period. Meanwhile, this study used relevant journals, books, theses, and published financial statements of manufacturing companies listed in IDX during 2016 -2020 to collect the data.

Data Analysis Technique
This study uses the index ratio analysis technique to the company's financial statement data as the research sample. The index ratio calculation is used as a benchmark to determine whether companies are classified as manipulators or non-manipulators. Companies are categorized as manipulators or nonmanipulators based on their M-score obtained. The steps used in conducting data analysis are as follows: 1. Analysis of the company's financial statements using the Beneish ratio index to obtain the Mscore (formula in

RESULTS AND DISCUSSION
The data in this study is analyzed in several steps, comprising the analysis of financial statements using the Beneish ratio index to obtain the M-Score, determining whether companies are categorized as manipulators or non-manipulators using certain terms, and calculating the amount of percentage companies categorize as a manipulator, non-manipulator, or gray companies.

Result of Beneish M-Score and Categorization of Company
The result of the computation of each ratio index is used to generate the M-Score with a defined formula. After that, the Beneish M-Score will determine whether companies are categorized as manipulators if the M-Score is > -2.22 or nonmanipulators if the M-Score is < -2.22. Therefore, according to the formula and certain terms, the Beneish M-Score resulting from the research sample is presented in Table 3. Those result data are summarized in the following Table 4.

Calculate the Percentage of Each Category
Using a certain formula, the percentage of each category is described in the following Table 5.

Manipulator Companies
According to the analysis using the Beneish M-Score model conducted on 42 manufacturing companies listed in Indonesian Stock Exchange (IDX) in 2016 -2020, 8 companies were detected as having indications to commit fraud in their financial statements in 2016. In 2017 and 2018, the number of manufacturing companies that indicate committed financial statement financial fraud increased to 18 and 20 companies. Meanwhile, in 2019 it decreased to 11 companies and increased again to 13 companies in 2020. Sequentially from 2016 -2020, the percentage of companies that are indicated to have fraud on their financial statement is as much as 19.05%, 42.86%, 47.62%, 26.19%, and 30.95%.

Non-Manipulator Companies
According to analysis using the Beneish M-Score model conducted on 42 manufacturing companies listed in Indonesian Stock Exchange (IDX) in 2016 -2020, 34 companies were detected as not manipulating their financial statements in 2016. In 2017 and 2018, it decreased to 24 and 22 companies. Meanwhile, in 2019 it increased to 31 companies, and in 2020 decreased again to 28 companies. Sequentially from 2016 -2020, the percentage of companies that are categorized as non-manipulator is as much as 80.95 %, 57.14%, 52.38%, 73.81%, and 66.67%.

Gray Companies
According to the analysis using the Beneish M-Score model conducted on 42 manufacturing companies listed in Indonesian Stock Exchange (IDX) in 2016 -2020, from 2016 until 2019 there are no manufacturing companies that are categorized as gray companies. Meanwhile, in 2020, one research sample company has a Beneish M-Score value of -2.22 and was categorized into a gray company which is PT Wilmar Cahaya Indonesia Tbk.

CONCLUSIONS AND SUGGESTIONS Conclusions
According to the analysis, results, and discussion, it is concluded that in 2016, 8 research samples were indicated as manipulators, 34 were categorized as nonmanipulator, and none were categorized into a gray company. In 2017, 18 research samples were indicated as manipulators, 24 were categorized as nonmanipulator, and none were categorized as a gray company. In 2018, 20 research samples were indicated as manipulators, 22 were categorized as nonmanipulator, and none were categorized as a gray company. In 2019, 11 research samples were indicated as manipulators, 31 were categorized as non-manipulator, and none were categorized as a gray company. Last, in 2020, 13 research samples were indicated as manipulators, 28 were categorized as nonmanipulator, and 1 was categorized as a gray company.

Suggestions
Indicated manipulators companies are advised to check the company's internal for the possibility of fraud that has escaped the supervision of internal and external auditors. In other words, it can be a warning or red flag for the occurrence of financial statement fraud. For investors or creditors, it is recommended to minimize the risk of investment loss by considering indicated manipulators companies. Then, for further researchers, it is recommended to use the wider companies sector to get a broad view of the research sample.