BANKRUPTCY PREDICTION ANALYSIS USING THE ZMIJEWSKI MODEL ( X-SCORE ) AND THE ALTMAN MODEL ( Z-SCORE )

DOI: 10.38035/DIJEFA Abstract: This study aims to identify and analyze the accurate models of Financial Distress in retail companies listed on the Indonesian Stock Exchange in 2014-2018 using the Zmijewski (X-Score) and Altman (Z-Score) Model. The sample used is 70. This study uses secondary data from the 2014-2018 annual financial reports. This study tested the hypothesis using the normality test and the Kruskal Wallis test or the difference test using SPSS version 26. The results of this study indicate that the Zmijewski (X-Score) model is the model that has the highest accuracy rate in predicting bankruptcy with an accuracy rate of 90%.


INTRODUCTION
Changing economic conditions will certainly have an impact on economic sectors, particularly the trade sector. Retail business is a business activity that includes selling goods or services directly to end consumers which are used for personal, family, or household needs (Utami, 2008).
With the existence of an online store, it hurts offline stores, if before the consumers shop for their needs at offline stores, now consumers have started to switch to online shop transactions. This is what causes people's attractiveness to offline stores or commonly called retailers to decline and not develop. The general chairman of the Indonesian Employers' Association (Apindo) Hariyadi B. Sukamdani said the continued sluggish trade performance was a reflection of weakening purchasing power or declining household consumption of Indonesians.
PT HERO Supermarket Tbk (HERO) throughout 2018 HERO has closed 26 outlets. On July 28, 2019, HERO closed 6 outlets under the Giant brand. According to HERO director, Hadrianus Wahyu Trikusumo, food retail in Indonesia has experienced increased competition in recent years due to changes in consumer spending patterns. Deputy Chairperson of Aprindo Tutum Rahanta of the Indonesian Retail Entrepreneurs Association (Aprindo) revealed and assessed that efficiency through closure is a natural thing in the retail industry scope to maintain the company's business.
The purpose of establishing a company in the long term is to maximize the value of the company through increasing the prosperity of the owner and shareholders, the higher the value of the company, the more prosperous the owner and shareholders are. The ability of a company to be able to maintain its business continuity to compete is determined by the company's performance. Some companies that cannot maintain their performance will lose to other companies that are more competitive. If the company cannot maintain and manage properly, it will result in a decrease in financial performance and even lead to bankruptcy. Bankruptcy, or the owner of another name bankruptcy, is the risk of business failure that will occur when the fair market value of the assets is less than the company's liabilities or debts (Sunarto, 2006). Bankruptcy is a situation where the company is no longer able to fulfill its obligations because the company has experienced a lack of funds to run its business so that the purpose of establishing the company cannot be achieved. Bankruptcy analysis is very important because it can assess indications of company bankruptcy, whether a company is threatened with bankruptcy or not. Information regarding finances is used by people who are at the same time as people who are at an early age. So that the damage and even the parties to the worst conditions are still very poor (Dirman Angela, 2020).
The research that has been conducted (Huda, Paramita, and Amboningtyas, 2018) is entitled Financial Distress Analysis using Altman, Springate, and Zmijewski Models in Retail Companies Listed on the IDX 2013-2017. The results of his research, namely according to the Altman model, 4 companies were experiencing financial distress, the Springate and Zmijewski models, 2 companies were experiencing financial distress. Zmijewski's model is the model with the highest level of accuracy and has the lowest error rate.
There are differences in the results of other research conducted (Anggraeni and Sitorus, 2019) with the title Comparative Analysis of the Potential Bankruptcy of Retail Companies in Indonesia with the Altman Z-Score, Springate, and Zmijewski models. The results showed that eleven companies experienced bankruptcy for the period 2012-2017 based on the Altman Z-Score, Springate, and Zmijewski models. The eleven companies, namely: CENT, GOLD, HERO, KOIN, LPPF, MAPI, MIDI, MPPA, RIMO, SKYB, and TRIO. Then, it is known that there are significant differences in results between the results of these studies.

Financial Statements Definition of Financial Statements
In the opinion of Hery (2016). "Financial reports are basically the result of an accounting process that can be used as a tool to communicate financial data or company activities to interested parties". According to the opinion of the Indonesian Accounting Association (IAI) (2009) "financial reports are part of the financial reporting process which includes balance sheets, income statements, changes in financial position reports, notes and other reports as well as explanatory material that is an integral part of financial statements". Bankruptcy Definition of bankruptcy Sembiring (2016) suggests that bankruptcy is a failure that occurs within a company when the company experiences an economic failure (economic distressed) and financial failure (financial distressed). Economic failure (economic distressed), which means that the company's revenue is no longer able to cover its costs or that its profit rate is lower than the cost of capital. Financial failure (financial distressed) means that the company fails to fulfill one or more conditions in its debt provisions (technical default), and the company is unable to fulfill its obligations at a predetermined time (technical insolvency).

Bankruptcy prediction
Bankruptcy can be predicted long before the company goes bankrupt. Therefore, bankruptcy cannot be detected in a short time. However, the time used is usually two to five years as a tolerance limit for decreased performance to detect possible company bankruptcy (Triharyanti, 2018 in Bimawiratma 2016). According to the Altman model, there are 4 companies that experience financial distress, the Springate and Zmijewski models, there are 2 companies that experience financial distress. Zmijewski's model is the model with the highest level of accuracy and has the lowest error rate. Adamko, and Chutka (2019) Company bankruptcy and its prediction in conditions of globalization. The accuracy rate of the Fulmer model is up to 98%. By analyzing 60 companies, half of the companies were successful and half of them indicated financial problems. Only predicts bankruptcy, does not test model accuracy. Ditasari, Triyono, and Sasongko (2019). There are differences in clarifying financial difficulties between the Altman and Springate models, Altman and Grover, Altman and Zmijewski, Springate and Grover, Springate, and Zmijewski models. Meanwhile, there is no difference in the clarification of financial difficulties between the Grover and Zmijewski models. Comparison of Altman, Springate, Zmijewski, and Grover Models in Predicting Financial Distress on Companies of Jakarta Islamic Index (JII) on 2013-2017.

RESEARCH METHODS
This study uses secondary data obtained from retail company financial statement data for 2014-2018 from www.idx.id. This study is intended to analyze the comparison of the prediction of Finacial Distress with the Zimjewski (X-core) model and the Altman (Z-Score) model of the 2015 -2017 study and which is the best model used in predicting financial distress in retail companies.

Zmijewski model
Zmijewski used a probit model to predict bankruptcy, Elviani, Simbolon, Riana, Khairani, Dewi, and Fauzi (2020). An empirical test is conducted on a financial distress model that requires a sample of companies that experience financial pressure and companies that do not. The equation for the function of the Zmijewski model is as follows: X-Score = -4.3 -4.5X1 + 5.7X2 -0.004X3 Where: X1 = Net Income / Total Assets X2 = Total Liabilities / Total Assets X3 = Current Assets / Current Liabilities Zmijewski model categories, namely: a. If X> 0 the company is predicted to go bankrupt, b. If X <0 the company is predicted not to go bankrupt. Altman model Fitriyanti, and Irni (2015) in Sari, (2018) suggest that the Altman model bankruptcy analysis is one of the tools used to predict the level of bankruptcy of a company by calculating the value of several ratios and then entering it into a discriminant equation. Altman's model discriminant equations are as follows (Hanafi and Halim 2009 (2013), the accuracy rate shows the percentage of the model in predicting the company's condition correctly based on the entire sample. The following is a formula for calculating the level of accuracy and type of error: Tingkat Akurasi = number of predistions is correct x 100% Number of samples In addition to the level of accuracy, this study also analyzes the percentage of error types. According to Prihadi (2010), each model always has the possibility of incorrect predictions and a difference in the level of 100% accuracy. The prediction tool is said to be true if the predicted and actual are the same, while an error occurs when the predicted and actual ones are not the same. Errors arising from prediction tools consist of:    (2020) In table 4. the results of the normality test using the One-Sample Kolmogorov -Smirnov test, the sig value is more than 0.05, which means that the data is normally distributed, namely the Zmijewski model is 0.200, while the Altman model has an abnormally distributed value of 0.022.
Because the data were not normally distributed, a different test was performed using the Kruskal Wallis test. The Kruskal Wallis test is an alternative test for the One Way ANOVA test. Because the data were not normally distributed, the One Way ANOVA test was discontinued, so it used the Kruskal Wallis test. The results of the Kruskal Wallis test are shown in Table 5. In Figure 5 the results of the Kruskal Wallis test, the sig value shows the value of 0.000, which is the sig value <0.05, which means that there is a significant difference between the two bankruptcy prediction models. 1. Model Zmijewski  Table 7 that the results of the calculation of the accuracy of the Altman model are 74.29%, for type I error is 0%, type II error is 2.86%, while the gray area is obtained 22.86%.
The Altman model that is predicted to go bankrupt is PT Matahari Putra Prima Tbk (MPPA), while it is predicted to be in the gray category of PT Catur Sentosa Adiprana Tbk There are seven companies predicted to experience bankruptcy from the four models, namely ECII, AMRT, CSAP, HERO, KOIN, MAPI, and MPPA. 2. Zmijewski model (X-Score) is the most accurate bankruptcy prediction model compared to Altman model (Z-Score). The accuracy value of the Zmijewski model (X-Score) is 90% with a type I error of 0%, while a type II error is 10%. Suggestion 1. There are limitations in this study, namely that it only uses two bankruptcy prediction models and the period for which it is carried out is only five years. Therefore, the researchers suggest that further research can use other models such as the Sringate model, Ohlson, Fulmer, Zavgren, CA-Score, and other bankruptcy prediction models. And extend the research period so that the sample used is also increasing. 2. For companies, to be able to consider the results of the Zmijewski model to minimize or avoid the risk of bankruptcy.