IPO審查階段的公司盈餘管理初探：以中國大陸之國營企業及民營企業之比較為例 - 政大學術集成

全文

(1)National Chengchi University in Taiwan Department of Economics, College of Social Science Master’s Thesis. A comparative analysis of pre-IPO earnings management between 治 SOE and NSOE in 政大 China. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Advisor: Shinn-Shyr Wang, Wen-Chieh Lee Author: Takashi Miura. 2016/05/31.

(2) 2. Abstract: This study investigates an appropriate accrual-based model in detecting earnings management (EM) of Chinese firms undergoing initial public offering (IPO) process. The mainstream literatures show that state-owned enterprise (SOE) has involved with less earnings management (EM) than non-state-owned enterprise (NSOE) in pre-IPO market from 2003 to 2009. The reason is that SOE could borrow money from bank, compared to NSOE. However, SOE has been proven to conduct stronger EMs in pre-IPO market during 2011 to 2013 by manipulating with property plant and equipment (PPE), the flows of account receivable (AR) and long-term debt (LTD). Besides, I also suggest a new. 政治大. accrual-based model that could better fit into the unique features of Chinese firms in their process of IPO application.. 立. ‧. ‧ 國. 學. Key Words: Earnings management, State-owned enterprise, Non-state-owned enterprise, IPO market, Property plant and equipment, Account receivable, Long –term debt.. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(3) 3. Table of Contents I.. INTRODUCTION ........................................................................................................... 5. II.. LITERATURE REVIEW .................................................................................................. 7 Review of earnings management in China .......................................................................... 7 Research method in detecting earnings management .......................................................... 7 Intertwined relationship between Chinese companies and government ................................ 8 Accounting items relating to discretionary accruals ............................................................ 9. III. HYPOTHESIS DEVELOPMENT ................................................................................... 10. 政治大 Application of the accrual model in the developing counties .............................................. 10 立. Previous studies on basic regression model ....................................................................... 10. The accrual model with Chinese characteristic ................................................................. 11. ‧ 國. 學. Factor of earnings management in Chinese pre-IPO companies ........................................ 12. ‧. IV. STATISTICAL ISSUES .................................................................................................. 13 MEASURING DISCRETIONARY ACCRUALS .............................................................. 15. V.. y. Nat. sit. Data Selection .................................................................................................................. 15. n. al. er. io. Decomposition of Total Accrual ........................................................................................ 15. i n U. v. Accrual-Based Model ....................................................................................................... 16. Ch. engchi. i.. The modified Jones model ........................................................................................... 16. ii.. Yoon (2006) model ..................................................................................................... 16. iii.. The new model ....................................................................................................... 18. VI. EXPERIMENTAL DESIGN ........................................................................................... 19 Sample Construction ........................................................................................................ 19 i.. Utilized Sample .......................................................................................................... 19. ii.. Identification of SOE .................................................................................................. 19. Data Analysis ................................................................................................................... 19 Difference Test for Earnings Management ........................................................................ 20 VII. THE EMPIRICAL RESULTS ......................................................................................... 21.

(4) 4. CONCLUSIONS AND IMPLICATIONS ............................................................ 35. 立. 政治大. 學 ‧. ‧ 國 io. sit. y. Nat. n. al. er. VIII.. Ch. engchi. i n U. v.

(5) 5. I. INTRODUCTION Motivation: A straight thought held by operational firms is to maximize profits by enhancing revenues and lowing costs in the competitive market. Thus firms listed in the stock trading markets would have superfluous incentives to maintain very good performance in their periodically published financial statements. For the good figures in sales and costs control would convince the market to have high evaluation in their good standing of operational. 政治大. status. One arguable action taken by firms is to involve in the earnings management that manipulate several core accounting courses to beautify firms’ overall performance in a. 立. gray zone that legal action would be difficult to regulate. In general, accrual items such. ‧ 國. 學. as account receivable, account payable and inventory are decomposed into discretionary accrual (DA) and non-discretionary accrual (NDA) over the investigation of EM.. ‧. Artificial increase of earning created by DA is regarded as EM. Higher ranked management are undoubtedly weaving the bright stories demanded by the associated EM.. sit. y. Nat. Evidence revealed by Benmelech (2008) and Qiang (2005) show that CEO has incentive to conceal bad news for future performance and manipulate earnings that will consist with. io. n. al. er. investor’s expectation to sell more shares in post-IPO market. Besides, many studies. i n U. v. assumed that company appropriates higher earnings by using accrual accounting items in. Ch. engchi. pre-IPO market. Consequently, many alternative accrual-based models were created and applied for EM studies. We can date back to the work done by Dechow (1995) that the modified Jones model is the most effective accrual-based model in detecting EM. Many researches focusing on EM detection employed the modified Jones model. However, Dechow (1995) also warned that accrual models are unlikely to detect EM of economically plausible magnitudes. In addition, McNichols (2001) pointed that the accrual models are likely to suffer from serious omitted variables problem because the omitted variables represent noise which reduces power and creates type II error. However, none of the prior EM research focused on the necessity of model specification for preIPO Chinese companies. If the model is misused for EM study, it will induce wrong results. Therefore, it is necessary for EM study to specify the use of the accrual model..

(6) 6. In this study I investigate a new accrual-based model and the factor of EM for pre-IPO companies. At first, Aaker (2010) identified the unique “inside information” into the accrual process for a specific industry. It allows to explain accruals into managed and unmanaged components. Next, Cheng (2015) revealed that companies classified into SOE and NSOE showed different magnitude of EM in China. It is because SOE is supported by government as well as state-owned bank (SOB). In addition, SOE hires local auditor who provides low quality of audit Wang (2008). According to these studies, EM study needs to be compared across industries and company’s type. The rest of paper is organized as follows. First, I provides the hypothesis. 政治大. development. Secondly, I test three different accrual-based model. Thirdly, I conduct difference test to find which accounting items affect the magnitude of DA in both SOE. 立. and NSOE. Fourthly, I report the empirical results. Finally, I present the conclusion for. ‧. ‧ 國. 學. io. sit. y. Nat. n. al. er. this study. Ch. engchi. i n U. v.

(7) 7. II. LITERATURE REVIEW Review of earnings management in China Study of EM is ubiquitous all over the world. The purpose of those studies is to estimate the value of IPO companies and avoid high investment risk. In China Experts often employed the modified Jones model to investigate earnings management for IPO companies. The method is also supported by Dechow (1995) which stated that the model. 治政大(2015) reported that CEO has method has also been-used for some applied research. Cai 立 incentives to manage earnings, especially around IPO years. Jerry (2014) examined that is the most effective in detecting EM, compared to other accrual models. Recently, their 1. 2. ‧ 國. 學. 506 companies with larger managed accruals performs better on the first trading.3 Cheng (2014) found that the magnitude of EM is different between SOE and NSOE because. ‧. SOE can access more loans from banks. There studies indicated that Chinese IPO companies have managed earnings while undergoing IPO process, and especially SOE. y. Nat. sit. holds the informational superiority to investors. Therefore, Xu found that Chinese. al. er. io. domestic IPO issuers did took advantage of the over optimism of investors and. n. successfully sold overpriced initial shares during 2010 to 2011.. Ch. engchi. i n U. v. Research method in detecting earnings management Dechow (1995) reported that all the accrual-based model may not be very well to detect EM. In this respect the modified Jones model appeared to be the most powerful among all the models despite its limited efficiency in detecting EM. Thus, many experts started to utilize the modified Jones model and applied the results for further researches. 1. Cai (2015) uses data from 2011 to 2012 and applied the cross-section modified Jones model.. 2. Jerry (2014) uses data from 1998 to 2003 and applied the modified Jones model.. 3. Cheng (2014) uses data from 2003-2009 and applied the cross-section Jones model..

(8) 8. However, they also mentioned that their model cannot extract economically plausible results for EM test. Aaker (2010) revealed that the Jones specified models are very sensitive with respect to firm performance, but inventory accruals alone provided considerable increase in power of model. Yoon (2006) found that the modified Jones model is less effective in Korea. In addition, Islam (2011) and Alareeni (2014) examined that the modified Jones model does not exercise well in Bangladesh and Palestine. Therefore, these experts argued that the use of the modified Jones model is not appropriate model and, therefore, they applied Yoon (2006) model to detect EM. However, Alareeni (2014) stated the Yoon (2006) model is not effective when their results in Palestine compared to the results of Islam (2011) in Islam.. 立. 政治大. ‧. ‧ 國. 學. Intertwined relationship between Chinese companies and government Chinese companies developed under unique financial environment. Panel A. Nat. sit. y. introduces the historical flow of Chinese financial deregulation by Hujita (2014) from. io. er. Panel A of Appendix 1. It indicates that Chinese financial deregulation was constructed step by step since 1980. However, even though China has started to establish strong. n. al. Ch. i n U. v. regulation and open their financial market, the relationship between companies and. engchi. government created serious financial issue since then. Qian (2008) found that Chinese SOE is more likely to hire small local auditors.4 Chen (2010) stated that Chinese SOB is more supportive for SOE and also, they found that the relationship between companies and SOB prompts conservative accounting as companies get more loans. Cheng (2014) mentioned that SOE has less EM because of easier access for long-term debt (LTD) from SOB. These studies showed that although huge financial issue was generated by government intervention, they still remain the relationship. Chinese government. 4. Chen (2010) uses data from 2001-2006..

(9) 9. Accounting items relating to discretionary accruals According to the PwC China Assurance Partner, Doing business and investing in China, it stated that either CAS 2006 or ASBE is accepted for delivering financial. statement in China. These two standards require to use different accounting method, such as cost method and impairment method. And choosing the difference method allows companies to manipulate earnings. Ball (2008) and Armstrong (2008) found that strong monitoring and auditing can restrict EM in the developed countries. On the contrary,. Cheng (2015) they argued that China’s stock market exhibits typical features of an. 政治大. emerging equity market: ineffective internal and external monitoring, weak legal system EM, a low level of investor protection, and an underdeveloped institutional environment.. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(10) 10. III.. HYPOTHESIS DEVELOPMENT. Previous studies on basic regression model Cheng (2015) uses the cross-sectional modified Jones model to estimate the DA. The model is introduced by Defond (1994) who assumed that the magnitude of NDA in the same industry is the same. However, the assumption is not seemingly correct. First, each company has different products and holds different scale of asset. Therefore, the magnitude of the variances for revenue will vary in the cross-sectional modified Jones model. Secondly, although the industry classification allows us to investigate more details. 政治大 industry after the classification. Thirdly, the model is potentially lack of explanation 立 power with only three independent variables even though it is verified theoretically about peer groups companies, the number of observations will be different across the. ‧ 國. 學. Hujino (2009). And hence, application of other models should be considered for the research of EM. Since CEO is able to manage credit sales to control revenue, the. ‧. modification version of Jones model is more suitable, according to Dechow (1995). The model deducts account receivable from revenue and focus on the credit sales in the event. y. Nat. al. er. io. sit. period results from EM.. v. n. Hypothesis 1: The modified Jones model is not effective detecting EM in China.. Ch. en. hi. i n U. g c in the developing counties Application of the accrual model Recently, Ball (2008) and Armstrong (2008) reported that strong monitoring and auditing can restrict EM in the UK and the U.S. However, weaker regulation as well as loose auditing would breeds the vigorous EMs in the market. In China Wang (2008) reported that SOE hires local auditors who provide lower-quality of auditor’s check. And Cheng (2015) argued that the Chinese financial regulation is still under developing. Therefore, SOE may be able to manage earnings. However, the modified Jones model may not have economically plausible in detecting EM by Dechow (1995). In fact, the model is mostly applied for developed countries. It means that it cannot be applicable under strong monitoring and market regulation. Yoon (2006) introduced their model is.

(11) 11. more effective than the modified Jones model in Korea because their prior research found that income-increasing firms frequently employ non-cash revenues including assetdisposition gains, increases in accounts receivables and increases in inventories. Incomedecreasing firms generally employ on-cash expenses including bad-debt expenses, assetdisposal losses and increases in accounts payables. Yoon (2006) model was used in Korean with these accounting items which include features of Korean companies. In addition, Islam (2011) and Alareeni (2014) applied Yoon (2006) model in Bangladesh and Palestine, respectively. From these results, Yoon (2006) models catch some characteristic of developing countries.. 政治大. Hypothesis 2: Yoon (2006) model is not effective detecting EM in China.. 立. ‧ 國. 學. The accrual model with Chinese characteristic Because Yoon (2006) stated that their model especially focused on unique. ‧. characteristic of Korean companies, therefore Alareeni (2014) found that Yoon (2006) model does not show better goodness of fit in Palestine. McNichols (2001) pointed that. Nat. sit. y. the accrual models are likely to suffer from serious omitted variables problem because. io. er. the omitted variables represent noise which reduce power and create type II errors. From this perspective, it will be necessary for the accrual model to include specific features of. n. al. Ch. i n U. v. Chinese companies. Baolong (2015) stated that when Chinese company appropriate. engchi. account receivable, inventory, fixed asset and intangible asset, the price of accounting item depends on accounting method either Chinese Accounting Standard (CAS 2006) or the New Accounting Standard for Business Enterprises (ASBE) to. Particularly, the impairment test and cost method is different between CAS 2006 and ASBE. ASBE allows company to choose the First In First Out (FIFO), the Last In First Out (LIFO), the weighted average cost method or the specific identification method to assign the accrual cost of inventories. But CAS 2006 does not allow to use the LIFO. For the impairment test ASBE allows to reverse the price, but CAS 2006 does not allow to reverse the price. These accounting methods affect the price of particular accounting items. And it will be possible for Companies to manage earnings by switching the method. Therefore, accrual-.

(12) 12. based model should include the change of account receivables, inventories and impairment in each years and consider those effect for extracting exact magnitude of EM.. Hypothesis 3: The new model is not effective detecting EM in China.. Factor of earnings management in Chinese pre-IPO companies Cheng (2015) found that SOE accesses more bank loans in comparison with NSOE. Chen (2010) found that there is a negative relationship between CEO’s incentive. 治政 their earnings due to outside support. However, their大 investigation is based on the 立might not be suitable in detecting EM in China if the model modified Jones model, which. of the SOE and bank loans. Their finding suggests that SOE have less incentive to manage. ‧ 國. 學. does not have enough effectiveness. Therefore, it is necessary to investigate which accounting items affects the magnitude of EM with the most suitable accrual-based model.. ‧. Hypothesis4: SOE has less incentives to manage earnings because of bank. n. al. er. io. sit. y. Nat. loans.. Ch. engchi. i n U. v.

(13) 13. IV.. STATISTICAL ISSUES. This section elaborates on the potential misspecification problems with the current accrual-based models due to the lack of relevant variables, mismeasurement of collected raw data and the wrong classification of data. Fist, two variables of the modified Jones model is not plausible for the study of Chinese IPO companies. (1) The lagged total asset is a stock variable, but total accrual is flow variable. (2) The PPE accounts for the non-current accrual, which expect to explain non-discretionary accrual. However, PPE itself is also the stock variable, therefore, it. 政治大 factors suggests the lack of explanation power in the Modified Jones model. 立. should be replaced by other feasible variable to explain non-discretionary accrual. Above. ‧ 國. 學. Secondly, I suggest that the Modified Jones model can be applied in any kinds of data; however. Yoon (2006) model can be used for panel data only.. ‧. Yoon (2006) model uses the panel data analysis with between data when Yoon. y. Nat. and Miller applied it for detecting earnings management in Korea because the model. al. er. io. (ΔEXP-ΔPAY)/REV.. sit. shows multicollinearity with the high correlation between (ΔREV-ΔREC)/REC and. n. v i n Where REV = net sales revenue,C REC receivables, h e= ntrade h i U EXP= sum of cost of goods c g sold and selling and general administrative expenses excluding non-cash expenses, PAY= trade payables and Δ=change operator. The panel data analysis is corrected over cross sectional data and times series data.. Baltagi (2008) and Hsiao (2003) pointed that the advantage of panel data analysis is (1) Panel data can eradicate multicollinearity, increase the degree of freedom and improve the unbiasedness. (2) it can control the individual diversity and let us know the common effect in data. In other hand, the negative aspects of panel data analysis is to affect the quality of statistics, for example, the feature of times series data will be minimize if the analysis uses between data. From this perspective Yoon (2006) model does not consider the time effect on their analysis..

(14) 14. VIF (variable influence factor) or eigenvalue of the correlation is used to detect the multicollinearity by Paul (2006). Multicollinearity will be recognized if the independent variables are highly correlated or perfectly correlated. Because the variable influence is overlapped on dependent variable if some variables in multi-regression are highly correlated. Therefore, multicollinearity is crucial matter when the study need to detect pure DA in accrual-based model.. 𝑉𝐼𝐹𝑗 = 𝐶𝑗𝑗 = (1 − 𝑅𝑗2 )−1. Where the 𝑅𝑗2. 政治大 is the coefficient of 立determination obtained when one of the variable 𝑋. 𝑗. ‧ 國. 學. regressed on the remaining j-1 independent variables.. is. If the variable 𝑋𝑗 is hardly linearly independent on other j-1 independent variables, the. ‧. VIF will be larger. Klein, (1962) suggests that “alternative criterion that if R exceeds R2 of the regression model. In this vein, if VIF is greater than 1/(1-R²) or a tolerance value. io. sit. y. Nat. is less then (1-R²), multicollinearity can be considered as statistically significant.”. n. al. er. If the test need to get rid of the multicollinearity it needs simply to omit one of. i n U. v. the high correlated independent variables from the regression model and test VIF.. Ch. engchi. However, it creates the omitted variables bias on the model interpretation. It is necessary to add the other variable to substitute the omitted variable in the model..

(15) 15. V. MEASURING DISCRETIONARY ACCRUALS. Data Selection The listed company’s financial data is retrieved from Chinese Stock Market and Accounting Research (CSMAR) database. Company’s data consists of Shanghai stock exchange and Shenzhen stock exchange where each stock market is divided into A share and B share, respectively. Panel B shows the total number of listed company’s data during. 政治大 The data includes only non-financial company and excludes some company that enters 立 in the IPO during 2010-2014. Each industries are classified by CSMAR code and. 2011-2013 is 1663 and each financial data is quarterly accumulated during a fiscal year. 1. ‧ 國. 學. separated into 17 groups. The codes and the results for the classification are shown in the Panel B of Appendix 1.. ‧. North American financial data consists of 2834 listed-companies collected via. sit. y. Nat. Compustat. The data is also classified into 14 groups, which are consistent with Chinese industries groups in order to compare two countries. However, there are three industries. io. n. al. i n U. and 2GICS. It can be seen from the Panel C of Appendix 1.. Ch. e. i. er. are not shown on the Panel C due to the different classification method between CSMAR. v. h Accrual n of g cTotal Decomposition The accruals accounting is a simple method to display earnings growth on financial statement. The method in decomposing accruals into discretionary and nondiscretionary accruals is used to generate more accurate conclusion from arbitrary. Financial company’s financial statement style and accounting item differs from other industries. Because high EM is observed around IPO period thereby, the data excluded some company that entered IPO from 2009 to 2012.. 1. GICS sector and group codes are used for the industry classification in North America. 2.

(16) 16. management of accruals in general. In my research the total accrual (ACC) is calculated by net income minus operating cash flow.. Accrual-Based Model i.. The modified Jones model Dechow (1995) revealed that the change of sales will include discretionary earnings in the Jones model. Because total sales is consisted by pure sales earnings and accrued account. Therefore, deducting account receivables from sales is needed to explain non-discretionary accruals since all independent variables should have power to explain the non-discretionary accruals in total accruals. The inverse lagged. 政治大 the depreciation cost as the non-current accruals. 立 1 𝑇𝐴𝑖𝑡−1. + 𝛼2. (∆𝑅𝐸𝑉𝑖𝑡 −∆𝑅𝐸𝐶𝑖𝑡 ). 𝑃𝑃𝐸𝑖𝑡 𝑇𝐴𝑖𝑡−1. + 𝜀𝑖𝑡. REV is equal to Sales.. io. al. er. sit. y. 𝑇𝐴𝑖𝑡−1. + 𝛼3. ‧. 𝑇𝐴𝑖𝑡−1. = 𝛼0 + 𝛼1. Nat. Where. 𝐴𝐶𝐶𝑖𝑡. 學. (1). ‧ 國. total asset variable indicates the size of individual companies and PPE accounts for. n. v i n lessC receivable h e n gincyear h i 𝑡 −U1 scaled by total assets at. (∆𝑅𝐸𝑉𝑖𝑡 − ∆𝑅𝐸𝐶𝑖𝑡 ) represents revenue in year 𝑡 less revenue in year 𝑡 − 1 minus the receivable in year 𝑡 1.. 𝑡−. 𝑃𝑃𝐸𝑖𝑡 = gross property plant and equipment in year 𝑡 scaled by total assets at 𝑡 − 1. 𝑇𝐴𝑖𝑡−1 =total asset in year 𝑡 − 1. 𝛼0 =constant variable and others are coefficients.. ii.. Yoon (2006) model Except for the sales transactions, Yoon (2006) model considers operating accounts and non-operating accounts in detecting EM. Because the model assumed that Korean companies will be intended not only to delay payment but also to.

(17) 17. increases frequency of gathering receivable when they want to manipulate earnings. They revealed that Korean companies are more likely to use the operating accounts if their income increases. On the contrary, the income decreasing companies are intended to use non-current accounts. 1 They decomposed current accruals into receivable, inventory, payable and other current accruals as well as non-current accruals into depreciation expense, bad debt expense,. retirement benefits expenses,. losses from the disposition of assets or redemption of liabilities, gains from the disposition of assets or redemption of liabilities and other non-current accruals. 2 Yoon (2006) mentioned that the model has multicollinearity due to the high correlation between first and second independent variables. It is shown on the table 2. 治政大is substituted into impairment cost of goods sold and the third independent variable 立 loss and gain in this study. = 𝛼0 + 𝛼1. (∆𝑅𝐸𝑉𝑖 −∆𝑅𝐸𝐶𝑖 ) 𝑅𝐸𝑉𝑖𝑡. + 𝛼2. (∆𝐸𝑋𝑃𝑖 −∆𝑃𝐴𝑌𝑖 ) 𝑅𝐸𝑉𝑖. + 𝛼3. (𝛥𝐼𝑀𝑃𝑖 ) 𝑅𝐸𝑉𝑖. + 𝜀𝑖. ‧. Where. 𝐴𝐶𝐶𝑖. 𝑅𝐸𝑉𝑖. 學. (2). ‧ 國. and 3. Because of the unavailable data in China the second EXP variable include only. Nat. sit. y. (∆𝑅𝐸𝑉𝑖𝑡 − ∆𝑅𝐸𝐶𝑖𝑡 ) represents revenue in year t less revenue in year t − 1 minus. io. al. er. the receivable in year t less receivable in year t − 1 scaled by revenue at t − 1.. n. v i n C h the payable in year less expense in year t − 1 minus e n g c h i U t less payable in year t − 1. ∆𝐸𝑋𝑃𝑖𝑡 − ∆𝑃𝐴𝑌𝑖𝑡 represents expense including only cost of goods sold in year t scaled by revenue at t − 1.. (𝛥𝐼𝑀𝑃𝑖𝑡 ) represents the change of the impairment losses and gains in year t.. 1. In my research, depreciation expense is included in the model as a Non-current accrual because other non-current accruals are less available from research data. 2. According to the VIF and correlation result from the table, multicollinearity must be considered in Yoon and Miller model. The strong correlation between sales and cost of goods sold can be seen. It may be a serious issue to apply this model. By Paul (2006), the model need to be omitted one of the highly correlated variables from the model because you cannot know the relationship between dependent variable and independent variables while including multicollinearity..

(18) 18. The new model Yoon (2006) model is created for the study of Korean companies. Figure 1 shows that the change of impairment loss and gain is quite different between 2011~2012 and 2012~2013 in China. Moreover, the change of inventory is larger than North America. It is because the price reverse is implemented under different financial regulation system. If the company uses CAS 2006, they are no longer able to reverse the impairment loss and gain. On the contrary, if the company uses ASBE, they can reverse the price from the impairment loss to the fair value. Secondly, the change of inventory is shown much higher level that can be seen from Figure 1 because a company can choose the FIFO, the LIFO, the weighted average cost method or the. 政治大 But the LIFO is not allowed under the CAS 2006. On the balance sheet data, 立 specific identification method to assign the actual cost of inventories under ASBE.. inventories are measured at the lower of cost and net realizable value”, according to. 學. ‧ 國. Balong (2015). Thus, including both accounting items should be considered for the new model for the study of EM in China.. ‧. In addition Bernard (1996) and McNichols (2001) mentioned that “EM by. Nat. sit. y. working capital accruals receivable, inventory and some non-operating components should be considered in the change of sales. It will be crucial to deduct necessary. io. er. iii.. variables when the study investigates a particular industry such as personal computer,. n. al. Ch. i n U. v. oil and dresser industry when these components caused large part of earnings. engchi. management.” Especially, manufacturing industry occupy the largest number of IPO companies in China, absence of those working capital accruals components for EM will cause the omitted variable problem.. (3). 𝐴𝐶𝐶𝑖𝑡 𝑅𝐸𝑉𝑖𝑡. = 𝛼0 + 𝛼1. (∆𝑅𝐸𝑉𝑖𝑡 −∆𝑅𝐸𝐶𝑖𝑡 ) 𝑅𝐸𝑉𝑖𝑡. + 𝛼2. ∆𝐼𝑀𝑃𝑖𝑡 𝑅𝐸𝑉𝑖𝑡. + 𝛼3. ∆𝐼𝑁𝑉𝑖𝑡 𝑅𝐸𝑉𝑖𝑡. + 𝜇𝑖𝑡. Where ∆𝐼𝑀𝑃𝑖𝑡 represents impairment loss and gain in year t less impairment loss and gain in year t − 1 ∆𝐼𝑁𝑉𝑖𝑡 represents inventory in year t less inventory in year t − 1.

(19) 19. VI.. EXPERIMENTAL DESIGN Sample Construction. i.. Utilized Sample Chinese pre-IPO accounting item data is collected from the financial statement of individual companies. Panel B shows 213 SOEs and 209 NSOEs and total 422 pre-IPO companies’ data.. Identification of SOE. 政治大. The classification of SOE is determined by the stock holder’s holding ratio.. 立. If the holder’s holding ratio excess 40% and they are associated with. 學. government, the company is regarded as SOE. Cheng (2015) classified SOE. ‧ 國. if the company’s ultimate owner is a government officer or state-owned. ‧. companies or local government. In the Companies Act of Republic of China if CEO is a government officer or the holder occupies more than 50% of. Nat. sit. y. stock, the company is regarded as SOE.. io. Shareholding rate of companies can be seen in the financial statement [the. n. al. section: the basic situation of issuer].. Ch. e nAnalysis gchi Data. er. ii.. i n U. v. The magnitude of EM will be. (4). 𝐷𝐴𝑖𝑡 =. 𝐴𝐶𝐶𝑖𝑡 𝑅𝐸𝑉𝑖𝑡. − [𝛼̂0 + 𝛼̂1. (∆𝑅𝐸𝑉𝑖𝑡 −∆𝑅𝐸𝐶𝑖𝑡 ) 𝑅𝐸𝑉𝑖𝑡. + 𝛼̂2. ∆𝐷𝐸𝑃𝑖𝑡 𝑅𝐸𝑉𝑖𝑡. + 𝛼̂3. Where 𝛼̂0 , 𝛼̂1 , 𝛼̂2 𝑎𝑛𝑑 𝛼̂3 are an estimated coefficient from listed companies. 𝐷𝐴𝑖𝑡 represents a discretionary accrual in industry i and in year t.. ∆𝐼𝑁𝑉𝑖𝑡 𝑅𝐸𝑉𝑖𝑡. ].

(20) 20. Difference Test for Earnings Management Yoon (2006) investigated mean differences in order to know the discretionary level for individual components between income-increase and income- decrease companies. The regression results allow to observe the systematic differences and determine which accounting items are statistically significant on DA. This study utilizes their idea to detect the factors of EM between SOE and NSOE. Furthermore, Chen (2010) argued that there is negative relationship between bank loans and CEO’s incentives. And. 政治大. Cheng (2015) found that the negative relationship in Chinese IPO companies and interpreted that the relationship refers less EM in SOE. Therefore, I added long-term debt. 立. (LTD) and short-term debt (STD) for the difference test to check the relationship between. ‧ 國. DA = 𝛼0 + 𝛼1 ∆𝐴𝑅 + 𝛼6 ∆𝑜𝑡ℎ𝑒𝑟 𝑟𝑒𝑐 𝑎𝑐𝑐𝑟𝑢𝑎𝑙𝑠 + 𝛼2 ∆𝐴𝑃 +. ‧. (5). 學. EM and bank loans. The model for difference test is as follow.. 𝛼6 ∆𝑜𝑡ℎ𝑒𝑟 𝑝𝑎𝑦 𝑎𝑐𝑐𝑟𝑢𝑎𝑙𝑠 + 𝛼3 ∆𝐼𝑀𝑃 + 𝛼4 ∆𝐼𝑁𝑉 + 𝛼5 𝑠𝑡𝑑𝑒𝑏𝑡 +. y. Nat. er. io. al. sit. +𝛼7 𝑝𝑝𝑒 + 𝛼8 𝐿𝑇𝐷 + 𝜀. n. ∆𝐴𝑅=account receivable in year t less account receivable in year t-1;. Ch. i n U. v. ∆𝑜𝑡ℎ𝑒𝑟 𝑅𝐸𝐶 𝑎𝑐𝑐𝑟𝑢𝑎𝑙𝑠=other receivable accruals in year t less other actuals in year t-1;. engchi. ∆𝐴𝑃=account payable in year t less account payable in year t-1; ∆𝑜𝑡ℎ𝑒𝑟 𝑃𝐴𝑌 𝑎𝑐𝑐𝑟𝑢𝑎𝑙𝑠=other payable accruals in year t less other actuals in year t-1; ∆𝐼𝑀𝑃=impairment loss and gain in year t less impairment loss and gain in year t-1; ∆𝐼𝑁𝑉=inventory in year t less inventory in year t-1; stdebt=short-term debt in year t; LTD= long-term debt divided by liability: 𝑝𝑝𝑒=property, plant and equipment in year t. ※ Above regression is standardized by Revenue, except for the long-term debt..

(21) 21. VII. THE EMPIRICAL RESULTS Table 1 shows that each model has different goodness of fit with adjusted R squared and t-ratio at p<0.05 level of confidence. The data is collected from CSMAR database and X, Y, Z represent independent variables for each models. I conducted regression test with between data of panel data. As you can see from Table 1, the result of the modified Jones model (C9, B, D, E,K) and Yoon (2006) model as well as The new model (C1, C4, C5, C7, F) describe higher goodness of fit (Adjusted R squared>50%) with more than one independent variable is statistically significant at p<0.05 level of confidence.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(22) 22. Table1 Regression tests for comparative analysis of alternative models. The results are based on a sample of 3326 firm-years. Modified Jones model. Yoon (2006) model. The new model. Industry. n. 𝑋1. 𝑋2. 𝑋3. adjR2. (C1) food (C2) textile (C4) paper (C5) chemical (C6) electronics (C7) metal (C8) machinery (C9) medicine (A) farm (B) mining. 138 92 62 340 160 292 582 198 50 124. 3.84 1.24 1.51 2.91 -1.42 0.84 9.9 5.27 1.61 -1.91. 8.26 0.57 -0.75 -2.94 2.24 -1.02 2.07 17.17 0.3 -0.96. 0.353 0.146 0.05 0.494 0.05 0.444 0.15 0.794 0.012 0.912. 7.63 3.42 -5.68 -4.97 -5.14 -90.08 -8.8 -0.65 -2.59 0.13. -7.36 -2.17 5.31 0.85 6.6 27.29 9.72 2.36 2.86 0.01. 10.57 -0.9 4.03 -5.9 -9.45 -2.95 -7.97 0.5 0.48 2.24. (D) utilities. 164. -0.3. 15.64. 0.7. -1.17. 1.66. -1.81. (E) construction (F) transportation (G)information (H) retail sale (J) real estate (K) public facility. 98. -1.04. -5.7 -3.2 -0.93 2.02 -2.8 -6.09 -6.03 -0.16 -0.95 11.56 18.74 -1.48. 27.38. 1. -0.07. -0.13. -4.43. 0.147. 146. 1.23. -0.97. -0.93. 0.011. 2.32. 1.17. -12.07. 0.505. 4.33. -12.28. 1.26. 0.505. 196 270 302. 1.91 -0.73 2.09. -6.34 -3.37 -1.45. 3.12 -2.05 2.64. 0.206 0.051 0.035. -3.65 2.7 -2.12. 3.52 -4.08 1.09. 2.61 -3.21 1.11. 0.104 0.138 0.009. -1.63 -1.36 -2.13. -2.95 -6.42 1.07. 2.09 5.09 -6.54. 0.067 0.165 0.13. 112. 0.78. -3.14. 2.3. 0.539. -3.5. 3.78. -0.98. 0.098. -0.84. -0.07. 1.64. 0.003. 𝑌3. 𝑍1. 𝑍2. 𝑍2. adjR2. 0.828 0.103 0.657 0.815 0.436 0.973 0.189 0.029 0.117 0.114. -10.04 2.69 -6.47 -9.87 0.8 -40.5 0.16 3.37 -0.33 0.25. 10.97 -1.46 4.96 -74.71 -7.78 0.77 -5.64 0.21 -0.87 2.35. 10.52 6.44 -3.77 55.1 -0.32 2.7 0.23 8.24 6.03 0.68. 0.868 0.358 0.591 0.981 0.28 0.907 0.057 0.26 0.419 0.118. 0.017. 0.44. -1.64. 0.88. 0.005. -1.18. -4.72. 5.01. 0.327. 政治大. er. io. sit. y. ‧. Nat. al. 𝑌2. 學. ‧ 國. 立. 𝑌1. adjR2. n. 𝑋1 = 1/TAi,t-1; 𝑋2 = (ΔREVi,t-ΔRECi,t)/TAi,t-1; 𝑋3 = PPE/TAi,t-1 𝑌1 = (ΔREVi,t-ΔRECi,t)/REVi,t; 𝑌2 = (ΔEXPi,t-ΔPAYi,t)/REVi,t; 𝑌3 = (ΔIMPi,t)/REVi,t 𝑍1 = (ΔREVi,t-ΔRECi,t)/REVi,t; 𝑍2 = ΔIMP/REVi,t; 𝑍3 = ΔINVi,t/REVi,t Ho: the each independent variable is not significant. H1: the each independent variable is significant. If t-ratio implies that Ho holds mean that independent variable does not have significant power to explain the dependent variable. Since the defree of freedom is more than 29, the coefficient variable will be significant with 2.045>tratio at p<0.05 level of confidence.. Ch. engchi. i n U. v.

(23) 23. The effectiveness in detecting EM is heavily affected by the selection of independent variable. If the model eliminates the overlapped effect among independent variables, it will provide the pure magnitude of EM. Therefore, the correlation effect on dependent variable needs to be tested by VIF, which allows to observe whether the model includes multicollinearity or not. If the model shows multicollinearity, independent variables are highly correlated. Table 2 indicates that Yoon (2006) model has higher possibility of multicollinearity, but other models from Table 2. In addition this finding is also stated in the Yoon (2006). They argued that multicollinearity is caused by the high correlation between 𝒀𝟏 and 𝒀𝟐 .. 立. 政治大 Table 2. 𝑿𝟑 2.34 1.53 0.427 0.573. Mean 1.94. 𝒀𝟏 4.38 2.09 0.229 0.771. 𝒀𝟐 4.52 2.13 0.221 0.779. 𝒀𝟑 1.08 1.04 0.93 0.07. Mean 3.32. 𝒁𝟏 1.02 1.01 0.98 0.0203. 𝒁𝟐 1.01 1 0.993 0.0074. 𝒁𝟑 1.01 1.01 0.987 0.0131. Mean 1.01. sit. 𝑿𝟐 2.32 1.52 0.431 0.569. The new model (obs=3326). ‧. 𝑿𝟏 1.16 1.08 0.864 0.136. Nat. Variable VIF SQRT VIF Tolerance R². Yoon (2006) model (obs=3326). y. Modified Jones model (obs=3326). 學. ‧ 國. Comparative VIF test for overall regression test for alternative models.. n. al. er. io. 𝑋1 = 1/TAi,t-1; 𝑋2 = (ΔREVi,t-ΔRECi,t)/TAi,t-1; 𝑋3 = PPE/TAi,t-1 𝑌1 = (ΔREVi,t-ΔRECi,t)/REVi,t; 𝑌2 = (ΔEXPi,t-ΔPAYi,t)/REVi,t; 𝑌3 = (ΔIMPi,t)/REVi,t 𝑍1 = (ΔREVi,t-ΔRECi,t)/REVi,t; 𝑍2 = ΔIMP/REVi,t; 𝑍3 = ΔINVi,t/REVi,t. Ch. engchi. i n U. v. In Table 3 I implemented correlation test for Yoon (2006) model and the result points out that there is a high positive correlation between 𝒀𝟏 and 𝒀𝟐 . This result tells us that Yoon (2006) model includes overlapped effect, therefore, it is not reliable for detecting EM in China..

(24) 24. Table 3 Correlation test for Yoon (2006) model (obs=3326). 𝒀𝟏. 𝒀𝟐. 𝒀𝟏. 1. 𝒀𝟐. 0.874. 𝒀𝟑. 1. -0.0855 -0.196 1 𝒀𝟑 𝑌1 = (ΔREVi,t-ΔRECi,t)/REVi,t; 𝑌2 = (ΔEXPi,t-ΔPAYi,t)/REVi,t; 𝑌3 = (ΔIMPi,t)/REVi,t. North America data is collected from GICS of Compustat and Chinese listed. 政治大. company’s financial data is retrieved from CSMAR. The total number of North America listed companies are 2834 and they were classified into 14 industries and 1663 Chinese. 立. listed companies were classified into 17 industries. The classification code is different. ‧ 國. 學. between GICS and CSMAR. This study uses the simulation to specify industry with group code and business type. Figure1 shows that bar 1 represents the change of accounting. ‧. item in 2012 and bar 2 represents the change of accounting item in 2013. According to the graph, the change rate of receivable and payable is similar between China and North. Nat. sit. y. America, but the change of inventory shows higher rate in China, compared to North. io. er. America. The change of impairment cannot be compared because the impairment test employs different method in both countries. North America companies uses IFRS which. n. al. i n U. v. does not allow them to reverse the price unless the industry has unique characteristic. For. Ch. engchi. example, the inventories are held by producers of agricultural and forest products, minerals and mineral products. Therefore, the sign of impairment appears to be negative on the financial statement of North American companies and much less companies appropriates impairment loss. While North America does not show enough data on impairment, the change of impairment dramatically change in Chinese companies during 2011 to 2013..

(25) 25. Figure 1 Time Series for the variable change of accounting items, each variable is standardized by Revenue. Year 1 represents the change from 2011 to 2012 and Year 2 represents the change from 2012 to 2013. Sample consists of 2834 firm-year selected from GICS and 1663 firm-year selected from CSMAR. North America ΔAccount Receivable. China ΔAccount Receivable. 政治大. 立. ‧ 國. 學. North America Δ Account Payable. China Δ Account Payable. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. North America Δ Impairment loss and gain. Only (200/total 2834) is available from North America database. Their impairment is attributed to the goodwill.. i n U. v. China Δ Impairment loss and gain.

(26) 26. North America Δ Inventory. China Δ Inventory. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(27) 27. Since the modified Jones model and Yoon (2006) model do not consider direct effect of impairment and inventory, the model is susceptible to catch the unique characteristic of accounting items of Chinese companies. Table 4 and Table 5 display the regression result from three accrual-based model. The modified Jones model shows better goodness of fit in (C5, D, E, K) in 2012. Yoon (2006) model and the new model shows better goodness of fit in (C1, C4, C6, C7, C8, F) in 2012. The modified Jones model shows better goodness of fit in (C7, C9, B) in 2013. Yoon (2006) model shows better goodness of fit in (C1, C4, C5, B, H) and The new mode shows better goodness of fit in (C1, C4, C6, C7, C8, F) in 2013. At a first sight it looks like Yoon (2006) model and the new model is the same result as Table 1. However, the regression was tested in 2012 and. 政治大 2012 and 2013. As the result, modified Jones model, Yoon model and the new model are 立. 2013, separately. Table 6 and 7 indicates that every model includes multicollinearity in. required to being omitted the collinearity to obtain enhanced result of DA. C1, C2, C4,. ‧ 國. 學. C5, C6, C7, C8, C9, F, J from the new model and D from the Modified Jones model in 2012 as well as C1, C4, C5, C7, A, E, F, H, J from the new model and C8, C9 from the. ‧. modified Jones model in 2013 is used to calculate DA. Consequently, Hypothesis 1 (The. y. Nat. modified Jones model is not effective detecting EM in China) is rejected and Hypothesis. sit. 2 (Yoon (2006) model is not effective detecting EM in China) is not rejected and. al. er. io. Hypothesis 3 (The new model is not effective detecting EM in China) is rejected. These. n. v i n characteristic of NDA when we C apply for particular industries and year in China. h ethem ngchi U results tell us that the modified Jones model and the new models are able to catch unique. However, the use of them depends on what type of data we use. For example, Yoon (2006). model needs to be tested with panel data due to the multicollinearity. Also, there should exist other accounting items which can enhance the model explanatory power in each model..

(28) 28. Table4 Comparison tests for EM based on alternative models to measure DA. Each model estimated coefficient which is extracted from Chinese listed-companies in 2012. Sample of 1663 firm-year is collected from CSMAR. Modified Jones model in2012 Yoon (2006) model in 2012 The new model in2012 𝑋1. 𝑋2. 𝑋3. (C1) (C2) (C4) (C5) (C6) (C7) (C8) (C9) (A) (B) (D) (E) (F) (G) (H) (J) (K). 69 46 31 170 80 146 291 99 25 62 82 49 73 98 135 151 56. 71928167* 3522013 1.06E+08 -6994109 -1.4E+07 17665636* -16576679*** 9209719** 49541595 -12462771** 5140941 -5144216 128132849*** 1242692* 2338227 6265176 14910097. -0.43*** -0.20* -0.13 -0.10** -0.39 0.04 -0.15*** 0.07 -0.02 0.05 -0.10*** -0.10 -0.065 -0.22*** -0.14** -0.19 -0.10. 0.14*** 0.01 -0.19 0.03 0.66* 0.03 0.09*** 0.18* -0.00 0.06** 0.22*** 0.23*** -0.13 0.05* -0.01** -0.15** 0.02. adj. R2 0.34 0.10 0.00 0.90 0.08 0.08 0.27 0.13 0.04 0.24 0.87 1.00 0.26 0.23 0.12 0.07 0.70. 立. ‧ 國. N. 𝑌1. 𝑌2. 𝑌3. -0.76** -0.68* -1.67 -0.68*** -1.15*** -2.42*** -0.39*** -0.00 -1.21** -0.43 -0.19 0.09 0.28 -0.38** 0.10** -0.18*** -0.75**. -1.44*** 1.37*** 1.58 0.71** 1.50*** 2.38*** 0.60*** 0.54 1.47** 0.52 0.23 -0.14 0.05 0.40* -0.11* 0.03 1.12**. 3.99*** -0.21 18.18*** -1.15*** -1.43*** 1.55*** -1.28*** 1.46 0.35 -0.69 -0.04 -0.57* -1.01*** -0.50 -0.31* -0.81*** -2.63. 政治大. adj. R2 1.00 0.33 0.88 0.50 0.95 0.99 0.60 0.05 0.31 0.00 0 0.06 0.72 0.05 0.16 0.16 0.15. 學. I. 𝑍1. 𝑍2. 𝑍3. -0.45 0.08 -0.14*** -0.13 0.27* -2.63*** 0.04 0.73*** -0.05 0.01 -0.04 -0.11 0.36 -0.10 0.02 -0.16*** -0.15. 4.44*** -0.59** 21.52*** -1.31*** -1.46*** 2.74*** -1.24*** 1.09 -0.16 -0.37 -0.35 -0.42 -1.02*** -0.06 -0.51** -0.81*** -2.32. 1.23*** 0.60*** -0.39 0.40 0.41 4.13*** 0.33*** 1.34*** 0.65** -0.03 0.39 0.37* 0.18 -0.05 0.02 -0.01* -0.26. adj. R2 1.00 0.44 0.87 0.47 0.92 0.98 0.55 0.32 0.28 -0.05 0.01 0.17 0.73 0.00 0.12 0.17 -0.01. P-value < 0.1, 0,05, 0.01 level of confidence is denoted by *, **,***, respectively.. ‧. Table5. a l adj. RC h. n. Modified Jones model in2013 I. N. 𝑋1. 𝑋2. 𝑋3 *. (C1) (C2) (C4) (C5) (C6) (C7) (C8) (C9) (A) (B) (D) (E) (F) (G) (H) (J) (K). 69 46 31 170 80 146 291 99. 54107992.6*** 15029491 40715909 91000067.2** -1.1E+07 -7288832 641846726*** 23929017.5***. 25 62 82 49 73 98 135 151 56. -8035625 38070216.8* -7276114 -6297191 -107739189* 194177.8 -9951783 29734624 -3111500. 0.032 -0.03 -0.02 0.11 0.02 0.23 0.09*** 0.17 -0.03*** -0.10 0.22* 0.03 0.14** 0.05 0.56*** 0.13 -0.08. -0.16 -0.08 0.07 -0.34* -0.2*** -1.6*** -0.04 -0.18 0.15** -0.07 -0.19 -0.03 -0.3*** -0.11 -0.45 -0.13 -0.16*. 2. 0.23 0.15 0.12 0.07 0.04 0.59 0.39 0.92 0.06 0.99 0.06 0.13 0.05 0.21 0.02 0.09 0.10. er. io. sit. y. Nat. Comparison tests for EM based on alternative models to measure DA. Each model estimated coefficient which is extracted from Chinese listed-companies in 2013. Sample of 1663 firm-year is collected from CSMAR. Yoon (2006) model in 2013. i v adj. 𝑌 𝑌 𝑌n R U i e h n 0.17 0.13 -0.21 g c 0.95 1. 2. 2. ***. **. 0.48***. -0.50***. -0.99*** -1.95** -1.22*** 0.46** -1.67*** -0.28. 0.98** -1.24 1.24*** -0.31 2.29*** 0.50 0.50 -0.38* 0.27 -0.06 0.39 0.46** -2.09*** 5.69* 0.55. -0.22 0.30* -0.05 -0.08 0.74 -0.39** 1.60*** -9.84*** -0.32. 3. 0.30 -2.04* -1.16*** -0.05 -1.09*** -1.22*** -0.97* -0.89 0.26** -1.0 -1.24*** -1.10*** 1.48** -4.34*** 6.91 0.30. 0.31 0.79 0.91 0.24 0.46 0.18 0.06 0.05 0.55 0.02 0.21 0.31 0.17 0.56 0.13 0.03. The new model in2013 𝑍1. 𝑍2. 𝑍3. adj. R2. -0.3*** 0.13 -0.32* -0.5*** -0.6** 0.2*** -0.24 0.05. 0.14*** 0.21 -2.18* -1.95*** 0.01 -1.07*** -0.71* -1.0** -0.87 0.22* -0.87 -1.28*** -1.10*** 1.14** -4.62*** -0.90 0.73. 0.74*** 0.40*** -0.54* 0.98*** 0.35 0.06 0.03 0.71*** 0.75*** -0.09 -0.05 0.50*** 0.07 1.05*** 0.37*** -2.06*** 0.39*. 0.96 0.38 0.76 0.99 0.09 0.45 0.01 0.31 0.54 0.51 0.01 0.47 0.30 0.32 0.50 0.84 0.05. 0.00 0.03 0.14 -0.10 1.1*** -0.19* -0.5*** 3.06** -0.07. P-value < 0.1, 0,05, 0.01 level of confidence is denoted by *, **,***, respectively..

(29) 29. Table6 Variable Influence Factor represents the level of multicollinearity. If the VIF>1/(1-R²) the multicollinearity is statistically significant on the independent variable. Test is conducted in 2012. The sample of 1663 firm-year is collected from CSMAR database.. VIF3. 1.3 1.2 1.6 2.7 1.1 1.1 1.4 1.0 1.0 1.2 1.4 56.0 1.1 1.1 1.5 1.0 36.0. 3.8 1.2 1.8 1949 1.0 1.8 2.8 1.0 1.1 1.6 1.8 2295 1.1 1.0 1.0 1.0 528.0. 3.7 1.0 1.6 1961 1.1 1.9 2.3 1.0 1.1 1.4 2.2 2399 1.0 1.1 1.5 1.0 450.8. 1.6 1.2 1.1 10.4 1.1 1.1 1.4 1.2 1.2 1.4 7.7 non 1.4 1.3 1.2 1.1 3.6. 立. n. al. 𝑌3. VIF1. VIF2. VIF3. 2.5 5.4 5367 16.6 7.3 2.7 12.5 2.5 18.2 9.1 4.4 7.8 1.7 4.7 8.2 1.2 2.7. 26.7 5.7 5362 17.5 7.3 2.8 11.3 2.5 19.1 9.6 4.4 8.1 1.5 4.5 7.0 1.2 2.7. 21.9 1.2 2.3 1.3 1.1 1.4 1.4 1.0 1.3 1.2 1.0 1.2 1.2 1.1 1.5 1.0 1.0. 𝑍1. 𝑍2. 𝑍3. VIF1. VIF2. VIF3. 2.0 1.2 1.3 1.7 1.1 6.0 1.5 1.1 1.2 1.0 1.0 1.3 1.2 1.1 2.7 1.0 1.0. 5.3 1.2 1.0 1.4 1.1 1.5 1.5 1.0 1.7 2.6 1.0 1.1 1.2 1.0 2.5 1.0 1.1. 7.4 1.1 1.3 1.5 1.1 18.3 1.2 1.1 1.5 2.6 1.0 1.3 1.0 1.0 4.9 1.0 1.0. (1 − 𝑅²). 250.0 1.6 9.0 2.0 20.0 125.0 2.5 1.1 1.7 1.1 1.0 1.1 3.8 1.1 1.2 1.2 1.3. 政治大. y. VIF2. (1 − 𝑅²). 𝑌2. sit. VIF 1. 𝑌1. The new model in 2012. 1. ‧. 𝑋3. io. 25 62 82 49 73 98 135 151 56. 𝑋2. Yoon (2006) model in 2012. 學. 69 46 31 170 80 146 291 99. 𝑋1. Nat. (C1) (C2) (C4) (C5) (C6) (C7) (C8) (C9) (A) (B) (D) (E) (F) (G) (H) (J) (K). n. ‧ 國. I. 1. er. Modified Jones model in 2012. i n U. v. 1 (1 − 𝑅²). 250.0 1.9 8.5 1.9 12.8 41.7 2.2 1.5 1.6 1.0 1.1 1.3 3.9 1.0 1.2 1.2 1.1. T ratio indicates if the value is more than 1.6, the mean value of each independent variables are significantly no-relationship. Variable Influence Factor (VIF) shows how multicollinearity has increased the instability of the coefficient estimates (Freund 2000: 98). Put differently, it tells you how “inflated” the variance of the coefficient is, compared to what it would be if the variable were uncorrelated with any other variable in the model. Klein (1962) suggests alternative criterion that if R exceeds R2 of the regression model. In this vein, if VIF is greater than 1/(1-R²) or a tolerance value is less then (1-R²), multicollinearity can be considered as statistically significant.. Ch. engchi.

(30) 30. Table7 Variable Influence Factor represents the level of multicollinearity. If the VIF>1/(1-R²) the multicollinearity is statistically significant on the independent variable. Test is conducted in 2013. The sample of 1663 firm-year is collected from CSMAR database. Modified Jones model in2013. VIF1. VIF2. VIF3. 1.14 1.22 1.18 1.34 1.3 2.24 1.02 4.83 1.21 226.4 1.04 1 1.16 1.02 1.04 1.04 1.09. 1.1 1.02 1.11 5 1.28 5.82 1.1 4.9 1.28 9.89 1.24 1.02 2.08 1.01 1.04 1.03 1.12. 1.11 1.21 1.3 4.69 1.1 7.04 1.08 1.05 1.07 259 1.26 1.02 1.93 1.01 1.01 1.01 1.18. (1 − 𝑅²). 1.35 1.26 1.27 1.09 1.08 2.49 1.66 12.35 1.21 71.43 1.10 1.22 1.09 1.31 1.04 1.12 1.18. 𝑌2. 𝑌3. VIF1. VIF2. VIF3. 1015 2.52 3.87 42.1 2.56 6.91 3.3 2.97 7.2 31.33 2.69 3.2 3.12 3.4 42.19 1.14 2.38. 1021 2.51 6.43 20.35 2.62 6.9 3.24 2.89 7.03 27.17 2.73 3.26 3 3.45 38.99 1.13 2.27. 1.2 1.02 2.75 20.27 1.08 1.03 1.07 1.08 1.08 7.25 1.04 1.13 1.11 1.03 1.62 1.02 1.15. 𝑍1. 𝑍2. 𝑍3. VIF1. VIF2. VIF3. 1.09 596 1.02 3.23 6.47 1.13 1.06 1.08 1.21 7.35 1.05 1.11 1.11 1.12 1.21 1.28 1.06. 1.08 1.14 1.03 3.41 4.79 1 1.05 1.02 1.08 8.14 1.01 1.11 1.1 1.07 1.76 1.05 1.05. 1.01 598 1.01 4.64 3.13 1.13 1.07 1.06 1.13 1.4 1.05 1.01 1 1.19 1.62 1.28 1.01. (1 − 𝑅²). 21.74 1.55 5.26 11.24 1.36 1.89 1.23 1.10 1.20 2.32 1.06 1.34 1.50 1.24 2.30 1.18 1.09. 政治大. sit. 立. 1. 𝑌1. y. 𝑋3. ‧. 𝑋2. The new model in 2013. 學. 69 46 31 170 80 146 291 99 25 62 82 49 73 98 135 151 56. 1. 𝑋1. Nat. (C1) (C2) (C4) (C5) (C6) (C7) (C8) (C9) (A) (B) (D) (E) (F) (G) (H) (J) (K). n. ‧ 國. I. Yoon (2006) model in 2013. 1 (1 − 𝑅²). 25.64 1.71 4.63 100.00 1.15 1.87 1.03 1.50 2.48 2.13 1.04 2.02 1.48 1.52 2.05 6.21 1.12. n. al. er. io. T ratio indicates if the value is more than 1.6, the mean value of each independent variables are significantly no-relationship. Variable Influence Factor (VIF) shows how multicollinearity has increased the instability of the coefficient estimates (Freund 2000: 98). Put differently, it tells you how “inflated” the variance of the coefficient is, compared to what it would be if the variable were uncorrelated with any other variable in the model. Klein (1962) suggests alternative criterion that if R exceeds R2 of the regression model. In this vein, if VIF is greater than 1/(1-R²) or a tolerance value is less then (1-R ² ), multicollinearity can be considered as statistically significant.. Ch. engchi. i n U. v. Figure 2 reports the bar graph of DA. The vertical line denotes the magnitude of DA and horizontal line denotes the name of industry. Also, Figure 2 shows that the DA bar between SOE and NSOE looks similar in 2012 and 2013. The data table beneath the graph allows to see the median of DA (C1, C2, C6, F, J) in 2012 and (C4, C5, C8, E, F, E) in 2013 are greater than NSOE. The most interesting here is that the bar location is opposite during two periods. For instance, bar graph in 2012 is asymmetric with bar graph.

(31) 31. in 2013. Although only (C1, C4, C5, C7, C8, F) can be compared between two periods, the asymmetric results can be seen these industries.. Figure 2 Each Graph represents the level of DA for SOE and NSOE. Sample is retrieved from IPO companies financial statement.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(32) 32. According to the Table 8, the PPE in 2012, LTD and AR in 2013 is statistically significant on DA. However, the regression results of SOE holds multicollinearity in 2013. Therefore, it should be retested to eliminate multicollinearity. The goodness of fit for SOE in 2012 is quite high, which is indicated by the adjusted-R squared (98%). In addition, the regression result of the adjusted-R squared (17.5%) of SOE in 2013 is comparatively better than NSOE.. Table 8 The results of difference test regressing DA on accrual and non-accrual items. Sample is retrieved from IPO companies’ financial statement.. Nh 1/(1-R²) adj. R2. 1.01 1.01. (10906.293) 165. al. n. LTD. 2.41. io. PPE. 1.23. Nat. STD. 1.90. Ch. 1.30. OAR. 1.17. AP. 1.91. OAP. 1.74. IMP. 1.17. INV. 1.20. Std. 1.07. PPE. 1.32. LTD. (57407.122) 210. e n gN c h i U 1.16. 0.102. adj. R2. 55.56 0.980. y. INV. 1.20. AR. ‧. IMP. 1.36. 1.54. 學. OAP. 立. 2.20. 政治大. SOE da2013 -8569.644** (4275.695) -3998.726 (4505.481) -630.076 (5738.316) 736.029 (6149.970) 17982.125 (34161.185) 10192.627 (13918.281) -68.780 (1497.597) 0.117 (0.268) -8.019e+09***. sit. AP. NSOE da2012 1.066*** (0.236) 0.325 (0.245) -1.141*** (0.356) -0.609 (0.406) -1.143 (1.308) 0.426 (0.481) -0.079 (0.087) 0.000 (0.000) 4201.734. er. OAR. VIF 1.53. ‧ 國. AR. SOE da2012 1.376 (10.594) 4.726 (11.283) -0.544 (11.621) -4.655 (11.089) 3.831 (89.418) -0.884 (2.932) -2.083 (3.642) 0.090*** (0.001) 596.833. v n i(1.629e+09). VIF 1.79 1.29 2.14 2.33 1.28 1.39 1.69 1.08 1.08. NSOE da2013 -11259.607** (5568.284) -7426.250 (4875.135) -4406.336 (7382.317) -28502.223 (17654.909) 40685.591* (20657.039) 2445.107 (6418.264) 2354.602 (1990.260) 0.346 (1.335) 1.547e+10. VIF 1.43 1.19 1.55 10.26 1.08 19.31 8.64 1.01 1.02. (6.767e+10) 197. 152 1.21 0.175. 1.09 0.044. OAR = other account receivable, AP = account payable, OAP = other account payable, IMP = impairment loss and gain, INV = inventory, STD = short-term debt, PPE = property plant and equipment, and LTD = long-term debt. *All variable are standardized by revenue, except for LTD. *LTD is standardized by Liability. *P-value < 0.1, 0,05, 0.01 level of confidence is denoted by *, **,***, respectively..

(33) 33. In order to check relationship of each independent variable on the regression results, table 4 tested correlation. Table 9 shows PPE is positively correlated with DA at (99%) in 2012 and Table 5 shows LTD and AR is negatively correlated with DA at (-42%) and (-27%) in 2013. Compared to NSOE, SOE controls DA by PPE, AR and LTD. This result of LTD is consistent with Cheng (2015), but only 64 SOEs appropriate LTD in 2013 and also the regression test includes multicollinearity. Additionally, short-term debt is positively correlated with payable in 2012 and 2013 and LTD is positively correlated with Account Receivable in 2013.. AP. OAP. IMP. 1.00 -0.54 0.11 0.25 0.39 0.02 0.04. 1.00 0.04 0.01 -0.68 0.01 0.01. 1.00 -0.16 -0.07 0.00 -0.02. INV. STD. PPE. LTD. 1.00 -0.04 0.00. 1.00 0.00. 1.00. 學. n. al. sit. y. 1.00 -0.06 0.01 0.01. er. 1.00 0.33 0.01 0.07 -0.02 0.02 -0.03 0.00. ‧. 1.00 0.21 0.32 0.01 0.48 0.00 -0.05 -0.06 0.08. io. 1.00 -0.05 -0.02 0.02 0.01 0.01 0.00 -0.05 0.99 0.00. OAR. Nat. DA AR OAR AP OAP IMP INV STD PPE LTD. AR. ‧ 國. DA. Table治 9 政 Correlation test on DA for SOE大 in 2012. 立. i n U. Ch. v. Table10 i in 2013. e n g cforh SOE Correlation test on DA DA AR OAR AP OAP IMP INV STD PPE LTD. DA 1.00 -0.27 -0.08 -0.13 -0.01 -0.02 0.06 0.04 0.02 -0.42. AR. OAR. AP. OAP. IMP. INV. STD. PPE. LTD. 1.00 0.10 0.50 -0.03 0.40 0.06 -0.02 0.00 0.22. 1.00 0.28 0.07 0.10 -0.02 -0.14 0.25 -0.02. 1.00 -0.46 0.21 0.26 0.30 -0.02 0.05. 1.00 -0.02 -0.49 -0.62 0.03 0.02. 1.00 0.19 0.02 -0.04 -0.03. 1.00 0.29 -0.02 -0.02. 1.00 -0.05 -0.08. 1.00 -0.02. 1.00. Table 11 reports that LTD and AR is statistically significant at p<0.05 level of confidence for SOE without multicollinearity. Table 12 shows that LTD (-49%) and AR.

(34) 34. (-51%) are negatively correlated. On the contrary, difference test for NSOE does not show better result. Coefficient of LTD positively related with DA for NSOE. Moreover, independent variables are not statistically significant. It may be because independent variables are not directly correlated with DA. For the additional information about LTD, 34 companies gained LTD and 25 companies paid LTD from 2012 to 2013. Other 8 companies does not change LTD from 2012 to 2013. It suggests that Hypothesis 4 (SOE has less incentives to manage earnings because of bank loans.) is rejected in 34 companies because they borrow money and decreased the magnitude of DA in 2013.. Table 11 SOE with LTD. N １/(1-R²) adj. R2. 1.12 1.17. Nat. R²,. 1.01. VIF 1.61 1.57 1.01. 1.63. 1.01. ‧. LTD2013. 1.11. DA2013 -205.8 (-1.20) 1455.9 (0.96) 0.543 (0.03) 0.00594 (0.11) 7055.1 (0.34) 66 0.026 -0.056. 1.03. y. PPE2013. VIF 1.40. 學. INV2013. 立. 政治大. 1.06. sit. IMP2013. DA2013 -23276.9** (-3.44) 21150.1 (0.37) 3271.5 (0.14) -129.7 (-0.07) -6.44827e+09** (-3.15) 64 0.386 0.333. ‧ 國. AR2013. NSOE with LTD. al. n. Table 12 SOE in 2013 DA AR IMP INV PPE LTD. er. io. P-value < 0.1, 0,05, 0.01 level of confidence is denoted by *, **,***, respectively.. C h AP e n g cIMP hi. DA. AR. 1.00 -0.52 0.04 0.06 0.14 -0.49. 1.00 -0.05 0.01 -0.31 0.35. 0.01 0.05 -0.03 -0.04. 1.00 -0.10 0.29 -0.17. DA. AR. AP. 1 -0.18 0.08 0.02 0.02 0.01. 1 0.27 -0.04 0.01 0.04. 0.12 -0.04 0.00 0.12. iv n U INV. PPE. LTD. 1.00 -0.17 0.03. 1.00 -0.05. 1.00. IMP. INV. PPE. LTD. 1 0.00 0.03 0.00. 1 0.00 0.00. 1 -0.01. 1. Table 13 NSOE in 2013 DA AR IMP INV PPE LTD.

(35) 35. VIII. CONCLUSIONS AND IMPLICATIONS. In this paper we found that the specific accrual-based model needs to be applied for each year and each industry. In other words, the model sensitively reflects the unique features of each industry as well as the change of accounting item’s value. The above mentioned problems should be carefully taken care of in order to accurately detect EM. Therefore, panel data analysis may not be feasible for EM study because it will dismiss the discretional change of accounting item’s value. And hence, two different models such as the modified Jones model and the new model performed better in particular industry. 政治大 other accrual-based model. Consequently, I found that SOE has conducted stronger EM 立 in pre-IPO market during 2011 to 2013 by manipulating with PPE, the flow of AR and and year. Especially, the new model showed more appropriate performance, compared to. ‧ 國. 學. LTD. In addition, only 34 companies reported that they lent loans from bank and other 25 companies paid loans back from 2012 to 2013.. ‧. These findings can be implicated as following three points. First, the new model. y. Nat. is not theoretically verified, but the use of appropriate model conducts different results. io. sit. from prior researches. 10We thus need to suggest a new solid model specification to clarify. al. er. the EM in China. Secondly, Chinese SOE undergoing IPO process manipulates earnings. n. v i n Ch between SOE and NSOE is not different, e n gtherefore, c h i USOE camouflages earnings by. because they are not exposed under strict audit and monitor. The magnitude of EM. manipulating accrual items. Finally, Chinese central bank appreciated interest rate after 2010 and it caused higher deposit rate as well as lower investment rate. Therefore, I assume that some SOE needed to increase their earnings to obtain bank loan before they enter IPO market..

(36) 36. References. [1] Klein, L. R. (1962). An introduction to econometrics.* Prentice-Hall. New Jersey, 280p, 62-64. [2] DeFond, M. L., & Jiambalvo, J. (1994). Debt covenant violation and manipulation of accruals. Journal of accounting and economics, 17(1), 145-176. [3] Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1995). Detecting earnings management. Accounting review, 193-225. [4] Chaney, P. K., & Lewis, C. M. (1995). Earnings management and firm valuation under asymmetric information. Journal of Corporate Financ1995 e, 1(3), 319-345.. 政治大. [5] Bernard, V. L., & Skinner, D. J. (1996). What motivates managers' choice of discretionary accruals?. Journal of Accounting and Economics, 22(1), 313-325.. 立. [6] Freud, R. J., & Littell, R. C. (2000). SAS system for regression. Sas Institute.. ‧ 國. 學. [7] McNichols, M. F. (2001). Research design issues in earnings management studies. Journal of accounting and public policy, 19(4), 313-345.. ‧. [8] Baltagi, B. (2008). Econometric analysis of panel data. John Wiley & Sons.. sit. y. Nat. [9] Hsiao, C. (2003). Analysis of panel data, 2nd. Cambridge: Cambridge University Press. Kose, MA, ES Prasad and ME Terrones (2003), Financial Integration and Macroeconomic Volatility, IMF Staff Papers, 50, 119-142.. al. er. io. [10] Roychowdhury, S. (2006). Earnings management through real activities manipulation. Journal of accounting and economics, 42(3), 335-370.. n. v i n C h P. (2006). Earnings [11] Yoon, S. S., Miller, G., & Jiraporn, management vehicles for i U & Accounting, 17(2), 85Korean firms. Journal of Internationale Financial n g c hManagement 109.. [12] Wang, Q., Wong, T. J., & Xia, L. (2008). State ownership, the institutional environment, and auditor choice: Evidence from China. Journal of accounting and economics, 46(1), 112-134. [13] Ball, R., & Shivakumar, L. (2008). Earnings quality at initial public offerings. Journal of Accounting and Economics, 45(2), 324-349. [14] Hujino Yutaka, (2009). New problems and present situation of the accrual-based model. Saint Paul’s University Economic Study volume 62.3, p95-p112 (藤野裕. (2009). 裁量的会計発生高推定モデルの現状と新たな課題: モデルが仮定する条件の現実妥当性について. 立教経済学研究, 62(3), 95-112.) [15] Berger, A. N., Hasan, I., & Zhou, M. (2009). Bank ownership and efficiency in China: What will happen in the world’s largest nation?. Journal of Banking & Finance, 33(1), 113-130..

(37) 37. [16] Armstrong, C., Foster, G., & Taylor, D. J. (2008). Earnings management around initial public offerings: a re-examination. Rock Center for Corporate Governance Working Paper, (23). [17] Chen, H., Chen, J. Z., Lobo, G. J., & Wang, Y. (2010). Association between borrower and lender state ownership and accounting conservatism. Journal of Accounting Research, 48(5), 973-1014. [18] Aaker, H., & Gjesdal, F. (2010). Do Models of Discretionary Accruals Detect Actual Earnings Management via Inventory? A Comparison of General and Specific Models. A Comparison of General and Specific Models (June 24, 2010). [19] Benmelech, E., Kandel, E., & Veronesi, P. (2008). Stock-based compensation and CEO (dis) incentives (No. w13732). National Bureau of Economic Research. [20] Islam, M. A., Ali, R., & Ahmad, Z. (2011). Is modified Jones model effective in detecting earnings management? Evidence from a developing economy. International Journal of Economics and Finance, 3(2), 116.. 立. 政治大. ‧ 國. 學. [21] Paul, R. K. (2006). Multicollinearity: Causes, Effects and Remedies. IASRI, New Delhi.. ‧. [22] Alareeni, B., & Aljuaidi, O. (2014). The modified jones and yoon models in detecting earnings management in Palestine Exchange (PEX). International Journal of Innovation and Applied Studies, 9(4), 1472.. sit. y. Nat. [23] Shen, Z., Coakley, J., & Instefjord, N. (2014). Earnings management and IPO anomalies in China. Review of Quantitative Finance and Accounting, 42(1), 69-93.. al. er. io. [24] Cheng, C. A., Wang, J., & Wei, S. X. (2015). State ownership and earnings management around initial public offerings: Evidence from China. Journal of International Accounting Research, 14(2), 89-116.. n. v i n C h progress and U [25] Tetsuo Hujita, (2014). Financial challenges in China. The Japan i e h n c g Research Institute, United, Pacific Rim Business Information 14, No.55. （藤田哲雄.. (2014). 中国の金融改革の進展と課題. Rim: 環太平洋ビジネス情報, (55), 62-79.） [26] Cai, Y., & Zhang, L. (2015, May). The relation between equity incentives and earnings management in China. In Information Engineering and Education Science: Proceedings of the International Conference on Information Engineering and Education Science (ICIEES 2014), Tianjin, China, 12-13 June, 2014 (Vol. 5, p. 29). CRC Press. [27] PwC China Assurance Partner, Doing business and investing in China. Retrieved Feb 15 2006, from: http://www.pwccn.com/webmedia/doc/634940150734265198_iic_full.pdf. (Global Annual Review 2015 (PWC). p146-p160) [28] Xu, W. Chinese domestic IPO over-issuance..

(38) 38. Appendix 1 Historical flow of Financial deregulation in China Hujita, (2014) :Panel A Reform of Chinese financial market 1980 Separation of finance and government 1984 Renmin bank introduce the deposit reserve ratio operation 1985 Credit Creation Plan formulated by National planning commission, Ministry of Finance and Renmin Bank 1995 Central Bank and Commercial Bank law was enacted ① Ensuring the dependence of monetary policy ② Eliminating influence of the local government on the blanch of Renmin Bank ※ Monetary policy was supervised by government officials ※ Big four Bank was separated from Renmin Bank ※ Mostly loan was available for SOE 1995~1997 30%-50% loan was regarded as bad debt 1998 270billion RMB was injected for Four big commercial bank 1999~2000 1.4trrilion RMB bad debt was separated into four asset management company (AMC) 2001 Entering in the WTO ① Chinese government’s agenda shows more liberalization of interest rates ② Less restriction on ownership takeovers and mergers and acquisitions ③ Greater freedom of operational and geographical scope in the banking industry ※①➁③ explained in the paper of Berger (2009) 2002 Big four bank prompted demutualization and management structure, corporate governance and financial status was improved 2003 Accepting foreign financial institution for fund management and international competitiveness 2005~2007 Big three banks were listed on stock market 2010 Chinese agricultural bank was listed and world largest IPO was conducted in this year ~2013 More than 3 % of difference between lending ratio and deposit interest ratio 2013 Lending ratio deregulation. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(39) 39. Panel B Code that this study uses. CSMAR industry. A. Faming, Forestry, Animal Husbandry, and Fishing. IPO companies. Total Number of IPO company. Total Number of listed company. SOE. NSOE. 25. 0. 3. 3. 62. 5. 2. 7. 69. 7. 10. 17. A01 A03 A05 A07 A09. B. Mining. CSMAR Code. B01 B03 B05 B07 B09 B50 C01 C03 C05. Food and Beverage Textile, Apparel, Fur and Leather. C1 C2. C11 C13 C14. 46. 5. 13. 18. Paper and Allied Products; Printing. C4. C31 C35 C37. 31. 8. 6. 14. C5. C41 C43 C47 C48 C49 C51 C55 C57. 170. 29. 6. 35. 80. 18. 14. 32. 146. 16. 25. 41. 291. 59. 94. 153. C6. ‧ 國. Information Technology Wholesale and Retail trades Real estate. Nat. Public Facilities, public service Total Number. C9. C71 C73 C75 C76 C78 C81 C85. 99. 26. 17. 43. D. D01 D03 D05. 82. 10. 1. 11. E. E01 E05. 49. 13. 9. 22. F. 73. 5. 4. 9. G. F01 F03 F05 F07 F09 F11 F19 F21 G81 G83 G85 G87. H. H01 H03 H11 H21. 135. 7. 4. 11. J. J01 J05 J09. 151. 1. 1. 2. K. K01 K21 K30 K32 K34. 4. 0. 4. 213. 209. 422. 學. Transportation, postal service. C8. y. 立. Medicine and Biological Utilities Construction, repairment. C61 C65 C67 C69. 98. 56. 1663. io. n. al. er. Vehicle, machine. 政治大. C7. ‧. Electronic products Metal, nonmetal and new material. sit. Chemical, Oil Product. Ch. engchi. i n U. v.

(40) 40. The industry classification and comparison between CSMAR and GICS Panel C. C6. 4510,4520,4 530. C7. 1510. C8. 2510. C9. 3510,3520. B F G H J. 1510 2030 5010 2550 4040. K. 2530. Energy. 151050. Paper and forest products Textiles, Apparel and Luxury goods Oil, gas and chemical. 252030 101010,101020, 151010. Software, Services, technology hardware, equipment, Semiconductors and equipment Material. 政治大. Automobiles and components Health care equipment, pharmaceutical biotechnology and life sciences Material Transportation Telecommunication Services Retailing Real Estate Consumers services. 立. 151020,151030. 151040. 253010. Nat. io. n. al. Ch. engchi. Construction Materials and Packaging. Metals and Mining. y. 1010. Name. sit. C5. Industry. er. 2520. ‧ 國. C4. Compustat GICS Name Food, staples retailing, Beverage, Tobacco Materials Consumer Durable & Apparel. ‧. Group 3010,3020 1510. 學. CSMAR Industry C1 C2. i n U. v. Hotels, Restaurants and Leisure.

(41)