China Listed Companies' Financial Statements Malpractice Study

After nearly twenty years of development, China capital market has made a sight progress. But along with it, many problems must be paid attention, the common and serious one is accounting information distortion, witches essence reason is the fraudulent practices in financial statements. So this paper mainly studies the problem of the fraudulent practices in financial statements, and on the basic of learning the reasons and main methods of the fraudulent practices, I try to structure the model to detect the fraudulent practices through empirical analysis.The present article is made up of three parts. The first part is the first chapter, mainly discussing the significance and goal of the fraudulent practices study, the study method. The second part is the second and the third chapters, which theoretically analyze the reasons and main methods of the fraudulent practices. It happens when having incentive and good outside opportunities at the same time, in our country, incentives of some enterprises are dealing with performance appraisal, getting the offer from financial organizations, obtaining the qualification of being on the market, even some politics goal and so on; the opportunities contain imperfect system, week accounting surveillance etc. The methods of fraudulent financial statements in our enterprises are still on a lower lever, mainly fictionalize or conceal transaction fact, and so on.The third part is the forth and the fifth chapters, the part structure the model of detecting fraudulent practices through empirical analysis. This thesis, sampling 46 listed companies discovered fraud engagements (contain 91 samples), and the control samples selected under some standard. First through the samples description the thesis investigates the current status of fraudulent financial statements, finding that the fraudulent type is keeping in overstating asserts or earnings. There are 71 samples belong to overstating asserts and 20 sample belong to overstating earnings, in the other side, fictionalizing sales and unrecorded expenses are universal. Then the thesis choose 22 variables to engage in the following empirical analysis. In empirical analysis, first practice mean test and median test on 22 variables between the fraud and non-fraud samples, and find fraud companies have higher ratio of sales in current assets, proportion of receivables to current assets or total assets, and debt ratio; have lower ratio of adjusted operational cash flow in total assets or total shares, and proportion of fixed assets to total assets. Second, develop and test a logistic regression model that estimates the likelihood of overstated assets or earning in financial statements, conditioned on the presence or absence of several variables. The final model keeps three significant variables, they are ratio of sales in current assets, proportion of receivables to current assets and debt ratio. Last, through the study, to the obtaining enlightenment give some suggestion.