| Under the background of the State Council's promulgation of establishing the joint incentive of trustworthy and joint punishments of dishonesty system,and under the background of the "13th Five-Year Plan" put forward by Zhejiang Province,the problem of dishonesty was pushed to the public opinion and academic research.Among them,the blacklist construction has attracted much attention,so that credit can be quantified become a new bright spot.Taxation is the key field in the blacklist of dishonesty.Therefore,it is very important to study the enterprises which lose credit.These tax illegal enterprise why will enter blacklist? What are their credit characteristics? What internal and external factors will encourage them to produce dishonesty behavior? Is there any discrepancy in the degree of default? If there are differences,what are the influencing factors of this degree of difference and how are they affected? Different levels of breach of faith enterprises,the probability of default should be how to measure? To this end,this paper takes the problem of credit default as the starting point,analyzes the credit characteristics of the enterprises with bad credit,discusses the influencing factors of the illegal degree and establishes the model of default probability,calculate the accuracy of the model predictions.This paper mainly consists of theoretical analysis and empirical analysis of two parts.Theoretical analysis,the data mining theory,data mining theory based on credit features and three-dimensional credit evaluation theory as the starting point,the tax credit business characteristics of feature extraction and analysis,and the degree of default of credit evaluation.On the basis of data mining process,this paper uses SPSS 17.0 to study the factors influencing the degree of illegal taxation and the probability of default model.The empirical research is based on the data of 60 major tax blacklist companies published on Zhejiang credit platform.The empirical results show that the P-value of the parallel-line test is more than 0.05,which indicates that it is appropriate to use the ordinal multiple logistic regression analysis method.The P value of the chi-square test(Pearson,deviation)is 0.921 or 0.954,which indicates that the model has a high degree of goodness of fit.The value of McFadden is 0.341,indicating that the independent variable can explain which group the dependent variable belongs to,and the value of Nagelkerke is 0.564,indicating that the model can explain the variance of 56.4%.In addition,the accuracy rate of the regression model is 62.5%,the accuracy of the tax law is 52.63%,81.82% and 70% respectively.In the analysis of the factors influencing the degree of illegal taxation,the level of regional economic development,the level of regional financial development,the duration of the enterprise,the nature of the industry and the nature of ownership have a significant impact on the extent of the loss of trust.The level of local revenue and the sex of the legal representative have no significant effect on the degree of illegal taxation.From an external macroeconomic point of view,when a region's economic development level = 2,that is,when the economy is more developed level,will significantly reduce the degree of corporate tax law.When an area of financial development level = 2,that is,when the level of financial development is high,will significantly improve the enterprise's tax law.From the internal characteristics of the enterprise,the duration of the enterprise and the degree of dishonesty are positively correlated.Enterprises in the industry for the wholesale and retail trade will significantly reduce the degree of corporate tax law.The nature of the ownership of a limited liability company would significantly reduce the degree of tax lawbreaking.According to the theoretical analysis and empirical results,this paper puts forward the corresponding policy recommendations to the enterprises,the public credit information platform,the tax law enforcement agencies and the government.The enterprises should improve their management ability and credibility,pay attention to the relevant information of the industry,According to the stage of the life cycle to develop appropriate business strategy,improve the internal governance structure,improve the management and staff legal awareness and expertise in finance and taxation;The public credit information platform should give full play to the platform integration of resources,information gathering,publishing information,to provide the role of the exchange of data,strengthen the cross-regional,cross-sectoral,cross-field credit information sharing,attention to bad faith "blacklist" column construction,and urge the law enforcement agencies to improve the efficiency of the release of information loss;The tax law enforcement agencies should assume the functions of credit supervision,warning and punishment,improve the tax credit information in the public Credit information platform on the publicity efficiency;The Government should guide the society to establish the good credit consciousness,establish the joint incentive of trustworthy and joint punishments of dishonesty system,and actively cultivate credit culture.The theoretical value of this paper is to open up a quantitative analysis path for the study of tax credit loss,enrich the theoretical system and framework of platform economics,and provide the theoretical basis for credit investigation.The practical application value is in favor of giving play to the role of credit in allocating resources in the market and effectively reducing the transaction cost of the market.;In favor of the public credit information platform,the law enforcement departments,the field of dishonesty data exchange,sharing,play the role of public credit information platform,more adapt to the Internet economy and social credit system construction of credit information platform requirements;Which is conducive to providing credit decision support for tax authorities and government departments. |
PDF Full Text Request