Research On The Evolution And Strength Measurement Of China’s Shadow Banking Regulation Policy

After the international financial crisis in 2008,shadow banking has developed rapidly in China,its scale has expanded rapidly,its product diversification continues to increase,and the supervision of shadow banking has gradually increased.Existing researches on shadow banking focused on the measurement of shadow banking,there is less research on supervision policy,and there are few quantitative models of shadow banking supervision policy.Existing research on quantifying policy intensity is mostly used as an intermediate variable to study on other variables.Graphic analysis such as cluster analysis and simple negative word frequency analysis are often used to analyze the strength of policies.Compared to financial policy research,there are relatively few quantitative researches on supervision policies.Not only quantitative research can assist financial institutions in predicting future regulatory intensity,and also can provide a data research basis for regulatory agencies;at the same time,it can provide new perspectives for future theoretical research.It can also be used to extend the domain of the same analysis method in different fields.Therefore,this article selects the shadow banking related regulatory policies as the research object,and uses different methods for quantitative analysis of the policy content.This article collects the shadow banking supervision policies of China’s financial regulators from 2008 to 2019,and quantifies the intensity of every policies in different years through three methods:negative word frequency analysis,text sentiment tendency analysis,and PMG model.Among them,text sentiment tendency analysis methods include ROST model and sentiment analysis model.The characteristics of shadow banking supervision are as follows:Since the beginning of 2008,the scope of supervision of shadow banking has been continuously expanded.The characteristics of supervision have gradually shifted from the supervision of shadow banking activities to the supervision of the characteristics of shadow banking.Supervision has phases of transformation,but on the whole,it presents a dynamic change process from a single supervisory body to an uneven balance to a comprehensive supervision.This article uses different policy quantitative models for analysis of policy intensity.The main work and conclusions are:Firstly,this article uses the collected shadow banking supervision policies and uses different policy text analysis methods to analyze 2008-2019 shadow banking policies one by one.The results show that different policy quantitative models have similar conclusions and have strong convergence.Secondly,although the results of different policy quantification models are similar,the differences are also obvious:the negative word frequency statistics are only a simple count of the number of negative words,and do not classify different words;in the policy text analysis The intensity of the policy is directly related to the number of shadow banking policy supervisions in the year,and there is no analysis of the propensity of the policy;in the text sentiment analysis model,different sentiment analysis dictionaries will directly lead to differences in sentiment analysis results,and the accuracy will vary greatly,we can correct the accuracy by training the sentiment dictionary.Therefore,for further research in the future,text sentiment tendency analysis is more meaningful.Third,in the study of quantitative policy research,the indicators used in the quantitative analysis of financial policy in the past were mostly keywords,and graphic analysis such as cluster analysis was used as an intermediate research variable or method to study the effect of policy.At the same time,subject-based analysis lacks the shortcomings of global analysis,and it is more one-sided to use it only as an intermediate variable.The calculation in this paper provides a good theoretical basis for subsequent theoretical and practical research.