Decision-making Of P2P Investors, Borrowers Information And Words Signaling

Due to the information asymmetry and huge credit risk in the trading process,it has been difficult for a large group of small enterprises and hundreds of millions of individuals with poor credit records to get loans from formal financial institutions for a long time.Network lending is a typical form of Fintech and has transformed from brutal growth to standardized operation.With the help of the credit authentication and information disclosure mechanism,the network lending platform works as the information intermediary which gathers and transmits messages.Among many tpyes of information,the descriptive text provided by the borrower is an important kind of soft information.It contains some credit quality signals,such as the borrower’s personal quality and social cognition,and will make influence on the lender’s decision-making.Based on the data of Renrendai,the P2P lending platform which develops fastest in China now,this paper collects 388522 loans which are from May 2010 to December 2016 by the web crawler and artificial collection.On the basis of the asymmetric information theory,signal theory and transaction cost theory,this paper mainly explores the following five questions.First,under the condition where the P2P lending platform cannot guatantee the truthfulness of descriptive texts,can the signals from different kinds of words in the narratives effectively alleviate the financing difficulties of the long tail group?Second,how do the borrower’s demographic characteristics affect the function mechanism of word signals on lender’s decision?Third,reputation is the borrower’s another important soft information.How does the reputation influence the function mechanism of word signals during the process of information gathering and transmission?Fourth,is there any interaction effect between the word signals and borrowers’ hard information during the process of word signals’ alleviation on financing contraints?Fifth,can the word signals effectively restrain the price distortions and improve the information transparency of P2P lending market?Based on the Chinese word segmentation and large data sample,the empirical results show the following conclusions.First,the signals of the different kinds of words in descriptive texts provided by P2P borrowers have significant impacts on lenders’decision-making.To be specific,the proportion of positive words and the proportion of financial words are positively related with the borrowing success rate.The proportion of negative words,the proportion of strong modals and the proportion of weak modals are negatively related with the success rate.Second,when the borrowers are of different gender,education levels and income levels,the word signals of their descriptive texts have significantly different influences on lenders’ decision-making.And the difference of age almost has no effect on the word signals.Third,this paper finds that the borrower’s reputation has obvious alternative effect on the influence of word signals.The borrower’s hard information weakens the influence of word signals and plays the different role in sharing the borrowing cost with the word signals.Finally,the quality signals of different kinds of words are partly effective in the lenders’ decision-making.To be specific,the signals of financial words are effective and they are identified correctly by investors.The signals of strong modals are also effective but they are not recognized correctly by lenders.Other categories of words are not the effective signals of borrowers’ high credit quality.Based on the conclusions of empirical researches,this paper puts forward the corresponding proposals and suggestions for borrowers,lenders and platforms in P2P lending,which aims to promote the long-term and healthy development of network lending.