Improving the predictability of business failure of supply chain finance clients by using external big dataset

The purpose of this paper is to help the financial institutions improve the predictability of business failure of supply chain finance (SCF) clients with the use of external big data set.

A prediction model for the business failure of SCF clients was built upon different theoretical perspectives. Logistic regression method was deployed to test the model.

The authors develop a model that illustrates several key determinants to predict the probability of business failure of SCF clients based on several theoretical perspectives. The results show that taxable sales revenue, frequency of making value added tax (VAT) payment, number of counterparty for VAT invoice issuance, frequency of VAT invoice issuance and firm age are negatively correlated with business failure of SCF clients while the VAT paid and industry clockspeed are positively correlated with their business failure.

This paper shows how financial institutions can effectively leverage the external information sources through “unconventional” predictor variables in order to reduce the credit risks associated with business failure of SCF clients.

This paper is one of the first to focus on the potential use of financial big data set from external sources to improve of predictability of financial institutions on the business failure of SCF clients. In addition, this paper is a pivotal study on the financial client risk assessment based on taxpaying behaviors, tax amount, firm and industry characteristics.