Probability causal inference of interest rate fluctuations: Evidence from private credit in emerging markets

Abstract

Preventing risks in the private credit market is an important component of maintaining financial stability, especially in emerging markets. As loan amounts serve as both the cause and the effect of credit interest rates, through the weighted path convergence design of conditional probability and joint probability, this study establishes a probabilistic causal inference framework of loan amounts and interest rates to explore the mechanism of interest rate fluctuations in China's private credit market, market behaviour of the allocation of private loan amount, and the design of policies. The analysis provides several results. (1) The fluctuation of private credit interest rate occurred during the outbreak of the epidemic (2019Q1-2020Q1), the credit interest rate was mainly controlled by the loan amount, and the interest rate fluctuation mainly originated from private credit market behaviour. In the early stage of the post COVID-19 economic recovery (2020Q3-2021Q3), the credit interest rate is mainly controlled by non-loan amount factors, and the interest rate fluctuations occurred mainly because of the non-market behaviour of private credit, which is related to government policy support and the increase in low-interest loans from financial institutions. (2) As a result of market behaviour, for the 19.54 %–22.50 % of high interest rates on borrowing, the probability of high interest rates due to the loan amount increased by 40.48 %. (3) By policy design, reallocating the risky loan amount, the interest rate can be reduced by up to 7.36 % on a highlight-interest rate day. The reallocation of loan amounts for different security methods has varying effects on interest rates reductions, and there may even be ineffective allocations. The findings provide decision support for financial regulators in emerging markets to stabilise interest rate volatility in private credit markets.