信用风险传染效应与外溢冲击研究

Research on Credit Risk Contagion Effects and Spillover Impacts

本文采用前沿的连通性频域分解方法(frequency domain decomposition),准确测度中国产业债券在短期、中期和长期的风险传染效应,并就信用风险在不同行业间的传导关系与冲击强度展开对比分析。同时,分别根据上下游产业、债项评级等不同债券分类,构建信用风险关联网络,深入剖析中国债务风险传染中的关键节点与薄弱环节。此外,进一步考察了信用风险溢出强度对经济金融变量的作用方向与驱动力度。研究表明,尽管在短期信用风险传染对中国金融系统产生了相对可控的外生影响,且对宏观经济的冲击较为有限,但随着时间的推移,市场中期、长期的流动性逐步趋紧,融资难度显著提升,信用风险传染对经济景气指数、社会融资、企业投资等宏观基本面产生了显著的负面影响。在此基础上,本文为应对信用风险外溢冲击提出了相关建议,从而有助于稳妥处置化解重大风险隐患、健全金融有效支持实体经济的长效机制。

Serving the real economy is the purpose and origin of finance. However, since 2020, both domestic and international environments have become increasingly complex, driven by which enterprises have suffered significant deterioration in profitability and a sharp increase in debt repayment pressure, and have frequently been exposed to credit risk. More importantly, investors will become more risk averse as the credit risk exposure expands continuously in the bond market, resulting in a structural tightening of social financing, severe volatility in the financial market and considerable pressure on the macroeconomy, which has provoked great concern of Chinese governments. Therefore, at this stage, it is obviously of great academic value and practical significance to thoroughly explore the contagion effects of corporate bond risks, accurately analyze the spillover impact of credit risks on the financial system and macroeconomy, and comprehensively review their heterogeneous mechanisms and impact intensities in the short, medium, and long term.The spillover effects of credit risks on various economic and financial variables have been hardly considered empirically under a unified framework in the existing literature. Despite the significant differences in the risk contagion effects and impact intensities measured based on various frequencies, prior research often neglects the heterogeneous frequency response of risk to shocks,making it difficult to accurately capture the correlations among different assets or markets. Additionally, Granger causality in mean ignores the potential dependence in the conditional tail distribution of sequences, thus possibly leading to significant biases in the causal inference among financial variables.In view of this, this paper employs the newly developed connectedness measure with frequency decomposition(Baruník &K?ehlík, 2018) to accurately quantify the short-, medium-, and long-term risk contagion effects of industrial bonds in China. We also conduct comparative analyses on the transmission relationships and impact intensities of credit risks among different industries.Meanwhile, we build a credit risk linkage network based on different bond classifications such as supply chain positions and bond ratings to investigate the key nodes and weak links of the credit risk contagion in China. Furthermore, Granger causality in quantiles and factor-augmented vector autoregressive model are applied to examine the direction and magnitude of the impact of credit risk spillover on economic and financial variables. On this basis, we yield policy implications for coping with credit risk shocks and strengthening the long-term mechanism of financial support to the real economy.The empirical analysis contributes to the following findings.(1) Significant heterogeneities are detected in the credit risk contagion effect in China during different periods.(2) The credit risk network is closely linked with risk propagating via complicated channels in China.(3) The risk interconnectedness among various industrial bonds increases massively over time, indicating that the aggravating contagion of the industrial bond risk might become a hidden danger for China's financial system as an outcome of the materialization of credit risk.(4) In the short term, on the one hand, the causality link between the spillover intensity of credit risk and indices of middle and small-sized enterprises' confidence and consumer sentiments is evident, as the accumulation of risks leads to significant pressure on market participants' expectation and severe erosion of market confidence. On the other hand, the spillover of credit risk has a considerable impact on financial indicators such as interbank offered rates.(5) In the long term, as credit risk accumulates continuously, its spillover impact is likely to be further amplified through financial channels, which will cause serious adverse impact on both the supply and demand sides of financing. This will tighten the market liquidity margin and make the financing environment deteriorate for small and medium-sized enterprises, thereby causing significant negative impacts on macroeconomic fundamental indicators such as the economic prosperity index, investment, and financing.Based on the above conclusions, this paper proposes that we should improve the pre-control system for debt risk, steadily promote the classified management of the credit risk of industrial bonds, and build a long-term credit risk prevention and control system. It also emphasizes the need to strengthen the foresight and initiative of expectation guidance work and boost market confidence simultaneously from both the supply and demand sides.