Credit Default Risk and Market Risk Premium of Chinese Corporate Bonds

This thesis presents a comprehensive analysis of credit risk in China's corporate bond market. Utilising data on all Chinese corporate bonds issued by publicly traded companies before September 2020 and actual default cases, the research investigates credit risk from the perspectives of forecasting corporate bond defaults, measuring credit default risk, measuring credit default risk, and exploring the determinants of bond risk premiums at the individual level and excess bond premium at the market level. To measure credit default risk and forecast corporate bond defaults in China, this research applies three types of models: accounting-based Altman models, the Merton distance-to-default (DD) model, and the default hazard model. Our findings confirm the robustness of three original Altman models in predicting Chinese corporate bond defaults within a one-year horizon. Re-estimating and revising these models significantly enhance predictive power and extend the forecasting horizon to three years. The three Altman models are robust across estimation methods, with the Z'-score model being most effective in China's bond market. Additionally, our results show the DD as the solution of the Merton model is not a sufficient statistic for measuring default risk, and its discriminatory power is sensitive to the solution's derivation approach. The structural form of the Merton model offers more information for default forecast than DD. Building on these findings, the research develops a hybrid default hazard model that combines financial ratios from the Altman models with variables derived from the Merton-DD model, outperforming the other models in predictive accuracy. Lastly, determinants of the market price of corporate bond risk premium are investigated using a dual-level credit spread modelling approach and credit spread decomposition technique, accounting for China's corporate bond market's unique characteristics. Credit spread magnitudes are determined by bond and firm-specific factors, and the implicit government guarantee for state-owned enterprises significantly reduces credit risk premiums for these companies. In the meantime, dynamics of the monetary policy tools, Treasury term structure and economic growth are significant determinants of excess bond premium at the market level.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2023

Author dc:creator

• Zhang, Sixia

Advisor dc:contributor.advisor

• Arestis, Philip

• bond risk pricing
• corporate bond default
• credit default risk
• credit spread puzzle
• risk modelling

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/69c49baf-2d1d-4ec2-822a-5510c8d37ef3/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.101791

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/357779