Estimation of Default Probability by Three-Factor Structural Model

by Cho-Hoi Hui of the Hong Kong Monetary Authority,
Chi Fai Lo of the Chinese University of Hong Kong, and
Ming Xi Huang of the Chinese University of Hong Kong

October 10, 2002

Abstract: This paper develops a three-factor structural model for estimating probability of default. The model incorporates the stochastic asset value of a corporate, liability and risk-free interest rate with time-dependent model parameters. A corporate defaults when its leverage ratio increases above a predefined default-triggering level. Using average market data for corporates with different external credit ratings, the three-factor model is capable of producing the term structure of probabilities of default for rated corporates, that are broadly matched with the average default rates of the corresponding ratings reported by Standard & Poor's. The numerical results show that the estimation of probability of default is sensitive to the time-dependent target leverage ratio. The three-factor model can be applied to the estimation of probability of default under the internal ratings-based approach proposed in the New Basel Capital Accord.

JEL Classification: G21, G28, G13.

Keywords: Credit Risk Models, Credit Rating, Default Probability, Bank Regulation.

Download paper (273K PDF) 14 pages