Abstract

The Advanced-IRB banks should be able to demonstrate to the regulatory supervisors that the long-run exposure at default (EAD) and downturn (or stress) EADs are validated with their own data on historical exposures. This paper proposes an empirical stress EAD model that is driven by a risk driver of the credit conversion factor (CCF) and by a systematic factor governing the cyclical effects of EAD and PD of facilities. Stress EAD is then computed as the expected value of EAD conditional on a particular value of the risk driver of a facility and on the value of a systematic factor that achieves, say, a 99.9% confidence level for a desired A or BBB+ rating. The reference data set (RDS) studied in the paper covers the corporate credit card exposures from 2001:Q1 to 2005:Q4. The most intriguing aspect of this RDS is that the sample period covers the severe 2003 credit card distress witnessed in Korea, which lenders it an excellent candidate for developing and validating the stress EAD models. The empirical evidence would provide ample opportunity to a better understanding of the meaning of Φ ?1 (.999) in the New Capital Accord.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call