Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/86734 
Erscheinungsjahr: 
2010
Schriftenreihe/Nr.: 
Tinbergen Institute Discussion Paper No. 10-004/2
Verlag: 
Tinbergen Institute, Amsterdam and Rotterdam
Zusammenfassung: 
We determine the magnitude and nature of systematic default risk using 1971{2009) default data from Moody's. We disentangle systematic risk factors due to business cycle effects, common default dynamics (frailty), and industry-specific dynamics (including contagion). To quantify the contribution of each of these factors to default rate volatility we introduce a new and flexible model class for factor structures on non-Gaussian (defaults) and Gaussian (macro factors) data simultaneously. We find that all three types of risk factors (macro, frailty, industry/contagion) are important for default risk. The systematic risk factors account for roughly one third of observed default risk variation. Half of this is captured by macro and financial market factors. The remainder is captured by frailty and industry effects (in roughly equal proportions). The frailty components are particularly relevant in times of stress. Models based only on macro variables may both under-estimate and over-estimate default activity during such times. This indicates that frailty factors do not simply capture missed non-linear responses of defaults to business cycle dynamics. We also find significant differences in the impact of crises on defaults at the sectoral level, implying frailty as well as contagion may play a role in systematic default clustering. Finally, we show that the contribution of frailty and industry factors on top of macro factors is economicallysignificant for assessing portfolio risk.
Schlagwörter: 
systematic default risk
credit portfolio models
mixed-measurement dynamic factor model
frailty-correlated defaults
state space methods
dynamic credit risk management
JEL: 
G21
C33
Dokumentart: 
Working Paper
Erscheint in der Sammlung:

Datei(en):
Datei
Größe
774.6 kB





Publikationen in EconStor sind urheberrechtlich geschützt.