The Journal of Credit Risk (1–33) Volume 6/Number 2, Summer 2010
- Shumway (2001) argues that the use of static models with multi-period data leads to estimates which are biased, inconsistent and inefficient and proposes the use of discrete time hazard model
- B.Headd (2003) finds that only one-third of new businesses (33%) closed under circumstances that owners considered unsuccessful
- Hudson (1987) studying UK companies between 1978-1981 find that young companies form the majority of the liquidated companies and that a company needs at least 9 years to be regarded as established. Point out that newly formed company is most likely to have a honey moon period of around 2 years before being in real risk
- Use total asset values to control for company size
- Rather than using industrial sector dummy variables use weight of evidence variable which expresses the previous years’ sector failure rate
- Addition of non-accounting (qualitative) data to the basic Z-score model significantly improves the classification performance
- The late filing of accounts is associated with a higher probability of failure. (variable no cash flow statement)
- Subsidiaries are less likely than non-subsidiaries because they have access to the financial and other resources of the parent company and can survive poor financial performance for longer than non-subsidiaries
- There is a non-linearity between probability of insolvency and size as measured by asset value
- Control of industry factor is significant and picks up the effects of average sector level failure
- Having liquidity and cash is associated with lower PD
- Smaller subsidiaries are not supported by parent in the same way as larger ones
- The addition of qualitative information improved accuracy by 13% (AUC 0.67->0.76)
- Can update qualitative information frequently to better re-estimate the PD over time