relationship between PD and recovery rate

[shanlane]

There is something inconsistent in the way Hull says these two are related.

At the beginning of either Ch 22 or 23 he says that as default rate increases, Recovery rates decrease.

Then, when talking about pricing CDSs, he says that the PD is appx proportional to 1/(1-RR). This says the opposite. As RR increases, PD seems to increase. Rs RR increases, (1-RR) decreases and 1/(1-RR) increases.

Am I missing something?

Thanks!

Shannon

 

[ChadWOB]

Without referencing any text, the relationship to me requires additional input. What type of industry does the business operate in? Holding PD constant, a company with more tangible assets would likely have a higher recovery rate and lower LGD vs. a service based company.

Intuitively, the PD and RR don't have an "apples to apples" comparison. I'm sure empirically they do, but I would think the nature of the business would influence recovery rates more than PD would.

 

[Aleksander Hansen]

While, in the original Merton (1974) framework, an inverse relationship between PD and RR exists, the credit risk models developed during the 1990’s treat these two variables as independent. The currently available and most used credit pricing and credit VaR models are indeed based on this independence assumption and treat RR either as a constant parameter or as a stochastic variable independent from PD. In the latter case, RR volatility is assumed to represent an idiosyncratic risk which can be eliminated through adequate portfolio diversification. This assumption strongly contrasts with the growing empirical evidence - showing a negative correlation between default and recovery rates – that has been reported in the previous section of this paper and in other empirical studies. This evidence indicates that recovery risk is a systematic risk component. As such, it should attract risk premia and should adequately be considered in credit risk management applications.
Empirical results, especially demonstrated by historical record high levels of recovery in the extreme benign credit environment of 2004-2007 and then the opposite credit market turmoil and high defaulted, low recovery environment of 2009, show the potential cyclical impact as well as the supply and demand elements of defaults and recoveries on LGD. Finally, we feel that the microeconomic/financial attributes of an individual issuer of bonds or loans combined with the market’s aggregate supply and demand conditions can best explain the recovery rate at default on a particular defaulting issue.

"Default Recovery Rates and LGD in Credit Risk Modeling and Practice"
Edward I. Altman