I'm having trouble grasping this concept; I don't see the relevance of the explanation given in the text (Gregory, Counterparty Credit Risk and CVA) either.
When expected exposure and probability of default are positively correlated (wrong way risk), the exposure conditional on default is higher than if the two were independent. This part makes sense. But, according to Gregory (below), the conditional exposure will be higher for a smaller PD than it would for a larger PD.
Gregory offers an explanation, that exposure will decline because for higher quality counterparties the increase in PD is unexpected. I don't see how this affects it though. Why does an unexpected increase in PD cause exposure to fall? Why would exposure increase more with a smaller probability of default than it would with a larger one?
From John Gregory:
We can see that with 50% correlation, wrong-way risk approximately doubles the EE whilst with 50% correlation the impact of right-way risk reduces it by at least half. This is exactly the type of behavior expected: positive correlation between the default probability and exposure increases the conditional expected exposure (default probability is high when exposure is high), which is wrong-way risk. Negative correlation causes right-way risk.
Let us look into this simple model in a bit more detail. Consider now the impact of counterparty default probability on the EE with wrong-way risk. Figure 15.2 shows the EE using three different hazard rates, indicating that the exposure decreases as the credit quality of the counterparty also decreases. This result might seem at first counterintuitive but it makes sense when one considers that for a better credit quality counterparty, default is a less probable event and therefore represents a bigger surprise when it comes.