A prudent loss given default estimation for mortgages. II

Bogie Ozdemir and Emma Huang

Abstract

When the possibility of another house price correction continues to be a concern, it is important to accurately estimate loss given default (LGD) for mortgage portfolios and be able to stress test those portfolios effectively. In this paper we introduce a prudent methodology to do so. The methodology builds on the earlier Journal of Risk Model Validation paper by Ozdemir, which provides an accurate estimation of “workout” LGD by directly modeling the house value at default by incorporating market effects and potential appraisal biases. Improving on it, in this extension the probability of “curing” and the probability of full payment after default (“exit”) are explicitly modeled with respect to the inverse of the loan-to-value (LTV) ratio. This makes the LGD estimation more accurate and risk sensitive, and particularly suitable for stress testing purposes. While the methodology is presented in a mortgage setting, it can easily be applied to all types of secured lending where the collateral value relative to exposure at default (ie, the LTV) is likely to influence the potential outcomes following the default.