Credit risk models for mortgage loan loss given default

Arguably, the credit risk models reported in the literature for the retail lendingsector have so far been less developed than those for the corporate sector,mainly due to the lack of publicly available data. Having been given access to adataset on defaulted mortgages kindly provided by a major UK bank, this workfirst investigates the Loss Given Default (LGD) of mortgage loans with thedevelopment of two separate component models, the Probability of Repossession(given default) Model and the Haircut (given repossession) Model. They are thencombined into an expected loss percentage. Performance-wise, this two-stageLGD model is shown to do better than a single-stage LGD model (which directlymodels LGD from loan and collateral characteristics), as it achieves a better Rsquarevalue, and it more accurately matches the distribution of observed LGD.We next investigate the possibility of including macroeconomic variables intoeither or both component models to improve LGD prediction. Indicators relatingto net lending, gross domestic product, national default rates and interest ratesare considered and the interest rate is found to be most beneficial to bothcomponent models. Finally, we develop a competing risk survival analysis modelto predict the time taken for a defaulted mortgage loan to reach some outcome(i.e. repossession or non-repossession). This allows for a more accurateprediction of (discounted) loss as these periods could vary from months to yearsdepending on the health of the economy. Besides loan- or collateral-relatedcharacteristics, we incorporate a time-dependent macroeconomic variable basedon the house price index (HPI) to investigate its impact on repossession risk. Wefind that observations of different loan-to-value ratios at default and differentsecurity type are affected differently by the economy. This model is then usedfor stress test purposes by applying a Monte Carlo simulation, and by varying theHPI forecast, to get different loss distributions for different economic outlooks