Predict Loss Given Default

Business Problem

Mortgage loans are typically secured on the borrower’s property, which serves as collateral for the loan. Before approval, mortgage loans are assessed for credit risk and loss in the case of default. These two risk metrics help financial institutions include in their cash flows their expected losses; a regulatory requirement which must be maintained on a continuous basis. In the case of non-collateral loans, once the borrower defaults, the financial institution will likely incur a direct loss on the amount owed to them. However, in case of mortgage loans, the financial institution can get back the entire/partial amount owed by selling the collateral. Being that financial institutions take on millions if not billions of dollars of risk in their mortgage portfolio, it is imperative that they reduce their risk and uncertainty by predicting in advance the loss they will incur given defaults.

Intelligent Solution

AI can help your organization make predictions of loss given default on your mortgages. With advancements in predictive technology, financial institutions can now create more accurate predictions on their loss given defaults (LGD) with historical default data. Being that Expected Loss is the product of Probability of Default and Loss Given Default, or EL = Pd x LGD, the accuracy by which we can predict LGD is a core component of forecasting loss. This information can help your mortgage writing team more precisely judge risk in your mortgage loan portfolio, and assists your organization in setting loan loss reserves, forecasting losses, and maintaining capital adequacy.