We present an analysis of the sensitivity of household mortgage probabilities of default (PDs) and loss given default (LGDs) on unemployment rates, house price growth, interest rates, and other drivers. A structural micro-macro simulation model is used to that end. It is anchored in the balance sheets and income-expense flow data from about 95,000 households and 230,000 household members from 21 EU countries and the U.S. We present country-specific nonlinear regressions based on the structural model simulation-implied relation between PDs and LGDs and their drivers. These can be used for macro scenario-conditional forecasting, without requiring the conduct of the micro simulation. We also present a policy counterfactual analysis of the responsiveness of mortgage PDs, LGDs, and bank capitalization conditional on adverse scenarios related to the COVID-19 pandemic across all countries. The economics of debt moratoria and guarantees are discussed against the background of the model-based analysis.