Modeling Context in FHA Mortgage Terminations: The Impact of Local Violence, Natural Disasters, and Climate on Individual Loan Behavior

• In Event: Everyday Concerns: Housing and Transit Policy Causes and Consequences

Abstract

Scholarship on mortgage delinquency and default predominantly focuses on the micro and macroeconomic determinants. While this body of work is valuable, it tends to overlook a critical influence on individual behavior and decision-making: neighborhood effects. In this study, we shift the focus to the neighborhood context, investigating the impact of local violence, natural disasters, and temperature fluctuations on loan decision-making for FHA-backed mortgages.
Our findings affirm the importance of borrower characteristics, mortgage details, and macroeconomic indicators in forecasting loan performance. However, neighborhood-level variables also emerge as robust predictors of an individual's loan behavior.

In this work, we leverage a novel dataset. By cross-walking six publicly available datasets, along with FHA loan performance data, we have assembled a comprehensive, multi-tiered panel to support our analyses. This paper examines the concurrent mortgage termination risks associated with 90-day delinquency and prepayment for single-family fixed-rate FHA mortgages originating between 2011 and 2019.

We run multinomial logit models to test the relationship between (1) violent crime, (2) natural disasters, and (3) dramatic temperature fluctuation and their respective impact on loan prepayment and delinquency.

This comprehensive study offers valuable insights for policymakers and practitioners in the realm of housing finance at both the local and national levels.

Authors

• S Hawke, HUD

• Wenzhen Lin