Do Large-Scale Refinancing Programs Reduce Mortgage Defaults? Evidence from a Regression Discontinuity Design

Abstract

In 2012, the Federal Housing Administration (FHA) reduced fees to refinance FHA-insured mortgages obtained before -- but not after -- a retroactive deadline. We use a natural experiment to study how reduced mortgage payments affect default rates. Using a regression discontinuity design, we find that reducing payment size by 1 percent lowers conditional default rates by 2.75 percent. Evidence suggests that those effects are larger for borrowers with negative equity and lower credit scores. We estimate that the policy will prevent more than 35,000 defaults of FHA-insured mortgages, saving FHA nearly $1 billion in present-value terms.