Author: Yingqi XuPublisher:
ISBN:
Category :
Languages : en
Pages : 0
Book Description
This dissertation examines mortgage curtailment, an important household deleverage channel that has received little attention in the literature. In the first chapter, using Fannie Mae single-family loan performance data, I show that more than 20% of mortgage borrowers opt to curtail their mortgage payments in a given quarter if their mortgage was originated after 2009. These borrowers pay an additional USD500, which represents over 50% of their monthly mortgage payment and 15% of their income. Mortgage origination conditions matter more than mortgage lifetime events in determining curtailment. Mortgages originated after 2009 have a higher propensity for mortgage curtailment compared to those originated before 2009, at 22% versus 14%. Observable mortgage characteristics account for 60% of this difference, with higher income, higher credit scores, and lower debt-to-income ratios positively correlated with a higher curtailment propensity. Moreover, borrowers use curtailment as an alternative to refinancing when refinancing is not cost-effective in reducing interest costs, as shown by the negative correlation between mortgage rate incentive and curtailment propensity. In the second chapter, I investigate borrowers' mortgage curtailment behavior after refinancing through the HARP program. On average, borrowers who refinance through HARP pay USD223 more per month than before they enter the program, which is equivalent to 18.7% of the monthly required mortgage payment. Additionally, higher income, higher credit scores borrowers, and mortgages with lower origination LTV ratios tend to contribute more. To address the endogeneity issue, I use the quasi experiment setup of the HARP program's eligibility condition. The results suggest that HARP eligibility leads borrowers to curtail USD37.7 more than ineligible borrowers, with high original LTV borrowers being more likely to contribute more. Households increase curtailment while both income effect and intertemporal substitution effect are at play, indicating that they are more sensitive to changes in liquidity than to changes in mortgage rates.
