The effect of changing mortgage payments on default and prepayment: Evidence from HAMP resets

Abstract

The Home Affordable Modification Program (HAMP) is a government-sponsored program to reduce the monthly mortgage payments of borrowers who are in danger of default. After five years of below-market interest rates, HAMP interest rates jump predictably, increasing annually in increments of up to one percentage point until they reach a pre-determined market rate. We identify the causal effect of increasing interest rates – and with them, monthly payments – on default (as well as on delinquency transitions and prepayment) with an event study design, comparing default rates immediately before and after rate reset times for loans that do and do not reset. Since the size of the interest rate reset is a kinked function of the difference between the subsidized interest rate and the market interest rate, we also identify the effect using a regression kink design (RKD). We find that a one percentage point rate increase led to a roughly 20 percent increase in the default hazard (e.g., from 0.6 percent per month to 0.72 percent per month). Our estimates imply that the payment reduction associated with a one percentage-point decline in the mortgage rate has a similar default-reducing effect as principal reduction in HAMP’s Principal Reduction Alternative (as estimated by Scharlemann and Shore (2016)) at much lower cost. JEL Codes: G21 (Mortgages), R30 (Real Estate Markets, General) ∗Views and opinions expressed are those of the authors and do not necessarily represent official positions or policy of the Office of Financial Research or the U.S. Department of the Treasury. The authors would like to thank Karen Dynan, Stathis Tompaidis, and seminar participants at the Office of Financial Research, the University of Southern California, and Brookings. The authors are grateful for assistance with HAMP data from Emily Haeckel, Mark McArdle, and Aaron Ferguson. †Office of Financial Research, U.S. Treasury Department. The author declares that she has no relevant or material financial interests that relate to the research described in this paper. Email: therese.scharlemann@ofr.treasury.gov ‡Georgia State University, Department of Risk Management and Insurance. The author declares that he has no relevant or material financial interests that relate to the research described in this paper. Email: sshore@gsu.edu