A Dynamic Model of Reverse Mortgage Borrower Behavior

Jason R. Blevins, Wei Shi, Donald R. Haurin, and Stephanie Moulton.

Abstract. We carry out an empirical analysis of the Home Equity Conversion Mortgage (HECM) program using a unique and detailed dataset on the behavior of HECM borrowers from 2006–2012. We semiparametrically estimate a structural, dynamic discrete choice model of borrower behavior. To avoid bias in predictions and welfare analysis, our estimator is based on a new identification result for models with distinct, but inter-related terminating actions that allows us to identify the entire utility function without the need to impose an ad hoc normalization. The estimated utility functions provide insights about the factors that influence HECM refinance, default, and termination decisions. We quantify the trade-offs involved for modifications to the HECM program that would impose income and credit requirements. Our results indicate that these modifications would be effective in reducing undesirable HECM outcomes, at the expense of excluding some borrowers, and we quantify the welfare losses due to restricting access to the HECM program. We also investigate how shocks to housing prices and income and improvements in credit scores affect HECM outcomes and we estimate the corresponding changes in household welfare. In particular, our results shed light on how much households value the house price insurance component of the HECM program.

Keywords: reverse mortgages, mortgage default, senior housing, dynamic discrete choice, identification, semiparametric estimation.

JEL Classification: R21, G21, C25, C61.

BibTeX Record:

  author       = {Jason R. Blevins and Wei Shi and
                  Donald R. Haurin and Stephanie Moulton},
  title        = {A Dynamic Model of Reverse Mortgage Borrower Behavior},
  year         = {2016},
  type         = {Working Paper},
  institution  = {Ohio State University}