Size of home, homeownership, and the mortgage interest deduction

Abstract

This paper offers an empirical test of the effect of the mortgage interest deduction (MID) on both the extensive (own vs. rent) and intensive (size of home) housing purchase margins. Using state level differences in MID availability to identify, I examine this relationship using standard ordinary least squares, instrumental variables, regression discontinuity, and sample selection estimation techniques. I find the MID to be responsible for a 10.9–18.4% increase in the size of home purchased, but that no relationship exists between the MID and home ownership. These results imply an elasticity of home size with respect to changes in user cost between −1 and −1.4.