This study examines mortgage termination behavior at the household lev
el, including fixed-rate mortgage terminations due to call, default, a
nd mobility behavior. It exploits information in the American Housing
Survey to examine national cross sections of 6119 and 4160 owner house
holds. While accounting for the influence of interest rate movements i
n motivating termination-the call option influence-the study uses two
sequential panels to investigate the modifying effect of household fin
ancial, demographic, and location characteristics upon termination beh
avior. Although we find the effect of the call option (systematic effe
ct) to be highly significant in explaining terminations in a declining
interest rate environment, we also find that demographic and location
al effects (diversifiable) are potentially at least as significant. Th
is result suggests that incomplete diversification over the nonsystema
tic factors of demography and location could cause significant variati
on in termination rates. (C) 1997 Academic Press.