Borrowing constraints and access to owner-occupied housing

Abstract

Cross-sectional data from the 1983 Survey of Consumer Finances (SCF) enable us to identify two groups of households: those whose housing tenure status is unaffected by borrowing constraints, and those whose housing tenure status may be affected by borrowing constraints. Using these data, a bivariate probit model is estimated to evaluate the joint probability of whether families prefer to live in owner-occupied housing and whether borrowing constraints affect access to owner-occupied housing. Findings indicate that borrowing constraints have a significant negative effect on homeownership rates. Indeed, if borrowing constraints had not been binding in our sample period (the early 1980s), ceteris paribus, owner-occupancy rates in the United States would have risen from 64.5% to nearly 73%. Moreover, borrowing constraints appear to have a disproportionate effect on the ability of younger families and non-white families to own a home, consistent with popular perceptions.