Household Debt Quintiles: Explaining Changes 1983–1989

Abstract

Households' changes in consumer debt from 1983–89 were examined using the panel dataset of the Survey of Consumer Finances. Logit models of the odds of an increase and a decrease in debt quintile revealed four factors hypothesized by Bryant (1990) to be symmetrically related—household size, respondents marital status, inflationary expectations, and time preferences. Asymmetrical effects of other variables, including life cycle consumption needs, present resources, expected future resources, and preferences were found for either the odds of an increase or decrease in consumer debt, but not both. Households' access to credit markets and changes in economic status between 1983 and 1989 also significantly affected the odds of change in consumer debt status, albeit not consistently across models. Household level change in consumer debt and the factors that affect it are more complex than hypothesized.