This study investigates the relationship between credit availability and household consumption using a novel approach to separate credit demand and supply. We find that a deterioration in local-bank health reduces household consumption, with the strongest effects occurring for households that are more likely to need credit—especially those experiencing a negative income shock and having limited liquid assets. The main contributions of the study are the use of an arguably exogenous measure of local-bank health and multifaceted indicators of constrained households. Our findings contribute to the discussion of the linkages between the financial sector and real economic activity.

Rising house prices affect household spending by either loosening a household's lifetime budget constraint (pure wealth effect) or the household's borrowing constraint so that consumption rises toward the level implied by the consumption Euler equation (borrowing collateral effect). The empirical findings in this paper are consistent with house price appreciation affecting household spending through the borrowing collateral channel and not the pure wealth effect channel. The consumption of potentially borrowing constrained households increases between [dollar]0.06 and [dollar]0.18 per dollar increase in their housing equity, while the consumption of unconstrained households is little changed.