Housing Wealth and Consumption Growth: Evidence from a Large Panel of Households

Abstract

This paper uses a large panel dataset that tracks the housing wealth and spending behavior of 12,793 individuals in Hong Kong to study how housing wealth affects household consumption. Housing wealth is measured based on the repeated-sales price indices of nine districts in Hong Kong, which are estimated using government registry of all the 900,000 repeated housing transactions. Consumption is measured by credit card charges provided by the five largest credit card issuers in Hong Kong. These detailed data make it possible to identify consumption responses to housing wealth based on time-series variation within individual households (through household fixed effects). Further, rich variations across households help pin down the mechanism of the observed consumption sensitivity.

I find a significant effect of housing wealth on consumption. Such an effect seems to be on account of a reduction in precautionary saving, as opposed to relaxation of borrowing constraints. In particular, consumption sensitivity exists even in the absence of refinancing and occurs only among less leveraged households and younger households who behave like "buffer stock" consumers (Gourinchas and Parker, 2002). Finally, consumption does not respond to predictable changes in housing wealth, which is inconsistent with liquidity constraints. The results highlight the importance of housing wealth in influencing consumption even in the absence of refinancing and relaxation of credit constraints.

Keywords: Housing wealth, Consumption, Liquidity constraints, Precautionary saving, Credit cards, Life cycle

JEL Classification: E21, G21