An appraisal of the wealth effect in the US: evidence from pseudo-panel data

How does household wealth influence consumption? The empirical evidence brought so far by the literature is unclear, mostly because of the low quality of the data more readily available: aggregate data, cross sections and panel datasets lacking important variables all present major shortcomings for a proper analysis of the wealth effect. The aim of our paper is to contribute to the appraisal of the wealth effect using a new, accurate dataset, and employing a proper estimation technique. We perform a pseudo-panel analysis for the USA (1989-2007) combining information from the Consumer Expenditure Survey and the Survey of Consumer Finances. We divide between durables and non durables consumption, and we also investigate the roles of the different components of household wealth, both gross and net. Our estimates indicate that there is a significant tangible wealth effect (between 3 and 6 cents per dollar), confirming the economic importance recognized by some of the previous empirical literature. On the contrary, financial wealth seems to have no significant effects on consumption, even when debt considerations are included in the analysis. In addition, the wealth effect seems to matter more for older households, for which both the house of residence and the rest of the real estate properties positively affect consumption.