Another look at the spending response to the 2001 income tax rebates

This article revisits the spending response to the 2001 US tax rebates by focussing on two key aspects of how tax policy researchers use the Consumer Expenditure Survey (CEX). These two attributes, which are often overlooked, are as follows: the measures used for consumption and the ‘outlier’ criteria applied to the data. First, I reproduce the results in Johnson et al. (2006 Johnson, DS, Parker, JA and Souleles, NS. 2006. Household expenditure and the income tax rebates of 2001. American Economic Review, 96: 1589–610. [Crossref], [Web of Science ®] , [Google Scholar]), which (using the CEX) concluded that households immediately spent 20–40% of their rebates on nondurable consumption goods. Then, I show how making two changes – both of which are relied upon in the literature – affects their results. These adjustments reduce the estimated magnitude of the rebate's impact by as much as 100%.