A Reinvestigation on the Cointegration of Consumption, Wealth, and Labor Income

john qi zhu
DOI: https://doi.org/10.2139/ssrn.758684
2005-01-01
SSRN Electronic Journal
Abstract:Lettau and Ludvigson (2001a, 2001b)'s striking finding that short term deviations from a cointegrating relationship among aggregate consumption, wealth, and labor income can forecast future asset returns has been a locus of interests, as well as scrutinies, of many researchers. Particularly, Rudd and Whelan (2002) question the empirical presence of cointegration relationship of aggregate consumption, wealth, and labor income, on which Lettau and Ludvigson rest their whole argument. Using disaggregate individual level data from Consumer Expenditure Survey, this paper studies the proposed cointegration relationship among consumption, wealth, and labor income from a perspective of household level data. We construct variables to be consistent with underlying consumer budget constraints for representative households with limited participation in asset markets. The paper convincingly reconciles the conflicting views between Lettau/Ludvigson and Rudd/Whelan, and provides mild to strong supporting evidence for the existence of cointegration relationship between consumption, wealth, and labor income.
What problem does this paper attempt to address?