Consumption and habits : evidence from panel data

The purpose of this paper is to test for the presence of habit formation in consumption decisions using household panel data. We apply the test proposed by Meghir and Weber (1996)to a Spanish panel data set in which households are observed for up to eight consecutive quarters. This temporal dimension is crucial, because it allows us to take into account time invariant unobserved heterogeneity across households (‘fixed effects’) and, therefore, to investigate whether the relationship between current and past consumption reflects habits or heterogeneity. Our results confirm the importance of accounting for fixed effects when analysing intertemporal consumption decisions allowing for time non-separabilities.The first and the second authors acknowledge research funding from the Spanish DGES, Grant BEC2002 04294 C02 01 02Publicad