Consumption and habits : evidence from panel data

Abstract

The purpose of this paper is to test for the presence of habit formation in consumption decisions using household panel data. We use the test proposed by Meghir and Weber (1996) and estimate the within -period marginal rate of substitution between commodities, which is robust to the presence of liquidity constraints. To that end, we use a Spanish panel data set in which households are observed up to eight consecutive quarters. This temporal dimension is crucial, since it allows us to take into account time invariant unobserved heterogeneity across households ("fixed effects") and, therefore, to investigate if the relationship between current and past consumption reflects habits or heterogeneity. Our results conf irm the importance of accounting for fixed effects when analyzing intertemporal consumption decisions allowing for time non-separabilities. Once fixed effects are controlled for and a proper set of instruments is used, the results yield supporting evidence of habit formation in the demand system of food at home, transport and services.