Integrability of demand accounting for unobservable heterogeneity: a test on panel data

Mette Christensen

No W07/14, IFS Working Papers from Institute for Fiscal Studies

Abstract:

In recent years it has become apparent that we must take unobservable heterogeneity into account when conducting empirical consumer demand analysis. This paper is concerned with integrability (that is, whether demand is consistent with utility maximization) of the conditional mean demand (that is, the estimated demand) when allowing for unobservable heterogeneity.

Integrability is important because it is necessary in order for the demand system estimates to be used for welfare analysis. Conditions for conditional mean demand to be integrable in the presence of unobservable heterogeneity are developed in the literature. There is, however, little empirical evidence suggesting whether these conditions for integrability are likely to be met in the data or not. In this paper we exploit the fact that the integrability conditions have testable implications for panel data and use a unique long panel data set to test them. Because of the sizeable longitudinal length of the panel, we are able to identify a very flexible specification of unobservable heterogeneity: We model individual demands as an Almost Ideal Demand system and allow for unobservable heterogeneity by allowing all intercept and slope parameters of the demand system to be individual-specific. We test the conditions for integrability of the conditional mean demand of this demand system. We do not reject them. This means that the conditional mean demand generated by a population of consumers with different preferences described by different Almost Ideal Demand systems is consistent with utility maximization. Given that integrability is not rejected, we conclude by an comparing the estimated demand system elasticties and welfare effects from a model with no heterogeneity (which is the model that would usually be estimated from cross sectional data) to those obtained from our heterogeneous model.

We find that the homogeneous model severely overestimates income elasticities for luxury goods and that the welfare effects from the heterogeneous model exhibit a large amount of heterogeneity, but deviate with only a few percentage points from the homogeneous model at the mean.

Date: 2007-09-21
New Economics Papers: this item is included in nep-upt
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (9)

Downloads: (external link)
http://www.ifs.org.uk/wps/wp1407.pdf (application/pdf)
Our link check indicates that this URL is bad, the error code is: 404 Not Found (http://www.ifs.org.uk/wps/wp1407.pdf [301 Moved Permanently]--> https://www.ifs.org.uk/wps/wp1407.pdf [302 Found]--> https://ifs.org.uk/wps/wp1407.pdf)

Related works:
Working Paper: Integrability of Demand Accounting for Unobservable Heterogeneity: A Test on Panel Data (2007)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ifs:ifsewp:07/14

Ordering information: This working paper can be ordered from
The Institute for Fiscal Studies 7 Ridgmount Street LONDON WC1E 7AE

Access Statistics for this paper

More papers in IFS Working Papers from Institute for Fiscal Studies The Institute for Fiscal Studies 7 Ridgmount Street LONDON WC1E 7AE. Contact information at EDIRC.
Bibliographic data for series maintained by Emma Hyman ().