Envy and Habits: Panel Data Estimates of Interdependent Preferences
| DOI | http://doi.org/10.1111/obes.12111 |
| Date | 01 August 2016 |
| Published date | 01 August 2016 |
443
©2015 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 78, 4 (2016) 0305–9049
doi: 10.1111/obes.12111
Envy and Habits: Panel Data Estimates of
Interdependent Preferences*
Francisco Alvarez-Cuadrado†, Jose Maria Casado‡ and
Jose Maria Labeaga§
†McGill University, 855 Sherbrooke Street West, Montreal H3A 2T7, CA USA
(e-mail: francisco.alvarez-cuadrado@mcgill.ca)
‡Banco de Espa˜na, Alcal´a 48, Madrid 28014, Spain (e-mail: jmcasado@bde.es)
§UNED, Paseo Senda del Rey 11, Madrid 28040, Spain (e-mail: jlabeaga@cee.uned.es)
Abstract
We estimate the importance of preference interdependence from consumption choices.
Our strategy follows the literature that tests the constraints imposed by optimality on the
evolution of individual consumption.The introduction of habits and envy places additional
restrictions on the evolution of the optimal consumption path. We use a unique data set to
test these restrictions. Our estimates suggest that, if one defines utility over consumption
services, a large fraction of these services is relative, with one third of the weight placed
in the consumption of the reference group and another third placed in the agent’s past
consumption.
I. Introduction
The assumption that preferences are separable across time and across households is standard
in the economic literature, but it is not particularly appealing. Indeed, social scientists have
long stressed the relevance of habit and status-seeking as being important characteristics
of human behaviour. In our discipline origins of this proposition can be traced as far back
as Smith (1759) and Veblen (1899), although it was not until the works of Duesenberry
(1949), Pollak (1976) and Ryder and Heal (1973) that an effort was made to provide these
ideas with some micro-theoretic foundations. The subsequent literature has associated two
types of reference consumption levels to these non-separabilities. The first is an internal
criterion based on the individual’s own past consumption levels. This case is often referred
to as characterizing ‘habit formation’or ‘intrinsic habit’, Rozen (2010). The second is based
JEL Classification numbers: C23, D12, D91
*The authors are grateful for helpful comments from two anonymous referees and seminar participants at the
SAEe 2011, the EEA/ESEM 2012, the NASM 2013 and the Banco de Espa˜na. We also acknowledge financial
support from the Ministerio de Ciencia y Tecnologia under grant ECO2012-39553-C04-01. Additionally, Alvarez-
Cuadrado acknowledges financial support from the Social Sciences and Humanities Research Council under grant
SSHRC 430-2013-000424. Viewsexpressed in this paper are those of the authors and should not be attributed to the
Banco de Espa˜na or the Eurosystem.
444 Bulletin
on an external criterion, expressed in terms of the consumption of some outside reference
group, typically the average consumption of the community, or the overall economy. This
is often referred to as ‘envy’ as inVarian (1974), ‘catching up with the Joneses’ as inAbel
(1990), ‘keeping up with the Joneses’ as in Gali (1994), ‘status’as in Cor neo and Jeanne
(2001), ‘jealousy’ as in Dupor and Liu (2003) or ‘rivalry’ as in Bruni and Porta (2005) or
‘consumption externalities’ as in Liu and Turnovsky (2005).
A large body of empirical work investigatesthe impor tance of habit formation for con-
sumption behaviour.1The point of departure in any of these studies is an Euler equation
derived under a preference specification that allows for temporal interdependencies.Then,
a linearized version of this equation is estimated using time-series data on consumption
and asset returns. Using UK data, Osborn (1988) introduces a consumption specification
that allows for seasonal variation and habit persistence finding that the habit coefficients
are jointly significant. Ferson and Constantinides (1991) find evidenceof habit persistence
dominating durability at monthly, quarterly and annual frequencies. Fuhrer and Klein
(2006) and Fuhrer (2000) relying on a utility function that assigns relative weights to both
current consumption and an internal benchmark find that 80% of the weight should be
attached to the latter. Dynan (2000) uses panel data, specifically food consumption from
the Panel Study of Income Dynamics, and finds no evidence of habit formation at the
annual frequency. As Carrasco, Labeaga and David L´opez-Salido (2005) point out this
result could be a consequence of unobserved heterogeneity across households, and show
that, after controlling for fixed effects, food consumption and services exhibit habit forma-
tion. Browning and Collado (2007) find evidence of habit formation in certain categories
of goods, such as, food out of home, alcohol and tobacco, although they conclude that
the intertemporal dependence is not sufficiently strong to make composite consumption
significantly habit forming. Finally, following the revealed preference tradition, Crawford
(2010) characterizes a set of identifying restrictions for the habit formation model. His
results suggest that the introduction of habit formation in the standard discounted utility
model improves its explanatory power considerably, virtually to the point where 100% of
the micro-data are perfectly rationalizable if one allows intertemporal complementarities
for many goods.
At the theoretical level, envy has been introduced to rationalize several departures from
the predictions of the standard paradigm that assumes preferences that are separable across
households. Abel (1990) and Gali (1994) rely on interpersonal comparisons to account for
the excess return on equity. Carroll et al. (2000) explore the implications of relative con-
sumption for the process of capital accumulation. Alonso-Carrera, Caballe and Raurich
(2007) study the impact of interpersonal comparisons in an economy displaying dynastic
altruism. Akerlof andYellen (1990) present a model of worker behaviour where individual
effort does not only depend on the workers’own wage but also on the wagereceived by their
coworkers.Liu and Turnovsky (2005) explorethe impact of envy on labour supply choices.
Wendner and Goulder (2008) find that the marginal excess burden from taxation is lower
1The introduction of habits in the standard consumption model induces agents to adjust slowly to permanent
income shocks and this helps rationalizing the reported excess smoothness puzzle. See, for instance, Campbell and
Deaton (1989). Furthermore, habit formation has been used to reproduce the hump-shaped response of aggregate
spending to monetary shocks (Fuhrer (2000)), the link between savingand g rowth(Car roll, Overlandand Weil, 2000)
and to improve the empirical fit of business cycle models (Boldrin, Christiano and Fisher, 2001).
©2015 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial