Price and Income Elasticities of Demand for Housing Characteristics in the City of Barcelona

Abstract

Garcia J. and Raya J. M. Price and income elasticities of demand for housing characteristics in the city of Barcelona, Regional Studies. This paper estimates price and income elasticities of demand for housing characteristics using the information of properties appraised in the city of Barcelona (1998–2001). It employs a two-stage approach, estimating hedonic price equations for different districts in order to estimate subsequently the corresponding demand equations. The results allow an analysis to be made of the complementarity or substitutability relationships among the characteristics analysed, and also to catalogue these characteristics as goods. By knowing the price, cross and income elasticities of housing characteristics, policy recommendations can be made about the type of housing units that are the most desirable to be subsidized.

Garcia J. et Raya J. M. Les élasticités par rapport au prix et au revenu de la demande de caractéristiques particulières du logement à Barcelone, Regional Studies. A partir des données sur l'immobilier à Barcelone (de 1998 à 2001), cet article cherche à estimer les élasticités par rapport au prix et au revenu de la demande de caractéristiques particulières du logement. On emploie une façon à deux temps qui estime des équations de prix hédonistes pour divers districts afin d'estimer par la suite les équations de demande correspondantes. Les résultats permettent d'analyser jusqu'à quel point les caractéristiques analysées sont complémentaires et substituables, et de classer aussi ces caractéristiques comme biens. En connaissant le prix, les élasticités croisées et les élasticités par rapport au revenu des caractéristiques du logement, on peut faire des recommandations de politique quant aux unités de logement les plus susceptibles d'être subventionnées.

Modèles hédonistes Elasticités Demande de logement

Garcia J. und Raya J. M. Preis- und Einkommenselastizität der Nachfrage nach Wohnungsmerkmalen in der Stadt Barcelona, Regional Studies. In diesem Beitrag schätzen wir die Preis- und Einkommenselastizität der Nachfrage nach Wohnungsmerkmalen mit Hilfe von Informationen über die in der Stadt Barcelona im Zeitraum von 1998 bis 2001 bewerteten Merkmale. Zum Einsatz kommt ein zweistufiger Ansatz, bei dem zunächst die hedonischen Preisgleichungen für verschiedene Bezirke geschätzt werden, um anschließend die entsprechenden Nachfragegleichungen zu schätzen. Die Ergebnisse ermöglichen eine Analyse der Komplementaritäts- und Ersetzbarkeitsbeziehungen zwischen den untersuchten Merkmalen; ebenso lassen sich diese Merkmale als Güter katalogisieren. Bei einer Kenntnis der Preis-, Kreuz- und Einkommenselastizität von Wohnungsmerkmalen lassen sich Empfehlungen für die Politik formulieren, was die Art der Wohnungseinheiten anbelangt, deren Subvention am wünschenswertesten ist.

Hedonische Modelle Elastizitäten Wohnungsnachfrage

Garcia J. y Raya J. M. Elasticidades de precios e ingresos en la demanda de características de la vivienda en la ciudad de Barcelona, Regional Studies. En este artículo calculamos las elasticidades de precios e ingresos en la demanda de las características de la vivienda usando información sobre las propiedades valoradas en la ciudad de Barcelona (1998–2001). Empleamos un enfoque bifásico, estimando las ecuaciones de precios hedónicos para los diferentes distritos a fin de calcular posteriormente las correspondientes ecuaciones de la demanda. Los resultados permiten un análisis de las relaciones de complementariedad o sustituibilidad entre las características analizadas; también podemos catalogar estas características como mercancías. Al conocer las elasticidades cruzadas, de precios y de ingresos de las características de las viviendas, pueden realizarse recomendaciones políticas acerca del tipo de viviendas que se deberían subvencionar.

Modelos hedónicos Elasticidades Demanda de viviendas

Keywords:

• Hedonic models
• Elasticities
• Housing demand

JEL classifications:

• C30
• R21

Acknowledgements

The authors acknowledge funding through the Project Number ECO2008-06395-C05-01, financed by the Ministry of Technology and Science.

Notes

Action by the Spanish public sector in terms of housing policy is described in depth by Rodríguez López (1990), López García (1992), Pareja and San Martín (1999), and García Montalvo and Mas (2000). In particular, Pareja and San Martín (2002) describe action by the Spanish public sector affecting the decision as to which tenure system individuals should adopt.

For a detailed analysis of Andrew Court's work and the reasons why his technique was neglected for more than two decades, see Goodman (1998).

The hedonic technique rests on modern consumer choice theory, according to which the consumer derives utility not directly from the good, but from its characteristics (Lancaster, 1966).

As examples of this literature, see, for example, Englund et al. (1999) and Chun et al. (1998).

As will be shown, sometimes socio-economic variables are also included in order to control for consumer heterogeneity or fiscal variables, with the aim of testing some hypothesis of public intervention.

The literature on labour supply models is a good example of the implications of non-linearity of the budget set in terms of estimation of leisure demand equations (labour supply), and also of the consequences of linearizing non-linear budget sets.

The most common solution to this endogeneity problem has been to use socio-demographic variables as instruments (Palmquist, 1984).

It should be stressed that the sample used is a selected sample, that is, a sample of potentially saleable properties; indeed, this is why they were valued. In this study, this has not been taken into account when estimating the hedonic price model. The evidence provided by Jud and Seaks (1994) shows that, in that particular case, the problem of sample selection had little effect on the results.

The price per square meter is a more correct specification of the dependent variable when the aim is to capture the variability in one market (Garcia et al., 2006). However, as Parsons (1986) pointed out, the correct specification of the dependent variable and the functional form is not important when the aim is variability between markets, as it is in the present paper. Moreover, the total value and a lineal specification avoid simultaneity and identification problems and make interpretation of the results in terms of implicit prices easier, since all the characteristics have a positive coefficient and, therefore, are goods.

Note that in Garcia et al. (2006), the dependent variable is the log of the price per square metre and the floor area appears in an explanatory variable with the definition:

where S is the floor area; and θ is a parameter that, if equal to zero, corresponds to a model with constant area elasticity.

In the case of Erekson et al. (1979), the expenditure on quality is obtained by the difference between the total value predicted by the hedonic equation, expenditure on floor area, and expenditure on location.

In Garcia et al. (2006), the contribution of location to explaining the variability of price per square metre is quantified as representing 53.58% of the explained variation.

In this second stage, the observations corresponding to District 4 (Les Corts) were removed because it was the least representative sample (347 observations) and because the price index calculated for some observations was negative.

Keeping the utility and the rest of the prices constant.

Keeping the expenditure and the rest of the prices constant.

The evidence provided is robust to alternative market segmentation (segmentation by year), which can be estimated because of the certain temporal dimension of the database. The estimation results and the elasticites are very similar between segmentations and can be obtained from the authors upon request.

For the econometric evidence, see Garcia et al. (2006).

Garcia et al. (2006) provided a detailed analysis with some conclusions on urban economics, explaining price differences between areas in terms of their characteristics.