Asymmetric behaviour of Spanish regional house prices.

Abstract In connection with the housing market, which is presently raising a great deal of concern among the general public, this paper investigates regional housing prices in Spain using variable co-integration techniques. It analyzes the asymmetric behavior in real house prices among the Spanish regions focusing on the study of the long-term relationships over time. This paper raises an important question of the national averages masking important regional asymmetries. Results indicate evidence of co-integration, which suggests a broad grouping of regions based on physical proximity or similar economic characteristics.

Keywords House prices . Unit root * Co integration Long-run equilibrium * Convergence

JEL C10 * E00 * 050 * R10

Introduction

Housing development prices in most countries appears to be a topic of major interest. In fact, the purchase of a house is one of the assets that individuals most invest in. Macroeconomic factors in general, and particularly monetary conditions, are important determinants of the increase in the price of this asset. This is why the time series econometric analysis of housing prices has a relatively long history.

Some recent empirical studies on house price determination can be found, among others, in Iacoviello (2000), who uses a structural vector autoregressive (SVAR) approach to identify the main macroeconomic factors behind fluctuations in housing prices in six European countries, Girourard and Blondal (2001), who compare housing developments in OECD countries and examine how houses influence consumption and residential investment, Meen (2002), who compares the behavior of data on real house prices in the USA and UK, Annet (2005), who analyzes the European area with panel regressions of sub-groups of countries based on common institutional characteristics and plans short to medium run equations, Tsatsaronis and Zhu (2004), who apply a vector autoregressive (VAR) model for 17 countries around the world, grouped on their mortgage finance structures, Terrones and Otrok (2004), who analyze dynamic panel regressions for 18 countries, and OECD (2005), which relates the housing prices from the OECD countries and their underlying determinants. Furthermore, in relation to a single country Cook (2003) applies an asymmetric unit root test studying the ripple effect of regional UK housing prices, MacCarthy and Peach (2004) suggest demand and supply equations for the United States real estate market, Nagahata et al. (2004), contribute to the real estate market literature with a panel co integration analysis of Japan, as well as the following papers that used the error correction method, including Abelson et al. (2005) in Australia, OECD (2006) in Ireland, Oikarinen (2005) in Finland, Hofman (2005) in the Netherlands and Jacobsen and Naug (2005) in Norway.

Most of these papers have been concerned with explaining national housing prices. Nevertheless, there are also several studies based on regional house prices, due to increasing researcher interest in this topic; see among others, MacDonald and Taylor (1993). who model the dynamics of interregional housing prices in Great Britain, focusing on long-term equilibrium relationships and examining shorter-term dynamics, and Schnure (2005), who obtains panel estimations for regional housing prices and short-run specifications for US house prices.