Modelling UK House Prices with Structural Breaks and Conditional Variance Analysis

Abstract

This paper differs from previous research by examining the existence of structural breaks in the UK regional house prices as well as in the prices of the different property types (flats, terraced, detached and semi-detached houses) in the UK as a whole, motivated by the uncertainty in the UK housing market and various financial events that may lead to structural changes within the housing market. Our paper enhances the conventional unit root tests by allowing for structural breaks, while including structural break tests strengthens our analysis. Our empirical results support the existence of structural breaks in the mean equation in seven out of thirteen regions of the UK as well as in three out of four property types, and in the variance equation in six regions and three property types. In addition, using a multivariate GARCH approach we examine both the behaviour of variances and covariances of the house price returns over time. Our results have significant implications for appropriate economic policy selection and investment management.

An empirical analysis of the Scottish housing market by property type

“…According to the results shown in Table for the mean equations, there is one single structural break in the average Scottish house price returns taking place in July 2007, as identified by the Bai–Perron test at a 0.10 significance level and further confirmed by both Chow's and Quandt–Andrews breakpoint tests at a 0.01 significance level. This result is somewhat consistent with Begiazi and Katsiampa () who using quarterly data found that there is a structural break in the fourth quarter of 2007 for the average Scottish house price index returns. Moreover, when examining each property type separately, it is clear that there is also a structural break taking place in July 2007 in the price returns of flats, while semi‐detached and terraced houses exhibit a structural break in September 2007.…”

“…(), who studied breakpoints in the US housing market, while authors who have investigated potential breakpoints in the UK house prices include Miles (), Zhang et al . () and Begiazi and Katsiampa ().…”

“…Again to the best of the authors’ knowledge, only Valadkhani et al . (), Morley and Thomas () and Begiazi and Katsiampa () have studied house price dynamics by property type. However, Valadkhani et al .…”

“…Furthermore, although accurate forecasting could be affected by structural breaks (Lee et al, 2006;Rapach and Strauss, 2008), there is rather limited literature on structural break analysis of house price dynamics, especially prior to modelling conditional volatilities. Among few authors who have investigated potential breakpoints of house prices are Chien (2010), who examined breakpoints in the Taiwan housing market, and Karoglou et al (2013) and Canarella et al (2012), who studied breakpoints in the US housing market, while authors who have investigated potential breakpoints in the UK house prices include Miles (2015), Zhang et al (2017) and Begiazi and Katsiampa (2019).…”

“…Nevertheless, previous studies of UK housing markets have focused mainly on England and Wales, often excluding Scotland. Exceptions include the studies of Maclennan and Tu (), Miles (, ) and Begiazi and Katsiampa (). Interestingly, while Maclennan and Tu (), in their study of UK cities, included Glasgow, as its house prices continued to rise, similar to the rest of Scotland, and moved in the opposite direction of Bristol and Luton during 1989–1993, Miles () found that Scotland exhibited the least variable returns in his sample.…”