Estimating local repeat sales house price indices for British cities
| DOI | https://doi.org/10.1108/14635780210416255 |
| Date | 01 February 2002 |
| Pages | 36-58 |
| Published date | 01 February 2002 |
| Author | Chris Leishman,Craig Watkins |
| Subject Matter | Property management & built environment |
JPIF
20,1
36
Journal of Property Investment &
Finance, Vol. 20 No. 1, 2002,
pp. 36-58. #MCB UP Limited,
1463-578X
DOI 10.1108/14635780210416255
ACADEMIC PAPERS
Estimating local repeat sales
house price indices for
British cities
Chris Leishman
Department of Building Engineering and Surveying,
Heriot-Watt University, Riccarton, Edinburgh, UK, and
Craig Watkins
Centre for Property Research, Department of Land Economy,
University of Aberdeen, Kings College, Aberdeen, UK
Keywords Housing, Pricing strategy, Estimating, Repeat buying, United Kingdom
Abstract This paper argues that the methods of constructing house price indices for UK
markets lag behind those employed in Europe, Australasia and North America. This is
particularly evident in terms of the range and level of technical sophistication of the index
construction methodologies. Importantly, the paper argues that the absence of reliable house price
indicators undermines the decision-making ability of policy makers and investors operating in
urban housing markets. The paper suggests that this can, in part, be remedied by the construction
of a system of local house price indices for British cities. The empirical research presents the first
UK application of the repeat sales method to UK data. Indices are constructed for four cities and
a range of diagnostic tests are used to establish the reliability and accuracy of the indices as a
means of monitoring house price change. The research concludes by suggesting that the methods
used here should be tested further on data from major metropolitan regions in England and
Wales.
Introduction
In the last decade there has been considerable debate in the UK conc erning
the nature and construction of house price indices and their accuracy (see
Earley, 1996; Nicol, 1996; RICS, 1998; Leishman, 2000). House price indices are
intended to provide a representative measure of price movements for defined
housing markets (Maclennan, 1977). They provide an important summary of
market performance because housing is a highly heterogeneous good for
which there is no single observable market price. Within a housing market
demand is unlikely to rise and fall uniformly for all types of housing over
time. Instead, it is likely that there are times when the price of properties of a
certain type, or in a certain sub-location, will be growing faster than the price
of other types of private housing or properties in other sub-locations within
the market.
The research register for this journal is available at
http://www.emeraldinsight.com/researchregisters
The current issue and full text archive of this journal is available at
http://www.emeraldinsight.com/1463-578X.htm
This research was supported by the RICS Education Trust and the Carnegie Trust for the
Universities of Scotland. The authors would also like to thank the Land Value Information Unit
(LVIU) at the University of Paisley who provided the data used in the analysis. Responsibility
for any errors lies with the authors.
Academic
papers: House
price indices
37
At present in the UK there is no completely reliable measure of the level of
house prices or the rate at which prices are changing. The most widely
consulted measures are the indices constructed by the Nationwide and Halifax.
In recent years these have sent conflicting signals about the magnitude and
pace of change in prices. This has had significant implications for a range of
actors in the housing market including private investors and public policy
makers. Nicol (1996) suggests, for example, that policy makers were
misinformed about market performance and exacerbated the problems of the
last housing market recession. It is further argued that because of the relatively
small sample of transactions available to the institutions, these indices, which
are published for regional and national levels, are too highly aggregated. At
present it is argued that the indices are being distorted by the differential
performance of the housing market in the South East.
In this paper we argue that if we are to have useful information on the
performance of a housing market or markets then we require indicators that are
not distorted by the differential performance of sub-sectors and sub-locations
within the market. This requires that we construct a system of local house price
indices. This, however, is not without a number of technical and practical
difficulties.
In the USA a voluminous literature (including special issues of the AREUEA
Journal and the Journal of Real Estate Finance and Economics) has debated and
tested the comparative merits of alternative index construction methods. The
availability of large-scale datasets for most major metropolitan areas in the
USA has provided researchers with considerable resources on which to test out
technical debates.
In the UK, however, researchers have had little opportunity to contribute to
this literature. A major practical constraint on this strandof housing research has
been the paucity of data. Costello and Watkins (2000) review the UK literature
and highlight that the institutional arrangements for data collection compare
unfavourably with those in the USA, Australia, Canada, Sweden and other parts
of Europe. Consequently, index construction work has tended to be based on the
data available from a single institution (Fleming and Nellis, 1994), survey data
collected to test out a local environmental or transport policy issue (see, for
example, Forrest et al., 1992; Henneberr y, 1999) or as king prices av ailable from
local agents (Cheshire and Sheppard, 1998). The method used, despite its widely
documentedlimitations, has beenpredominantly the hedonic technique.
In this paper, however, we note that the data compiled by the Land Registry
lends itself better to repeat sales methods rather than the hedonic approach. To
date, however, this method has not been used in the UK. Against this policy
and technical background, the paper aims to assess whether the repeat sales
method can be employed in using Land Registry data to develop a system of
local house price indices. Specifically, the research seeks to do this by
investigating the construction of local repeat sales indices and undertaking an
evaluation of the performance of the indices. The empirical part of the paper is
thus based on a pilot study of four city housing markets. The pilot study
To continue reading
Request your trial