A comparative analysis of Scottish property rents
| DOI | https://doi.org/10.1108/14635780010338326 |
| Pages | 332-351 |
| Published date | 01 June 2000 |
| Date | 01 June 2000 |
| Author | Michael White,Daniel Mackay,Kenneth Gibb |
| Subject Matter | Property management & built environment |
JPIF
18,3
332
Journal of Property Investment &
Finance, Vol. 18 No. 3, 2000,
pp. 332-351. #MCB University
Press, 1463-578X
Received June 1999
Revised February 2000
ACADEMIC PAPERS
A comparative analysis of
Scottish property rents
Michael White
Department of Land Economy, University of Aberdeen, Scotland
Daniel Mackay and Kenneth Gibb
Department of Urban Studies, University of Glasgow, Scotland
Keywords Property markets, Rent, Modelling
Abstract This paper uses time series econometric techniques to model regional property rents
in order to build a picture of the distinctiveness and commonality of the Scottish property sector.
Data used comes from a series stretching from 1970-1998 and allows Scotland's market
performance (in terms of rents) in each of the three main property sectors to be benchmarked
against a selective comparison of other UK regions. In doing so, we pay particular attention to the
statistical properties of the time series used, applying tests of data stationarity and cointegration
to develop a reduced form model of rents comprising both demand and supply-side variables. The
paper develops a predictive approach to property rents based on the autoregressive moving
average (ARMA) methodology. Initial within-sample predictive power is reasonably high. The
implications of our results for a better understanding of the Scottish property market, as well as
the more general modelling, are sketched out.
Introduction
Property markets, a natura, are exemplars of the need to disaggregate national
markets down to regional or urban scales. In the UK, time series data is often
available at the standard regional planning level (with separate data for
London) and this means that it is possible to focus on a specific region; in our
case, Scotland, and compare aspects of its property market by sector with other
regions and with Great Britain as a whole.
In this paper, time series data from 1970-1998 are used to compare and
contrast the properties of property rent performance in the three main sectors
(office, retail and industrial). In this way, it is possible to begin to characterise
the Scottish property market in terms of its distinctive and common features.
The paper seeks to identify key factors explaining property rents by sector in
Scotland and in other benchmark regions. It does this by applying tests of
stationarity and cointegration in order to develop a reduced form model of rents
encapsulating both demand and supply-side factors. Another purpose of the
paper, however, is to pay close attention to the statistical properties of the
series and to apply appropriate tests in terms both of explanation and,
The research register for this journal is available at
http://www.mcbup.com/research_registers/jpif.asp
The current issue and full text archive of this journal is available at
http://www.emerald-library.com
This paper is part of a project co-funded by the University of Glasgow and Richard Ellis St.
Quintin to examine the role of property in the Scottish economy.
The authors would like to thank two anonymous referees for helpful comments. All views
and errors expressed are their own.
Academic
papers: Analysis
of Scottish rents
333
subsequently, forecasting. For reasons discussed, an empirical methodology
based on predicting property rents by autoregressive moving average (ARMA)
methodology is employed. Fairly robust (within-sample) predictive results are
generated which are discussed.
The structure of the paper takes the following form. After the introduction,
there is a short description of the data series, plotting regional rents in the three
sectors from 1970-1998. This is followed by a brief literature review of previous
work on commercial property rents, focusing on equilibrium adjustment
processes and other specification issues. The paper then sets out the basic
single equation model of rents and reports tests of the variables used for unit
roots. This has significant regional variation. The paper then proceeds to report
the explanatory power of the models used to explain rents in each of the three
sectors for a cross-section of regions, including Scotland, London the South
East and Great Britain as a whole. Again, there are important divergences in
results across space and sector. The penultimate section constructs and reports
on an ARMA model of property rents for the same set of regions and sectors.
The final section of the paper discusses the key themes emerging from this
paper in terms of Scotland's property sector, econometric questions and future
issues for research.
What has happened to rents in commercial property markets?
Various measures of property market performance are available, including
capital growth, income growth and yields. Here we focus on rents. Figures 1 to
3 show the real rental performance of the difference property types. In each
case, 1990 is given a value of 100. The rent data are based on valuers' estimates
and as such may be subject to smoothing.
Figure 1.
Real office rents
To continue reading
Request your trial