An assessment of the impact of valuation error on property investment performance measurement
| Published date | 01 April 2001 |
| DOI | https://doi.org/10.1108/14635780110383695 |
| Date | 01 April 2001 |
| Pages | 139-157 |
| Author | G. Bowles,P. McAllister,H. Tarbert |
| Subject Matter | Property management & built environment |
Academic
papers:
Valuation error
139
Journal of Property Investment &
Finance, Vol. 19 No. 2, 2001,
pp. 139-155. #MCB University
Press, 1463-578X
Received March 1999
Revised July 2000
ACADEMIC PAPERS
An assessment of the impact
of valuation error on property
investment performance
measurement
G. Bowles
Department of Building Engineering and Surveying,
Heriot-Watt University, Edinburgh, UK
P. McAllister
Department of Land Management and Development,
The University of Reading, Reading, UK, and
H. Tarbert
Department of Accounting and Finance, Glasgow Caledonian University,
Glasgow, UK
Keywords Valuation, Property management, Sampling, Theory,
Investment prope rty portfolio
Abstract Analyses the effect of valuation error on the implied precision of investment
performance measurement of property assets. A prerequisite for measuring the absolute or
relative performance of commercial property investments is that valuations provide a reliable
proxy for prices. However, there are conceptual and empirical grounds to suggest that uncertainty
is inherent in the valuation process. This is primarily due to the structure of the commercial
property market and the techniques and guidelines of the property valuation process. Sampling
theory is used to measure portfolio valuation error confidence bands for hypothetical property
investment portfolios based on different assumptions concerning assumed levels of valuation
error, size of portfolio and number of measurement time periods. It is concluded that, for the
majority of investment portfolios, property investment performance measures will include some
uncertainty and thus the property fund manager should be sceptical of the implied precision in
reported measures of return.
Introduction
Investment performance measurement for equities and bonds is relatively
straightforward since it is based on actual transaction prices that are readily
observable. In contrast, valuations are used as a proxy for obtainable
transaction prices in the measurement of investment performance for
commercial property assets. Hence a necessary pre-condition for accurate
property investment performance measurement is that commercial property
valuations are a valid proxy for transaction prices. If this is not the case, then
portfolio managers are subject to potential divergence between their reported
and actual investment performance at any given point in time.
The research register for this journal is available at
http://www.mcbup.com/research_registers
The current issue and full text archive of this journal is available at
http://www.emerald-library.com/ft
JPIF
19,2
140
It now seems to be increasingly accepted by the property profession that the
valuation process is prone to uncertainty. The recent RICS Information Paper
(1997) states that ``.. . the valuer and most informed users of the valuation
recognise that there will be a degree of uncertainty attached to the figure
provided'' (RICS, 1997, p. 26). The implications of this uncertainty may be
considerable because investment performance measurement provides an input
into the process of making property investment decisions at a number of levels:
.disposal or retention of individual property assets;
.the performance of the portfolio manager;
.geographical and sectoral asset allocation decisions; and
.the performance of the property portfolio in relation to various
benchmarks.
The following sections focus on discussion and quantification of the potential
sources of error in these measures and the implications for subsequent
analysis. Although there are significant methodological problems when
conducting empirical research into valuation uncertainty, a substantial body of
research exists which would support an a priori expectation that valuers will
differ in setting market value for the same commercial property asset and also
that valuations will vary from transaction prices. The review of existing
evidence indicates that valuations are likely to contain a margin of stochastic
error and thus, at a given point in time, portfolio managers will be uncertain as
to whether the reported performance reflects actual performance. Subsequent
application of sampling theory indicates that aggregate measures of
commercial property portfolio investment return which are based on valuations
will also contain a margin of stochastic error and thus the implied accuracy in
precisely reported portfolio performance measurements will be spurious.
Valuation uncertainty
The use of the term ``error'' in the context of valuation uncertainty is not meant
to imply that observed differences between either valuations and valuations or
valuations and market prices are due to systematic misjudgements or mistakes
by valuers. Statistical error is merely the term used to quantify the deviation of
a measurement from its true value. In the quality assurance literature, it is
recognised that ``except in straightforward counting situations, there will
always be a difference between the measured value of a variable and the true
value'' (Jefferies, 1997, p. 1). Further, this error can be decomposed into two
forms. Inaccuracy can only be assessed after the transaction and implies a fixed
positive or negative error that is the same if the measurement is repeated.
Alternatively, imprecision (valuation variance in this context) is an ex ante
measure of a positive or negative error that varies randomly every time the
measurement is made. Thus, evidence which suggests that different valuers
place a different value on the same property does not necessarily imply
inaccuracy. Rather it implies imprecision. Given the trading environment and
To continue reading
Request your trial