The long‐run diversification attributes of commercial property

Pages354-373
DOIhttps://doi.org/10.1108/14635780210435047
Published date01 August 2002
Date01 August 2002
AuthorW.D. Fraser,C. Leishman,H. Tarbert
Subject MatterProperty management & built environment
JPIF
20,4
354
JournalofPropertyInvestment&
Finance,Vol.20No.4,2002,
pp.354-373.#MCBUPLimited,
1463-578X
DOI10.1108/14635780210435047
ACADEMICPAPERS
Thelong-rundiversification
attributesofcommercial
property
W.D.Fraser
PaisleyBusinessSchool,UniversityofPaisley,Paisley,UK
C.Leishman
DepartmentofBuildingEngineeringandSurveying,
Heriot-WattUniversity,Riccarton,Edinburgh,UK,and
H.Tarbert
DepartmentofAccountingandFinance,GlasgowCaledonianUniversity,
Glasgow,UK
KeywordsDiversification,Cointegration,Commercialproperty
AbstractCorrelationcoefficientsmeasuringthehistoricalrelationshipsofreturnsoncommercial
propertyandbothequitiesandconventionalgiltsappeartobelow.Conversely,thecorrelation
betweengiltsandequitiesappearstoberelativelyhigh.Thisimpliesthatpropertyprovides
diversificationbenefitstoamixedassetportfoliodominatedbyequitiesandgilts.However,thereis
somedebateastothereliabilityofthesecorrelationsandproperty'sdiversificationbenefits.Inthis
paperweuseGrangercausalitytestsandcointegrationtechniquestodemonstratethatthereisno
long-runrelationshipbetweenpropertyreturnsandthoseofeithergiltsorequities.Thisconfirms
thediversificationbenefitsofincludingpropertyinamixedassetportfolio.
Introduction
Itisconventionaltoregardtheinclusionofcommercialpropertyinvestmentsin
amulti-assetinvestmentportfolioascontributingtoitsdiversification
potential.Thereturnsfromcommercialpropertyinvestmentdonotappearto
behighlycorrelatedwiththereturnsfromeithergiltsorequities.Low
correlationsbetweenthereturnsfromthealternativeassetclassesthat
constituteaninvestmentportfolioimplythatitisrelativelydiversifiedsinceto
constructaportfoliousingassetswhosereturnsdonothistoricallymove
togetherhelpsmitigatetheriskofasignificantdownturnintotalportfolio
returns.Asidefromtheempiricalevidence,whichisconsideredbelow,there
areanumberofapriorireasonstosupposethattheinclusionofcommercial
propertyinamulti-assetportfoliowilladdtoitsdiversification.
Intuitively,onemightexpectalowcorrelationbetweenconventionalgilts
andproperty,astheformerisinflationproneandthelatterisgenerallyviewed
asaninflationhedge.Incontrast,onemightexpectapositivecorrelation
Theresearchregisterforthisjournalisavailableat
http://www.emeraldinsight.com/researchregisters
Thecurrentissueandfulltextarchiveofthisjournalisavailableat
http://www.emeraldinsight.com/1463-578X.htm
The authors gratefully acknowledge the comments of anonymous referees. Responsibility for
error and omissions remains with the authors.
Academic papers:
Commercial
property
355
between equities and property, as both are driven by economic activity and
corporate earnings. In fact, all three asset classes are similarly influenced by
trends in interest rates. Falling interest rates will tend to raise returns and
rising interest rates reduce returns, and gilt yields provide a measure of the
opportunity cost of investment capital relevant to the pricing of both equities
and property.
However, the differential movement of returns over the economic cycle is
likely to provide an important explanation for the correlations between the
three asset classes. Returns from gilts have tended to lead the cycle in the UK
whereas property tends to be a late cycle performer. Inflation is the principal
determinant of returns from long-dated gilts and, typically, a fall in inflation
has been a precursor to an economic upturn. Conversely, a sharp rise in
inflation has preceded deflationary measures which have heralded a recession.
Hence, gilt prices and returns have tended to peak quite early in an economic
upturn and trough in a recession.
Equities also tend to lead the cycle, but lag the performance of gilts.
Primarily, share prices reflect expectations for corporate profits. They respond
to evidence of economic activity while (like gilts) anticipating the next cyclical
turning point. Thus prices have tended to peak before the top of the cycle and to
bottom before the cyclical trough, in each case lagging the trend in gilts.
Traditionally, a recovery in gilts was seen as a necessary condition for a
recovery in equities.
The timing of property's performance over the economic cycle seems to be
less consistent than gilts and equities (Key et al., 1994) but evidence suggests a
tendency towards performance late in the cycle, lagging that of both gilts and
equities. Primarily, returns from property are driven by trends in rental
growth. Rental values closely reflect economic activity at peaks but have a
tendency to lag at troughs. This is particularly the case with industrial and
office property where a glut of vacant space arising during a recession can
overhang the letting market in the following recovery phase, thereby delaying
growth despite an upturn in demand.
The cyclical dynamics of the three asset classes are also influenced by the
supply side; gilts by the PSBR and equities by new issues which can suppress
share prices in times of stock market boom. However, the supply side is more
important in the case of property due to the development cycle created by the
long time lags inherent in major property development projects. This cycle can
react with the economic cycle to generate occasional property boom/bust
scenarios such as those experienced in the early 1970s and, more recently, in
the late 1980s/early 1990s.
The boom/bust tendency of the property market increases the risk of
property investment, but may tend to reduce its correlation with other
investment classes. Nevertheless, it is interesting to note that the period of
strongest correlation between the three asset classes since reliable records
began in the UK occurred during the property market slump of late 1973-1975.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT