In order to reduce the risk in matching long-term real liabilities, institutions require assets which provide a hedge against inflation. A number of authors have considered the role of real estate as an inflation hedge. In most cases they found a positive correlation29 between real estate values and the rate of inflation. However, given current economic conditions, as compounded by structural oversupply, the maintenance of real estate's historical relationship with inflation is now in question. The debt used to finance real estate in the 1960s and 1970s was typically long-term and fixed rate. During the inflationary late 1970's, this benefited borrowers. Property owners not only paid a reduced, often negative, real interest rate, but also benefited from interest payments being tax deductible and increasing property appreciation from demand pressures. However, lending policies have changed. Commercial mortgages are being written over shorter time periods, in many cases at interest rates variable during the term, and are underwritten to protect lenders from unexpected inflation. As Dohrmann [1995, p. 89] has argued, this further undermines real estate as a hedge against inflation.

In the US,30 Fama & Schwert [1977a], inter alios, found that real estate provided a good hedge against expected and unexpected inflation over the period 1953-1971. Hartzell, Hekman & Miles [1987] followed Fama & Schwert's methodology, and provided evidence that diversified real estate portfolios were a complete hedge against inflation over the periods 1973-1983 and 1978-1983. Webb [1990] supported this conclusion, finding a positive correlation between real estate and inflation over the period 1947-1986.

In the UK, Limmack & Ward [1988] found evidence that real estate offers a partial hedge against inflation, particularly in its expected component. Estimating expected inflation using Fama's [1975] methodology and an auto-regressive integrated moving-average model ("ARIMA"), their results suggested that real estate is a hedge against expected inflation. However, its unexpected component is less certain.31 These results were supported by Brown [1991b], who employed Fama & Schwert's [1977a] model and Miles [1996], whose evidence suggests that commercial property is a very imperfect hedge against unanticipated inflation.

Rosen & Smith [1983] argued that long-run real estate asset values are a function of construction costs and land use planning. With construction costs and domestic inflation highly correlated, real estate provides a good hedge against inflation. In the short run, however, asset values are affected by economic circumstances e. g. vacancy rates, and may not provide short-term protection.32

Barkham, Henry & Ward [1996] supported this argument, suggesting that in the long-run real estate returns and inflation are cointegrated i.e. that real estate provides a long-term hedge against inflation. However, in the short-term they found that real estate may not provide a good match for inflation. Matysiak, Hoesli, MacGregor & Nanthakumaran [1996] agree, suggesting that on an annual basis between 1963-1993, there is no evidence that property returns systematically provide hedging characteristics against either expected or unexpected inflation.

Therefore, the fall in institutional investment horizons - see section 2.3.1 - coupled with the perception that real estate has weakened as a hedge against short-term inflation, may have reduced its attraction to fund managers.


29For example, see Goslings & Petri [1991].

30References include, Brueggerman, Chen & Thibodeau [1984], Fogler [1984], Ibbotson & Siegel [1984], Zerbst & Cambon [1984], Sirmans & Sirmans [1987], Liu, Hartzell, Grisson & Greig [1990], and Wurtzebach, Mueller & Machi [1991].

31Results which were broadly in line with those of Peel & Pope [1985].

32In a similar vain, Schofield [1996] confirms that the inflation hedging characteristics of real estate are influenced by the level of yield, the level of gearing and number of years between reviews. Also see Rosen [1984] and Giliberto [1992b].