Printer Friendly

The investment characteristics of real estate in other countries.

Many researchers have suggested that the addition of real estate to financial asset investments provides significant gains in portfolio performance. (1) Specifically, studies have shown that real estate has historically provided rates of return that are comparable to those of common stocks and bonds. At the same time, real estate has demonstrated a lack of positive return correlation with these financial instruments. Thus, the addition of real estate to a financial asset--based investment portfolio should provide diversification gains in the form of risk reduction. In addition, real estate typically provides an investor with a better-than-average hedge against inflation compared with the traditional forms of financial investment.

Data limitations in the previous studies restrict the generality of the conclusions regarding the investment characteristics of real estate. (2) First, previous analyses were conducted almost exclusively with U.S. data. Institutional incentives and restrictions (e.g., tax laws, farm subsidies, investment rules, and related statutes and structures) as well as cultural attitudes and geography vary among countries and may affect real estate investment attributes. Second, reliable real estate data is generally limited to post 1970. Only with the rise in popularity of real estate investment trusts (REITs) in the late 1960s have institutional real estate portfolios grown large enough to yield meaningful return data. Pension funds, prompted into real estate diversification by the Employee Retirement Incom Security Act (ERISA), have also been important sources of data. These data, however, are also very recent, dating back only to the passage of the law in 1974.

The first of these two important data limitations is addressed in this article by investigating the performance characteristics of real estate investments in other countries. Specifically, the hypothesis that the investment and diversification attributes of real estate that prevail in the United States also exist for domestic investors of other developed countries is empirically tested. Acceptance of the hypothesis would additionally provide some evidence on the time specificity of results obtained by previous real estate studies. If real estate investment attributes are found to be reasonably consistent over a wide range of environmental situations and economic conditions, the hypothesis logically supports the generality of these performance characteristics through time as well as space. Rejection of this hypothesis, however, would suggest that real estate asset performance is highly responsive to some combination of the location-specific and time-specific characteristics of institutional environment, culture, and geography.

The performance of real estate investments in the United Kingdom and Japan from 1973 to 1987 is considered and compared with the results of U.S. real estate investments during the same period. These nations were selected as examples of developed countries in Europe and the Far East, respectively. For each country, we examine the historical returns on a variety of domestic investment media, both financial and in real estate. Inter-investment correlation matrices are presented to compare the potential for risk reduction resulting from mixed-asset diversification in the three nations. The correlation of domestic asset returns with the domestic inflation rates is also presented.

Two efficient investment frontiers are constructed for each country. The first frontier is found using traditional financial assets only. The second frontier is found using both financial and real estate assets. The magnitude of the potential improvement in risk-return portfolio performance from real estate diversification is then estimated for all three countries. All efficient frontiers were generated using the method developed by Elton, Gruber, and Padberg. (3)

The investment media considered in each country include four traditional financial assets: common stock, long-term corporate bonds, long-term government bonds, and treasury bills. Also included are three real estate assets: farmland, commercial real estate, and residential housing. Residential real estate is not included for the United Kingdom because these data are generally unavailable. (4)


Mean annual rates of return and standard deviations are calculate for the seven investment types in the United States, seven investment types in Japan, and six investment types in the United Kingdom over the period 1973--1987. All returns are measured in terms of the domestic currency, are pre-tax, and do not include transaction costs.

For U.S. assets, all annual returns came from Ibbotson Associates. (5)

Data for the calculation of annual returns and standard deviations on Japanese financial assets came from Y. Hamao's article "Japanese Stocks, Bonds, Bills, and Inflation, 1973--1987." (6) Japanese real estate capital gains were provided by the Japanese Real Estate Institute (JREI), a non-profit, independent organization that conducts real estate research. For commercial and residential property, JREI measures the time series movement of real estate prices in 140 cities throughout Japan. JREI also tracks agricultural land prices in approximately 1,600 Japanese cities, towns, and villages.

Unfortunately, no reliable estimates of Japanese real estate operating income returns were found. To compensate for this deficiency, we assumed an annual operating income of 3% on all types of Japanese real estate. (7) The total returns calculated using this assumption for Japanese farmland, commercial real estate, and residential real estate seemed reasonable in relationship to the total returns on Japanese financial assets. Specifically, in the United States and the United Kingdom the mean annual return on real estate was typically lower than the mean annual return on common stocks and generally comparable to or slightly above the mean annual return on long-term bonds. Use of a 3% Japanese real estate operating income is based on the assumption that this relation is also true in Japan. Naturally, however, this procedure introduces several possible errors to the analysis. First, it ignores variation in year-to-year total returns caused by variations in year-to-year operating income. Second, it ignores the possibility that different types of Japanese real estate may yield different levels of operating income. Finally, the 3% level may vary considerably from reality.

Annual returns on all U.K. financial assets (stocks, bonds, and bills) were provided by Barclays de Zoete Wedd of London. Annual returns on British farmland came from the British real estate investment firm, Savills. Savills, in cooperation with the Investment Property Databank (IPD), holds the investment details of approximately 80% of the institutionally owned agricultural real estate in the United Kingdom. Returns on British commercial real estate came from two sources. For the period 1973--1980, data provided by the British real estate investment firm Jones Lang Wootton (JLW) were used. After 1980, JLW pooled the data from its clients with those of Healey and Baker, Hillier Parker, and Richard Ellis to obtain a broader representation of the market. At the end of 1988, this total portfolio contained over $4 billion in gross property value.


Results of the mean and standard deviation analyses are summarized in Table 1. A brief examination of these results shows that the risk-return relationship among financial assets were similar in all three countries. Common stock consistently provided the highest rates of return and exhibited the highest variance. Corporate bonds performed slightly better than long-term government bonds in the United States and the United Kingdom. In Japan, the yields of these instruments were virtually identical. In the United Kingdom and Japan, government bonds were slightly more risky than corporate bonds. In the United States, the reverse was true. Treasury bills in the United Kingdom and the United States and the comparable Japanese short-term interest rates (STIR) [8] provided the lowest returns

[TABULAR DATA OMITTED] with lowest risk. Overall, these results were almost exactly as expected.

The risk-return relationships of real estate returns, however, demonstrated far less uniformity. In Japan, farm, commercial, and residential real estate all showed similar annual yields and nearly identical levels of risk. In the United States, we found similar annual rates of return, but farm real estate exhibited nearly three times the risk associated with commercial and residential real estate. U.K. real estate showed the greatest diversity, with commercial real estate providing roughly twice the yield of farm real estate, but only one-third the risk.

Comparing real estate with financial assets, our results confirm the results of previous studies in that all types of U.S. real estate have provided favorable rates of return relative to traditional financial assets. Assuming a reasonable level of 3% operating income on all types of Japanese real estate, our results would indicate that this was also true in Japan. In the United Kingdom, although commercial real estate performed well, farm real estate performed very poorly, both in terms of yield and risk.



The ability of a particular asset to reduce the risk of a portfolio is determined by the correlation of its returns with the returns of the other assets in the portfolio. The less positive the correlation, the greater the contribution of an asset in terms of risk reduction.

Correlation coefficients were calculated for the U.S., Japanese, and British asset returns to estimate the potential for risk reduction provided by real estate in each of these three countries. The results are presented in Tables 2, 3, and 4, respectively. Also shown are the correlation coefficients between the returns of each asset and the equivalent of the consumer price index (CPI) in each country, which serves as a proxy for inflation.

Initially we note that the correlation




coefficients for the U.S. assets, shown in Table 2, were consistent with other studies. (9) In general, all types of U.S. real estate showed low or negative correlation with U.S. common stock and bonds. It is interesting that returns from U.S. Treasury bills exhibited high positive correlation with commercial real estate, virtually no correlation with residential real estate, and negative correlation with farm real estate. Finally, all three types of U.S. real estate showed high positive correlation with the U.s. CPI, implying that they shall served as a relatively good hedge against inflation. This was certainly true in comparison to common stock and bonds, which showed strong negative correlation with the CPI.

Japanese correlation coefficients, shown in Table 3, if anything more emphatically demonstrate the benefits of real estate diversification. All groups of Japanese real estate showed high negative correlation with Japanese common stock and bonds. Strangely, the correlation relationships between the Japanese short-term interest rates and the various types of Japanese real estate were almost exactly reverse of those found in the United States. The STIR showed strong positive correlation with farmland, very little correlation with residential real estate, and moderately negative correlation with commercial real estate. As in the United States, Japanese real estate did offer a hedge against inflation, although different types of real estate provided different levels of protection. While Japanese farmland was highly correlated with the Japanese CPI (0.73), commercial and residential real estate were only moderately correlated (0.39 and 0.27, respectively).

The situation in the United Kingdom, shown in Table 4, was quite different, however. In the United States and Japan, all types of real estate showed moderate to strong negative correlation with bonds. In the United Kingdom, real estate showed moderately positive correlation to British bonds. Correlation coefficients between British common stock and British real estate were comparable to those found with U.S. assets. In another surprise finding, U.K. real estate showed virtually no correlation with the British CPI. Thus, the correlation coefficients indicate that British real estate offered no hedge against domestic inflation.


Figures 1, 2, and 3 graphically present the efficient frontiers generated for U.S., Japanese, and British investors, respectively. Tables




5, 6, and 7 provide the compostions of the optimum portfolios calculated to produce these frontiers.

The results for the U.S. efficient frontiers, shown in Figure 1, are generally consistent with results found by Webb, Curcio, and Rubens, who used a similar method in their article "Diversification Gains from Including Real Estate in Mixed-Asset Portfolios." (10) Their study considered only one point on the efficient frontier, however, and covered a slightly different time period. As expected, U.S. real estate improved U.S. asset portfolio performance at all but the highest risk level. The largest gains were found at the low and moderate risk levels, gradually evaporating as risk increased.

At the lowest risk level, the optimum U.S. portfolio was composed of almost 50% real estate. Real estate ultimately occupied over 90% of the optimum portfolio in the moderate risk range. Commercial real estate was clearly the best performer, followed by residential real estate and then farmland.

The gains available to Japanese investors in Japanese real estate, as seen in Figure 2, closely paralleled those found in the U.S. market. Gains were substantial at the low and moderate risk levels, decreasing gradually as portfolio risk increased. Also, as in the United States, a substantial portion of the optimum Japanese portfolio was directed toward real estate investments at most risk levels. The major difference between U.S. portfolios and Japanese portfolios was in the composition of the real estate. While commercial real estate was best in the United States, farm and residential real estate played the most important role in Japan.

For the United Kingdom the findings are quite different, as illustrated in Figure 3. The addition of real estate to a British investor's portfolio provided essentially no significant gains in efficiency. British farm real estate never entered the optimum portfolio. Commercial real estate entered the optimum portfolio at all risk levels, but had little impact on total portfolio performance.


Many researchers have empirically demonstrated that during the 1970s and 1980s, U.S. real estate generally provided U.S. investors with significant diversification gains through portfolio risk reduction. This article investigates whether investors in other countries enjoyed similar benefits from investing in domestic real estate. Specifically, diversification gains were estimated for British and Japanese investors when domestic real estate was added to a portfolio containing only domestic financial assets. The results of the study are mixed. Assuming a reasonable operating income of 3% for Japanese real estate, Japanese investors derived diversification gains that closely parallel those available to U.S. investors in U.S. real estate. No significant benefits in terms of mean-variance efficiency were found, however, for British investors in British real estate.

It was further determined that real estate investments respond differently in each country to changing economic conditions. In the United States, all types of real estate returns demonstrated high positive correlation with the domestic CPI indicating that they all provide a reasonably good hedge against inflation. Neither British commercial nor farm real estate returns showed any substantial correlation with the British CPI. Japanese farm real estate returns were also highly correlated with the Japanese CPI, but Japanese commercial and residential real estate returns showed only moderate positive correlation with inflation.

The various types of real estate responded differently as well to changes in short-term interest rates depending on the country. In the United States, commercial real estate returns were highly correlated with the returns of U.S. Treasury bills. Residential real estate returns were virtually uncorrelated with short-term interest rates, and farm real estate returns showed moderate negative correlation. In Japan, the reverse was true. Farm real estate returns were positively correlated with short-term interest rates, residential real estate showed no correlation, and commercial real estate was negatively correlated. Finally, as in the United States, British farm real estate returns were negatively correlated with British treasury bill returns. Commercial real estate, however, showed little correlation with short-term British interest rates.

In sum, unlike the financial assets that demonstrate reasonably consistent patterns of return, risk, and correlation in all three countries, real estate assets show substantial diversity. this diversity suggests that real estate markets are influenced by forces that are seperate and distinct from those influencing the financial markets. Local conditions appear to have a profound impact on real estate asset performance. Thus, many of the perceived benefits associated with real estate assets may be country specific rather than characteristics inherently linked with this type of asset.

Further research on the general portfolio contribution of real estate is clearly warranted. Determination of the cause-and-effect relationships in real estate asset investment and diversification performance is of critical importance. Evaluation of country-specific factors affecting real estate returns may provide valuable insights on this issue.

(1) Pioneering this theory was H.C. Friedman, "Real Estate Investment and Portfolio Theory," Journal of Financial and Quantitative Analysis (6: 1971): 861-874.

(2) The most significant studies are summarized in G.S. Sirmans and C.F. Sirmans, "The Historical Perspective of Real Estate Returns." Journal of Portfolio Management (13: 1987): 22-31: and R.H. Zerbst and B.R. Cambon, "Real Estate: Historical Returns and Risks," Journal of Portfolio Management (10: 1984): 5-20.

(3) E.J. Elton, M.J. Gruber, and M.W. Padberg, "Simple Criteria for Optimal Portfolio Selection," Journal of Finance (31: 1976): 1341-1357.

(4) In general, British residential real estate was not a viable vehicle for investment during the period of interest. British laws were so totally biased in favor of residential tenants that it was virtually impossible for residential landlords to earn a reasonable rate of return. As a result, fully 70% of the residential real estate in Great Britain is owner-occupied with the large minority of the remaining 30% being public sector.

(5) Ibbotson Associates, Inc., Chicago, Illinois.

(6) Y. Hamao, "Japanese Stocks, Bonds, Bills and Inflation, 1973-1987," Journal of Portfolio Management (15: 1989): 20-26.

(7) This estimate is consistent with M.A. Hines, "Investing in Japanese Real Estate" (Westport, Conn.: Quorum Books, 1987).

(8) Japanese Government Bills offer interest rates that are artificially held low. Thus, Hamao uses the Japanese "call money rate" through November 1977 and, thereafter, the "Genseki rate" as the best proxy for the riskless rate. Herein, we refer to it as the Japanese short-term interest rate (STIR).

(9) R.G. Ibbotson and L.B. Siegel, "Real Estate Returns: A Comparison with Other Investments," American Real Estate and Urban Economics Association Journal (12: 1984): 219-242.

(10) J.R. Webb, R.J. Curcio, and J.H. Rubens, "Diversification Gains from Including Real Estate in Mixed-Asset Portfolio," Decision Sciences (19: 1988): 434-452.

Alan J. Ziobrowski, PhD, is an assistant professor of finance at Lander College in Greenwood, South Carolina. He received his PhD from Kent State University and has previously published articles in The Journal of Real Estate Research and Global Finance Journal.

Richard J. Curcio, PhD, is a professor of finance at Kent State University in Kent, Ohio. He received his PhD, from Pennsylvania State University. Dr. Curcio has published articles in the Journal of the American Real Estate & Urban Economics Association. The Journal of Real Estate Research, and numerous finance publications.

The authors gratefully acknowledge the assistance of the Japanese Real Estate Institute of Tokyo, Japan, and the British firms Savills, Healey & Baker, Jones Lang Wootton, Hillier Parker, Richard Ellis, and Barclays de Zoete Wedd, all of London, England.
COPYRIGHT 1992 The Appraisal Institute
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ziobrowski, Alan J.; Curcio, Richard J.
Publication:Appraisal Journal
Date:Apr 1, 1992
Previous Article:Valuation of fractional interests in real estate limited partnerships - another approach.
Next Article:Global real estate services.

Related Articles
Report: institutional players need new approach.
Strong investment opportunities in Latin America seen, as radical change is experienced in Europe due to EMU.
Fixed income alternatives in a low interest rate environment. (Special Advertising Section).
Bricks and mortar and beyond: in a time of worldwide economic uncertainty, some investors are searching for safer options for their capital. Property...
ISKAN announces major real estate project in Oman.
Global MENA Ijarah Real Estate Fund announces distribution of Dividend.
Global MENA Ijarah Real Estate Fund announces distribution of dividend.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters